Citation
Examining the efficiency and equity of solid waste service production at the city level

Material Information

Title:
Examining the efficiency and equity of solid waste service production at the city level
Creator:
Davis, Mark William ( author )
Language:
English
Physical Description:
1 electronic file (322 pages). : ;

Subjects

Subjects / Keywords:
Refuse and refuse disposal ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Abstract:
Across the United States city-level service delivery methods can vary greatly - from fragmented and overlapping to consolidated service arrangements. "New Urbanism" literature often argues that the consolidated models are most efficient and most equitable. "Local Public Economy" literature often favors polycentric arrangements noting that fragmented and overlapping governmental jurisdictions can instill competition and leading to more efficient programs. This dissertation explores the fragmentation and overlap discussion by researching the city-level production of solid waste (trash) collection services and asks the question: How does the organization of solid waste collection at the city level relate to: (1) fiscal efficiency, (2) environmental efficiency, and (3) equity of service production? The city-level organization of solid waste collection services are analyzed by comparing three service production arrangements: monocentric (consolidated and no overlap), franchise zones (fragmentation yet no overlap), and polycentric (fragmented and overlapping). This dissertation combined in-depth case study research completed on eight cities with populations greater than 50,000. The analyses found a demarcation between cities utilizing the polycentric service production model and cities utilizing the franchise zone and the monocentric service production models. Among the cities surveyed and measured on a per household basis, the franchise zone and the monocentric cities performed at statistically significant levels of high efficiency (both in terms of fiscal efficiency and environmental efficiency) than the polycentric cities. The findings in relation to equity also indicated superior service production by the franchise zone and monocentric service production models for two of equity's three measures. In terms of level of service, all three service production models performed largely equally and thus equitably. In terms of breadth of service and in terms of cost of service, the franchise zone and the monocentric cities outperformed the polycentric cities surveyed and were more equitable in their service delivery than the polycentric cities.
Thesis:
Thesis (Ph.D.)--University of Colorado Denver. Public affairs
Bibliography:
Includes bibliographic references.
System Details:
System requirements: Adobe Reader.
General Note:
School of Public Affairs
Statement of Responsibility:
by Mark William Davis.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
903402069 ( OCLC )
ocn903402069

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

EXAMINING THE EFFICIENCY AND EQUITY OF SOLID WASTE SERVICE PRODUCTION AT THE CITY LEVEL by MARK WILLIAM DAVIS M P A Indiana University, Bloomington, 2009 B S Washington University in Saint Louis, 1992 A thesis submitted to the Fac ulty of the Graduate School of the University of Colorado in partial fulfillment o f the requirements for the degree of Doctor of Philosophy Public Affairs 201 4

PAGE 2

2014 MARK WILLIAM DAVIS ALL RIGHTS RESERVED

PAGE 3

! "" This thesis for the Doctor of Philosophy degree by Mark William Davis has been approved for the Public Affairs Program by Tanya Heikkila, Dissertation Chair Paul Teske Examination Chair Christopher Weible Anu Ramaswami Lisa Skumatz July 25, 2014

PAGE 4

! """ Davis, Mark William (Ph.D., Public Affairs) Examin in g the Efficiency and Equity of Solid Waste Serv ice Production at the City Level Thesis directed by Associate Professor Tanya Heikkila ABSTRACT Across the United States city level service delivery methods can vary greatly from fragmented and overlapping to consolidated service arrangements. "New Urbani sm" literature often argues that th e consolidated models are most efficient and most equitable "Local Public Economy" literatur e often favors polycentric arrangements noting that fragmented and overlapping governmental jurisdictions can instill competiti on and leading to more efficient programs. This dissertation explores the fragmentation and overlap d iscussion by researching the city level production of solid waste (trash) collection services and asks the question: How does the organization of solid wa ste collection at the city level relate to: (1) fiscal efficiency, (2) environmental efficiency, and (3 ) equity of service production? The city level organization of solid waste collect ion services are analyzed by comparing three service production arrang ements: monocentric (consolidated and no overlap), franchise zones ( fragmentation yet no overlap ), and polycentric ( fragment ed and overlap ping) This dissertation combined in depth case study research completed on eight cities with a large N regression da ta analyses of survey data from 102 United States cities with populations greater than 50,000. The analyses found a demarcation between cities utilizing the polycentric service production model and cities utilizing the franchise zone and the monocentric s ervice production models. Among the cities surveyed and measured on a per household basis, the franchise zone and the monocentric cities performed at statistically significant levels

PAGE 5

! "# of higher efficiency (both in terms of fiscal efficiency and environment al efficiency) than the polycentric cities. The findings in relation to equity also indicated superior service production by the franchise zone and monocentric service production models for two of equity's three measures. In terms of level of service, all three service production models performed largely equally and thus equitably. In terms of breadth of service and in terms of cost of service, the franchise zone and the monocentric cities outperformed the polycentric cities surveyed and were more equitab le in their service delivery than the polycentric cities. The form and content of this abstract are approved. I recommend its publication. Approved: Tanya Heikkila

PAGE 6

! # DEDICATION This dissertation is dedicated to my parents, Jack and Angie Davis and to the memory of Richard Rubin. To my parents for their ceaseless love and support of me t hrough my childhood, education, and professional career. To Richard Rubin for helping me to discover my life passion as an educator and academic.

PAGE 7

! #" A CKNOWLEDGEMENTS I would like to acknowledge and r ecognize my University of Colorado Denver dissertation committee members Tanya Heikkila, Christopher Weible, Paul Teske, Anu Ramaswami and Lisa Skumatz To Dr. Heikkila for being the best dissertation advisor any Ph.D. s tudent could ever hope to have the honor of working with and for her sharing of the deep learning she received from the Ostrom School To Dr. Weible, for his deep thinking, hard questions, and intellectual rigor. To D ean Teske, for his polycentric resear ch and his unwavering support of my academic achievement s To Dr. Ramaswami, for making me a part of her groundbreaking work in urban sustainability and sustainable urban infrastructure systems. To Dr. Lisa Skumatz, for her deep understanding of integrat ed solid waste management and statistics.

PAGE 8

! #"" TABLE OF CONTENTS CHAPTER I. INTRODUCTION.. 1 Provision Versus Production.. 2 Introduction to Key Literature Reviewed.. 3 Current Literature Paradi gm: The Polycentric Monocentric Continuum... 5 Fragmentation and Consolidation.. 7 Overlap and No Overlap 7 Expanded Paradigm: Monocentric and Polycentric... 10 Expanded Paradigm: Franchise Zones a nd Functional Consolidation.. 10 Research Design and Defining Key Terms 15 Dissertation Roadmap 17 Key Dissertation Findings and Significance.. 19 II. REVIEW OF THE LITERATURE 20 Consolidation: Unified without Overlap 23 Polycentricity: Fragmentation and Overlap... 24 More Focus on the Spectrum, Less Focus on the Extremes.. 27 III. RESEARCH DESIGN.. 31 Research Questio n and Hypotheses.. 31 Hypothesis 1.1: Fiscally Efficient. 31 Hypothesis 1.2: Fiscally Efficient. 32 Hypothesis 2: Environmentally Efficient.. 33 Hypothesis 3: Equitable. 33

PAGE 9

! #""" Thick Description Via Case Studies.. 34 Overview of Research Design 35 Methods: Case Study Research Design.. 35 Methods: Large N Study Research Design 42 Variables and Indicators . 44 Independent Variable. 44 Dependent Variables.. 46 Fiscal Efficiency 48 Environmental Efficiency.. 48 Equity 52 Control Variables ... 53 Data Analysis and Models. 62 Efficiency Models. 62 Fiscal Efficiency Model for Hypothesis 1.1. 62 Fiscal Efficiency Model for Hypothesis 1.2. 62 Environmental Efficiency Mo del for Hypothesis 2... 63 IV. COMPARATIVE ANALYSIS ACROSS EIGHT CASE STUDY CITIES.. 64 Introduction to Case Studies 64 Comparative Information Across the Eight Case Studies. 67 Thick Description Across Case Studies . 77 Program Changes Designed to Increase Efficiency.. 78 Other Efficiency Related Items of Note 81 Contrasting Two Cases: Efficiency Versus Equity... 83

PAGE 10

! "$ Program Changes Designed to Increase Equity 85 Other Equity Related Items of Note. 86 Program Changes and Citizen Satisfaction... 87 Conclusions from Case Studies.. 90 Fiscal Efficiency 92 Environmental Efficiency . 93 Equity . 95 V. INDIVIDUAL CASE STUDIES... 97 The City of Austin, Texas. 98 Introduction to Austin 98 Introduction to Austin's Solid Waste Collection .. 98 Austin Trash Collection. 99 Austin Recyclables Collection... 100 Austin Organics Collection... 102 Key Findings: Fiscal Efficiency, Environmental Efficiency, and Equity . 103 Fiscal Efficiency 103 Environmental Efficiency. 103 Equity 104 Discussion Beyond the Efficiency Hypotheses. 105 Fiscal Efficiency 105 E nvironmental Efficiency.. 106 The City of Denver, Colorado 107

PAGE 11

! $ Introduction to Denver... 107 Introduction to Denver's Solid Waste Collection.. 107 Denver Trash Collection 108 Denver Recyclables Collection. 112 Denver Organics Collection.. 114 Key Findings: Fiscal Efficiency, Environmental Efficiency, Equity 116 Fiscal Efficiency 116 Environmental Efficien cy.. 116 Equity . 118 Discussion Beyond the Efficiency Hypothesis .. 121 Denver's In house Efficiency Analyses 121 Fiscal Efficiency 122 Environmental Efficiency . 123 The City of Fresno, California... 123 Introduction to Fresno... 123 Introduction to Fresno's Solid Waste Collection.. 123 Fresno Trash Collection. 124 Fresno Recyclab les Collection.. 125 Fresno Organics Collection... 126 Key Findings: Fiscal Efficiency, Environmental Efficiency, Equity 126 Fiscal Efficiency 126 Environmental Efficiency.. 126 Equity . 128

PAGE 12

! $" Discussion Beyond the Efficiency Hypotheses . 129 Fiscal Efficiency 129 The City of Knoxville, Tennessee .. 130 Introduction to Knoxville .. 130 Introduction to Knoxville's Solid Waste Collection . 131 Knoxville Trash Collection ... 132 Knoxville Recyclables Collection . 132 Knoxville Organics Collection .. 133 Key Findings: Fiscal Efficie ncy, Environmental Efficiency, Equity 134 Fiscal Efficiency 134 Environmental Efficiency .. 134 Equity . 135 Discussion Beyond the Efficiency Hypotheses . 136 Fiscal Efficiency 136 The City of Indianapolis, Indiana .. 137 Introduction to Indianapolis .. 137 Introduction to Indianapolis's Solid Waste Collection .. 137 Indianapolis Trash Collection 138 Indianapolis Recyclables Collection . 140 Indianapolis Organics Collection .. 142 Key Findings: Fiscal Efficiency, Environmental Efficiency, Equity 142 Fiscal Efficiency 142 Environmental Effic iency .. 143

PAGE 13

! $"" Equity . 144 Discussion Beyond the Efficiency Hypotheses . 146 Fiscal Efficiency 146 The City of Phoenix, Arizona 148 Introduction to Phoenix . 148 Introduction to Phoenix's Solid Waste Collection 148 Phoenix Trash Collection .. 149 Phoenix Recyclables Collection 150 Phoenix Organics Collection . 151 Key Findings: Fi scal Efficiency, Environmental Efficiency, Equity 151 Fiscal Efficiency 151 Environmental Efficiency .. 152 Equity . 153 Discussion Beyond the Efficiency Hypotheses . 153 Fisca l Efficiency 154 Environmental Efficiency .. 155 The City of Colorado Springs, Colorado ... 156 Introduction to Colorado Springs .. 156 Colorado Springs Trash Collection ... 15 6 Colorado Springs Recyclables Collection . 158 Colorado Springs Organics Collection .. 159 Key Findings: Fiscal Efficiency, Environmental Efficiency, Equity 159 Fiscal Efficiency 159

PAGE 14

! $""" Environmental Efficiency .. 160 Equity . 161 Discussion Beyond the Efficiency Hypotheses . 161 Environmental Efficiency .. 161 The City of Fort Collins, Colorado 162 Introduction to Fort Collins ... 162 Introduction to Fort Collins's Solid Waste Collection .. 162 Fort Collins Trash Collection 163 Fort Collins Recyclables Collection .. 164 Fort Collins Organics Collection ... 165 Key Findings: Fiscal Efficiency, Environmental Efficiency, Equity 166 Fiscal Efficiency 166 Environmental Efficiency .. 166 Equity . 168 Discussion Beyond the Efficienc y Hypotheses . 168 The Special Case of Fort Collins ... 169 Fiscal Efficiency 169 Environmental Efficiency .. 170 VI. RESULTS FROM LARGE N DATA ANALYSIS... 171 Research Question and Hypotheses... 171 Hypothesis 1.1: Fiscally Efficient. 171 Hypothesis 1.2: Fiscally Efficient. 171 Hypothesis 2: Environmentally Efficient .. 172

PAGE 15

! $"# Hypothesis 3: Equitable. 173 Roadmap for Chapter VI 174 Sample Data and Descriptive Statistics. 175 Sample Dataset and Sample Population 175 Descriptive Statistics. 178 Independent Variables ... 178 Dependent Variables.. 179 Control Variables... 185 Efficiency Hypotheses Analyses 192 Regression Analyses.. 192 Fiscal Efficiency Models... 192 Fiscal Efficiency Model for Hypothesis 1.1.. 194 Fiscal Efficiency Model for Hypothesis 1.2.. 199 Fiscal Efficiency Discussion Beyond the Research Hypothesis for Polycentric Cities.. 204 Testing Regression Mod el Fit Against Case Study Cities 206 Environmental Efficiency Models. 209 Environmental Efficiency Discussion Beyond Research Hypothesis. 219 Equity Hypothesis Analysis... 221 Equity Hypothe sis Discussion... 221 Equity Discussion Beyond Research Hypothesis.. 223 Conclusions: Zooming Out from the Hypothesis Level 226 VII. CONCLUSIONS 227

PAGE 16

! $# Key Findings.. 227 Key Findings in Relation to Research Hypotheses... 228 Practical Significance 232 Key Findings from Case Study Thick Description 233 Contributions to the Literature... 235 Expanded Theoreti cal Paradigm: New Service Production Arrangements 235 New Service Production Performance Measures.. 239 New Insights to Old Questions.. 240 Theoretical Limitations.. 242 Research Limitations.. 243 Future Research.. 246 Conclusions 249 REFERENCES 252 APPENDIX A. KEY DEFINITIONS IN ALPHABETICAL ORDER.. 261 B. SURVEY I NSTRUMENTS.. 264 C. SAMPLE POPULATION OF CENTRAL CORE CITIES... 294

PAGE 17

! $#" LIST OF TABLES TABLE 3.1 Tasks and Timeline for Case Studies . 36 3.2 Independent Variable Operationalization .. 45 3.3 Depen dent Variables Concept, Indicators and Units, and Data Sources 46 3.4 Combined Flow Diagram of this Dissertation's Research Design. 47 3.5 Dependent, Independent, and Control Variables for Fiscal Efficiency Models and Environmental Efficiency Mode l... 59 4.1 Interviews for Case Study Cities 65 4.2 Key Demographic Information for Case Study Cities... 68 4.3 Comparative Service Delivery Information... 69 4.4 Comparing Fiscal Efficiency .. 70 4.5 Comparing Environmental Efficiency... 73 4.6 Comparing Equity.. 76 4.7 Comparing Efficiency Enhancing Measures.. 79 4.8 Denver's Varied Trash Collection Methods.. 81 4 .9 Summary of Fiscal Efficiency Hypothesis Findings. 92 4.10 Ranking Case Study Cities by Route Based Environmental Efficiency 93 4.11 Ranking Case Study Cities by Total Environmental Efficiency 94 4.12 Ranking Case Study Cities by Di sposal Based Environmental Efficiency 95 5.1 Summary of Austin Case Study Findings. 104 5.2 Summary of Denver Case Study Findings 117 5.3 Summary of Fresno Case Study Findings .. 128

PAGE 18

! $#"" 5.4 Summary of Knoxville C ase Study Findings . 135 5.5 Summary of Indianapolis Case Study Findings . 144 5.6 Summary of Phoenix Case Study Findings 153 5.7 Summary of Colorado Springs Case Study Findings . 160 5.8 Summary of Fort Co llins Case Study Findings .. 167 6.1 Survey Cities by Geographic Region and Service Production Model... 177 6.2 Cost of Trash Collection Service Production. 180 6.3 One sample t test Results for Cost of Trash Collection Servic e Production: Polycentric, Franchise Zones, and Monocentric 181 6.4 One sample t test Results for Cost of Trash Collection Service Production: Polycentric, Franchise Zones, and Monocentric Public, Monocentric Private 181 6.5 Environmental Efficiency of Trash Collection Service Production... 183 6.6 One sample t test Results for Environmental Efficiency of Trash Collection Service Production 184 6.7 One sample t test Results for Route onl y Environmental Efficiency of Trash Collection Service Production. 184 6.8 One sample t test Results for Disposal only Environmental Efficiency of Trash Collection Service Production. 184 6.9 Unionization of Trash Collec tion Workforce 188 6.10 City Funding Mechanism for Solid Waste Collection... 189 6.11 Comparing Number of Cities with City level Solid Waste Goals Against Service Production Models and Breadth of Services Offered... 191 6.12 Dependent, Independent, and Control Variables for Fiscal Efficiency Models and Environmental Efficiency Model... 192 6.13 Linear Regression Analysis Runs for Fiscal Efficiency Hypothesis 1.1 p values for Independent and Control Variables ... 195

PAGE 19

! $#""" 6.14 Regression Analysis for Fiscal Efficiency Hypothesis 1.1 197 6.15 Zero order and Partial Correlations for Fiscal Efficiency.. 198 6.16 Linear Regression Analysis Runs for Fiscal Efficiency Hypothesis 1.2 p values for Independent and Control Variables... 200 6.17 Regression Analysis for Fiscal Efficiency Hypothesis 1.2 202 6.18 Zero order and Partial Correlations for Fiscal Efficiency.. 203 6.19 Comparing Fiscal Efficiency Against Other F actors for Polycentric Cities 205 6.20 Testing Regression Model Fit Against Case Study Actual Findings. 207 6.21 Testing Regression Model Fit Against Case Study Actual Findings: Differences Between Predicted and Actual Versus Standard Deviation... 208 6.22 Linear Regression Analysis Runs for Environmental Efficiency Hypothesis 2 Significance Levels for Model, Independent, and Control Variables.... 210 6.23 Regression Analysis for Environmental Efficiency Hypothesis 213 6.24 Zero order and Partial Correlations for Environmental Efficiency... 214 6.25 Regression Analysis for Environmental Efficiency Hypothesis Utilizing Route only Environmental Efficiency as the Dependent Variable 216 6.26 Zero order and Partial Correlations for Route only Environmental Efficiency. 217 6.27 Regression Analysis for Environmental Efficiency Hypothesis Utilizing Disposal only Environmental Efficiency as the Dependent Variable 218 6.28 Su mmary Data for Tonnage Per Household Data for 30 Cities: Landfill Disposal, Recycling, and Organics Bound for Composting.. 220 6.29 Disposal Environmental Efficiency of All Landfill Disposal Versus Landfill Disposal, Recycling, and Organics Compos ting.. 221 6.30 Equity Across Multiple Cities and Across Service Production Models 223 C.1 Sample Population of Central Core Cities Over 50,000 in Population.. 294

PAGE 20

! $"$ LIST OF FIGURES FIGURE 1.1 Polycentric Monocentric Continuum .. 6 1.2 Expanded Service Production Arrangements Polycentric, Monocentric, Franchise Zones, and Functional Consolidation 9 1.3 Expanded Service Production Arrangements Including Examples for Each Quadrant 12 1.4 Solid Waste Collection Specific Examples of Service Production Arrangements 14 3.1 Schematic of Boundaries of Environmental Efficiency Measure.. 52 4.1 Case Study Cities and Service Production Models 6 6 4.2 Comparing Fiscal Efficiency.. 71 4.3 Comparing Environmental Efficiency... 74 6.1 Hypothesis 1.1 Fiscal Efficiency Findings. 182 6.2 Hypothesis 2 Environmental Efficiency Findings. 185 6.3 Large N Equity Compared on an Across Multiple Cities Basis. 225 7.1 Independent Overlap and Fragmentation Two by Two. 237 7.2 Solid Waste Collection Specific Two by Two Example... 241 7.3 Fiscal Efficiency, Environmental Effic iency, and Equity Findings Across All Three of the Service Production Models.. 250

PAGE 21

CHAPTER I INTRODUCTION This dissertation will provide insights into how differences in the organization of solid waste collection processes utilized in the United States at the city level impact the efficiency and equity of solid waste service delivery T h is inquiry focuses on the research question: How does the organization of solid waste collection at the city level relate to: (1) fiscal efficiency, (2) environmental efficiency, and (3) equity of service production? This dissertation will specifically e xplore this question from the perspective of single family household solid waste collection services delivered at the city scale. Colloquially referred to as trash collection, garbage collection, or rubbish removal solid waste collection today also often includes recyclables collection (trash bound for recycling rather than disposal) and organics collection (yard waste and / or food waste bound for composting rather than disposal). While solid waste collection is considered a required 1 service for urban r esidents in the United States, the collection of recyclables and compostables is by no means universal. Bohm, Folz, Kinnaman, and Podolsky (2010) estimated that 51% of the United States population is served by 8 817 municipal curbside recycling programs. Organics collection bound for composting is even less common, with an estimated 3 260 composting programs nationally (Bohn, Folz, Kinnaman, & Podolsky, 2010 ). !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! While not all cities provide solid waste collection services, there is an expectation in these cities that residents will procure this service directly and / or properly manage their trash in a legal fashion.

PAGE 22

! 2 Provision Versus Production Cities in the United States make provisions (V. Ostrom, 2008; Oake rson, 1999) for the collection of solid waste; however the execution of service production can vary in its delivery. The service production can be delivered directly by the city or delivered via a contractor. In other cases, cities place the procurement of services into the hands of individual residents. This dissertation explor es the production side of service delivery T his service production in the solid waste case can most simply be defined as the physical delivery of trash collection services Loc al public economy literature provides a clear distinction between the provision of a public service (or public good) and the production of a public service (or public good). Provision is defined as th e decisions surrounding what public goods and services will be provided and how they will be produced. This can include private activities that require regulation, how and how much revenue is to be raised, and it arranges for how the production of a public good or service will be conducted. Provision nearly always is solely the purview of government (Oakerson, 1999; McGinnis, 2011). Production is defined as the act of producing a good or physically delivering a service. Production may be conducted by government or by private entities acting as government's agent (Oakerson, 1999; McGinnis, 2011). The distinction between provision and production was first advanced by Richard A. Musgrave (1959), more fully developed by Vincent Ostrom, Charles M. Tiebout, & Robert Warren (1961), and then further refined by Ron Oakerson (1999). Because this dissertation centers on service production with trash collection as the research frame, the following example utilizes solid waste collection services to illustrate the difference between provision and production In many citi es trash collection is

PAGE 23

! 3 considered to be a basic service that should be provided by the city. In these cases there are generally two primary variations for trash service: (1) Government itself delivers the collection service. That is, government makes pro visions for solid waste services and is directly involved in the production of the service ; or (2) Government contracts with a third party (most often a private sector firm) to deliver the collection service within a designated service area. That is, gove rnment makes provisions for the service, but a third party delivers production of the service. In these cities residents must use the trash service as provided to them, there is no "shopping the market" for the best deal or the best service. There are, h owever, some cities that do not deliver trash collection as a basic service. Instead they place this in the hands of their local citizens, usually in concert with a codified anti litter or mandatory trash service requirements of some kind. In this case t he government's "provision" is to require residents to make their own arrangements for production. In these cities individuals contract directly with a service provider, in contrast to the earlier examples, the resident is free to "shop the market" for th e best price or the best service. A consequence of this added choice is that multiple companies may serve the same street and / or household. Introduction to Key Literature Reviewed This dissertation looks at the organization of single family household so lid waste collection "Organization" is defined via how these services are delivered. Specifically, is the service production overlap ping or non overlapping and is the service production fragmented or consolidated. (The terms overlap, non overlap, fragm ented, and consolidated will each be defined in more detail later in this chapter.) The two literatures that frame city level services in terms of overlap and fragmentation are : (1) t he theory of

PAGE 24

! 4 "local public economy" (McGinnis, 2011; V. Ostrom, Bish, & E. Ostrom, 1988 ; Heikkila, 2006 ); and (2) the theory of "new regionalism" (Lyon & Lowery, 1989a and 1989b ; Rusk, 1993; Pierce, Johnson, & Hall, 1993; Downs, 1994; Seamon & Feiock, 1995; Orfield, 1997). Both "new regionalism" and "local public economy" lit eratures are concerned with how the organization of public services affects outcomes, including the efficiency and equity of services The theory of l ocal public economy is built largely on theories drawn from the field of political economy (McGinnis, 2011 ). According to this literature, in a local public economy there is neither a pure market economy nor a pure political hie rarchy (Aligica & Boettke, 2011; Aligica & Tarko, 2013) Instead, local public economy draws on elements found both in a private sec tor context and a public sector context (Aligica & Boettke, 2011; Aligica & Tarko, 2013; McGinnis, 2011). This concept that local public economies are neither market s nor hierarch ies, in concert with a focus on the functions of local governments comprise s the central tenants of the theory of local public economy (V. Ostrom, Bish, E. Ostrom, 1988; Oakerson, 1999). 2 The theory of new regionalism argues that performance is improved in unified and consolidated city regions and is built from the fields of p olitical science and welfare economics (Swanstrom, 2001). The theory explores how unified and consolidated city regions, characterized by local government cooperation within the city regions, can !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! # It should be noted that t he th eory of local public economy is imbedded within the institutional analysis and development ( IAD ) framework (Kiser & E. Ostrom, 1982; E. Ostrom, 2005). In brief, the IAD framework addresses how institutions emerge, change, and perform over time. Within th e context of the IAD framework the term "institutions" is not defined as buildings or physical structures, but instead refers to the rules, norms, and strategies agreed upon by the individual members of a group via collective action (E. Ostrom, 2005; E. Os trom, 2011).

PAGE 25

! 5 reduce negative externalities caused by fragmented and over lapping government structures, and ensure the greatest equity in service delivery possible across the city region (Frisken & Norris, 2001). Characteristics of this consolidated model include: local government cooperation within the city regions, reduced ne gative externalities caused by fragmented and overlapping government structures, and improved equity in service delivery across the city region (Lyon & Lowery, 1989b; Frisken & Norris, 2001). Current Literature Paradigm: The Polycentric Monocentric Con tinuum As noted above, the service production being researched will be explored from the perspective of whether the service production is overlapping or non overlapping and is the service production fragmented or consolidated. The terms "monocentric" and "polycentric" come from a key proposition within the theory of local public economy However, it is important to note this dissertation utilizes a more precise definition of each of these terms than the definition s utilized within th is literature. Within local public economy literate, the distinction is solely between monocentric ity ( characterized by non overlap and consolidation) and polycentricity (characterized by overlap and fragmentation) (V. Ostrom, 2008; V. Ostrom, Tiebout & Warren, 1961). Figure 1 .1 below is a visual representation of this continuum. As will be discussed further later in this chapter, this dissertation decouples the concepts of fragmentation and overlap. Thus service production may be described in four categories instead of two: fragmented and overlapping, fragmented yet non overlapping, consolidated and non overlapping, or consolidated yet overlapping The following section will define fragmentation and consolidation and note how they will be utilized within this dissertation; a fter this the

PAGE 26

! 6 next section will define overlap and no overlap and note how they will likewise be utilized within this dissertation. Figure 1 .1 : Polycentric Monocentric Continuum ! ! ! Polycentric: Fragmented and Overlapping Monocentric: Consolidated and No Overlap

PAGE 27

! 7 Fragment ation and Consolidation Fragmentation is defined as mul tiple differentiated units. For fragmented provision these units must be multiple governmental units having jurisdiction over some form of service provision 3 F or fragmented service production, these units must be mul tiple producers and these multiple p roducers each need to be delivering the service (Oakerson, 1999; Frisken & Norris, 2001; Oakerson & Parks, 1988, 2011) Consolidated is defined as a singular integrated unit. For consolidated provision th is unit is one single governmental unit that has jurisdiction over some form of service provision. For consolidated service production, this unit must be one singular producer delivering the defined service (Aligica & Boettke, 2011; V. Ostrom, 2008; Oakerson, 1999). It is important t o note that within the literature the terms "monocentric" and "consolidated" are often considered to be interchangeable, however within this dissertation the two have distinct definitions Monocentric will be defined later in this chapter Overlap and No O verlap Overlap i s defined as the real or potential overlying, compounded co occurrence of service provision or production by various entities within the same geographic area For real overlapping provision the r e must be multiple governmental units having jurisdiction ov er some form of service provision and (at a minimum) at least two of these multiple governmental units are exercising their right to execute this service provision in the same geographic region as another unit. For potential overlapping provision there mu st be multiple governmental units having jurisdiction over some form of service !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! $ It is important to note that fragmentation does not have a geographic element. The geographic element is key to defining overlap and non overlap.

PAGE 28

! 8 provision however one or none of these multiple governmental units are exercising their right to execute this service provision in the same geographic region as other units. F or real overlapping service production there must be multiple producers of the service and (at a minimum) at least two of these producers are exercising their right to deliver this service in the same geographic region as another unit. For potential overl apping service production there must be multiple producers of the service however one or none of these producers are exercising their right to deliver this service in the same geographic region as other units. No overlap (or non overlap) is defined as the complete lack of overlying, compounded co o ccurrence of service provision or production within the same geographic area. Within any defined geographic area there is a single service provider and / or single service producer. For non overlapping service provision there must be only one governmental unit exercising its right to execute service provision within a defined geographic region. For non overlapping service production there must only be one service producer delivering a defined service within a d efined geographic region. By decoupling fragmentation from overlap, the former literature paradigm of a polycentric monocentric continuum transforms into a 2x2 diagram. Figure 1.2 below is a representation of this diagram. Via this decoupling this disser tation has identified two additional service arrangements. Fragmentation is represented in the left two quadrants of this figure. By contrast, consolidation is represented in the right two quadrants of this figure. Overlap is represented in the upper tw o quadrants of this figure. For the upper left quadrant, that is both fragmented and overlapping, this quadrant is termed "polycentric" and the overlap is real For the upper right quadrant, where there is the

PAGE 29

! 9 potential for overlap yet service is, in fac t, consolidated, this quadrant is termed "functional consolidation". No overlap is represented in the lower two quadrants. For the lower right quadrant, that is both consolidated and non overlapping, this quadrant is termed "monocentric". For the lower left quadrant, that is fragmented yet non overlapping, this quadrant is termed "franchise zone". Figure 1.2: Expanded Service Production Arrangements Polycentric, Monocentric, Franchise Zones, and Functional Consolidation ! ! ! Polycentric Franchise Zones Functional Consolidation Monocentric Overlap No Overlap Consolidated Fragmented ! !

PAGE 30

! 10 Expanded Paradigm: Monocentric and Polycentric Both t he local public economy literature and this dissertation define a service that is non overlapping and consolidated as a m onocentric arrangement (McGinnis, 2011). A common example of a monocentric service produc tion arrangement would be a citywide water utility delivering drinking water to customers within a city. Both t he local public economy literature and this dissertation def ine a service that is both overlapping and fragmented as a polycentric arrangement ( McGinnis, 2011). A common example of a polycentric service arrangement would be policing services delivered by multiple agencies in an overlapping geographic region. For example a State Police Officer, a County Sheriff and a City Police Officer could a ll patrol the same street and all have jurisdiction on said street Expanded Paradigm: Franchise Zones and Functional Consolidation This dissertation contributes to the academic literature by its expansion of the literature surrounding fragmentation versu s consolidation and overlap versus non overlap. By decoupling fragmentation from overlap, t his dissertation has identified two additional service arrangements. A third arrangement that is non overlapping yet fragmented is defined as the franchise zone ar rangement. In a franchise zone scheme, citywide there will be multiple service producers and multiple non overlapping geographic service delivery zones, however within any one given zone there will only be one single service producer. An example of franc hise zones would be a city breaking itself into zones for a particular service delivery, for example snow removal, and then bid ding out these non overlapping franchise zones to different contractors for service delivery

PAGE 31

! 11 A fourth arrangement that is over lapping yet consolidated is defined as the functional consolidation arrangement. In a functional consolidation arrangement, multiple government units each having jurisdiction for a particular service provision and / or service production, will all collec tively and voluntarily consolidate under a single service delivery mechanism for a specific service Thus while this quadrant has the potential of overlap the mutually agreed upon consolidation has precedence in this quadrant. Also, in relation to the ov erlap, it is important to note these same government entities maintain their individual autonomy in other service provision and / or production areas and likewise, their defined geog raphic boundaries will remain intact. An example of functional consolida tion would be the many mass transit authorities in city regions across the United States. These mass transit authorities have a geographic boundary larger than a single city, provide a consolidated service to the region, yet in terms of other jurisdiction al authorities t he local governments within this city region maintain their individual boundaries and individual autonomy. Figure 1.3 below displays the expanded visualization of the polycentric monocentric continuum, including the examples discussed with in the above two sections.

PAGE 32

! 12 Figure 1.3: Expanded Service Production Arrangements Including Examples for Each Quadrant For th e case of solid waste collection the franchise zone arrangement does indeed occur. However, the functional con solidation arrangement does not occur legally. This is because for solid waste collection that is a consolidated yet overlapping arrangement would represent a cartel which is illegal in the United States. W hile still a "functional ! ! ! Polycentric Franchise Zones Functional Consolidation Monocentric Overlap No Overlap Consolidated Fragmented Example: Law enforcement: city police, county sheriff, s tate police. Multiple fragmented agencies each have overlapping service production. Example: Snow Removal Franchise Zones Creating service delivery zones within a city that each can be bid to private sector contractors on a competitive basis. Example: Regional Mass Transit Districts Multiple government agencies, each with jurisdiction within a city region, consolidate under a single service producer. Example: City Water Utility Pa rticular service provider is the sole producer of the service within the designated service area.

PAGE 33

! 13 consolidation" in that the cartel arrangement is consolidated in the sense that each individual household within the city is only delivered service via one producer and is overlapping in the sense that the city as a whole is provided the service via multiple service producers the cartel arrangement is certainly not in the public's best interest Thus while conceptually still within the functional consolidation quadrant, this is certainly a dubious special case of solid waste collection and functional consolidation as the foll owing paragraph will highlight. Historically, New York City, New York was the most infamous example of this cartel service delivery model. Prior to the 1990s, solid waste collection in New York City was largely a mafia organized venture (Cowan & Century, 2003 ; Savas, 2005 ). Although there were the appearances of a free market, in fact the mafia would in advance decide which property belonged to whom for waste collection. Thus, when a customer would call for service, it was pre determined via mafia coll usion which solid waste company would provide the customer the low bid Additionally, in researching this dissertation and in completing previous solid waste related research, the author has not identified an legal arrangements specifically related to tras h collection that would be qualify as a "functional consolidation" arrangement. Thus, a s this is a dissertation is in the realm of public policy research and not criminal justice research, the dissertation is specific to the three legal solid waste collec tion schemes in the United States. F or this dissertation s olid waste collection schemes analyzed at the city level will fall into one of three primary categories: non overlapping and consolidated (monocentric), non overlapping yet fragmented (franchise zon es ), and overlapping and

PAGE 34

! 14 fragmented (polycentric). Figure 1 4 presented below is a visual representation of the solid waste collection specific example including the cartel arrangement Figure 1. 4 : Solid Waste Collection Specific Exa mple of Service Production Arrangements ! ! ! Polycentric Franchise Zones Cartel (Illegal) Monocentric Overlap No Overlap Consolidated Fragmented Private Sector Service Production Example: Colorado Springs, CO Public and Private Sector Service Production Example: Indianapolis, IN Private Sector Service Production Historical Example: New York City, NY "Exclusive Franchise" Public and Private Sector Service Production Example: D enver, CO

PAGE 35

! 15 In a city that provides multiple solid waste collection services (e.g., trash collection, recyclables collection, and organics collection), these services are by no means necessarily provided under the same arrangem ent. As a hypothetical example, it is possible to have a city with polycentric trash collection, franchise zone recyclables collection, and monocentric organics collection. Research Design and Defining Key Terms Finally, this introduction chapter concludes with a brief summary of the dissertation's research design and an outline of the results chapters The dissertation's research questi on is: How does the organization of solid waste collection at the city level relate to: (1) fiscal efficiency, (2) environmental efficiency, and (3 ) equity of service production? Fiscal efficiency and equity are commonly utilized measures for program asse ssment within the local public economy literature (Oakerson, 1999) and new regionalism literature (Swanstrom, 2001; Lyon & Lowery, 1989a, 1989b). This dissertation adds environmental efficiency as a third measure. These terms are defined as follows : Effic iency: The basic definition of e fficiency is the ratio of service outputs to service inputs. "The criterion of efficiency dictates that choice which produces the largest result from the given application of resources" (Simon, 1965, p.179). For this study efficiency will be been divided into two measures: fiscal efficiency (the traditional measure of efficiency) and environmental efficiency (a new measure of efficiency) Fiscal efficiency is specific to providing the most service for the least cost (E. Os trom, 2011; Oakerson, 1999; V. Ostrom, Bish, E. Ostrom, 1988; Savas, 1977, 1987, 2005).

PAGE 36

! 16 Environmental efficiency is specific to providing the most service with the smallest environmental footprint (Hillman & Ramaswami, 2009; Chavez & Ramaswami, 2011). An environmental footprint is measured in terms of quantifiable pollutants or quantifiable environmental impacts. In developing this new measure, t his dissertation utilize s greenhouse gas es (GHG) as the measure of environmental efficiency. The units for th is measure are carbon dioxide equivalents (CO 2e 4 ) This environmental efficiency measure is built from measures developed by Anu Ramaswami and her colleagues cross disciplinary work in relation to sustainability and urban systems (i.e. Chavez & Ramaswami, 2011; Davis & Weible, 2011 ; Hillman & Ramaswami, 2009; Hillman, Janson, & Ramaswami, 2011; Kennedy et al, 2010; Ramaswami et al, 2008; Ramaswami et al, 2011; and Ramaswami et al, 2012 ) to measure an urban area's GHG emissions footprint The environmental e fficiency measu re combines the GHG potential triggered by (1) the solid waste collection service production method utilized and (2) a measure of the GHG potential of the ultimate fate of the materials: landfill disposal, waste to energy, recycling, or comp osting. Combining these two measures of GHG potential provides a total citywide footprint of solid waste collection. This citywide footprint total is then divided by the number of households serviced to produce a "per household" GHG potential number as th e final environmental efficiency measure Thus t he equation for environmental efficiency can simply be stated as: !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 4 The term "CO 2e reads as "carbon dioxide equivalents". While carbon dioxide is the primary greenhouse gas in the Earth's atmosphere, there are other greenhouse gases. The other two primary greenhouse gases are methane (CH 4 ) and nitrous oxide (N 2 O). However each molecule of a greenhouse gas has a different "warming potential" on the climate. T hus CO 2e is based lined at an impact of 1 ton of CO 2 one molecule of methane has a "CO 2e of 19 molecules of CO 2 (EPA, 2012) and nitrous oxide has a "CO 2e of 281 molecules of CO 2 (EPA, 2012). Thus the term "CO 2e translates all the varied greenhouse gas es into one standardized measure.

PAGE 37

! 17 [(Route CO 2e ) + (Disposal CO 2e )] / (# of Household s Served ) Equity: Equity is defined as fairness, impartiality, or equality in service (Sal amon, 2002 ; Oakerson, 1999 ) In this solid waste specific case this translates as how evenly the level of service the breadth of service, and the cost of service is provided across a city. Does everyone receive an equal level of service or are there pock ets of uneven service delivery? Do some communities receive better or worse service than other communities? Do certain communities pay more or less for their service? In gathering data for this project a large N survey and quantitative statistical anal ysis of central core cities 5 in the United States each with a population greater than 50,000 A ssociat ed with the large N analysis, eight qualitative case studies of specific cities were also conducted The large N had a sample of 429 United States citie s that were surveyed. The final response rate from the survey was 102 cities (23.8%). The eight case study cities were among the 102 cities responding to the survey. Both the case studies and the large N analysis explored the three research questions ti ed to the fiscal efficiency, the environmental efficiency, and the equity of the solid waste service delivery in relation to the service delivery method being monocentric, franchise zone, or polycentric in its delivery Dissertation Roadmap There are a to tal of seven chapters for this dissertation. Chapter II presents a review of relevant academic literature. Chapter III presents a more detailed discussion of this dissertations research design. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 5 Appendix C provides a full list of these cities.

PAGE 38

! 18 Chapter s IV and V present the results from the eight case st udy cities. These case studies represent "diverse cases" as defined by Gerring (2007). Diverse cases are utilized to be representative of (in an overarching sense) the full variation of the population. That is to say, the cases represent all relevant ca tegories. There are four monocentric case studies: Austin, TX; Denver, CO; Fresno, CA; and Knoxville, TN. There are two franchise zone case studies: Indianapolis, IN and Phoenix, AZ. Lastly, there are two polycentric case studies: Colorado Springs, CO a nd Fort Collins, CO. In addition to representing the breadth of cases, the cases provide further depth and thickness in their description than the large N statistical analysis alone in particular, adding an over time analysis to the dissertation's findin gs. Chapter V I of this dissertation is a presentation of the statistical analys e s for the 102 cities that responded t o the large N survey ( N sampled = 429) a response rate of 23.8% O rdinary least squares (OLS) regression models were developed and util ized to analyze the dependent variables of fiscal efficiency and environmental efficiency. T he regression models were run on the trash collection service. (That is to say, the fiscal efficiency and environmental efficiency of recyclables collection servi ce and organics collection service were not tested in these models.) D ata analysis method s were also developed for the dependent variable of equity. The determinant to differentiate between equitable and inequitable service delivery w as completed via a w ithin city comparison of three measures: cost of service, level of service production, and breadth of service production. Variation in these service measures on a within city basis was an indication of inequities.

PAGE 39

! 19 Chapter VII is the conclusions chapter and presents a further synthesis of the data analyses from Chapters IV, V, and VI and presents overall conclusions from the research project, including contributions to the academic literature and key findings from both academic significance and practical sig nificance perspectives. Also included within Chapter VII are theoretical limitation, practical limitations, and research limitations of the project. The chapter also includes a discussion of potential future research. Key Dissertation Finding s and Sign ificance This research has significance to managers and researchers of city level urban infrastructure as it provides insights into the performance of the institutional arrangements that govern the production of solid waste collection. Specifically, this dissertation finds that where a city utilizes either a monocentric service production model ( a single sole producer for solid waste collection service production either via direct city service or via a sole private contractor) or a franchise zone service p roduction model, these arrangements are a more efficient form of service delivery than when a city requires individual residents to make their own arrangements for solid waste collection from the contractor of their choic e via a polycentric service product ion model. This dissertation will now turn to Chapter II This chapter is a review of relevant a cademic literature

PAGE 40

! 20 C HAPTER II REVIEW OF THE L ITERATURE The provision and production of public goods and services is arguably the most important function of l ocal government entities. V. Ost rom, Bish, and E. Ostrom (1988) describe local government (including government's contracted producers) as the mechanism for both the provision and the production of local services and public goods. Yates (1977) describes u rban government as ultimately a mechanism for service delivery, citing the examples of fire protection, police protection, garbage collection, and public education. Analyzing the performance of service provision and service production enhances the study o f provision and production of public goods at the local level Thus this literature review and this dissertation examin e service production through the lens of performance utilizing established and defined measures of effectiveness 6 efficiency, and equi ty. A local public economy must be viewed as distinct and separate from the state level unit of analysis. This concept of a "public economy" was intended to realize two goals: (1) to correct the false notion that "public" automatically equated with "the s tate" and (2) to clearly differentiate the concept of a "public economy" from that of a "market economy" (E. Ostrom, 2010). In other words, "to show that it is possible to have systems that are neither markets nor states, and which preserve the autonomy a nd the freedom of choice of the individual" (Aligica & Boettke, 2011, p. 37). A s previously noted, a !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! % While effectiveness is not included as a performance measure for this dissertation, it is commonly paired with efficiency and equity in other studies. Effectiveness i s defined as the successful delivery of a service. Basically an effective program would mean the production of a service meets the provision goals for that service. Cost is not a factor for effectiveness, it is simply a measure of was the job being well done or not (Oakerson, 1999; Dietz, Dolsak, E. Ostrom & Stern, 2002).

PAGE 41

! 21 specific subset of the local public economy, and the focus of this dissertation is the public service industry (V. Ostrom & E. Ostrom, 1991; McGinnis & E Ostrom, 2012). A public service industry involves those organizations (public, private, or not for profit) that are actively engaged in some recognizable domain of the provision or production of public goods and services (e.g. solid waste collection, p olicing, education, drinking water delivery, or wastewater collection) or the management of common pool resources (e.g. a forest or an underground aquifer) (McGinnis, 2011). In a federalist system like the United States there is always some degree of gov ernmental overlap (Elazar, 1987). In the study of local public economy a considerable amount of pen and paper has gone toward the debate surrounding the performance of consolidated and non overlapping public services versus fragmented and overlapping loca l public services (Keating, 1995 ; E. Ostrom, 1972 ). Often the fundamental question has centered on wh ich is more effective and efficient: c onsolidated city services or fragmented city services overlapping city services or non overlapping city services? (V Ostrom, Tiebout & Warren, 1961; Hawley & Zimmer, 1970; V. Ostrom, 1972; E. Ostrom, Parks & Whitaker, 1978; Lind, 1997; Oakerson, 1999). More recently, the question of equity has been added to this debate (Oakerson, 1999; Savitch & Vogel, 2009). On one side of the debate are scholars from the "new regionalism" school of thought, which favors consolidation and non overlap (Lyon & Lowery, 1989a; Rusk, 1993; Pierce, Johnson, & Hall, 1993; Downs, 1994; Seamon & Feiock, 1995; Orfield, 1997). A common "new r egionalism" argument is that service duplication leads to government waste (Hawley & Zimmer, 1970; Lind, 1997). Research with findings that

PAGE 42

! 22 support consolidation improving the scope and quality of municipal services include: Erie, Kirlin, & Rabinovitz, 19 72; Benton & Gamble, 1984; and Seamon & Feiock, 1995. "New regionalism" argues that performance is improved in unified and consolidated city regions. Characteristics of this consolidated model include: local government cooperation within the city regions, reduced negative externalities caused by fragmented and overlapping government structures, and improved equity in service delivery across the city region (Lyon & Lowery, 1989b; Frisken & Norris, 2001). Another side of the debate has found that the provi sion of public goods and services at the city region scale are often better in a polycentric setting. Their argument being that polycentric arrangements promote competition and competition increases efficiency and effectiveness (V. Ostrom, Tiebout & Warre n, 1961; V. Ostrom, 1972; E. Ostrom, Parks, & Whitaker, 1978; Oakerson, 1999; E. Ostrom, 2000). Examples of scholarship favoring polycentricity arguments include studies of policing (i.e. E. Ostrom, Parks & Whitaker, 1978) and studies of education (i.e. T eske, Schneider, Minstrom, & Best, 1993). These scholars come from the field of political economy research, usually focused on "public choice" (Tiebout, 1956; V. Ostrom, Tiebout, & Warren, 1961; V. Ostrom, 1972a; V. Ostrom, 1972b; Oakerson, 1999; McGinnis & E. Ostrom, 2012). "Public choice" is a theory built from classic and welfare economics (Frisken & Norris, 2001; Aligica & Boettke, 2011 Aligi c a & Tarko, 2013; Heikkila, 2006 ). It should be noted that "public choice" scholarship has a number of variant s. For the purposes of this dissertation, the Indiana University branch of "public choice" will be the branch drawn upon (McGinnis & E. Ostrom, 2012 ; Aligica & Tarko, 2013 ).

PAGE 43

! 23 Both "new regionalism" scholars and "public choice" scholars use welfare econom ics as an underpinning theory, with stress on economic cost benefit analysis (Swanstrom, 2001). Both of these research frames play an important role in this dissertation Thus this literature review will provide a more in depth look at each area of schola rship with an eye toward how each measures performance, and whenever possible, specifically what their findings are in relation to measuring effectiveness, efficiency, and equity. The literature review will first look at the "new regionalism" approach to city region research and then will look at the "public choice" approach to local public economy research. Consolidation: Unified without O verlap The concept of consolidation presents a singular and unified government entity or service delivery scheme. The singular government authority has power over the jurisdiction and can control both the provision and the production of civic public goods within the given service area (Lyons & Lowery, 1989a, 1989b). Some "new regionalism" studie s point to increased eff iciency and effectiveness as a justification for consolidation (Hawley & Zimmer, 1970; Downs, 1994; Lind, 1997) As noted by Warner and Hefetz ( 2002, p.445) "Provision of public services by some form of regional government has been promoted on equity, ef ficiency and economic competitiveness grounds The strongest argument used by "new regionalism" scholars 7 however, is that most of their studies point to a consolidated structure reducing inequities by better matching needs and resources (Hawley & Zimme r, 1970; Downs, 1994; Lind, 1997; Feiock & Carr, 1997; Rusk, 1998; Savitch & Vogel, 2000) By contrast, this !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! & "N ew regionalism" scholars themselves make this "strongest argument" claim.

PAGE 44

! 24 literature argues that a fragmented structure increas es inequities in service delivery (Downs, 1994; Rusk, 1998; Savitch & Vogel, 2000, 2008; Sw anstrom, 2001; Warner & Hefetz, 2002). In addition to the effectiveness, efficiency, and equity arguments, s cholars who favor consolidation note the se three additional benefits from consolidation: ( 1 ) a better environment for long range planning and enhan ced planning capacity because "big picture" thinking is possible and encouraged (Feiock & Carr, 1997); ( 2 ) clarity in which local government should be turned to for a particularly good or service (Hawley & Zimmer, 1970; Downs, 1994; Lind, 1997); and (3 ) co nsolidated government structures linked with local level economic growth (Durning, 1989; Owen, 1992). Polycentricity: Fragmentation and Overlap In 1961, Vincent Ostrom, Charles M. Tiebout, & Robert Warren co authored an article titled, "The Organization of Government in Metropolitan Areas: A Theoretical Inquiry." The piece questioned the assumption that "the problem" at the local level was fragmented and overlapping metropolitan governments. "Duplication of services and overlapping jurisdictions are presu med . to be wasteful and inefficient. The proliferation of agencies and the fragmentation of authority are presumed to provoke conflict and create disorder and deadlock" (V. Ostrom, 2008, p.29). The key component of th eir journal essay (Ostrom, Tiebo ut, & Warren, 1961) was their defining and discussing the concept of polycentricity which is a system of local governments characterized by both fragmentation and overlap. Polycentric (or many centers) is defined by Ostrom, Tiebout, & Warren (1961) as nu merous overlapping governmental or service delivery jurisdictions where some combination of competition

PAGE 45

! 25 and / or cooperation between these "many centers" can result in increased effectiveness and efficiency in service provision and / or production. This f ragmentation and overlap can occur at varied levels of government, often with jurisdictions over the same area of service, and always with each entity being formally independent of the other entities (V. Ostrom, Tiebout, Warren, 1961; McGinnis, 1999; McGin nis & E. Ostrom, 2012). This polycentric order was an expansion and generalization of a proposition by Tiebout (1956) that individuals could simply "vote with their feet" when they disagreed with local level politics or services. In a polycentric order th is "vote" did not necessarily require a physical move since multiple overlapping service providers and / or producers were potentially delivering services to the same area (Tiebout, 1956; McGinnis, 1999; McGinnis & E. Ostrom, 2012). A polycentric order mus t meet two institutional conditions linked to their governmental order: "the first is the existence of multiple independent centers of authority; the second is that their independence must not be absolute" (Oakerson & Parks, 2011, p.154). This independenc e yet not absolute independence is built from overarching theories of federalism, where governmental entities are built on a series of vertical and horizontal checks and balances. Despite an individual government's autonomy (whether the entity be at the l ocal, state, or federal level), each entity still has responsibilities and requirements to other government entities Heikkila (2001, p. 50) notes, "This system of government can create a need for jurisdictions to coordinate the provision of goods and ser vices with other jurisdictions." This is true whether these entities are vertically differentiated (federal, state, local) or horizontally differentiated (entities at the same

PAGE 46

! 26 "level" of government but with overlapping jurisdictions of one kind or another ) (Elazar, 1987; V. Ostrom, Bish, E. Ostrom, 1988; Oakerson, 1999). While the local public economy literature states that "overlap" can incorporate both a spatial element and a function element (Oakerson, 1999) this dissertation, in clarifying definitions of overlap / non overlap and fragmentation / consolidation, defines overlap and non overlap first and primarily as a spatial element and secondarily in terms of there needing to be multiple governmental units (provision side) or multiple producers. This d issertation takes the stand that the "functional element" is not defined via overlap and non overlap but instead is defined via fragmentation and consolidation. Thus i f there are multiple operating units (either in terms of provision or production) this i s a fragmented structure (that may be overlapping or non overlapping) and if there is a single operating unit (either in terms of provision or production) this is a consolidated structure. Various empirical studies completed by public choice and / or loca l public economy scholars lend support to the polycentric proposition One of the most prominent of these studies looked at met ropolitan p olicing in the 1970s (E. Ostrom, Parks, & Whitaker, 1978). Th is large N study tested the theory of polycentric servi ce provision against consolidated service provision in relation to police service production. Their findings were that governments that represented smaller populations and had polycentric arrangements were more effective in their service production than g overnments that represented larger populations and had a monocentric arrangement for service production (E. Ostrom, Parks, & Whitaker, 1978; E. Ostrom, 2000). V. Ostrom (2008) noted that in a comparison among the best performing policing department s, the more fragmented metropolitan areas still performed better than their more consolidated counterparts. E.

PAGE 47

! 27 Ostrom noted, polycentric arrangements for any size department (small, medium, or large) generally outperformed cities with consolidated policing servi ces ( E. Ostrom, Parks, Whitaker, 1978; V. Ostrom, Bish, E. Ostrom, 1988; Tooten, 2010). S ubsequent studies have found similar findings in relation to the size and types of public schools in particular municipalities, including Teske, Schneider, Mintrom, & Best (1993). The general consensus among "public choice" scholars has been that fragmented metropolitan government, in the realm of service production / delivery, can be correlated with more effective and more economically efficient service delivery (Bis h, 1971; Bish & V. Ostrom, 1974, E. Ostrom, Parks, & Whitaker, 1978; Teske, Schneider, Mintrom, & Best, 1993; Frisken & Norris, 2001). More Focus on the Spectrum, Less Focus on the Extremes While both "new regionalism" scholars and local public economy s cholars offer quantitative analyses supportive of their arguments, this dissertation's literature review conclude s with the possibility that both groups of scholars are guilty of an excess focus on the extremes while ignoring nuanced arguments in between b oth. Much of the debate has focused on a simple continuum, with full polycentricity on one end and full consolidation on the other end. This, however, ignores that the fragmentation and overlap are independent of one another. For example, as discussed in the introduction chapter this researcher has identified two additional alternative service delivery models: (1) the "franchise zone" arrangement which is a fragmented yet non overlapping and (2) the functional consolidation arrangement which is consoli dated yet overlapping. Neither of these models fit neatly on the previously established polycentric monocentric continuum.

PAGE 48

! 28 Another more nuanced finding from the literature is Savas' (1977, 1987) research into solid waste collection (specifically, trash collection) that measured fiscal efficiency. Savas (1977, 1987) found the monocentric arrangement paired with an exclusive franchise to a private sector service producer was the most fiscally efficient service delivery model. Second to this in terms of fi scal efficiency was a monocentric arrangement where the municipality was the service producer. Savas (1977, 1987) found the least efficient model was a polycentric model Savas' 1977 stud y utilized a telephone survey method. Cities were selected for ph one interviews by meeting one of three criterion: (1) all cities with populations greater than 2,500 people were interviewed in 41 metropolitan areas geographically distributed across the United States, (2) all cities qualifying as "central core cities" ac ross the United States were interviewed, and (3) one half of the remaining cities in 159 metropolitan areas across the United States were interviewed. Savas (1977) found 3 7 percent of the cities relied on a n exclusive city operated arrangement. He termed this arrangement "municipal collection"; his definition for this arrangement matches with this dissertation's definition of a "monocentric public" arrangement. He likewise found 21 percent of cities relied on an exclusive private sector operated arrangeme nt He termed this arrangement "periodic competition / temporary monopoly"; his definition for this arrangement matches with this dissertation's definition of a "monocentric private" arrangement. Lastly Savas found 21 percent of the citie s utilized what he termed "contract collection"; his definition for this arrangement matches with this dissertation's definition of a "polycentric" arrangement. While Savas did not have a specific category for the "franchise zone" arrangement he mentions it within his

PAGE 49

! 29 re search merely as a special case and notes that cities utilizing this arrangement were very small in number and those cities were lumped into "periodic competition / temporary monopoly" (monocentric private) cities (Savas, 1977, 1987, 2005; Ostrom, Bish, an d Ostrom, 1988). The economies of scale question represents another possibility for further expansion of the two literatures. V. Ostrom, Bish, & E. Ostrom (1988) characterize economies of scale as a situation where the average cost of a good or service de creases as a larger number of goods are produced or as a larger number of consumers are delivered a service. Oakerson (1999) has postulated that performance of the institutional arrangem ents may be guided by economies of scale (specifically, how capital i ntensive versus how people intensive a particular service delivery scheme is ) rather than the polycentric monocentric continuum. Oakerson (1999) notes that the local public economy scholarship on the people intensive fields of policing and education hav e found polycentric arrangements to be more effective and efficient. However, Oakerson (1999) also notes that capital intensive environmental service deliveries such as delivering drinking water or the collection and treatment of wastewater show favorabil ity, in terms of efficiency, in the direction of consolidated arrangements. Oakerson's (1999) findings are of note for this dissertation as solid waste collection has features of both a people intensive and a capital intensive service delivery model. In researching costs of trash collection and recycling collection programs, Bohn, Folz, Kinnaman, and Podolsky (2010, p.865) find across the board "evidence of economies of scale in the collection and disposal of municipal solid waste". While Oakerson is ge nerally associated with local public economy scholars, he clearly states, "Neither . fragmentation nor .

PAGE 50

! 30 consolidation should be considered inherently good or bad" (Oakerson, 1999, p.63), but rather that appropriate degrees of each in relation to effectiveness efficiency and equity should be the consideration. This dissertation contributions to both the "new regionalism" literature and the "local public economy" literature by working to refine and expand performance measures of efficiency and e quity while taking a more nuanced approach to questions relating to the polycentric monocentric continuum as well as expanding upon the possible service configurations considered to be a part of this continuum. Chapter III will now turn to fully examining this dissertation's research design.

PAGE 51

! 31 C HAPTER III RESEARCH DESIGN Research Question and Hypotheses This dissertation's overarching research question is: How does the organization of solid waste collection at the city level relate to: (1) fiscal efficien cy, (2) environmental efficiency, and (3 ) equity of service production? This performance based research question has been broken into three hypotheses to better study the questions of fiscal efficiency, environmental efficiency, and equity. Both local pu blic economy literature and new regionalism literature make use of the measures of fiscal efficiency and equity. This research design endeavors to apply a carbon footprinting method (Ramaswami et al, 2011; Hillman & Ramaswami, 2010; US EPA WARM Model, 201 2; US DOE GREET Model, 2012) by which environmental efficiency can be logically added to these measures. The efficiency performance based measurements will be utilized as dependent variable s in regression analyses with service production models for solid w aste collection at the city level serving as these models independent variables For Hypothesis 1.1, delineated below, the service production models of monocentric, polycentric, and franchise zone s will be utilized and will serve as t he independent variab les in the regression analyses Hypothesis 1 .1 : Fiscally Efficient Hypothesis : The monocentric model will be most fiscally efficient. Justification: Lack of service production overlap and economies of scale for the monocentric model will produce the mos t fiscally efficient outcomes. Analysis Process: Ordinary least squares regression models

PAGE 52

! 32 Literature Grounding: Prediction based on Savas (1977, 1987) solid waste studies, where Savas found that a monocentric service delivery model was most efficient from a cost perspective. For Hypothesis 1.2, presented below, the fiscal efficiency hypothesis is further refined. The service production models of polycentric and franchise zones will again be utilized, however with Hypothesis 1.2 the monocentric service pr oduction model is broken into two arrangements: monocentric provision with private sector production (mono private) and monocentric provision with public sector production (mono public). Thus for Hypothesis 1.2 mono public, mono private, polycentric, and franchise zones will serve as the independent variables in the regression analyse s. Hypothesis 1 .2 : Fiscally Efficient Hypothesis: The monocentric mode l with private sector production will be most fiscally efficient. Justification: Lack of service product ion overlap competitive bidding by private sector service producers, and economies of scale for the monocentric model will produce the most fiscally efficient outcomes. Analysis Process: Ordinary least squares regression models Literature Grounding: Pre diction based on Savas (1977, 1987) solid waste studies, where Savas found that a monocentric service delivery model with private sector service provision was most efficient from a cost perspectiv e. 8 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! Savas (1987) full findings were: monocentric model with private sector production was most efficient, fol lowed closely by monocentric model with public sector production, and polycentric model was least fiscally efficient.

PAGE 53

! 33 Hypothesis 2: Environmentally Efficien t Hypothesis : Th e m onocentric model will be mo st environmentally efficient Justification: Lack of service production overlap will lead to a more environmentally efficient outcome because only a single trash truck, rather than multiple trash trucks, will service any one street in non overlapping service production situations. This overlap minimization reduces the multiplied environmental footprint (in terms of fuel consumption) from multiple trash trucks servicing the same streets. Analysis Process: Ordinary least square s regression models. Literature Grounding: Findings from urban area greenhouse gas (GHG) footprint studies by Ramaswami et al. ( E .g. Hillman & Ramaswami, 2 009; Ramaswami et al, 2011; Chavez & Ramaswami, 2011) suggest that minimizing service overlap may al so minimize GHG footprints. Hypothesis 3: Equitable Hypothesis : The m onocentric model will be most equitable. Justification: Citywide service production should lead to consistent service delivery across an entire city and eliminate service disparities from different zones of service and different service producers, allowing the monocentric model to deliver the most equitable service production.

PAGE 54

! 34 Analysis Process : Analysis of d ifferences in service production on a within city basis for cost of service, level of service, and breadth of service 9 Literature Grounding: Equity is currently an under researched realm 10 however the limited existing research does support this hypothesis 's assertion. DeHoog, Lowery, and Lyons (1991) found support for the advant ages of consolidated government based on equity grounds in service delivery and in a different study Lowery, Lyons, and DeHoog (1995) found "citizen satisfaction" with government service delivery was higher in monocentric models than polycentric models. Th ick Description Via Case Studies This dissertation's research design combined a case study research design of eight "diverse" (Gerring, 2007) case studies with a large N quantitative "variance study" (Van De Ven, 2007). The three hypotheses above have a t emporal element that is held to a single time period the se data are specific to the year 2012. Because of this "snapshot" limitation, the case study portion of this dissertation asks a number of questions to tease out information across multiple years. W hile th ese particular data are qualitative in nature rather than quantitative in nature and is limited to the eight case study cities rather than the broader cross section of the 102 cities in the large N quantitative analysis, it does still provide "thick description" insights from an across multiple years perspective rather !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! ( "Level of service" is defined in relation to frequency, ease of utilization by customers, and amenities that enhance/improve the trash c ollection services. "Breadth of service" is defined in relation to the suite of service s offered: trash only; t rash and recyclables collection; or trash, recyclables, and organics collection. 10 This observation is noted by Oakerson (1999) in relation to l ocal public economy research and is noted by Saha and Paterson (2008) in relation to sustainability research of the "3Es" of environment, economy, and equity their finding, among cities surveyed, was that equity was the least explored sustainability measur e.

PAGE 55

! 35 than the more limiting "one year" snap shot. The specifics of these thick description questions will be discussed in the case study research design section below. Overview of Research Design As mentioned above, t his dissertation combine s case study research and l arge N quantitative research The large N study started with a sample p opulation of 429 central core cities with populations greater than 50,000 in the United States. The ful l sample population of 429 cities was surveyed. From th ese 429 cities 128 cities responded to the survey, however not all surveys were complete enough to be utilized for analysis The re were ultimately 26 surveys eliminated because they were incomplete in relation to critical data for the fiscal efficiency dependent variables and critical data for environmental efficiency dependent variables. Ultimately 102 surveys contained sufficient data to conduct the required analyses across the three hypotheses ; t h us the analysis in this dissertation is based on 1 02 survey responses from the sample population of 429 central core cities a final usable response rate of 23.8% The quantit ative portion of the study allow ed for statistical analys e s whereas the case stu dies allow ed for more in depth "thick" descriptions of the three solid waste collection production methods (George & Bennett, 2005). This research design chapter will first discuss the case studies research method s in depth followed by an in depth descrip tion of the large N research method s Methods: Case Study Research Design Eight case studies w ere completed for this dissertation, with a minimum of two cases for each of the three service production models There are four monocentric case studies: Austin TX; Denver, CO; Fresno, CA; and Knoxville, TN. There are two

PAGE 56

! 36 franchise zone case studies: Indianapolis, IN and Phoenix, AZ. Lastly, there are two polycentric case studies: Colorado Springs, CO and Fort Collins, CO. The following narrative presents the iterative process by which the case studies were researched, developed, and written. Table 3.1, below, is a summary of the major events towards the research to write the case studies. Table 3.1: Task s and Timeline for Case Studies Task Case Study Citie s I nvolved Timeline Initial "Engaged Scholarship" background interviews. Colorado Springs, CO Denver, CO Indianapolis, IN Fall 2011 Survey instrument developed. Initial case study cities formally interviewed utilizing survey instrument. Colorado Springs, C O Denver, CO Indianapolis, IN Winter 2012 "Thick description" open ended research questions developed for case studies cities to supplement closed questions from the survey instrument. N/A Spring 2012 Interviews with 4 new case study cities utilizing bot h survey instrument and "thick description" open ended questions. Additional interviews of initial 3 case study cities specifically focused on thick description questions Austin, TX* Colorado Springs, CO** Denver, CO** Fort Collins, CO* Fresno, CA* India napolis, IN** Phoenix, AZ* Fall 2012 Final survey refinements completed in preparation for large N survey release. N/A Winter 2013 Survey list developed and contact information gathered for sample population of the large N survey. N/A Spring 2013 Online survey administered to sample population of large N cities N/A Fall 2013 Final case study city selected from pool of large N cities survey responses and final case study interviews conducted. Knoxville, TN* December 2013 Both survey and "thick descr iption open ended questions administered. ** Only "thick description" open ended questions administered.

PAGE 57

! 37 The initial three case studies were selected from a non random purposive sample to represent the three diverse cases (Gerring, 2007) to be researche d. They were selected based on this researcher's knowledge (from the review of previous studies) of these three cities (City of Denver Climate Action Plan, 2007; www.springsgov.com 2012; Savas, 1987, 2005; V. Ost rom, Bish, E. Ostrom, 1988). E ach case stud y represent ed one of the service production models: Denver, CO: Monocentric (Non fragmented and non overlapping) Indianapolis, IN: Franchise Zones (Fragmented yet non overlapping) Colorado Springs, CO: Polycentri c (Fragmented and overlapping) The development of both the interview questionnaire and the survey instrument were an iterative process. For the initial three case studies p ublic (city) officials from all cities were interviewed. For these three (and ult imately all eight case study interviews), Director level officials were interviewed. These individuals had direct operation and budget knowledge of the programs they were discussing. For Denver, three city officials were interviewed for background. For I ndianapolis, three city officials were interviewed for background. Background interviews for Denver and Indianapolis were conducted prior to the development of the interview questionnaire and were conducted in November 2011. These interviews were more of a discussion format where the researcher laid out his understanding of their solid waste collection system, the broad outline of his research design, and discussion of the city's specific progr ams ensued. Th ese background interviews followed an engaged s cholarship (Van De Ven, 2007) approach and helped inform the ultimate development of the research design including the interview questionnaire and the large N survey instrument.

PAGE 58

! 38 All of these interviews were supplemented with Interne t searches on each of t he city's websites, including documents freely and publicly available on line produced by or on behalf of the cities researched. Census data (2010) available on line was also utilized. The Colorado Springs case study was further supplemented by on line s earches of the private waste haulers' websites and a report on recycling in El Paso County (Skumatz et al, 2012). Because the initial interviews in 2011 were informal and background in nature, i n February 2012 one official from the City of Denver was the n formally in terviewed using the standardized interview questionnaire. Each of the original three individuals from Indianapolis were likewise formally interviewed in February 2012 using the standardized interview questionnaire. For Colorado Springs, one c ity official was interviewed in February 2012 using the standardized interview questionnaire. For Colorado Springs, because the city is not directly involved in trash collection or recyclables collection, the four largest private solid waste service produ cers identified in the city official interview and were then also interviewed in February 2012 Like with city official interviews, director level individuals from the private sector were sought out for the interviews, however, in some cases the interview er was directed instead to public outreach coordinators within the companies. In both cases, however, sufficient answers were obtained for successful completion of the interviews. The number of individuals serving at the director level for a particular ci ty who oversee solid waste collection services is generally a very small number of people

PAGE 59

! 39 Interviewees were generally "selected" by virtue of their assigned position or job title 11 For each of t he case study cities, two to three interviews among varied officials was a sufficient number of interviews to obtain a full picture of the solid waste collection services within any one city. Interviews from within the same city were compared for consistency between answers. The level of consistency was found to be very high. In cases of inconsistencies brief follow up phone calls always resulted in a clarification that lead to consistency between answers. Private waste hauler interviews for the two polycentric case study cities were also compared against the c ity official interviews from the same city for consistency and again the level of consistency was found to be high. The February 2012 interviews includ ed a standardized questionnaire asked of all interviewees (See Appendix B for this interview questionna ire). This data gathering stage was designed to achieve two goals: (1) the standardized questionnaire was a "test run" of what became the survey instrument for the large N data collection (See Appendix B for this survey instrument) and (2) the standardize d questionnaire allowed for the collection of critical information from the case study cities toward addressing the dissertation's three hypotheses Upon completion of the standardized questions, a number of additional open ended questions were asked. Th e additional open ended questions were designed to elicit more in depth responses based on previous answers and / or to clarify items from the previous questions. A fter the initial three city set of interviews, t hese open ended questions were refined into the following five open ended questions. These questions allowed for the incorporat ion of "thick description" elements into the case studies that looked at the research question and hypotheses across multiple !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! "" For example, job titles for interviewees included: Public Works Director, Solid Waste Director, Recycling Director, and Sustainability Director.

PAGE 60

! 40 years. This provided important added insight s to the case studies, as the large N was limited to a one year "snapshot" of data from 2012. (1) A backward looking temporal element Q uestion: Describe recent (within three years) changes to your collection system and perceptions of how this change / these changes have impacted the program's efficiency and equit y? (2) Likewise, interviewees were asked to describe the near future Question: Are there any pending changes (within three years) to your collection system ? Please discuss the decision process that went into making this change / these changes and if the decision can be tied to issues of efficiency and / or equity ? (3) Interviewees were asked to describe program elements Questions: Is there a program element you view as particularly efficient (fiscally or e nvironmentally) or particularly equitable or inequitable ? The same question was posed again asking the interviewee to answer from a "local citizen perception" perspective (4) For f iscal efficiency and environmental efficiency, interviewees were asked if th ey were aware of any causal or explanatory mechanisms. Question: How does your program define "efficiency", what represents "more efficient" conditions in both the fiscal and the environmental context for your program, how have you linked program elements to efficiency?

PAGE 61

! 41 (5) In particular, exploring the nuances of equity w as a key "thick description" element for final case studies. This was in part, because measuring equity proved challenging from a quantitative perspective. Question: Does your program have issues with citizen satisfaction in relation to service delivery, city regulation of service delivery, varied adoption of services made available to residents, and / or changes in progress to service delivery that may have perception implications ? Once re fined and finalized t hese five questions were then asked of the original three case study cities and were utilized when the remaining five case studies cities were interviewe d. The remaining five case study cities were also s elected via non random purpos ive sample While selected non randomly, the purposive sampling was done to ensure both geographic diversity and service delivery diversity. Like the initial three cities, these additional f ive cities also represented the three service production models: Austin, TX; Fresno, CA ; and Knoxville, TN are monocentric cities Phoenix, AZ is a franchise zone city and Fort Collins, CO is a polycentric city. The original city officials from for Colorado Springs, CO; Denver, CO, and Indianapolis, IN were again inte rviewed in October 2012, these interviews were specifically in regard to the five refined / finalized open ended questions. City officials from Austin, TX; Fort Collins, CO; Fresno, CA; and Phoenix, AZ were interviewed in October 2012 using both the stand ardized survey instrument and the standardized open ended questions. City officials from Knoxville, TN were interviewed in December 2013

PAGE 62

! 42 using both the standardized survey instrument and the standardized open ended questions. Methods: Large N Study Resear ch Design The sample population for this study could easily be all cities and towns in the United States. Th e sample population however, for this dissertation has been limited in a number of deliberate ways. In terms of what service delivery is being re searched th is dissertation is restricted onl y to service provided to single family households (as defined by the city being researched ). In terms of cities, the research design is limited to only include cities with a population greater than 50,000, and specifically central core cities in the United States excluding Alaska, Hawaii, Washington D.C., and territories / commonwealths. As of the 2010 Census, there are 719 cities and towns in the United States with a population over 50,000. Applying the centr al core city and lower 48 limitations br ought this number to 429 cities (Appendix C is a full list of the central core city sample population for this study). A central core city has been defined two ways: (1) the exclusion definition would be that any ci ty considered a suburb of another city would be excluded; (2) the inclusion definition would include a historical core municipality (the city around which the other cities grew), or a metropolitan area that contains only one single city 12 The ability to s urvey via the Internet with initial and follow up contacts via e mail moved s urveying the s ampl e population from a cost prohibitive perspective (for example, by mail surveys) to an affordable option. Thus, via the Internet based survey !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 12 In cases where the exclusion definition and the inclusion definition come into conflict, the inclusion d efinition is utilized. Thus for traditional "twin cities", such as Minneapolis St. Paul and Dallas Fort Worth, both cities are included in the study sample population.

PAGE 63

! 43 tool t he full sa mple population of cities was surveyed, rather than restrict ing the survey to a smaller sample The goal of the Internet survey (see Appendix B for the survey instrument) wa s to reach city officials responsible for solid waste operations or contract manage ment and achieve a high enough N across categories to ensure statistically significant analysis of all relevant categories. The project's survey utilized a tailored survey method, using multiple contacts / reminders, (Dillman et al, 2009 ; Folz, 1996 ). Th is survey's design worked to minimize all four of the most common sources of survey error: coverage, sampling, non response, and measurement (Dillman et al, 2009 ; Folz, 1996 ). Surveying the full sample population rather than a sample within the population minimized coverage and sampling errors Non response bias w as minimized by utilizing the tailored survey design method (Dillman et al, 2009 ; Folz, 1996 ) which includes multiple contacts / reminders (three contacts in total for this survey) and varied sur vey delivery as most appropriate to the individuals being surveyed (an example for this design is surveying city officials via the Internet and private waste haulers via phone surveys see next paragraph). Using Savas ( 1977, 1987) studies of solid waste co llection as a benchmark, provides further confirmation that the responses did not suffer from non response bias as the percent breakdowns by service delivery categories in this study provide a close match to the percentage breakdowns Savas found in his stu dies. Measurement error was minimized by the pre testing of the survey instrument 13 and via the initial seven case studies which served as further pilots of the survey !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 13 The researcher has five PhD students from University of Colorado Denver and 10 solid waste Directors from county solid waste management districts in the State of Indiana review the survey.

PAGE 64

! 44 For citi es where the polycentric model wa s utilized, a follow up survey (see Appendix B for the survey) was administered via phone and was sent to the largest (in terms of number of single family household customers) private waste haulers in each polycentric city. From the 102 completed surveys there were a total of 10 polycentric cities. For these cities there was a range of two to five follow up interviews required (based on how many citywide local contractors there were per city) for a total of 26 interviews with private waste haulers. The two criteria to determine if a hauler w as surv eyed were the following: (1) the hauler provide s service citywide and (2) the hauler should have at least 15% of the city's single family household collection market share. Both of these determination s were made via the city survey and were based on the b est professional judgment of the official completing the survey. Variables and Indicators Independent Variable The independent variable for this research design is the type of service production model utilized by a city for single family household solid wa ste collection which is categorized according to three types : a polycentric arrangement, a franchise zone arrangement, or a monocentric arrangement. Table 3. 2 below summarizes the underlying construct that defines each of these three service production ar rangements and summarizes what indicators will be utilized in data collection to identify this construct and what units of measurement will be utilized.

PAGE 65

! 45 Table 3. 2 : Independent Variable Operationalization Concept Underlying Constructs Indicators of Constr ucts Units of Measurement Data Source Monocentric Service Production No Overlap + No Fragmentation 1 Service Zone + 1 Service Producer in Zone # Of Service Zones Per City And # Of Service Producers Per Zone Survey Data: Questions 1.2, 1.3, 1.4, and 1.6 Franchise Zone Service Production No Overlap + Fragmentation >1 Service Zone + 1 Service Producer Per Zone # Of Service Zones Per City And # Of Service Producers Per Zone Survey Data: Questions 1.2, 1.3, 1.4, and 1.6 Polycentric Service Production Overlap + Fragmentation >1 Service Zone + >1 Service Producers Per Zone # Of Service Zones Per City And # Of Service Producers Per Zone Survey Data: Questions 1.2, 1.3, 1.4, and 1.6 For Hypothesis 1.1 this is a single category. For Hypothesis 1.2 this categor y is split into two categories. One where the single service provider was the city itself (Monocentric Public) and one where the single service provider was a private contractor employed by the city (Monocentric Private). The indicators of: (1) numb er of service zones and (2) number of service p roducers per service zone, allow ed the researcher to determine which service production model wa s utilized by each individual city. One service zone (and by default, one service producer per zone) is the key indicator of a monocentric arrangement. Having greater than one service zone, with only one service producer per zone is the key indicator of a franchise zone arrangement. When there is greater than one zone in a city and greater than one service producer per zone th is combination is the key indicator of a polycentric arrangement.

PAGE 66

! 46 Dependent Variables The dependent variables for this research design are: fiscal efficiency, environmental efficiency, and equity. Table 3. 3 below briefly summarizes each dep endent variable and data sources 14 with a more in depth discussion of each following in the narrative Table 3.4 then provides a combined flow diagram of this dissertation's research design, from research question, to hypotheses, independent variables and d ependent variables, and concluding on expected outcomes / expected relationships. Table 3.3 : Dependent Variables Concept, Indicators and Units, and Data Sources Dependent Variable / Concept Indicators (Units) Data Sources Fiscal Efficiency P er househol d cost of service production with lower cost as an indicator of higher efficiency (Dollars per household). Survey data : Questions 2.1 and 2.11 Environmental Efficiency Per household greenhouse gas footprint of service production. Lower CO 2e is an indicat or of higher efficiency. (CO 2e per household) Survey data : Questions 1.1, 2.1, and 2.10; GREET Model Analysis (US DOE, 2012) ; and WARM Model Analysis (US EPA, 2012) Equity Variance in the cost, level, or breadth of service across a city Multiple units: Variance in c ost to households: (Dollars per household) Variance in l evel of service production : ( High or low level of service production for trash collection ) Variance in b readth of service production : (Trash only or trash and recyc ling / trash recycling composting) Survey data : Questions 1.2, 1.3, 2.1, 2.3, 2.10, 2.11 3.1, 3.4, 3.8, 4.1, 4.4, 4.6, 4.10, 4.11, and 4.22 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! ") Data sources included: survey data, census data, and date derived from GREET and WARM models.

PAGE 67

! 47 Table 3.4 : Combined Flow Diagram of this Dissertation's Research Design Research Question Hypotheses Independent Variable (IV) Dependent Variable (DV) Expected Relationship / Expected Outcome How does the organization of solid waste collection at the city level relate to the efficiency and equity of service production? 1.1 M onocentric model will be m ost fiscally efficient. 1.2 Monocentric Private model will b e most fiscally efficient. 1.1: Monocentric Service Production Fiscal Efficiency Most efficient in terms of cost per household 1.2 Monocentric Private and Monocentric Public Fiscal Effic iency Monocentric private most efficient in terms of cost per household Franchise Service Production Fiscal Efficiency Franchise Zone and Polycentric will be less efficient in terms of cost per household than Monocentric Polycentric Service Product ion Fiscal Efficiency Franchise Zone and Polycentric will be less efficient in terms of cost per household than Monocentric M onocentric mode l will be more environmenta lly efficient than franchise zone and polycentric models Monocentric Service Productio n Environ mental Efficiency Most efficient in terms of GHG footprint Franchise Service Production Environ mental Efficiency Franchise Zone and Polycentric will be less efficient in terms of GHG footprint Polycentric Service Production Environ menta l Efficiency Franchise Zone and Polycentric will be less efficient in terms of GHG footprint Monocentric models will be most equitable. Monocentric Service Production Equity Monocentric will be most equitable Franchise Service Production Equity Franch ise zone and Polycentric will be less equitable than monocentric Polycentric Service Production Equity Franchise zone and Polycentric will be less equitable than monocentric Because the de pendent variables of fiscal efficiency, environmental efficienc y, and equity are all presented at the concept level, appropriate indicators of each concept

PAGE 68

! 48 were identified to measure the variables (Goertz, 2006). Each dependent variable, its underlying construct, its indicators, how the indicators w ere operationalize d into a measure, and the units of the measure will be discussed below in turn. Fiscal Efficiency The dependent variable of fiscal efficiency is measured as p er household cost of service production (Dollars per Household). For cities involved in service p roduction, this is obtained by taking the total cost of the trash portion of the solid waste collection program citywide and dividing this by the number of households served. For cities not involved in service production, the proxy measure is both a range and an average of service cost quotes from the largest surveyed (in terms of number of customers) private service producers in the city. Environmental Efficiency The dependent variable of environmental efficiency is a measure of per household environmen tal impact referred to as the "environmental footprint for services received. This environmental footprint is defined in terms of the greenhouse gas (GHG) potential triggered by (1) the solid waste collection service production method utilized and (2) a measure of the GHG potential of the ultimate fate of the materials: landfill disposal, waste to energy, recycling, or composting. Combining these two measures of GHG potential provides a total citywide footprint of solid waste collection. The units for

PAGE 69

! 49 GHG potential will be carbon dioxide equivalents (CO 2e ) 15 This citywide footprint is then divided by the number of households serviced, thus the units for environmental efficiency will be CO 2e per Household. The equation for environmental efficiency can simply be stated as: [(Route CO 2e ) + (Disposal CO 2e )] / (# of Household) The route based CO 2e calculation requires the following route based inputs: Total number of households, number of square miles in the city truck fuel source, and level of collectio n automation. These inputs then allow for a mathematically derived route footprint utilizing United States Department of Energy (US DOE) GREET Model (2012). The US DOE GREET model short for "Greenhouse Gases, Regulated Emissions, and Energy Use in Transpo rtation Model" was developed by the Argonne National Laboratory. The first version of GREET was released in 1996 and has been updated and peer reviewed as additional data become available regarding both greenhouse gases and emission characteristics. It i s a public domain model. This dissertation utilizes GREET's transportation based model from 2012. The model is a full life cycle analysis model T hat is the "footprint" model is a wells to wheels calculation meaning GHG emissions are calculated from th e point of resource extraction to the point of resource consumption / combustion. The model considers the fuel source utilized (diesel and / or natural gas), the engine type (in this case primarily !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! "* The term "CO 2e reads as "carbon dioxide equivalents". While carbon dioxide is the primary greenhouse gas in the Earth's atmosphere, there are other greenhouse gases. The other two primary greenhouse gases are methane (CH 4 ) and nitrous oxide (N 2 O). However each molecule of a greenhouse gas has a different "warming potential" on the climate. Thus CO 2e is based lined at an impact of 1 ton of CO 2 one molecule of methane has a "CO 2e of 19 molecules of CO 2 (EPA, 2012) and nitrous oxide has a "CO 2e of 281 molecules of CO 2 (EPA, 2012). Thus the term "CO 2e tr anslates all the varied greenhouse gases into one standardized measure.

PAGE 70

! 50 standard internal combustion or electric assisted hybrid internal combustion), and the vehicle / fleet characteristics (in this case trash trucks). The model takes into account a full suite of greenhouse gases in the calculati ng the output but the three primary gases considered are: carbon dioxide (CO 2 ), metha ne (CH 4 ), and nitrous oxide (N 2 0) (US DOE, 2012). Once the route based calculation is completed utilizing GREET a multiplier of 2.0 is utilized for collection frequencies of twice a week, a multiplier of 0.5 is utilized for collection frequency of every two weeks, and multiplier of 1.5 is utilized for each time a route is serviced by more than one collection producer. This multiplier is less than 2.0 to reflect that while there is inefficiency in collection route overlap, it is not a simple one to one ra tio as each additional collection producer along the route reduces the number of stops for each of the other producers and thus increases the individual producer's route efficiency in terms of a GHG footprint. The disposal based CO 2e calculation requires t he following disposal based inputs that were provided via survey responses : solid waste tonnage and disposal mechanism (such as: source reduction, recycling, combustion, composting, and landfilling) These inputs allow for a mathematically derived disposa l footprint utilizing United States Environmental Protection Agency's (US EPA) WARM Model (2012). The US EPA WARM model short for "Waste Reduction Model" was developed by the US EPA to "help solid waste planners and organizations track and voluntarily repo rt greenhouse gas (GHG) emission reductions" (US EPA, 2012). The first version of WARM was released in 1998 and has been updated and peer reviewed as additional data ha ve become available and new mat erial types have been added. Like

PAGE 71

! 51 GREET, it is a public domain model. This dissertation utilized the 2012 version (Version 12; Released February 2012) of the model (US EPA, 2012). The model utilizes a life cycle approach to calculate emissions. The model allows for calculations based on individual material types (such as an aluminum can or a newspaper) or for "mixed MSW" which is a weighted national average for municipal solid waste in the Unite d States. This dissertation utiliz ed the "mixed MSW" average for its footprint calculations. The model tracks varie d emission characteristics for five waste "disposal" scenarios: source reduction, recycling, combustion, composting, and landfilling. The model then provides an output in terms of greenhouse gases calculated in CO 2e For solid waste disposal, the two pri mary gases are carbon dioxide (CO 2 ) and methane (CH 4 ) (US EPA, 2012). To prevent double counting, the hauling distance portion of the EPA WARM will not be utilized as the transportation portion of the environmental footprint calculation will be determined utilizing the DOE GREET model. Both the US Department of Energy GREET model and the US Environmental Protection Agency WARM Model are peer reviewed tools dynamic and broad enough to provide a computation for environmental efficiency. Both utilize a lif e cycle analysis based approach to determining greenhouse gas footprints (US DOE, 2012; US EPA, 2012).

PAGE 72

! 52 Inclusions in Env ironmental Efficiency Measure: Route Based Footprint Disposal Footprint "Wells to wheels" fuel production footprint "Collection to d isposal" Fuel consumption along collection route and to final disposal facility "Final disposal" Footprint of ultimate fate of waste Exclusions from Environmental Efficiency Measure: Route Based Footprint Disposal Footprint Footprint of construction of collection vehicle Footprint of maintenance of collection vehicle (beyond fuel) Footprint of ultimate fate (disposal / recycling) of collection vehicle Footprint of materials prior to them becoming waste. (Initial production footprint and footprint over life of usage.) Figure 3.1: Schematic of Boundaries of Environmental Efficiency Measure Equity The dependent variable of equity is a measure of fairness, impartiality, and / o r equality in service production (Salamon, 2002) In this solid waste case, th is translates to three measures: (1) variations in cost per household on a within city basis (obtained from the fiscal efficiency measure); (2) variations in level of service production on a within city basis ("level of service production" including how i t is defined, is discussed in the following section); and (3) variations in the breadth of service production on a withi n city basis ( "breath of service production" including how it is defined, is discussed in the following section) Variation in any of these three measures on a within city basis is an indication of inequities. In the case of both "level of service" and "breadth of service" programs in transition from one service delivery method to another and programs defined by the city completing the survey as "pilot" in nature were excluded from the analysis of equity versus inequity. I n cases where inequit ies w ould have been detected, the original research plan was to conduct further analysis using descriptive statistics to determine if the inequit ies

PAGE 73

! 53 were tied to demographics such as income levels, race, and ethnicity and test if there wa s a statistically significant correlation between the inequities in the service delivery and the demographic differences between the various service areas. As wil l be discussed fully in Chapter V I however, inequities of significance were not found within the cities researched for this dissertation based on how equity was defined within this section Thus the above analysis was ultimately not required for this di ssertation. Despite this it is important to note that the differences in service production d o not have to be tied to income levels, race, and ethnicity for them to be considered of note. Any of the three measures identified above, on a within city bas is is an i ndicat ion of inequity in service delivery whether tied to demograph ics or not. Demographics simply add ed an additional layer of scrutiny to the research. Control Variables T hirteen control variables were identified and utilized in this disserta tion Twelve of the control variables were u sed in testing the hypothesized relationships between fiscal efficiency against monocentric, franchise zone, and polycentric service production and the hypothesized relationships between fiscal efficiency agains t monocentric public, monocentric private, franchise zone, and polycentric service production. Twelve of the control variables were used in testing the hypothesized relationships between environmental efficiency against monocentric, franchise zone, and p olycentric service production Th us eleven control variables were utilized in both models The control variable of w ork force unionization is utilized only in the fiscal efficiency model. The control variable of fuel source is only used in the environme ntal efficiency model. Table 3. 5 below summarizes the control variables, the rationale for

PAGE 74

! 54 including each control variable, a summary of the in dicators and units utilized, and which data collection tool w as utilized to identify the construct for each of t he t hirteen variables. In addition to this, two of these control variables are used as indicators (not controls) for the equity hypothesis. The two utilized in conjunction with equity are level of service production and breadth of service production. The level of service production is a control variable for both the fiscal efficiency hypothesis and the environmental efficiency hypothesis it adds a control for the level of service as compared against the cost and / or environmental footprint of the service It is also utilized as an indicator of equity / inequity for the equity hypothesis. For t he control variable of level of service production two levels of service have been defined: standard or high. A standard level of service production is defined a s either weekly or every two week trash collection and a minimum of monthly collection of bulky trash items. A high level of service production is defined as both categories meeting the minimum for standard and one or both categories providing an addition al level of service, including: a minimum of weekly collection of bulky trash items or more than once a week trash collection. The reason for the inclusion of this control variable is because as the level of service increases the cost of service also incr eases and the environmental footprint also increases. Thus a higher level of service production will result in a higher per household cost (lower fiscal efficiency) and a larger environmental footprint (lower environmental efficiency). The breadth of serv ice production is a control variable for both the fiscal efficiency hypothesis and the environmental efficiency hypothesis, it adds a control for the breath of service as compared against the cost and / or environmental footprint of the

PAGE 75

! 55 service. It is als o utilized as an indicator of equity / inequity for the equity hypothesis. For the control variable of breadth of service, cities are classified as (1) trash only collection or (2) trash and recyclables collection or trash recyclables organics collection service production. (It should be noted that no cities were identified that offered trash organics without recycling .) A trash only level of service production is defined as trash collection only, with no recyclables collection and no organics collection A trash and recyclables level of service production is defined as trash collection service combined with at least monthly collection of recyclables (but no organics collection). A trash recyclables organics level of service production is defined as tras h collection service combined with both recyclables collection and organics collection (both at least monthly). The reason for the inclusion of this control variable is because as the breadth of service decreases more waste is handled via the trash collec tion program, increasing the cost of administering this service. Thus a lower breadth of service production r esult in a higher per household cost (lower fiscal efficiency) and a larger environmental footprint ( lower environmental efficiency). The level of automation of the solid waste collection vehicles in a city is a control variable for both the fiscal efficiency hypothesis and the environmental efficiency hypothesis, it adds a control for the level of automation as compared against the cost and / or en vironmental footprint of the service. This control variable utilized a two point scale for level of automation: (1) fully automated and / or partially automated, and (2) fully manual. Solid waste trade journal s ha ve found a correlation that increasing au tomation deceases operating costs (increases fiscal efficiency) and decreases pollution

PAGE 76

! 56 emissions (increases environmental efficiency) (O'Brien, 2010, 2012; O'Brien & Johnson, 2012) The funding mechanism for service production is a control variable for bo th the fiscal efficiency hypothesis and the environmental efficiency hypothesis, it adds a control for the funding mechanism as compared against the cost and / or environmental footprint of the service. Studies have indicated that how individuals pay for their solid waste collection service s influence their behavior (Skumatz & Freeman, 2006; Folz & Giles, 2002) For example, Skumatz & Freeman (2006) quantified that pay as you throw rate schemes for trash collection over flat rate payment schemes had the e ffect of a statistically significant: decrease in trash disposal, increase in recycling and / or composting, and increase in source reduction (that is, the physical amount of total solid waste was decreased). The funding m echanism control was defined into two categories: (1) Flat Rate (i.e. the customer pays for the service via their taxes or the customer receives a flat rate bill for their trash service regardless of quantities set out for collection); and (2) Variable Rate (i.e. some form of pay as you t hrow (PAYT) where the customer pays based on volume or weight of material set out for collection). The control variable of less than 50,000 households is for both the fiscal efficiency hypothesis and the environmental efficiency hypothesis, it adds a contr ol for the size of the cities being studied. Large cities (in terms of population) are likely to have greater economies of scale than smaller cities and related lower service production costs ( higher environmental and fiscal efficiency). A city's definiti on of what constitutes a single family household, labeled "o ne is one is the first of two control variables related to the density of single family

PAGE 77

! 57 households within a city and is utilized in both the fiscal efficiency hypothesis and the environmental eff iciency hypothesis. The greater the density of single family households per square mile will increase both the fiscal efficiency and the environmental efficiency. One is one is based on how an individual city defines "single family households". One i s one is a "yes" (1) when the city definitio n is truly a single family household and "no" (0) when the definition is anything greater than one such as including duplexes, triplexes, etc. H ousehold density is the second of two control variables related to density of single family households within a city and is utilized in both the fiscal efficiency hypothesis and the environmental efficiency hypothesis. The greater the density of single family households per square mile will increase both the fiscal effi ciency and the environmental efficiency. Household density is a scale based control that is simply the number of single family households in a city divided by the number of square miles in the same city. The control variable of g oal is related to whether or not the specific city has a solid waste specific goal. Research indicates that city specific solid waste goals can influence the efficiency of collection programs ( Skumatz et al., 2013) The control variable of c ontainer d eposit law is an indicator of whether or not a particular city resides in a state with a container deposit law. This control variable is utilized in both the fiscal efficiency hypothesis and the environmental efficiency hypothesis as a proxy for the city being located within a partic ularly environmentally friendly state

PAGE 78

! 58 The control variable of y ard w aste b an is an indicator of whether or not a particular city resides in a state with a ban on the disposal of yard waste in landfills. This control variable is utilized in both the fisca l efficiency hypothesis and the environmental efficiency hypothesis as a proxy for the city being located within a particularly environmentally friendly state. The final control variable utilized in both the fiscal efficiency hypothesis and the environment al efficiency hypotheses is a g eographic region dummy. It utilizes the four Census defined geographic regions of the United States: West, Midwest, Northeast, and South This variable is included because the various regions of the United States have diffe rentiated costs for both the collection and disposal of solid waste. For example, the Northeast geographic region tends to have both the highest solid waste collection costs and highest solid waste disposal costs among these four geographic regions The control variable of work force unionization was included in the analysis to provide a control for workforce u nionization within the fiscal efficiency variable This is because work force unionization is often associated with higher labor costs that in tur n is associated with higher service production costs (lower fiscal efficiency) The truck fuel source was added as a control variable for the environmental efficiency model as the source of fuel for a trash truck has a significant influence on what the ul timate environmental footprint from the operation of the truck (Cannon, 2006). While most vehicles ended up being diesel trucks there is certainly a move in progress across the solid waste collection industry toward more diversified flee ts of vehicles in terms of truck fuel sources The other major fuel source category beyond diesel that was identified among responding cities was natural gas (either compressed or liquid). The

PAGE 79

! 59 reason for the inclusion of this control variable is because a diesel / natural gas mixed fleet wi ll have a lower environmental footprint (higher environmental efficiency) than a diesel only fleet. Table 3.5 below presents all dependent, independent, and control variables utilized within the fiscal efficiency models and environmental efficiency model. Table 3.5 : Dependent, Independent, and Control Variables for Fiscal Efficiency Models and Environmental Efficiency Model Label Signifying Data Source Type of Variable Utilized with Model(s) Fiscal Fiscal Efficiency ($ per Household per Year) Survey data: Questions 2.1 and 2.11 Dependent Fiscal Env Environmental Efficiency (CO 2e per Household per Year) Survey data: Questions 1.1, 2.1, and 2.10; GREET Model Analysis (US DOE, 2012); and WARM Model Analysis (US EPA, 2012) Dependent Envi ronmental Poly Polycentric Service Production Survey Data: Questions 1.2, 1.3, 1.4, and 1.6 Independent Fiscal and Environmental Fran Franchise Zone Service Production Survey Data: Questions 1.2, 1.3, 1.4, and 1.6 Independent Fiscal and Environmental Mo no Monocentric Service Production Survey Data: Questions 1.2, 1.3, 1.4, and 1.6 Independent Fiscal (1.1) and Environmental MonoPub Monocentric with Public Service Production Survey Data: Questions 1.2, 1.3, 1.4, and 1.6 Independent Fiscal (1.2)

PAGE 80

! 60 Label Signifying Data Source Type of Variable Utilized with Model(s) MonoPri M onocentric with Private Service Production Survey Data: Questions 1.2, 1.3, 1.4, and 1.6 Independent Fiscal (1.2) LSP 1 = High Level of Service Production 0 = Standard Level of Service Production Survey Data: Questions 2.3 and 2.10 Control Fiscal and Env ironmental BSP 1 = Low Breadth of Service Production 0 = Standard Breadth of Service Production Survey Data: Questions1.2, 2.1, 3.1, 3.4, 4.1, and 4.6 Control Fiscal and Environmental AUTO 1 = Manual Collection 0 = Automated Collection Survey Data: Ques tions 2.5 and 2.6 Control Fiscal and Environmental FUND 1 = Flat Rate Funding Mechanism 0 = Variable Rate Funding Mechanism Survey Data: Question 2.13 Control Fiscal and Environmental UNIO 1 = Unionized Work Force 0 = Non unionized Work Force Survey Data : Question 1.11 Control Fiscal FUEL 1 = Diesel only Fleet 0 = Diesel and Natural Gas Fleet Survey Data: Question 2.4 Control Environmental Less50K 1 = Less than 50,000 Households 0 = Greater than 50,000 Households Survey Data: Question 2.1 Control Fisca l and Environmental OneIsOne 1 = City Definition of a Single family Household is Truly a Single family Household 0 = City Definition of a Single family Household is more inclusive than just Single family Households (such as Duplexes, Triplexes, Etc.) Surv ey Data: Question 1.1 Control Fiscal and Environmental GOAL 1 = City has a Solid Waste Specific Goal 0 = City does not have a Solid Waste Specific Goal Survey Data: Questions 1.8, 1.9, & 1.10 Control Fiscal and Environmental ConDep 1 = City is within a S tate with a Container Deposit Law 0 = City is not within a State with a Container Deposit Law Based on state city is located within Control Fiscal and Environmental

PAGE 81

! 61 Label Signifying Data Source Type of Variable Utilized with Model(s) YWBan 1 = City is within a State with a Landfill Ban on Yard Waste Disposal 0 = City is wi thin a State without a Landfill Ban on Yard Waste Disposal Based on state city is located within Control Fiscal and Environmental HHDens Household Density = (# Households) per (# Square Miles Citywide) Survey Data: Question 2.1 And Census Data Control Fis cal and Environmental West Western United States Geographic Region: Arizona, California, Colorado, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, and Wyoming (U.S. Census) Based on state city is located within. Census defined region. Geog raphic Control Fiscal and Environmental Midwest Midwestern United States Geographic Region: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, and South Dakota (U.S. Census) Based on state city is located within Census defined region. Geographic Control Fiscal and Environmental NorEast Northeastern United States Geographic Region: Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont (U.S. Census) Based on state city is located within. Census defined region. Geographic Control Fiscal and Environmental South Southern United States Geographic Region: Alabama, Arkansas, Delaware, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Ok lahoma, South Carolina, Tennessee, Texas, Virginia, and West Virginia (U.S. Census) Based on state city is located within. Census defined region. Geographic Control Fiscal and Environmental

PAGE 82

! 62 Data Analysis and Models Ordinary least squares (OLS) regress ion models were developed and utilized to analyze the dependent variables of fiscal efficiency and environmental efficiency. A nalysis of equity was completed descriptively as discussed in the equity section earlier in this chapter This section will prese nt the two fiscal efficiency models and one environmental efficiency model Efficiency Models Ordinary least squares (OLS) regression models w ere run on the trash collection service (not recyclables collection and not organics collection). For the key independent variables, the polycentric service production serv e d as the omitted / baseline variable, with the variables for franchise zone service production and monocentric service production included in the models. Fiscal Efficiency Model for Hypothesis 1.1 Equation 1.1 below is the initial regression equation that was utilized to test Hypothesis 1.1. Multiple linear regressions were run to test the significance of the variables thus equation 1.1 represents the linear regression equation prior to any of the control variables being eliminated. FISCAL 1.1 ) = c + b 1 (D fran ) + b 2 (D mono ) + b 3 (D LSP ) + b 4 (D BSP ) + b 5 (D AUTO ) + b 6 (D FUND ) + b 7 (D UNIO ) + b 8 (D Less50K ) + b 9 (D OneIsOne ) + b 10 (D GOAL ) + b 11 (D ConDep ) + b 12 (D YWBan ) + b 13 (D HHDens ) + b 14 (D WEST ) + b 15 (D MIDWEST ) + b 16 (D SOUTH ) + e Equation 1.1: Initial Linear Regression Equation for Testing Hypothesis 1.1 Fiscal Efficiency Model for Hypothesis 1.2 Equation 1.2 below is the initial regression equation that was utilized to test Hypothesis 1.2. Multiple linear regre ssions were run to test the significance of the

PAGE 83

! 63 variables thus equation 1.2 represents the linear regression equation prior to any of the control variables being eliminated. This second fiscal efficiency model had monocentric service production split into two categories: monocentric service production by the city (MonoPub) and monocentric service production by a single private sector service producer (MonoPri). (FISCAL 1.2 ) = c + b 1 (D fran ) + b 2 (D MonoPub ) b 3 (D MonoPri ) + b 4 (D LSP ) + b 5 (D BSP ) + b 6 (D AUTO ) + b 7 ( D FUND ) + b 8 (D UNIO ) + b 9 (D Less50K ) + b 10 (D OneIsOne ) + b 11 (D GOAL ) + b 12 (D ConDep ) + b 13 (D YWBan ) + b 14 (D HHDens ) + b 15 (D WEST ) + b 16 (D MIDWEST ) + b 17 (D SOUTH ) + e Equation 1.2: Initial Linear Regression Equation for Testing Hypothesis 1.2 Environmental Efficiency Model for Hypothesis 2 Equation 2 below is the initial regression equation that was utilized to test Hypothesis 2. Multiple linear regressions were run to test the significance of the variables thus equation 2 represents the linear regression equation pr ior to any of the control variables being eliminated. (ENV) = c + b 1 (D fran ) + b 2 (D mono ) + b 3 (D LSP ) + b 4 (D BSP ) + b 5 (D AUTO ) + b 6 (D FUND ) + b 7 (D FUEL ) + b 8 (D Less50K ) + b 9 (D OneIsOne ) + b 10 (D GOAL ) + b 11 (D ConDep ) + b 12 (D YWBan ) + b 13 (D HHDens ) + b 14 (D WEST ) + b 15 (D M IDWEST ) + b 16 (D SOUTH ) + e Equation 2: Initial Linear Regression Equation for Testing Hypothesis 2

PAGE 84

! 64 CHAPTER IV COMPARATIVE ANALYSIS ACROSS EIGHT CASE STUD Y CITIES Introduction to Case Studies Eight case studies have been completed for Chapter IV and Ch apter V of this dissertation The eight case studies were selected from a non random purposive sample to represent the three diverse cases (Gerring, 2007) to be researched. Four monocentric (non fragmented and non overlapping) cities were studied: Austin, TX; Denver, CO; Fresno, CA; and Knoxville, TN Two franchise zone (fragmented yet non overlapping) cities were studied: Indianapolis, IN and Phoenix, AZ; and Two polycentric (fragmented and overlapping) cities were studied: Colorado Springs, CO and Fort C ollins, CO. Phone interviews were conducted for all of these case stud y cities. The interview included a set of standardized survey questions asked of all interviewees (see Appendix B for interview instrument) and a set of five standardized open ended que stions also asked of all interviewees (see Chapter III for open ended interview questions) The majority of information presented within these case studies is from these interviews. When information is from a source other than the interviews this source is explicitly cited. The initial cities interviewed were Colorado Springs, CO; Denver, CO; and Indianapolis, IN these cities were first interviewed starting in November 2011 and ending in February 2012 F ollow up interviews were conducted with these thre e cities in October 2012. Austin, TX; Fort Collins, CO; Fres n o, CA; and Phoenix, AZ were also interviewed in October 2012. Finally Knoxville, TN was interviewed in December 2013.

PAGE 85

! 65 Director level p ublic (city) officials were interviewed from all eight cit ies For the eight case studies a total of 17 city officials were interviewed. For the two polycentric cities, follow up interviews with the largest (in terms of number of customers) private waste haulers (Director level or public outreach liaison ) in ea ch city were also conducted. For the two polycentric case study cities a total of seven private waste hauler interviews were conducted. Both Table 4.1 and Figure 4.1 below note the eight case study cities in relation to the service production model they represent. Table 4.1 also includes the number of officials interviewed for each city. Table 4.1: Interviews for Case Study Cities City City Official Interviews Private Waste Hauler Interviews Monocentric Cities Austin, Texas 2 N/A Denver, Colorado 3 N/ A Fresno, California 2 N/A Knoxville, Tennessee 2 N/A Franchise Zone Cities Indianapolis, Indiana 3 N/A Phoenix, Arizona 2 N/A Polycentric Cities Colorado Springs, Colorado 1 4 Fort Collins, Colorado 2 3 Total Interviews 17 7

PAGE 86

! 66 Figure 4.1: Case Study Cities and Service Production Model s All of these interviews were supplemented with Internet searches on each city 's website, including documents freely and publicly available on line that were produced by the cities being research ed Particular attention was paid to publically available budget data 2010 C ensus data was also utilized. For polycentric case studies, research was ! ! ! Polycentric Franchise Zones Monocentric Overlap No Overlap Consolidated Fragmented Case Study Cities : Colorado Springs, CO Fort Collins, CO Case Study Cities : Indianapolis, IN Phoenix, AZ Case Study Cities : Austin, TX Denve r, CO Fresno, CA Knoxville, TN

PAGE 87

! 67 further supplemented by on line searches of the private waste haulers websites and in the case of Color ado Springs a report on recycling in El Paso County by Skumatz et al (2012). Th is chapter first presents comparative information across the eight case study cities with particular attention to this dissertation's three research hypotheses and the "thick d escription" questions specific to the case studies Chapter V then again presents case study data, however Chapter V presents the information as city specific vignettes for each of the eight case stud y cities Comparative Information Across the Eight Cas e Studies This section will present comparative information across the eight case study cities in relation to the dissertations three research hypotheses of fiscal efficiency, environmental efficiency, and equity. Table 4.2 below presents basic demographi c information across the eight cities. Table 4.3 below presents comparative information across the eight cities in terms of solid waste services delivered. Table 4.4 and Figure 4.1 below present findings related to the fiscal efficiency hypothesis for t h e case study cities. Table 4.5 and Figure 4.2 below present findings related to the environmental efficiency hypothesis. Table 4.6 below presents findings related to the equity hypothesis for the case study cities.

PAGE 88

! 68 Table 4.2: Key Demographic Informatio n for Case Study Cities City Service Production Type Population ** Single f amily Households Square Miles ** Austin, TX Monocentric City Service Production 790,968 184,000 298 Denver, CO Monocentric City Service Production 619,968 168,942 155 Fre sno, CA Monocentric City Service Production 494,665 107,000 112 Knoxville, TN Monocentric Mixed Private Contractor and City Service Production*** 178,874 60,000 104 Indianapolis, IN Franchise Zones 829,718 270,000 373 Phoenix, AZ Franchise Zones 1,445,632 354,729 517 Colorado Springs, CO Polycentric 419,427 128,958 195 Fort Collins, CO Polycentric 143,986 60,816 54 Data from this column are from interviews findings ** Data from this column are from 2010 Census data *** Private contractor deli vers trash collection service and recyclables collection service. City delivers organics collection service. While the diverse cases were selected primarily to present variation across the dissertation's three service delivery models of monocentric, fr anchise zone, and polycentric, these cities also present diversity in relation to geography across the United States and in terms of city size. Phoenix is the largest city researched both in terms of

PAGE 89

! 69 population and in terms of single family households ser ved. Fort Collins is the smallest city researched in terms of total population and Knoxville is the smallest city researched in terms of single family households served. Table 4.3: Comparative Service Delivery Information City (Number Households Served ) L evel of Trash Service Breadth of Service Level of Automation for Trash Service Funding Mechanism Monocentric Cities Austin, TX (184,000) Weekly Trash, Periodic Bulky Trash, Recyclable s Organics Automated Variable Rate (PAYT) Denver, CO (168,942) Weekl y Trash, Periodic Bulky Trash, Recyclables, Organics a s a limited pilot Ranging from Manual to Automated Flat Rate Fresno, CA (107,000) Weekly Trash, Periodic Bulky Trash, Recyclables, Organics Automated Variable Rate (PAYT) Knoxville, TN (60,000) Weekl y Trash, Periodic Bulky Trash and Organics, Recyclables as a limited pilot Manual Flat Rate Franchise Zone Cities Indianapolis, IN (270,000) Weekly Trash, Periodic Bulky Trash, Recyclables, Organics limited to fall leaf collection only Ranging from Manua l to Automated Flat Rate Phoenix, AZ (354,729) Twice Per Week Trash, Periodic Bulky Trash, Recyclables, Organics tied to Bulky Collection Automated Flat Rate Polycentric Cities Colorado Springs, CO (128,958) Weekly Trash, Bulky On Demand for Fee Trash, Recyclables Automated Flat Rate or Variable Rate (PAYT) ** Fort Collins, CO (60,816) Weekly Trash, Periodic Bulky Trash, Recyclables Ranging from Manual to Automated Variable Rate ( PAYT ) * PAYT is an abbreviation for "Pay As You Throw" ** Fee for service varies based on customer's contractor selection and payment package offered by selected contractor.

PAGE 90

! 70 Table 4.4 : Comparing Fiscal Efficiency City Total Program Cost (Combined Cost Trash and Recyclables Collection) Number of Households Served Fiscal Effi ciency (Cost per Household) Monocentric Cities Austin, TX $26,128,000 184,000 $142 Denver, CO $24,576,700 168,942 $145 Fresno, CA $16,692,000 107,000 $156 Knoxville, TN $5,673,906 60,000 for trash 20,000 for recyclables $124 Franchise Zone Cities Indianapolis, IN $56,060,413 270,000 $208 Phoenix, AZ $126,439,000 394,000 $322 Polycentric Cities Colorado Springs, CO N/A 128,958 $204 $372 Fort Collins, CO N/A 60,816 $146 $450 The eight case study cities lend support for the dissertation' s hypothesis that the monocentric model will be the most fiscally efficient. Three of the four monocentric cities deliver combined trash and recyclables collection services on a per household basis for less than the two franchise zone case study cities an d the two polycentric case study cities. The fourth monocentric city delivers combined trash and recyclables collection services on a per household basis for less than three of the four other cities (two franchise zone and one polycentric) and is slightly above the low range of one of the two polycentric cities, but well below the high range cost for both of the polycentric cities.

PAGE 91

! 71 Figure 4. 2 below presents a visual representation of th ese data, comparing the three service delivery models. Figure 4. 2 : Comparing Fiscal Efficiency ! ! ! Polycentric Franchise Zones Monocentric Overlap No Overlap Consolidated Fragmented Fiscal Efficiency : $146 $450 / Household / Year Fiscal Efficiency : $208 $322 / Household / Year Fiscal Efficiency : $124 $156 / Household / Year

PAGE 92

! 72 The eight case study cities lend support for the dissertation's hypothesis that the monocentric model will be most environmentally efficient. Both in terms of route based environmental efficiency and combin ed (route plus disposal) based environmental efficiency the case studies lend support for this hypothesis. One of the two polycentric cities, Colorado Springs, was less efficient environmentally efficient than all monocentric and franchise zone cities. T he other polycentric city, Fort Collins, was less environmentally efficient than most monocentric and franchise zone cities the exception s being Phoenix and Knoxville It should be noted that Phoenix is the only case study city with twice a week rather th an once a week trash collection and this addi tional service had an impact on Phoenix's environmental efficiency measure. Fort Collins was more environmentally efficient than Knoxville for the combined collection and disposal measure but was less environme ntally efficient than Knoxville for the route based environmental efficiency measure this finding is likely an indicator of Fort Collins success on the waste diversion front 16 Table 4.5 below provides a breakdown of the environmental efficiency measures a cross the eight case study cities and Figure 4. 3 below presents a visual representation of th e range of data comparing the three service delivery models. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! "% See the Fort Collins case study in Chapter V for a full discussion on the citywide waste diversion efforts.

PAGE 93

! 73 Table 4.5 : Comparing Environmental Efficiency Citywide Totals (CO 2e ) Per Household (CO 2e per House hold) City Route EE Disposal EE Combined EE Route EE Disposal EE Combined EE Monocentric Cities Austin, TX 49,680 123,280 174,800 0.27 0.67 0.95 Denver, CO 25,341 216,246 241,587 0.15 1.28 1.43 Fresno, CA 11,770 90,950 102,720 0.11 0.85 0.96 Knoxvill e, TN 18,600 72,600 91,200 0.31 1.21 1.52 Franchise Zone Cities Indianapolis, IN 64,800 10,800 54,000 0.24 0.04 0.20 Phoenix, AZ 197,000 807,700 1,004,700 0.50 2.05 2.55 Polycentric Cities Colorado Springs, CO 100,587 145,723 246,310 0.78 1.13 1.91 Fort Collins, CO 21,286 38,922 60,208 0.35 0.64 0.99 Indianapolis presents an interesting special case in terms of the disposal based environmental efficiency measure. It is the one case study city that utilizes a waste to energy facility for final di sposal rather than landfilling. The environmental efficiency measure ( utilizing EPA's WARM model ) gives waste to energy a far smaller (and in Indianapolis's case actually a negative) environmental footprint for disposal. This, in turn, gives Indianapolis both a disposal and a combined environmental efficiency that is smaller than the other case study cities.

PAGE 94

! 74 Figure 4. 3 : Comparing Environmental Efficiency The equity measure as developed in this dissertation found equity in service de livery for the eight of the case study cities The equity measure as developed in this dissertation was an analysis of equity on a within city basis This finding corresponds with a similar result among the large N survey cities (see Chapter VI ). While this finding ! ! ! Polycentric Franchis e Zones Monocentric Overlap No Overlap Consolidated Fragmented Environmental Efficiency : Route: 0.35 0.78 Disposal: 0.64 1.13 Total: 0.99 1.91 (Units are tons CO 2e / HH / Yr ) Envir onmental Efficiency : Route: 0.24 0.50 Disposal: 0.04 2.05 Total: 0.20 2.55 (Units are tons CO 2e / HH / Yr ) ! Environmental Efficiency : Route: 0.11 0.31 Disposal: 0.67 1.28 Total: 0.95 1.52 (Units are tons CO 2e / HH / Yr )

PAGE 95

! 75 is still of interest, it is likely also an indication that the equity measure, as de veloped for this dissertation, is a somewhat "blunt" measure. The case studies, however, were successful in terms of revealing some nuances in the area of equ ity. These nuances point to ways by which the cas e study cities could improve the equity in their service production and point s to ways a future equity measure could be refined to better measure inequities. A future, refined, equity measure could better measure inequities in four ways: (1) Look at i nequities in service utilization (three cities noted inequities in how their citizens utilized the services that were available to them ); (2) Look at i nequities created by pilot programs (two cities had unequal service delivery due new programs being in a pilot phase); (3) Look at c itizen s' perception s of inequities in service delivery (two cities noted citizen satisfaction differences because of different service delivery m ethods within the cities); and (4) Whi le all eight cities delivered service in an equitable manner, this measur e was executed on a within city basis. Expanding the equity measure to be an across multiple city comparison rather than a within city only comparison will likely elicit more interes ting equity results. To this end, the large N chapter presents both a discussion of equity as originally outlined i n this dissertation (the within city basis for comparison) and also includes a preliminary additional analysis utilizing the across multiple city basis for comparison. This across multiple city comparative analysis is supported by several observations among the case study cities: (1) T he two polycentric cities had a lower breadth of service (neither offered yard waste collect services) than the mono centric and the franchise zone cities ; and (2) The cost of service production is higher among the polycentric cities than among the monocentric cities

PAGE 96

! 76 Table 4.6 below provides both discussion and observations in relation to equity in service deliv ery (the original dissertation measure) and equity in service utilization (a measure garnered from case study data). Table 4.6 : Comparing Equity City Equity Hypothesis: Equity i n Service Delivery Equity in Service Utilization Discussion Monocentric Cities Austin, TX Equitable Inequities Noted Various neighborhoods have differentiated degrees of program adoption for recyclables and organics collection programs. Denver, CO Equitable + Perceived Inequities Noted Inequities Noted Demand for pilot organic s program exceeds availability Not all trash service is delivered in an identical manner: there is automated dumpsters collection, automated barrel collection, and manual collection. Some residents perceive this difference as an inequity in service delive ry. Various neighborhoods have differentiated degrees of program adoption for recyclables collection programs. Organics collection program is currently a pilot and demand for this service exceeds its availability. Fresno, CA Equitable Inequities Noted Various neighborhoods have differentiated degrees of program adoption for recyclables and organics collection programs. Knoxville, TN Equitable + Demand for pilot recycling program exceeds availability Recyclables collection program is currently a pilot a nd demand for this service exceeds its availability. Franchise Zone Cities Indianapolis, IN Equitable Perceived Inequities Noted Additional Fee for Recycling Not all trash service is delivered in an identical manner: there is automated and manual colle ction. Recyclables collection is only available on an additional fee for service basis. Phoenix, AZ Equitable No Inequities Noted No inequities noted via interviews. Polycentric Cities Colorado Springs, Equitable No Inequities No organics collection offered

PAGE 97

! 77 City Equity Hypothesis: Equity i n Service Delivery Equity in Service Utilization Discussion CO Noted Additional F ee for Recycling from some service providers For all but one hauler, recyclables collection only available on an additional fee for service basis. Fort Collins, CO Equitable No Inequities Noted No organics collection offe red. In defining the original hypothesis tied definition of "equity" it was decided by the researcher that pilot programs would not count as "inequitable" program delivery. Thick Description Discussion Across Case Studies The additional open ended "t hick description" questions were utilized among the case study citie s to accomplish two goals: (1) t o elicit more in depth responses than could be provided by the standardized and closed ended survey questions an d (2) t he questions addressed deficit s in th e Large N data. S pecifically, that the se data w ere from a single year "snap shot" of programs. The "thic k description" questions 17 allowed for the !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! "& The thick description questions asked of the case study cities specifi cally were: (1) Describe recent (within three years) changes to your collection system and perceptions of how this change / these changes have impacted the program's efficiency and equity? (2) Are there any pending changes (within three years) to your coll ection system? Please discuss the decision process that went into making this change / these changes and if the decision can be tied to issues of efficiency and/or equity? (3) Is there a program element you view as particularly efficient (fiscally or envir onmentally) or particularly equitable or inequitable? The same question was posed again, asking the interviewee to answer from a "local citizen perception" perspective. (4) How does your program define "efficiency", what represents "more efficient" condit ions in both the fiscal and the environmental context for your program, how have you linked program elements to efficiency? (5) Does your program have issues with citizen satisfaction in relation to service delivery, city regulation of service delivery, va ried adoption of services made available to residents, and/or changes in progress to service delivery that may have perception implications.

PAGE 98

! 78 interviewees among the case study cities to provide observations across multiple years from both a past focu s direction and a future focus direction. The following sections discuss case study findings that specifically came from these "thick description" questions. However, rather than walk through the questions one by one, these sections blend findings across the questions and across the case studies into the dissertations major topical areas of efficiency and equity. The final section will then discuss citizen satisfaction findings across the case studies in academic literature citizen satisfaction is often used as a proxy measure for equity (See DeHoog, Lowery, & Lyons, 1991 or Lowery, Lyons, DeHoog, 1995). Program Changes Designed to Increase Efficiency The eight case studies are largely a n accurate reflection of trends occurring within the solid waste col lection industry as a whole across the United States today. Many of these trends are occurring specifically to enhance efficiencies in the collection of solid waste. Three trends across the case studies are of note: (1) a trend toward the adoption of aut omated collection methods; (2) pilot programs utilizing natural gas as a fuel source, and (3) adoption of single stream as the recyclables collection method of choice. All of these trends are viewed by the industry as a whole to be cost saving measures th at enhance programs fiscal efficiency. Table 4.7 below summarizes the findings across the eight case studies in relation to these three trends and the narrative below the table discusses each in turn

PAGE 99

! 79 Table 4.7: Comparing Efficiency Enhancing Measures C ity Automated Collection? Natural Gas Powered Vehicles ? Single stream Recycling? Trash Recyclables Organics Monocentric Cities Austin, TX Yes Yes Transitioning ** Yes Yes Denver, CO Transitioning * Yes Yes No Yes Fresno, CA Yes Yes Yes Yes Yes Knox ville, TN No Semi No No Yes Franchise Zone Cities Indianapolis, IN Transitioning Yes N/A No Yes Phoenix, AZ Yes Yes N/A Yes Yes Polycentric Cities Colorado Springs, CO Yes Yes N/A No Yes Fort Collins, CO Transitioning Yes N/A No Yes Transition s tarted within past three years and in progress as of this writing. ** Transition anticipated to begin within the next three years. Automated trash collection is largely viewed by the waste industry as a measure that enhances collection efficiencies (i ncluding both costs and environmental efficiencies). Four of the eight case studies cities operate fully automated trash collection programs; three of the case study cities have begun transition to fully automated collection programs; and one city is util izing manual collection with no plans to transition at this time. For the three cities in transition, Denver, CO; Indianapolis, IN; and Fort Collins, CO all three noted their research indicated the transition would enhance the fiscal efficiency of their p rograms and this efficiency enhancement is what has l argely driven th e transition s (City Official Interviews, 2011, 2012, 2013). On the recycling front th is automation transition is complete. All eight cities utilize cart based fully automated or semi au tomated collection of recyclables and all

PAGE 100

! 80 eight cities utilize single stream recycling. Both of these are viewed by the waste industry to be the most efficient collection system for recyclable collection routes Cart based collection allows for fully aut omation of a route and reflects that recycling is "mainstream" among customers in that recyclable bins should be at a minimum of equal size to trash containers. Using Indianapolis as an example, their recycling program within the past three years moved away from a manual collection program that utilized 18 gallon collection bins to a fully automated collection system that utilized 65 gallon carts This program transition was completed in 2012 (City Official Interviews, 2011, 2012, 2013). On the organic s collection, front two cities utilize automated collection (Denver, CO and Fresno, CA), two cities utilize manual collection (Austin, TX and Knoxville, TN) and four cities d id not have comprehensive, ongoing, organics collection programs (Colorado Springs CO, Fort Collins, CO, Indianapolis, IN, and Phoenix, AZ). Austin officials noted in their interviews a plan to transition their manual and bag based collection program to an automated and cart based collection system by 2015. They explicitly noted two advantages to an automated and cart based collection system: (1) they could operate the program in a more fiscally efficient manner and (2) the transition would allow them to expand their organics collection program to include food wastes as well as the cu rrently collected yard wastes (City Official Interviews, 2011, 2012, 2013; Private Hauler Interviews, 2012). Austin, TX; Fresno, CA; and Phoenix, AZ are the t hree case study cities currently testing the feasibility of natural gas as a collection fleet f uel source All three cities are having favorable findings, all three noted that as of this writing the operating costs,

PAGE 101

! 81 maintenance costs, and fuel costs for natural gas vehicles were lower than for diesel powered vehicles. Likewise, all three noted tha t the emissions profile of a natural gas vehicle was far less polluting than a diesel powered vehicle. All three cities saw a strong fiscal efficiency and environmental efficiency argument in favor of natural gas powered vehicles. The City of Austin inte rviewees further noted that a mixed fleet of natural gas vehicles and diesel powered vehicles was a fleet more resilient to fuel price spikes from either natural gas or diesel (City Official Interviews, 2012). Other Efficiency Related Items of Note Of the eight case study cities, Denver, CO had the most unique trash collection program Interviews revealed three distinct trash collection methods utilized within the city for single family households : (1) Automated dumpster collection ( 65,000 households ) (2) Manual collection ( 47,990 households ) and (3) Automated collection utilizing 95 gallon barrels ( 54,890 households ) (City Official Interviews, 2011 & 2012 ). Table 4.8 below summarizes key aspects of these collection programs. Table 4.8: Denver's Varied T rash Collection Methods Collection Method Percentage of Citywide Households Served by this Collection Method Percentage of Total Collection Time Dedicated to this Collection Method Percentage of Total Citywide Collection Costs Percentage of C it ywide Tons o f Waste C ollected Automated Dumpster Collection 39% 47% 43% 47% Manual Collection 28% 19% 24% 19% Automated Cart Collection 33% 31% 33% 31% Bulky Collection Events 100%* 3% N/A ( Incorporated in Costs Above ) 3% A periodic "bulky item" collection prog ram supplements all three of these collection programs. This program is provided to residents of Denver once every 9 weeks on an ongoing and rotating basis. (City Official Interviews, 2011 & 2012; General Fund, 2011; Denver Master Plan, 2010; www.denvergov.org 2013).

PAGE 102

! 82 Denver has completed in depth analyse s of their current collection schemes in terms of efficiency measures. At first glance, the dumpster collection would appear to be most efficient, given that the stops for the trash truck are reduced approximately five fold (in that there is one dumpster per four to five houses). However, the city finds this efficiency is not realized as the dumpsters collect nearly twice the waste that similar communities ser viced by the automated barrel and manual programs collect. A large open container without a single family attached to it as "owner" appears to be an open invitation for illicit dumping. Instead, the city finds their automated barrel program to be the mos t efficient. The combination of an automated trash route in combination with lower trash per household translates into this collection program being most efficient. While the manual collection portions of the city collect volumes on a per household basis similar to the automated barrel portions of the city, these manual trucks requires crews of 2 individuals per truck; whereas both the automated dumpster collection and the automated barrel collection only require 1 individual per truck thus from a fiscal efficiency point of view the automated barrel collection is superior to the manual collection (City Official Interviews, 2011 7 2012; Denver Master Plan, 2010; www.denvergov.org 2012). City officials interviewed explicitly noted that Denver is moving in the direction of automated barrel collection. However, because of tight alleys in parts of the city, the geometry of older streets will preclude the full citywide adoption of this automated collection method. The eventual goal, however, is automated barrel collection wherever it can be accommodated (City Official Interviews, 2012; Denver Master Plan, 2010; www.denvergov.org 2012).

PAGE 103

! 83 For the currently manual collection househ olds, Denver estimates 37,200 households will be able to be converted to fully automated service and 15,400 households will need to be converted to semi automated service. For the currently dumpster collection households, Denver estimates 62,400 household s will be able to be converted to fully automated service and 8,800 households will need to be converted to semi automated service (Denver Master Plan, 2010). In addition to budget related efficiency concerns, the city is likewise interested in moving away from manual collection because of worker injury issues. In 2008, the manual collection program reported 26 workman's compensation injuries. By contrast, the combined collection programs of barrel and dumpster collection, both automated, reported only 8 workman's compensation injuries (City Official Interviews, 2011 & 2012; General Fund, 2011; Denver Master Plan, 2010; www.denvergov.org 2013). Interviews revealed Denver utilizes a fleet of almost 200 diesel powere d vehicles. Collection specific vehicles include: 21 automated barrel collection vehicles, 47 manual rear load vehicles, 31 automated dumpster side load collection vehicles, and 19 automated recyclables collection vehicles. An additional long term effici ency consideration for the city is that under the current suite of collection systems, none of the fleet vehicles can be used interchangeably. This forces the Solid Waste Management Agency to maintain four separate fleets of vehicles (City Official Interv iews, 2011 & 2012). Contrasting Two Cases: Efficiency Versus Equity The academic literature often discusses how different performance measures can come into conflict with one another. Two of the case study cities present an interesting

PAGE 104

! 84 contrast on this fr ont. Denver, CO has made a programmatic decision that places a higher value on efficiency than equity. Faced with the same decision, Austin, TX opted for a decision that placed the premium on equity over efficiency. When Denver created their citywide r ecyclables collection program, they required residents to subscribe to this service by calling the city and requesting a recycling bin this despite both the bin and the service being provided to residents free of direct charge. In discussing this decisi on with city officials, they noted that this "opt in" method forced residents to make a commitment to recycling. The city officials felt that if a resident was not going to recycl e they would likewise not "opt in" and the city could more efficiently opera te the recycling program because only those truly interested in recycling would actually opt in to the program. This "opt in", however, has resulted in a high degree of variation in program adoption on a neighborhood by neighborhood basis. Thus w h ile th e service is available in an equitable manner the adoption by residents is highly differentiated. This differentiation often follows socio economic divisions with the lowest income neighborhood generally having lower recycling adoption rates and higher i ncome neighborhoods generally having higher recycling adoption rates (City Official Interviews 2011 & 2012). By contrast, Austin deliberately requires residents to "opt out" o f recycling. Their default is for residents to automatically be provided a free of charge recycling bin and free of direct charge recyclables collection. City officials interviewed in Austin noted that they wanted to do whatever was possible to maximize recycling rates and their feeling was that an "opt out" rather than an "opt in" w as a solution that allowed efficien t adoption of their voluntary collection program but at the same time was more equitable

PAGE 105

! 85 than the "opt in". The city officials in Austin noted they already had lower recycling rates among low income and Spanish speaking households than higher income and English speaking households and they felt requiring an "opt in" simply created an additional barrier to a community they were already struggling to encourage to recycle (City Official Interviews 2012). Program Changes Desi gned to Increase Equity A s noted in the above section, some case study cities noted differences in the adoption of recycling and composting programs on a within city basis. This section continues the discussion started regarding Austin, TX and Denver, CO on this matter and adds findings from Fresno, CA to the discussion as well. Austin, TX officials noted in interviews that neighborhoods with a higher percentage of Latino residents and Spanish speaking residents tended to have lower recycling and organic s recycling rates. The city is working to address this disparity by targeting recycling education and marketing geared specifically to these communities / neighborhoods rather than by retooling the physical collection program itself (City Official Intervi ews, 2012). Denver, CO officials noted in interviews that lower income neighborhood tended to have lower recycling rates than higher income neighborhoods. Much like Austin, the city is working to address this disparity via targeted education campaigns rat her than via retooling the physical collection program (City Official Interviews, 2011 & 2012). Fresno, CA officials likewise noted in interviews this service utilization inequity. In Fresno a combination of the most affluent and least affluent neighborho ods were noted as least likely to utilize recyclables and organics collection services. For the low income

PAGE 106

! 86 neighborhoods, the city has been targeting stronger education programs to improve participation. For the affluent neighborhoods, there appears to b e a social norming opposed to recycling. The city has not as yet found an education based means to successfully address this lack of participation problem among the affluent Of these three programs, Fresno is the only city to "mandate" citizen participa tion in recycling and organics programs. Fresno, however, has not as yet resorted to fining residents for non compliance (City Official Interviews, 2012). Other Equity Related Items of Note Prior to beginning research for this dissertation the researcher made a decision that programs described by cities as "pilot programs" would not count as an inequity in service delivery. After completing research, two of the eight case study cities in fact did have programs described by them as "pilot programs". The C ity of Denver has been operating a pilot yard waste collection program for more than the past three years. The City of Knoxville has likewise been operating a pilot recyclables collection program for more than three years (City Official Interviews, 2011, 2012, 2013) In both cases, city officials interviewed noted there was a citizen perception of inequity created by these programs. Both programs had a cap on participation and both programs had reached this cap. Knoxville's pilot was citywide however D enver's pilot was neighborhood specific. Thus Denver's pilot created an added feeling of inequity among citizens in neighborhoods explicitly excluded from the pilot. While both programs have plans to expand to service all single family households citywid e 18 it is clear that when pilots operating over multiple years (in this case at least three years) a !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! "' Denver plans to begin taking their program citywide starting in 2015. Knoxville does not, as of this writing, have a clear timeframe for citywide expansion.

PAGE 107

! 87 feeling of inequity develops among citizens not served by the pilot (City Official Interviews, 2011, 2012, 2013) Program Changes and Citizen Satisfactio n Citizen satisfaction is often used as a proxy measure for equity (DeHoog, Lowery, & Lyons, 1991; Lowery, Lyons, DeHoog, 1995). A ll of the case study cities have done some form of formal or informal assessments of citizen satisfaction. This section will first discuss the case studies cities that have completed some kind of formal review of citizen satisfaction. This section will then conclude with a discussion of one very public failure on the citizen satisfaction front by one of the other case study ci ties. Among the case study cities four stood out as paying a very high level of attention to citizen satisfaction: Austin, TX; Denver, CO; Indianapolis, IN; and Fort Collins, CO A fifth worth mentioning briefly as well is Knoxville, TN. Among the case s tudy cities Knoxville's trash collection program is somewhat "old school" with a manual collection system and no coordination of collection days between trash, recyclables, and organics. With that said, city officials reported a high level of citizen sati sfaction with the collection program. This combined with a feeling within the city to go with what has proven to work leads to a program in Knoxville that delivers "tried and true" trash collection with no pending plans to make changes in the program at t his time 19 (City Official Interviews, 2013). Austin, TX; Denver, CO; Fort Collins, CO; and Indianapolis, IN have all completed some form of citizen satisfaction review and in completing these reviews have opted to make (or not make) program changes based o n citizen input. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! "( The exception being an eventual expansion of the pilot recycling program to a program serving all single family households in the city.

PAGE 108

! 88 Austin, TX sees their prog ram as being a progressive inte grated solid waste management system not merely a trash collection program. The feeling among Austin officials is that their local citizens demand this type of progressive program a nd they base many of their program revision and expansion plans on citizen surveys, citizen public input, and they program closely monitors citizen participation (City Official Interviews, 2012) Denver CO conducted a survey of residents in 2010 via a thir d party organization specifically to rate citizen satisfaction with solid waste collection services (Denver Master Plan, 2010) The survey had a 34% response rate with the following key findings: 78% of respondents rated refuse collection services as good to excellent, 66% of respondents rated recycling collection services as good to excellent, and 42% of respondents rated organics collection as good to excellent ( Denver Master Plan, 2010) The low satisfaction number for organics wa s not surprising given the program was a small pilot in 2010 when the survey was conducted. This, among other survey results, has largely guided the city toward planned expansion to a citywide organics collection program. As a part of an evaluation of their solid waste collecti on methods in 2008, Fort Collins, CO looked at a number of alternatives designed to increase the fiscal efficiency and effectiveness of collection programs. The alternative most seriously considered was to transition from the polycentric arrangement curre ntly in place to a franchise zone based arrangement (termed "districted collection" in city documents and reports to the city) (City Official Interviews, 2012; Trash Service Study, 2008).

PAGE 109

! 89 In 2010 Indianapolis, IN completed a major transition of their rec yclables collection program. This transition was driven, in part, by low citizen satisfaction with the old recyclables collection program. The previous fully manual collecti on program with dual stream 20 18 gallon bins was provided on a fee for service bas is and suffered from very low participation rates for a citywide program at the lowest point participation was at a slim 10% of city residents subscribing to the service. The new fully automated single stream 90 gallon cart collection program is, as of th is writing, enjoying subscription rates of approximately 50%. In 2009 the city council of Fort Collins directed city staff to investigate the creation of a pilot program in one area of the city utilizing a "districted collection" (franchise zone) arrangeme nt. As a part of this investigation the city when as far as completing a competitive bid process for this zone. The bid results indicated there was the potential of significant cost savings for residents within the zone with no change in level of service or breadth of service, however residents would lose the ability to choose which hauler they desired to receive service from. After considerable public input and debate the city council decided to maintain the current system in place and not execute the pi lot initiative. Further discussions about franchising trash and / or recyclables collection services were discontinued at that time (City Official Interviews, 2012). This case of non action by the city council of Fort Collins, based on citizen input, is in contrast to our final case study example where citizen input was largely ignored. While the service production in Fresno CA is currently (as of March 2014) an in house, city produced, monocentric service production model there has been significant !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! #, With a "dual stream" program one container is dedicated to accepted p aper products and one container is dedicated to accepted glass, plastic, and metal containers

PAGE 110

! 90 disc ussion and debate within the city in terms of moving to a different service production model ( www.fresno.gov 2014). From late 2012 through the summer of 2013 the city actively considered moving from city service prod uction to contract ed out service delivery by a single, citywide, private contractor (Copeland, 2013; Gerlat, 2013; Miller 2013). In December of 2012 the city council first voted to accept a bid from Mid Valley Disposal a Fresno based waste collection firm The city council cited that the new agreement would save the city $2.5 million on an annual basis. The contract was schedule to begin in March 2013 but was immediately challenged in courts by the union representing employees in the Solid Waste Manageme nt Division. The union described the contract as a "rushed" attempt at privatization (City Official Interviews, 2012; Gerlat, 2013; Copeland, 2013). When the council in March 2013 again voted to move forward with the privatization, the union and concerne d citizens opposed to the privatization successful ly o btain ed the number of signatures required to hold a voter referendum on the topic. The privatization option was narrowly defeated by voter referendum in the summer of 2013 (Miller, 2013). Conclusion s from Case Studie s The eight case study cities present a varied group of cities in terms of solid waste collection. In addition to representing all three of the service production models presented in this dissertation, the case study cities also represent a number of other diversities:

PAGE 111

! 91 A geographic diversity ( 5 Western Cities, 2 Eastern Cities, and 1 Midwest City) ; A service delivery diversity : from fully automated collect to fully manual collection, pay as you throw to flat rates, programs with and without organics collection option, etc.; and A diversity of public and private service delivery. Each of the eight case stud y cities offers valuable vignettes of information. Austin, TX is a solid waste collection program that sets a high bar in relation to bot h service delivery and waste diversion and does both to meet the demands of their local citizens. Denver, CO is a trash collection program in transition based on demands for greater fiscal efficiency in service delivery and an organics collection program in growth and transition based on local citizens demand for the service. Fresno, CA is a solid collection program facing internal and public support strife, but the dialog is built on demands for greater fiscal efficiency. Knoxville, TN is a "tried and t rue" program that plans to stick with what works, and it does so at the greatest fiscal efficiency of all 8 case study cities. Indianapolis, IN is a program that balances managed competition with quality service delivery. Phoenix, AZ is a program offerin g a higher level of service than other case study cities but for a directly related higher service cost. Colorado Springs, CO is a program delivering small government and private service delivery as demanded by their local citizens. Fort Collins, CO demo nstrates that managed competition, including government involvement in service provision private sector service production, and listening to local citizens can all be successfully merged into a quality service delivery.

PAGE 112

! 92 Fiscal Efficiency In t erms of t his dissertation's research hypotheses, the case studies each lend support to all three of the hypotheses The clearest support is for H ypothesis 1 fiscal efficiency. Table 4.9 below summarizes these finding in relation to the fiscal efficiency hypothes is. While the case studies also lend support in relation to environmental efficiency and equity, these findings require more of a narrative discussion of each Each is presented in turn below. Table 4.9: Summary of Fiscal Efficiency Hypothesis Findings City Number of Households Served Fiscal Efficiency (Cost per Household) Hypothesis 1.1: The monocentric model will be most fiscally efficient. Hypothesis 1.2: The monocentric model with private sector production will be most fiscally efficient. Mono centric Cities Austin, TX (Public Production) 184,000 $142 Denver, CO (Public Production) 168,942 $145 Fresno, CA (Public Production) 107,000 $156 Knoxville, TN (Private Production) 60,000 $124 Franchise Zone Cities Indianapolis, IN 270,000 $208 Phoenix, AZ 394,000 $322 Polycentric Cities Colorado Springs, CO 128,958 $204 $372 Fort Collins, CO 60,816 $146 $450

PAGE 113

! 93 Environmental Efficiency F or route based e nvironmental e fficiency, the monocentric cities were found to be most environmental ly efficient 21 Table 4.10 below summarizes th ese finding s and ranks the case study cities from most environmental efficient to least environmentally efficient in terms of routes. Table 4.10: Ranking Case Study Cities by Route Based Environmental Efficien cy Efficiency City Service Production Model Route EE (CO 2e per Household) Most Efficient City Least Efficient City Fresno, CA Monocentric 0.11 Denver, CO Monocentric 0.15 Indianapolis, IN Franchise Zone 0.24 Austin, TX Monocentric 0.27 Kno xville, TN Monocentric 0.31 Fort Collins, CO Polycentric 0.35 Phoenix, AZ Franchise Zone 0.50 Colorado Springs, CO Polycentric 0.78 For the combined environmental efficiency measure (route based environmental efficiency plus disposal based environm ental efficiency), the results are less certain but still lean in favor of the predict ion of the research hypothesis. Two of the four monocentric case study cities are within the "top four" most environmentally efficient cities (Austin and Fresno) A ll f our of these cities have a CO 2e of less than 1 ton per household per year. The other two cities in the top four, one is Polycentric (Fort Collins) and one is Franchise Zone (Indianapolis). The next two most efficiency cities are the remaining two monocen tric cities (Denver and Knoxville), both of these cities have a CO 2e of greater than 1 but less than 1.5 tons per household per year. The last two cities !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! #" However, Indianapolis, IN (a franchise zone city) was also within the range of the monocentric cities in terms of route based environmental efficiency.

PAGE 114

! 94 are the least environmentally efficient, both with CO 2e of approximately 2 tons per household per yea r. Table 4.11 below summarizes these findings and ranks the case study cities from most environmental efficient to least environmentally efficient in terms of total environmental efficiency. Table 4.11: Ranking Case Study Cities by Total Environmental Eff iciency Efficiency City Service Production Model Route EE (CO 2e per Household) Most Efficient City Least Efficient City Indianapolis, IN Franchise Zone 0.20 Austin, TX Monocentric 0.95 Fresno, CA Monocentric 0.96 Fort Collins, CO Polycentric 0.99 Denver, CO Monocentric 1.43 Knoxville, TN Monocentric 1.52 Colorado Springs, CO Polycent ric 1.91 Phoenix, AZ Franchise Zone 2.55 The findings for disposal based environmental efficiency were the least definitive in relation to the research hypothesis. Factors beyond what service production model a city utilize s likely drive this disparit y. For example, the most environmentally efficiency city is Indianapolis. This is primarily because Indianapolis disposal based environmental efficiency of 0.04 tons of CO 2e per household per year is driven by the waste going to a waste to energy facili ty rather than to a landfill 22 Likewise Austin's comparatively low 0.67 tons of CO 2e per household per year and Fort Collins' comparatively low 0.64 tons of CO 2e per household per year are both likely driven by those two cities aggressive !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! ## The EP A WARM model utilized to calculate greenhouse gas footprints (in combination with the formula in Chapter III) takes into account the avoided energy production from higher carbon alternatives versus the carbon actually released by waste to energy facilities (WARM, 2013). This is how a negative number is generated in this case.

PAGE 115

! 95 waste diversion efforts rather than their service delivery model. Table 4.12 below summarizes these findings and ranks the case study cities from most environmental efficient to least environmentally efficient in terms of disposal based environmental efficiency. Table 4. 12: Ranking Case Study Cities by Disposal Based Environmental Efficiency Efficiency City Service Production Model Route EE (CO 2e per Household) Most Efficient City Least Efficient City Indianapolis, IN Franchise Zone 0.04 Fort Collins, CO Polyc entric 0.64 Austin, TX Monocentric 0.67 Fresno, CA Monocentric 0.85 Colorado Springs, CO Polycentric 1.13 Knoxville, TN Monocentric 1.21 Denver, CO Monocentric 1.28 Phoenix, AZ Franchise Zone 2.05 Equity The e quity hypothesis as presented in Chapter III was Hypothesis 3: Monocentric models will be most equitable." This hypothesis looked at equity in three ways: cost of service, level of service, and breadth of service but it did so on a within city basis. Across all eight case study citie s, the finding was equity in all three categories on a within city basis. Thus this finding alone did not lend support to H ypothesis 3. However, when case studies are researched on an across multiple cities perspective, two of the three measures do indee d lend support to Hypothesis 3 On a cost of service basis the monocentric case studies cities have the lowest cost on a per household basis. The franchise zone case study cities and the polycentric case study cities both deliver services at a higher per household cost than the monocentric cities. In terms of breadth of service,

PAGE 116

! 96 the monocentric case study cities each present a more varied suite of services than franchise zone case study cities and polycentric case study cities. All four monocentric citi es offer trash collection, recyclables collection, and organics collection 23 By contrast, both franchise zone cities offered trash collection, recyclables collection, and organics collection but organics only on a very limited basis. Finally, both polyce ntric cities only offered trash collection and recyclables collection and neither offered any single family household organics collection. In conclusion, the case studies lend support to the dissertations research hypotheses, they expand the dissertations frame of research by providing perspectives across multiple years rather than only the large N 's single year snapshot perspective 24 and particularly for the equity hypothesis the case studies offer insights as to how future research could better measure eq uity at the city level. If the reader is interested in more of a vignette based approach to each case study city rather than comparative information across case studies, Chapter V again presents case study information but does so via a city specific, city by city, case study approach. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! #$ However, as previously noted Denver's organic program is a pilot and is at this time expanding to a citywide service. #) The perspectives, however, are limited to city specific insig hts and do not provide sufficient evidence to make predictions across multiple cities nor across different service production methods.

PAGE 117

! 97 CHAPTER V INDIVIDUAL CASE STUDIES While the information presented in this chapter is largely the same as the information presented in Chapter IV the formats differ between the two chapters. This Chapter presents in depth individual, case studies for each of the eight case study cities Thus each case study provides a vignette of each city's operation in relation to trash, recyclables, and organics collection for single family households. By contrast, Chapter IV presents this information in a comparative manner across all eight of the case study cities Within this chapter, m onocentric cities are presented first, in alphabetical order : Austin, TX; Denver, CO; Fresno, CA; and Knoxville, TN. F ranchise zone cities are presen ted next, likewise in alphabetical order : Indianapolis, IN and Phoenix, AZ. F inally polycentric cities are presented last also in alphabetical order : Colorado Springs, CO and Fort Collins, CO Within each in depth case information is presented in the fo llowing order: brief introduction to the city, an overview of the city's solid waste collection program, in depth discussion of the trash collection program, in depth discussion of the recyclables collection program, in depth discussion of the organic / co mposting collection program, key findings from the three research hypotheses of fiscal efficiency, environmental efficiency, and equity, and then each case study concludes with a qualitative discussion section in relation to efficiency and / or equity, and a brief case study conclusion.

PAGE 118

! 98 The City of Austin, Texas Introduction to Austin With a population of 790,390 (2010 census) and a square mile city footprint of 298 square miles the City of Austin is the 13 th most populous city in the United States. The city is a manager council system of municipal government with a six member city council plus the mayor who also serves as the seventh member of the city council. Austin is also the capital of the State of Texas. Introduction to Austin's Solid Waste Colle ction Interviews revealed t he City of Austin produces trash collection services, recyclables collection services and organics collection services citywide for single family households defined by the city as single family households (one unit, freestanding ) duplexes (two units), and triplexes (three units) within the city. Citywide this represents 184,000 units that receive solid waste collection. The service is delivered directly by city employees through Austin Resource Recovery, a utility enterprise d epartm ent that is solely funded via utility fee. Collection of trash is provided on a weekly basis. Collection of recyclables is provided every two weeks. Collection of organics is provided on a weekly basis. For trash, recyclables, and organics collec tion the citywide collection method falls under the definition of a monocentric arrangement because: the city itself delivers the service, the service is delivered across the full service area of the city, and there is no overlap in the service production (City Official Interviews, 2012)

PAGE 119

! 99 Austin Trash Collection According to city official interviews, there are 184,000 housing units in Austin that are both eligible for and actively serviced by city trash collection services. The vast majority of this coll ection service is fully automated utilizing side load collection vehicles. For particularly tight alleyway routes, or particularly tight dead end routes, semi automated rear load collection vehicles service the remaining routes (City Official Interviews, 2 012) The city issue s black barrels to all customers for trash collection. In conjunction with a pay as you throw rate structure, residents have a choice of size: 96 gallon, 64 gallon, 32 gallon, or 24 gallon containers While the barrels are provided f ree of charge, the customer's fee for service for the trash collection is based on the size of barrel the customer selects The collection serv ice is billed to customers on a monthly basis via a utility bill. 60% of the customers opt for the 64 gallon tr ash barrel. Automated collection represents 90% of the daily solid waste collection activities for the department. These routes require only one employee, a driver. The fully automated barrel collection program services 165,600 households (City Official Interviews, 2012). Semi automated collection is provided to the remainder of the city. These areas likewise are provided city issued barrels with customers selected their preferred size container and related fee for service. Semi automated collection rep resents 10% of the trash collection activities for the department. These routes require two employees: one laborer and one driver. The semi automated barrel collection program services 18,400 households (City Official Interviews, 2012).

PAGE 120

! 100 The Solid Waste Management Agency utilizes a fleet of roughly 200 vehicles utilizing a combination of rear load and side load trucks. At this time the fleet is approximately 30% natural gas and 70% diesel powered vehicles with plans to move closer to a 50 % fleet of natur al gas and 50% fleet of diesel powered vehicles (City Official Interviews, 2012) While the city is moving toward greater adoption of natural gas vehicles they indicated they do not want to go 100% natural gas so as to have flexibility between a natural g as fleet and a diesel fleet in case of fuel price spikes with either of the two fleets. With that said, they further noted the natural gas fleet with current energy prices is less expensive to operate on a per mile basis than the diesel fleet (City Offici al Interviews, 2012) The city notes they have greatly enhanced their route efficiencies (i.e. lower tota l operating costs) by paying close attention to their route structures and utilizing software to maximize route efficiencies ( i.e. savings generated by less total miles traveled and more efficient use of trucks and employees time) Savings have been able to enhance funding for other parts of the programs, such as a million dollar transfer from routes to reuse programs based on savings realized along col lection routes simply by improving operational efficiency (City Official Interviews, 2012). Austin Recyclables Collection City official interviews revealed that Austin delivers citywide rec yclables collection service, in house by city employees, and col lects approximately 55,000 tons of recyclables annually (52,236 tons in 2011). This service is citywide, single stream in

PAGE 121

! 101 nature (all recyclables together), and collects an extended mix of recyclables 25 Trash and recyclable are collected on the same day, with the one variation that trash collection is weekly and recyclables collection is every two weeks. The collection method utilizes the same configuration as trash collection with automated side load collection vehicles providing service to 90% of the c ity and semi automated rear load collection vehicles providing service to the remaining 10% of the city. The automated trucks operate with one employee, a driver; the semi automated trucks operate with two employees: a driver and a laborer. City issued b lue 65 gallon or blue 96 gallon barrels are utilized. 96 gallon is the default size and 65 gallons containers are issued by request of a household (City Official Interviews, 2012). City officials noted that r ecyclables collection is considered to be a base service and is automatically provided to all customers receiving trash collection services and is funded via the city utility fee covering trash. While participation is automatic (r esidents automatically receive a barrel from the city ) and free of di rect charge, Austin's recycling program is voluntary. While the program is voluntary, the city has calculated that on average, 85% of households set recycling barrels for collection on a regular basis (City Official Interviews, 2012) The majority of Aust in's recyclable are pro cessed at the Balcones Resources' Material Recovery Facility. The facility is currently operating under a 20 year contract with the City of Austin to proces s the bulk of the city's single family household curbside !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 25 Austin's extended mix includes: #1 #7 plastic bottles and plastic containers; aluminum cans, foil, and baking tins; cor rugated, paperboard, and paper bags; glass bottles and jars; office paper, wrapping paper, and junk mail; newspapers, magazines, catalogs, and phonebooks; steel cans and empty aerosol cans; and white goods (refrigerators and other bulky / primarily metal i tems).

PAGE 122

! 102 collected recyclab les as well as many commercial and other government customers. The facility processes single stream recyclables at the rate of 25 tons per hour (Resource Recycling, 2012) Austin Organics Collection City officials noted that Austin delivers citywide organi cs collection service, in house by city employees, and collects approximately 2 5,000 tons of organics annually ( 24,777 tons in 2011). This service is citywide and currently is limited to yard trimming s (grass, leaves, brush, and wood) with plans to add ve getable food scraps to the mix in 2015. Trash and organics are collected on the same day L ike tras h, organics collection is provided on a weekly basis. Organics collection is a semi automated collection process utilizing rea r load collection vehicles The semi automated trucks operate with two employees: a driver and a laborer. The city does not issue containers for this program; instead residents must set out their organics in paper bags that they must purchase from retail stores. The city is planni ng to switch to city issued 96 gallon carts to mirror the trash and recycling collection in 2015. This change will also allow the city to move from a semi automated collection service to a 90% automated collection service like that for both trash and recy cling (City Official Interviews, 2012). City officials noted that o rganics collection is considered to be a base service and is automatically provided to all customers receiving trash collection services and is funded via the city's utility fee covering trash. Austin's organics program is voluntary and free of direct charge (except that residents must purchase their own paper bags) (City Official Interviews, 2012)

PAGE 123

! 103 Key Findings: Fiscal Efficiency, Environmental Efficiency, and Equity Fiscal Efficiency Austin's operating costs for trash collection and recyclables collection in 2012 were $26,128,000 (City Official Interviews, 2012) Given 184,000 single family household units are served by the program, this translates to a fiscal efficiency of $142 per Household per Year. Environmental Efficiency In 2012 Austin collected approximately 126,497 tons of landfill bound waste annually (City Official Interviews, 2012) This translates to approximately 0.69 tons per household per year Austin's overall greenh ouse gas footprint for the trash collection program was calculated (utilizing the formula in Chapter III and the DOE GREET model) to be 49,680 tons CO 2e total (0.27 tons CO 2e per household per year) and the greenhouse gas footprint for landfill based dispo sal was calculated (ut ilizing the formula in Chapter III and the EPA WARM model) to be 123,280 CO 2e per household per year (0.67 tons CO 2e per household per year) for a combined calculated greenhouse gas footprint of 174,800 tons CO 2e Given 184,000 singl e family households units are served by the program, this translates to an environmental efficiency of 0.95 tons CO 2e per household per year. Table 5.1 below summarizes the key findings from the Austin case study in relation to this dissertation's key rese arch questions of fiscal efficiency, environmental efficiency, and equity in service delivery.

PAGE 124

! 104 Table 5.1 : Summary of Austin Case Study Findings City Fiscal Efficiency ($ Per Household per Year) Environmental Efficiency (CO 2e per Household per Year) Equi ty Austin, Texas $142 0.95 tons Equitable Average of All Monocentric Case Study Cities $142 1.22 tons Equitable Average of Both Franchise Zone Case Study Cities $265 1.25 tons Equitable Average of Both Polycentric Case Study Cities $175 $411 1.45 to ns Equitable Equity In terms of equity, city officials noted that Austin puts a strong premium on high quality and equitable service delivery to all customers via all three collection programs: trash, recyclables, and organics. City officials noted that trash collection can "make or break" elections and that there is a strong ethic, from the Mayor, other elected officials, and the officials within Austin Resource Recover, that services should always be provided on a high quality and equitable basis (City Official Interviews, 2012) In terms of this dissertation's equity hypothesis: (1) cost of service is "equal" citywide in that all residents are provided the same suite of pay as you throw options, (2) level of service is equal citywide weekly trash coll ection and periodic bulky item collection, and (3) breadth of service is equal citywide delivering trash, recyclables, and organics collection to al l single family households city wide. Based on th ese measures and based on interview comments, Austin progra m meets the equity hypothesis definition of e quitable in nature.

PAGE 125

! 105 It should be noted that the divide between automated collection and semi automated collection is made purely based on the geometry of the few routes w h ere full automation simply is not practi cal in nature (City Official Interviews, 2012). City officials noted in the interviews that while service is available in a fully equitable manner, they have noted variations in neighborhoods within the city having differentiated degrees of program adoptio n for recyclables and organics collection programs. In particular, it was noted that neighborhoods with a higher percentage of Latino residents and Spanish speaking residents tended to have lower recycling and organics recycling rates. The city is workin g to address this disparity by targeting recycling education and marketing geared specifically to these communities / neighborhoods (City Official Interviews, 2012). Discussion Beyond the Efficiency Hypotheses Each of the case study cities were asked a set of ope n ended questions (see Chapter III ) beyond the survey instrument questions (see Appendix B) asked of both case study cities and large N responding cities. This section highlights comments of note from these open ended questions. Fiscal Efficiency O ne arrangement noted as unique to Austin is that trash collection is automatically provided by the city to single family households, duplexes, and triplexes. Private waste haulers provide contracted waste collection service to multi unit housing of 11 uni ts and above. However, for multi unit housing in the range of 4 10 units the customers in these units have the choice to either contract with the city via Austin Resource Recovery for collection services or contract with a private waste hauler of their

PAGE 126

! 106 c hoice for collection services City officials noted that this unique arrange forces both the city and the private contractors to remain efficient in their service delivery as customers are likely to go with the lowest price for the service as long as the service is viewed as a quality service production (City Official Interviews, 2012) Environmental Efficiency Both the City of Austin and Austin Resource Recovery have specific waste reduction goals. The city's landfill waste diversion goal is 35% reductio n by 2015, 50% reduction by 2020, and 75% reduction by 2040 This goal was adopted by the City Council in 2009. With the first goal still several years out the city reported in 2012 a diversion rate of 38% for the single family residential collection ser vice (City Official Interviews, 2012) One unique program toward meeting the waste diversion goal is that Austin i s embarking in 2014 on a pilot effort to recycle one particularly problematic waste material: used mattresses. Used mattresses are highly pro blematic in landfills, they don't compact and the bedding material becomes a "leachate sponge" in the landfill setting ( SWANA, 2000 ) However, the mix o f multiple materials that make up the components of a mattress likewise makes them challenging to recyc le. Thus the city is initiating a collection, deconstruction, and recycling program uniquely suited to deal with mattresses. It is the first example of what the city hopes will be multiple programs by which the city will tackle difficult to recycle items as the city continues to ratchet up their waste reduction goals over time (City Official Interviews, 2012).

PAGE 127

! 107 The City of Denver, Colorado Introduction to Denver With a population of 619,968 (2010 census) and a square mile city footprint of 155 square mil es the City and County of Denver is the 23 rd most populous city in the United States (2010 census) The city is what "new regionalism" scholars would term a "consolidated" city and what "local public economy" scholars would term a "monocentric" city The city and county governments are fully unified and integrated and the city boundary and county boundary are one and the same. Denver is also the capital of the State of Colorado. Introduction to Denver's Solid Waste Collection This monocentric example of city county government is likewise translated into the monocentric service production of solid waste collection in the city. Interviews revealed that t he City and County of Denver pro duce trash collection services and recyclables collection services cityw ide for single family households (defined by the city as seven or fewer units) within the c ity. Citywide this represents 168,942 units (City Official Interviews, 2011 & 2012 ) that receive trash collection. The service is provided by the Solid Waste Manag ement Agency, which is a division of the City and County of Denver's Public Works Department. Collection of trash is provided on a weekly basis. Collection of recyclables is provided every two weeks. The citywide collection method falls under the defini tion of a monocentric arrangement because: the city itself delivers the service the service is delivered across the full service area of the city, and there is no overlap in the service production (City Official Interviews, 2011 & 2012 )

PAGE 128

! 108 While the deliver y of the service in terms of staffing is a monocentric arrangement, the c ity currently utilizes three different collection methods, but is moving in the direction of adopting the automated collection method that utilizes 95 gallon barrels provided by the c ity to residents as the standard citywide ( City Official Interviews 201 1 & 201 2; Denver Master Plan, 2010; www.denvergov.org 201 3 ). Denver Trash Collection Interviews revealed t he three trash collection methods in use within the city for single family households are: (1) Automated d umpster collection (2) Manual collection, and (3) Automated collection utilizing 95 gallon barrels (City Official Interviews, 2011) Interviewed officials noted that d umpster collection is provided in certain areas of the city that have alleyways particularly in more densely populated areas of the city This represents 47% of the daily solid waste collection activities for the department. In this case a dumpster (generally ranging fro m three cubic yard to eight c ubic yard container) is utilized by multiple households (generally four to five) and collection is automated utilizing collection vehicles generally with 30 cubic yards of capacity. About 12% of the area serviced by dumpsters has particularly narrow alleyways requiring smaller sized dumpsters and special operation considerations. As of 2012 there were 65,000 households serviced by the dumpster program. On any given day approximately 175 to 200 dumpsters are serviced represent ing approximately 650 to 700 households. Each route requires only one employee, a driver. This translates to 30 full time driver positions. In terms of budgeting, dumpster collection represents 43% of the city's collection costs and 47% of the tons coll ected (this was $6.46 million in 2009) ( City

PAGE 129

! 109 Official Interviews 201 1 & 201 2; General Fund, 2011; Denver Master Plan, 2010; www.denvergov.org 201 3 ). Interviewed officials also discussed the m anual collection tha t is provided in certain areas of the city. This represents 19% of the daily solid waste collection activities for the department. In these areas households provide their own trash containers (maximum 32 gallons) or trash bags (maximum 50 pounds) and cr ews manually load rea r load collection trucks with 25 cubic yards of capacity These routes require two employees: one laborer and one driver. This translates to 18 full time driver positions and 30 full time laborer positions. As of 2012 there were 47, 990 households serviced by the manual collection program. On any given day, from 500 to 800 households are serviced via this collection system. Approximately 30% of the manual service area is the portion of Denver with particularly narrow alleyways. Thi s portion of the city will be most difficult to move away from this manual collection system. In terms of budgeting, manual collection represents 24% of the city's collection costs and 19% of the tons collected (this was $3.69 million in 2009). In additi on to budget related efficiency concerns, the city is likewise interested in moving away from manual collection because of worker injury issues. In 2008, the manual collection program reported 26 workman's compensation injuries. By contrast, the combined collection programs of barrel and dumpster collection, both automated, reported only 8 workman's compensation injuries ( City Official Interviews, 201 1 & 201 2; General Fund, 2011; Denver Master Plan, 2010; www.denver gov.org 201 3). The interviewees then discussed Denver's a utomated collection program u sing city issued black 95 gallon barrels This service is provided in certain areas of the city

PAGE 130

! 110 utilizing side load collection vehicles with 30 cubic yards of capacity The side load vehicles are equipped with a robotic arm to automate this collection. This represents 31% of the daily solid waste collection activities for the department. These routes require only one employee, a driver. This translates to 17 full tim e driver positions. The automated barrel collection program services 54,890 households. On any given day, from 780 to 900 households are serviced via this collection system. In terms of budgeting, automated barrel collection represents 33% of the city's collection costs and 31% of the tons collected (this was $4.93 million in 2009). The city's internal analysis of these three collection schemes has reached the conclusion that the automated collection system utilizing the barrels is the most fiscally eff icient of the three ( City Official Interviews 201 1 & 201 2; General Fund, 2011; Denver Master Plan, 2010; www.denvergov.org 201 3). Interviewees further noted that in the automated barrel collection areas the city designates an "overflow collection week" once every three weeks. On this week residents may set out additional waste beyond the ir 95 gallon barrel, setting out either their own additional trash cans or bagging trash in additional trash bags ( City Official Interviews, 2011 & 2012; Denver Master Plan, 2010; www.denvergov.org 201 3). Interviewees noted that for all areas of the city, on an every nine weeks schedule, there is a designated "bulky item / large item coll ection week for larger / more difficult to dispose of items that will not fit in traditional dumpsters or barrels. These "overflow collections" and "bulky item / large item collections" represents 3% of the daily solid waste collection activities for the department. This special trash collection is serviced in a manual manner, with crews of two employees per route. This translates to 11 full time

PAGE 131

! 111 driver positions and 14 full time laborer positions ( City Official Interviews 2011 & 2012 ; Denver Master Plan, 2010; www.denvergov.org 2012) Denver has a stated goal of standardization of solid waste collection to the fully automated barrel service. This will require the procurement of both new barrels for residents and new automated collection vehicles for the Solid Waste Management Agency This standardization, however, has a caveat as some areas of Denver have exceptionally narrow alleyways. In these neighborhoods, residents will receive new carts and collection serv ice will be upgraded to semi automated vehicles to accommodate the narrow alleyways ( Denver Master Plan, 2010 ). For the currently manual collection households, Denver estimates 37,200 households will be able to be converted to fully automated service and 1 5,400 households will need to be converted to semi automated service. For the currently dumpster collection households, Denver estimates 62,400 households will be able to be converted to fully automated service and 8,800 households will need to be convert ed to semi automated service. As noted, these households will need to be converted to semi automated rather than fully automated service due to the narrow alleyways as the tight geometry of these alleyways simply will not allow the space for a fully autom ated collection program ( Denver Master Plan, 2010 ). Interviews revealed t he Solid Waste Management Agency utilizes a fleet of almost 200 diesel powered vehicles. Collection specific vehicles include: 21 automated barrel collection vehicles, 47 manual rear load vehicles, 31 automated dumpster side load collection vehicles, and 19 automated recyclables collection vehicles. An additional long term efficiency consideration for the city is that under the current suite of collection

PAGE 132

! 112 systems, none of the fleet v ehicles can be used interchangeably. This forces the Solid Waste Management Agency to maintain four separate fleets of vehicles (City Official Interviews, 2011 & 2012) Denver Recyclables Collection City officials noted in interviews that t he city's recyc ling program began in its current form in 2005 and reached full implementation in 2007. Denver delivers citywide recyclables collection service, in house by the Solid Waste Management Agency, and is currently (as of 2012) collect ing approximately 30,000 t ons of recyclables annually. This service is citywide, single stream in nature (all recyclables together), and collects an extended mix of recyclables 26 Trash and recyclable are collected on the same day, with the one variation that trash collection is w eekly and recyclables collection is every two weeks. The collection is provided via automated side load collection vehicles utilizing city issued purple 65 gallon or 95 gallon barrels (65 gallon is the default size, 95 gallons are issued by request of a h ousehold). City Officials noted i n the interviews that i t is more cost efficient to issue one 95 gallon container than two 65 gallon containers for a household requiring more capacity for their recyclables (City Official Interviews 2011 & 2012 ) The auto mated trucks operate with one employee, a driver. Like trash, r ecyclables collection is also funded via the city's general fund ( City Official Interviews 201 1 & 201 2; Denver Master Plan, 2010; www.denvergov.org 2 012). City officials noted Denver's recycling program is voluntary, free of direct charge, and subscription based. Residents must subscribe to the service by ordering a !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! #% Denver's extended mix includes: Aseptic cartons and waxed cartons; #1 #7 plastic bottles and plastic containers; aluminum; corrugated, paperboard, and paper bags; glass bottles and jars; office paper, wrapping paper, and junk mail; newspapers, m agazines, catalogs, and phonebooks; and steel cans and empty aerosol cans.

PAGE 133

! 113 barrel from the city. Thus in 2012, while all 168,942 households in Denver are eligib le for recyclables collection only 111,745 households have actually subscribed to the service, a participation rate of 66% ( City Official Interviews 201 1 & 201 2; www.denvergov.org 2012). This is trending upward; in 2009 the program had a participation rate of 55%. At that time the city conducted a participation audit and found of the 55% participating, these subscribers had a weekly set out rate of 79% and averaged 30 pounds per household per collection day (Denv er Master Plan, 2010). In terms of both fiscal and environmental efficiency, currently the route density for recyclables collection is not maximized. City officials noted that m andatory recycling is being considered as a remedy to this situation (City Off icial Interviews 201 1 & 201 2; Denver Master Plan, 2010; www.denvergov.org 2012). Based on barrel orders, the subscription rate varies greatly from community to community within Denver. The Stapleton community has the highest participation rate, at 93%; the Elyria Swansea community has the lowest participation rate, at 31% ( http://www.denvergov.org 2012). One could argue this observation one of two ways: (1) Service is equ itable in that it is provided citywide; or (2) Inequities are apparent in the adoption of recycling services. Interviewees noted that w ith the implementation of the citywide curbside recycling program there has been a noted shift in the waste stream pat terns for Denver. In 2004, 265,000 tons of solid wastes were disposed of by Denver and 15,700 tons of solid wastes were recycled. By contrast, in 2008, 220,000 tons of solid wastes were disposed of by Denver and 28,500 tons of solid wastes were recycled ( Denver Master Plan, 2010 ). By 2011 Denver's collection of recyclables had risen again to 31,018 tons

PAGE 134

! 114 of recyclables. This translates to 0.18 tons per household per year citywide or 0.28 tons per household per year for participating households (City Offi cial Interviews 201 1 & 201 2; www.denvergov.org 2013). Denver Organics Collection City official interviews noted that Denver currently operates a pilot organics collection program. This service is delivered in a l imited area within the city 27 collection is year round and weekly, collection is single stream in nature (all organics together), and collects materials categorized as food wastes, yard wastes, and compostable paper products. The collection is generally p rovided on the same day as trash collection services, however some of the routes in the trash dumpster areas are on a different day than trash collection. The collection is provided via automated side load collection vehicles utilizing city issued green 6 5 gallon barrels. The pilot o rganics collection is currently funded on a fee for service basis of $9.75 per month As of 2012, 2,300 single family households were participating in the pilot. While there are approximately 17,000 households within th is pi lot service area, because the program is a pilot in nature the maximum number of households that c an be served by the organics program in its current form woul d be approximately 3,000 single family households (City Official Interviews 201 1 & 201 2; Denver Master Plan, 2010; www.denvergov.org 2012). The Solid Waste Master Plan for Denver notes that the city is currently in the process of identifying funding and capital improvement planning with the intent of !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 27 In 2012 organics collection in Denver was limited to the following neighborhoods: Parkhill, Sloan's Lake, Washington Park, Whittie/City Park West/Five Points, Athmar Park, La Al ma, Greater Marlee, Stapleton, Goldsmith/Hampden/University Park, Wellshire, Country Club/Alama Placita/Speer, Cook Park/Virginia Village, and Hilltop Cranmer Park.

PAGE 135

! 115 bringi ng this service citywide via a five phase startup, similar to how the city initially began the citywide recyclables collection program (Denver Master Plan, 2010) City official interviews confirmed the intent of master plan is still moving forward (City O fficial Interviews, 2011 & 2012). The five phase startup will include the following steps: The first phase of this program will bring 20,000 households online via 6 new collection routes at a total program cost of $1,671,500 ; The second phase of the progr am will bring 40,000 households online via 10 new collection routes at a total program cost of $4,229,100; The third phase of the program will bring 15,000 households online via 4 new collection routes at a total program cost of $3,532,100; The fourth phas e of the program will bring 15,000 households online via 4 new collection routes at a total program cost of $4,088,500; and The fifth and final phase of the program will bring 10,000 households via 3 new collection routes at a total program cost of $4,273, 100 (Denver Master Plan, 2010). The first phase of the expansion beg an in 2014 ( www.denvergov.org 2014 ) with the program continuing the pilot's funding model utilizing a fee for service basis. Eligible households enrolling in the service have two payment options: quarterly payments of $29.25 every three months (totally $117.00 annually) or one annual payment of $107.00 for the entire year ( www.denvergov.org 2014)

PAGE 136

! 116 K ey Fin dings: Fiscal Efficiency, Environmental Efficiency, and Equity Fiscal Efficiency Based on budget data provided in interviews, Denver's operating costs for trash collection and recyclables collection in 2012 w ere $24,576,700 (City Official Interviews, 2011 & 2012) Given 168,942 single family household units are served by the program, this translates to a fiscal efficiency of $145 per Household per Year. Trash and recyclables collection are funded out of the city's general fund. The city's general fund is funded via a combination of taxes and fees. At this time there is not a solid waste specific fee for these two services. The largest share of monies come from a combination of sales taxes and use taxes, a much smaller share from property taxes, smaller y et from other taxes, and the remainder of general fund monies come from intergovernmental and general government funding / transfers (General Fund, 2011). Environmental Efficiency City official interviews noted Denver currently collects approximately 220,0 00 tons of landfill bound waste annually from single family households (City Official Interviews, 2011 & 2012) This translates to approximately 1.3 tons per household per year. However, the city officials noted in interviews some interesting generation variations among the three varied collection schemes, in 2008, the average household refus e generation rate was 1.60 tons per household per year for households served by the automated dumpster collection progr am. This compares to 1.24 tons per household p er year for households served by the automated barrel c ollection program and 0.85 tons per househol d per year for households served by the manual collection program. Given just under half of the city is served by the dumpster collection program, while the city is

PAGE 137

! 117 converting to the automated barrel collection program for fiscal efficiency reasons it is clear there will likely be an environmental efficiency benefit as well (City Official Interviews, 2011 & 2012; Denver Master Plan, 2010) Denver's overall gr eenhouse gas footprint for combined trash collection program across the three collection schemes was calculated (ut ilizing the formula in Chapter III and the DOE GREET model) to be 25,341 tons CO 2e total (0.15 tons CO 2e per household per year) and the gree nhouse gas footprint for landfill based disposal was calculated (ut ilizing the formula in Chapter III and the EPA WARM model) to be 216,246 CO 2e per household per year (1.28 tons CO 2e per household per year) for a combined calculated greenhouse gas footpri nt of 241,587 tons CO 2e Given 168,942 single family households units are served by the program, this translates to an environmental efficiency of 1.43 tons CO 2e per household per year. Table 5.2 below summarizes the key findings from the Denver case stu dy in relation to this dissertation's key research questions of fiscal efficiency, environmental efficiency, and equity in service delivery Table 5.2 : Summary of Denver Case Study Findings City Fiscal Efficiency ($ Per Household per Year) Environmental Ef ficiency (CO 2e per Household per Year) Equity Denver, CO $145 1.43 tons Equitable Average of All Monocentric Case Study Cities $142 1.22 tons Equitable Average of Both Franchise Zone Case Study Cities $265 1.25 tons Equitable Average of Both Polycentr ic Case Study Cities $175 $411 1.45 tons Equitable

PAGE 138

! 118 Equity In terms of this dissertation's equity hypothesis: (1) cost of service is equal citywide, (2) level of service is equal citywide weekly trash collection and periodic bulky item collection, and ( 3) breadth of service is equal citywide for trash collection and recyclables collection. While there is an inequity in the availability of organics collection this program is deemed by the city to still be a pilot program. The researcher, when initially defining the equity measure, decided that pilot program inequities would not count as inequitable service delivery Thus, b ased on th ese measures and based on interview comments, Denver's program meets the equity hypothesis definition of e quitable in natu re. With this noted, further discussion of Denver's programs reveal three more nuanced findings in relation to equity: (1) Some residents perceive inequities in service delivery because of the three different methods by which Denver completes their trash c ollection, (2) There are inequities in the utilization of recyclables collection program in Denver, and (3) While a pilot, the organics collection program is not currently equitable in that it has been selectively implemented only in certain parts of the c ity as of this writing. Each of these three issues will be discussed in turn below. This equity section will then conclude with a discussion of a survey Denver con ducted in relation to citizen s atisfaction. Citizen satisfaction has been used as a proxy measure for equity ( DeHoog, Lowery, & Lyons, 1991; Lowery, Lyons, DeHoog, 1995). Because the city utilizes three different collection methods for trash ( dumpsters with automated collection, barrels with automated collection, and manual collection), city of ficials noted in interviews that some residents p erceive these differentiated services as

PAGE 139

! 119 in equit able (City Official Interviews, 2011 & 2012). However, in terms of this dissertations research hypothesis on equity, the three varied trash collection schemes utilized by Denver were each de termined to be equivalent to each other in terms of cost to singe family households breadth of service, and level of service the three comparatives used to measure equity for this dissertation. City Official interviews no ted that while availability of recyclables collection is universal and equitable, there are i nequities in the utilization of recyclables collection. Different neighborhoods within Denver have vastly different levels of subscription to the service. While citywide subscription averages out at 66%, a neighborhood level analysis completed by Denver noted the lowest rate of subscription is the Elyria Swansea neighborhood at 31% (a low income and largely minority community neighborhood) and the highest rate of subscription is the Stapleton neighborhood at 93% (an upper income and largely white community neighborhood). While these are just two examples, th ey are mentioned here to highlight a trend noted by City Officials: higher income and higher white populati ons in a neighborhood had a tendency to translat e to higher recycling participation By contrast, lower income and higher minority populations in a neighborhood had a tendency to translate to lower recycling participations (City Official Interviews, 2011 & 2012; www.denvergov.org 2012). While t he recyclables collection program in Denver is offered citywide it should be noted that subscription to recyclables collection service is required (even though there is no f ee associated with the service) City officials note this i s a fiscal efficiency measure (i.e. only providing service to households that desire the service) (City Official Interviews, 2011 & 2012) However, it should be noted that two other case study ci ties

PAGE 140

! 120 that noticed similar inequities in recyclables collection have opted for a different default. For Austin officials noted that their default is instead providing the service and requiring an "opt out" phone call (City Official Interviews, 2012). The other city, Fresno, has mandatory recyclables collection and thus does not have an "opt out" option (City Official Interviews, 2012). Thus the conclusion for Denver is that while recyclables collection is available equitably the adoption of the service by residents displays signs of inequity. The organics collection program in Denver is at this time pilot in nature with plans to take the program citywide within a five year period, starting in 2014 As previously noted, this dissertation's author ( prior t o beginning data gathering ) decided that pilot programs would not count as "inequitable" Because of this, the current differential availability (as of 2014) of organics collection does not make Denver's program inequitable as defined by the equity hypoth esis for this dissertation. However, the reality is that at this time this particular solid waste collection service is only available to a limited portion of Denver's single family households. A second item that should be noted regarding this program is that at this time the program is being im pl emented utilizing a fee for service basis ( as of 2014 $29.25 every three months). This subscription fee will certainly impact subscription rates p articularly among low income families or families with minimal to no yard. Specifically looking at equity and citizen satisfaction, the C ity of Denver conducted a survey of residents in 2010 via a third party organization to rate citizen satisfaction with s olid waste collection services (Denver Master Plan, 2010) The su rvey had a 34% response rate with the following key findings: 78% of respondents rated refuse

PAGE 141

! 121 collection services as good to excellent, 66% of respondents rated recycling collection services as good to excellent, and 42% of respondents rated organics colle ction as good to excellent ( Denver Master Plan, 2010 ) The low satisfaction number for organics is not surprising given the program was a small pilot in 2010 when the survey was conducted. Discussion Beyond the Efficiency Hypotheses Each of the case study cities were asked a set of open ended q uestions (see Chapter III ) beyond the survey instrument questions (see Appendix B) asked of both case study cities and large N responding cities. This section highlights comments of note from these open ended questi ons. Denver's I n h ouse Efficiency Analysis The city has completed analys e s of their current collection schemes in terms of efficiency measures. At first glance, the dumpster collection would appear to be most efficient (both fiscally and environmentally), given that the stops for the trash truck are reduced approximately five fold (in that there is one dumpster per four to five houses). However, the city finds this efficiency is not realized as the dumpsters collect nearly twice the waste that similar com munities serviced by the automated barrel and manual programs collect. A large open container without a single family attached to it as "owner" appears to be an open invitation for illicit dumping, thus the increased waste collection in these areas effect ively cancels efficiencies realized by more limited numbers of stops. Instead, the city finds their automated barrel program to be the most efficient (both fiscally and environmentally) The combination of an automated trash route in combination with low er trash per household translates into this collection program being most fiscally and environmentally efficient. While the manual collection portions of the

PAGE 142

! 122 city collect volumes similar to the automated barrel portions of the city, these manual trucks re quires crews of 2 individuals per truck; whereas both the automated dumpster collection and the automated barrel collection only require 1 individual per truck thus from a fiscal efficiency point of view the automated barrel collection is superior to the m anual collection As mentioned previously, City officials noted that Denver is moving in the direction of automated barrel collection wherever possible in the city (City Official Interviews, 201 1 & 2012; Denver Master Plan, 2010; www.denvergov.org 2012). The city's future efforts are largely governed by the document "A Master Plan for Managing Solid Waste in the Mile High City" which was completed in October 2010. The overarching goals of the plan are: a solid w aste collection program that is fully automate while at a minimum maintaining current services and potential ly expand ing services, a program that empowers city residents to both recycle and compost, improves worker safety, reducing the city's greenhouse ga s footprint and increasing the operating life of the city's landfill via a goal to reduce landfill bound waste by 30%. The city and department operate under four guiding principles: efficiency and cost containment, good customer service, worker safety, an d environmental stewardship (City Official Interviews, 201 1 & 2012 ; Denver Master Plan, 2010; www.denvergov.org 2012). F iscal Efficiency The long term funding plan being discussed within the city is the potential adoption of a pay as you throw rate scheme for trash collection with recyclables collection being free of charge and eventually organics collection being expanded and likewise provided free of charge. City Official noted that there was a prevailing belief that this funding mechanism wi ll be both more fiscally efficient a nd more equitable

PAGE 143

! 123 because cost of trash service would be directly tied to trash produced. The plan would also provide an environmental efficiency benefit by creating a financial incentive favoring recycling o ve r disposal (City Official Interviews, 2011 & 2012; Denver Master Plan, 2010; www.denvergov.org 2012). Environmental Efficiency The city's waste master plan notes several goals that can be tied to environmental efficiency: (1) empower city residents to both recycle and compost, (2) reduc e the city's greenhouse gas footprint and (3) increas e the operating life of the city's landfill via a goal to reduce landfill bound waste by 30% ( D enver Master Plan, 2010 ) The City of Fresno, California Introduction to Fresno With a population of 494,665 (2010 census) and a square mile city footprint of 112 square miles the City of Fresno is the 34 th most populous city in the United States and the most populous inland city in California (2010 census) The city utilizes a modified strong mayor form of city government with a seven member city council. Introduction to Fresno's Solid Waste Collection Interviews with city official s and supporting publication s, note that t he City of Fresno produce d solid waste collection in hou se by city employees citywide for single family households (defined by the city as single family households (1 unit, freestanding), duplexes (2 units), and triplexes (3 units). This was true f or trash collection services recyclables collection services, and organics collection services. Citywide this represents 107,000 units that receive solid waste collection. The service is provided by the Department Public Utilities Solid Waste Management Division (a utility enterprise

PAGE 144

! 124 department) Collection of trash, recyclables, and organics are all provided on a weekly basis on the same service day. The current citywide collection method falls under the definition of a monocentric arrangement because : the city itself delivers the service, the service is delivered across the full service area of the city, and there is no overlap in the service production (City Official Interviews 2012 & 2013) The mission statement of the solid waste management divisi on is, "We, the employees of the Solid Waste Management Division, to ensure the preservation of your community's environment, collect and manage green waste, recyclables, and refuse in a professional, safe and efficient manner through teamwork, education a nd high quality services" ( www.fresno.gov 2014). Fresno Trash Collection City official interviews revealed t he trash collection method in use within the city for single family households is a fully automated collectio n system using city issued grey 96 gallon carts and side load collection vehicles with 30 cubic yards of capacity. The side load vehicles are equipped with a robotic arm to automate this collection and one employee a driver. Solid Waste and Recycling util izes a fleet that is basically 50% diesel powered and 50% natural gas powered (City Official Interviews, 2012 ). Interviewees note t rash, recyclables and organics collection are funded via a city utility fee. The fee is assessed on a pay as you throw basi s. That is to say, the size of collection containers selected by residents determine the solid waste fees paid by single family households. The base rate for trash collect ion is either $19.20 per month (for a 64 gallon trash cart) or $25.37 per month (fo r a 96 gallon trash cart). Thus the base rate can range from $230.40 annually to $304.44 annually. In either case, a 96 gallon

PAGE 145

! 125 recycling cart and a 96 gallon organics cart are automatically included with the base rate ( City Official Interviews, 2012 ); www.fresno.gov 2014). After initial three carts (for trash, recycle, and organics), a dditional carts are then available at the following rates: Ad ditional 64 gallon trash: $7.43 per month ; Add itional 96 gallon trash: $1 0.25 per month; Additi onal 96 gallon recycling: $3.74 per month; and Addit ional 96 gallon organics: $3.87 per month ( www.fresno.gov 2014). Fresno Recyclables Collection Interviewees noted that t he recyclable collecti on method in use within the city for single family households is a fully automated collection system using city issued blue 96 g allon carts and side load collection vehicles with 30 cubic yards of capacity. The side load vehicles are equipped with a robot ic arm to automate this collection and one employee a driver. Like trash, the recyclables collection vehicles are a fleet that is basically 50% diesel powered and 50% natural gas powered (City Official Interviews, 2012 ). Interview ee s noted th is service is citywide, single stream in nature (all recyclables together), and collects an extended mix of recyclables 28 Trash, recyclable, and organics are all collected on the same day and on a weekly basis. Fresno's recycling program i s mandatory and is tied to th e city's pay as you throw rate structure. Like trash, both !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 28 Fresno's extended mix includes: All paper, all cardboard, all plastics (except Styrofoam) all metals, and all glass bottles and jars.

PAGE 146

! 126 recyclables and organics collection are funded via a city utility fee (City Official Interviews, 2012). Fresno Organics Collection Interview ee s noted that t he organics collection method in use wit hin the city for single family households is a fully automated collection system using city issued green 96 gallon carts and side load collection vehicles with 30 cubic yards of capacity. The side load vehicles are equipped with a robotic arm to automate this collection and one employee a driver. Again, this fleet is basically 50% diesel powered and 50% natural gas powered (City Official Interviews, 2012 ). The city's organics program is citywide and collects a mix of both food waste (vegetable waste, plant based matter, and compostable paper products) and yard waste (grass, leaves, brush, and wood). As previously noted, all collects occur on the same day, all tied to the city's pay as you throw rate, and all are funded via a city utility fee (City Official Inte rviews, 2012 ). Key Findings: Fiscal Efficiency, Environmental Efficiency, and Equity Fiscal Efficiency Interviews and budget data noted Fresno's operating costs for their single family household trash collection and recyclables collection in 2012 were $16,692,000. Given 107,000 single family household units are served by the program, this translates to a fiscal efficiency of $156 per Household per Year. The city's total solid waste budget for fiscal year 2011 was $44,714,521 and for fiscal year 2012 was $47,927 ,890 (City Official Interviews 2 012 ; www.fresno.gov 2014).

PAGE 147

! 127 Environmental Efficiency Fresno achieved a 75% landfill diversion rate for their residential collection program in 2010. This was achieved beyon d a State mandated 50% landfill diversion rate and before the city council mandated 75% landfill diversion rate by 2012. The city council's next mandated goal is a 90% landfill diversion rate by 2030 (City Official Interviews, 2012). Fresno's greenhouse g as footprint for the trash collection program was calculated (ut ilizing the formula in Chapter III and the DOE GREET model) to be 11,770 tons CO 2e total (0.11 tons CO 2e per household per year) and the greenhouse gas footprint for landfill based disposal wa s calculated (ut ilizing the formula in Chapter III and the EPA WARM model) to be 90,950 CO 2e per household per year ( 0.85 tons CO 2e per household per year) for a combined calculated greenhouse gas footprint of 1 02,720 tons CO 2e Given 107,000 single famil y households units are served by the program, this translates to an environmental efficiency of 0.96 tons CO 2e per household per year. Table 5.3 below summarizes the key findings from the Fresno case study in relation to this dissertation's key research qu estions of fiscal efficiency, environmental efficiency, and equity in service delivery.

PAGE 148

! 128 Table 5.3 : Summary of Fresno Case Study Findings City Fiscal Efficiency ($ Per Household per Year) Environmental Efficiency (CO 2e per Household per Year) Equity Fres no, CA $156 0.96 tons Equitable Average of All Monocentric Case Study Cities $142 1.22 tons Equitable Average of Both Franchise Zone Case Study Cities $2 65 1.25 tons Equitable Average of Both Polycentric Case Study Cities $175 $411 1.45 tons Equitabl e Equity In terms of this dissertation's equity hypothesis: (1) cost of service is "equal" citywide in that all residents are provided the same suite of pay as you throw options, (2) level of service is equal citywide weekly trash collection and periodic bulky item collection, and (3) breadth of service is equal citywide delivering trash, recyclables, and organics collection to al l single family households city wide. Based on th ese measures and based on interview comments, Fresno's program meets the equit y hypothesis definition of e quitable in nature. In terms of equity, city officials noted in interviews that while services are available in an equivalent manner the utilization of these services is not always equitable. I t is a combination of the most affl uent and least affluent neighborhood s that least utilize recyclables and organics collection services. For the low income neighborhoods, the city has been targeting stronger education programs to improve participation. For the affluent neighborhoods, the re appears to be a social norming opposed to recycling. The city has

PAGE 149

! 129 not as yet found a n education based way to address this lack of participation problem among the affluent successfully D espite City Council ordinance mandated recycling and organics col lection programs the city has not as yet resorted to fining residents for non compliance (City Official Interviews, 2012). Discussion Beyond the Efficiency Hypotheses Each of the case study cities were asked a set of ope n ended questions (see Chapter III ) beyond the survey instrument questions (see Appendix B) asked of both case study cities and large N responding cities. This section highlights comments of note from these open ended questions. Fiscal Efficiency While the service production in Fresno is c urrently (as of March 2014) an in house, city produced, monocentric service production model there has been significant discussion and debate within the city in terms of moving to a different service production model. On the commercial side of solid waste collection, the city currently utilizes a franchise zone model with the city broken into two zones: the "North of Ashland" zone and the "South of Ashland" zone Ashland being a main east west street that fairly even divides the city. One discussion within the city was to move to a franchise zone model for single family household collecting. Under this model private waste haulers could compet itively bid on the t wo zones ; likewise, the solid waste management division itself could also competitively bid on t hese two zones At this time (as of March 2014), this option is no longer being actively pursued (City Official Interviews, 2012; www.fresno.gov 2014).

PAGE 150

! 130 The second option under consideration has been to maintain the monocentric service production model, but to contract the service out to a single, citywide, private contractor. This option was actively pursued by the city in 2013 and ultimately narrowly defeated by a voter referendum (Copeland, 2013; Gerlat, 2013; Mil ler 2013). In December of 2012 the city council voted to accept a bid from Mid Valley Disposal. (A Fresno based waste collection firm and the firm currently (as of 2014) holding the "South of Ashland" Commercial Waste Franchise Zone District ) The city c ouncil cited that the new agreement would save the city $2.5 million on an annual basis. The contract was schedule to begin in March 2013 but was immediately challenged in court by the union representing employees in the Solid Waste Management Division. The union described the contract as a "rushed" attempt at privatization (City Official Interviews, 2012; Gerlat, 2013; Copeland, 2013 ). In the spring of 2013 the union and concerned citizens opposed to the privatization were then successful in obtaining t he number of signatures required to hold a voter referendum on the topic. The privatization option was defeated in this referendum in the summer of 2013 (Miller, 2013). Thus, at this time (2014), t he delivery of trash, recyclables, and organics collectio ns services continues to be produced on an in house, monocentric service delivery basis by the Solid Waste Management Division ( www.fresno.gov 2014) The City of Knoxville, Tennessee Introduction to Knoxville With a population of 178,874 (2010 census) and a square mile city footprint of 104 square miles the City of Knoxville is the 128 th most populous city in the United

PAGE 151

! 131 States (2010 census) Knoxville is the central city in the Knoxville Metropolitan Area. Knoxville is not consolidated with the county it resides within, Knox County, and there are numerous additional municipalities within the Knoxville Metropolitan Area. This case study is focused specifically on the central core city of the metropolitan area. Introd uction to Knoxville's Solid Waste Collection Interviews with t he City of Knoxville officials reveal that the city delivers trash collection services and organics collection services citywide for single family households (defined by the city as four or fewe r units) W ithin the city trash collection is conducted by a private contractor, under the terms of a contract with the city The service is currently (as of 2014) delivered by the national waste management firm Waste Connections. This same contractor a lso provides recyclables collection services for the households included in the city's pilot recycling program. The organics collection program is conducted by city employees and is citywide in nature. Citywide 60,000 single family household units receiv e trash collection and organics collection; citywide 20,000 s ingle family household units receive recyclables collection Collection of trash is provided on a weekly basis. Collection of recyclables is provided every two weeks. Collection of organics is provided every two weeks. The citywide collection method falls under the definition of a monocentric arrangement because: the city has contracted with a single contractor for trash who citywide delivers the service, the service is delivered across the ful l service area of the city, and there is no overlap in the service production (Public Official Interviews, 2013)

PAGE 152

! 132 Knoxville Trash Collection A combination of semi automated and manual collection is provided across the city. Households provide their own trash containers ; per city ordinance these containers must have a lid and should be 32 gallons or less in size. Collection cr ews manually load or semi automated load the trash. The contractor utilizes a fleet of varied trucks and determines which is best to utilize along particular routes at their own discretion. The fleet is diesel powered. Collection routes require two employees: one laborer and one driver. Because the program is largely manual in nature, "bulky item / large item" collection is simpl y incorporated into the trash collection program and is thus available to residents on a weekly basis The city's trash collection program has been in place in its current form for over 30 years (Public Official Interviews, 2013 ). Knoxville Recyclables Collection Interviews noted that t he city's recycling program began in its current form in 2010. Knoxville's private waste contractor also delivers citywide recyclables collection service to residents. At this time 20,000 households are eligible for th e service and these households are distributed citywide. The program is currently collecting approximately 5,200 tons of recyclables annually. This translates to 0.09 tons per household per year citywide (based on 60,000 households) or 0.26 tons per house hold per year for participating household s (based on 20,000 household s ) 29 This is because w hile t his service is citywide it is still operating in a pilot mode and thus only one third of the households in the city are eligible to receive the service at thi s time. The program is s ingle stream in nature (all recyclables together), and collects a standard mix of !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! #( As noted earlier, there are 60,000 households citywide but only 20,000 households currently participate in the pilot recycling program.

PAGE 153

! 133 recyclables 30 on and every two week schedule. Trash and recyclable are generally not collected on the same day. The collection is provided via semi a utomated side load and rear load collection vehicles utilizing contractor issued 96 gallon carts. These carts are issued at no direct charge to the households. The semi automated trucks operate with one employee, a driver (City Official Interviews, 2013) Knoxville's recycling program is voluntary, free of direct charge, subscription based, and paid for via the city's tax base. Residents must subscribe to the service by ordering a cart In the first implementation round of the new program in 201 0 the c ity easily achieved 20,000 subscribing households and there is now a waiting list for households that desire to be added to the program a s it is expanded over time. (City Official Interviews 2013 www.cityofkno xville.org 2014). In terms of both fiscal and environmental efficiency, currently route density for recyclables collection is not yet maximized as only one in three houses is eligible for the service (City Official Interviews, 2013 ). Knoxville Organics C ollection City official interviews revealed that Knoxville delivers citywide organics collection service, in house by city employees, and collects approximately 35,000 tons of organics annually ( 34,641 tons in 201 2 ) from the 60,000 single family households within the city This service is citywide and currently includes yard trimming: grass, leaves, brush, and wood. Organics collection is considered to be a base service by the city, is voluntary, free of direct charge, and provided on an every two week sc hedule. Trash and !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 30 Knoxville's standard mix includes: Most (but not all) types of plast ic bottles and plastic containers; aluminum and steel cans; corrugated, paperboard, and paper bags; glass bottles and jars; office paper; newspapers; magazines; catalogs; and phonebooks.

PAGE 154

! 134 organics are collected on different service days. The program operates from February 1 through November 1 annually ( and is not active during November, December, and January). Organics collection is a semi automated collection process ut ilizing two employees per route : a driver and a laborer. The city does not issue containers for this program instead residents must provide their own containers or compostable bags that they must purchase from retail stores. This program has operated in its current form for 20 years (City Official Interviews, 201 3; www.knoxville.org 2014 ). Key Findings: Fiscal Efficiency, Environmental Efficiency, and Equity Fiscal Efficiency City official interviews noted that Knoxville's operating costs for trash collection and recyclables collection in 2012 were $5,673,906 (City Official Interviews, 2013) Given 60,000 single family household units are served by the trash collection program and 20,000 single f amily household units are served by the recyclables collection program, this translates to a fiscal efficiency of $124 per Household per Year. Environmental Efficiency City official interviews noted that Knoxville currently collects approximately 40,395 tons of landfill bound waste annually (City Official Interviews, 2013) This translates to approximately 0.67 tons per household per year. Knoxville's overall greenhouse gas footprint for their trash collection program was calculated (ut ilizing the formula in Chapter III and the DOE GREET model) to be 18,600 tons CO 2e total (0.31 tons CO 2e per household per year) and the greenhouse gas footprint for landfill based disposal was calculated (ut ilizing the formula in Chapter III and the EPA WARM model) t o be 72,600 CO 2e per h ousehold per year (1.21 tons CO 2e per household per year) for a

PAGE 155

! 135 combined calculated greenhouse gas footprint of 91,200 tons CO 2e Given 60,000 single family households units are served by the program, this translates to an environmental efficiency of 1.52 tons CO 2e per household per year. Table 5.4 below summarizes the key findings from the Knoxville case study in relation to this dissertation's key research questions of fiscal efficiency, environmental efficiency, and equity in service delivery. Table 5. 4 : Summary of Knoxville Case Study Findings City Fiscal Efficiency ($ Per Household per Year) Environmental Efficiency (CO 2e per Household per Year) Equity Knoxville, TN $124 1.52 tons Equitable Average of All Monocentric Case Study Cities $142 1.22 tons Equitable Average of Both Franchise Zone Case Study Cities $265 1.25 tons Equitable Average of Both Polycentric Case Study Cities $175 $411 1.45 tons Equitable Equity In terms of this dissertation's equity hypothesis: (1) cost of service is equal ci tywide, (2) level of service is equal citywide weekly trash collection and periodic bulky item collection, and (3) breadth of service is equal citywide for trash and organics collection. While there is an inequity in the availability of recyclables collec tion this program is deemed by the city to still be a pilot program as of this writing one in three households in the city are eligible for the service The researcher, when initially defining the equity measure prior to any data gathering decided that pilot program inequities

PAGE 156

! 136 would not count as inequitable service delivery. Thus, b ased on th ese three measures and based on interview comments, Knoxville's program meets the equity hypothesis definition of e quitable in nature. It can be argued that a recyc lables collection program serving one third of the city is beyond what could be termed a pilot program Thus it should be noted that this program is at a minimum in a phase that could be termed "temporary inequity" due to not having yet achieved full im plementation. Assuming continued expansion and addition of new households overtime this recycling program c ould grow into an equitable service delivery model. Discussion Beyond the Efficiency Hypotheses Each of the case study cities were asked a set of o pe n ended questions (see Chapter III ) beyond the survey instrument questions (see Appendix B) asked of both case study cities and large N responding cities. This section highlights comments of note from these open ended questions. Fiscal Efficiency City o fficials largely consider both the city trash collection program and the organics collection program each to be "tried and true" and there are not any pending plans to make changes in the program at this time (City Official Interviews, 2013 & 2014). The re cycling program, by contrast, is much newer and has yet to be fully implemented. Thus while it is available citywide, it is at this time (as of 2013) capped at 20,000 households. Thus the remaining two thirds of the single family households in

PAGE 157

! 137 Knoxville have not as yet had the opportunity to participate in this recyclables collection program (City Official Interviews, 2013 ). The City of Indianapolis, Indiana Introduction to Indianapolis With a population of 829,718 (2010 census) and a square mile city foo tprint of 373 square miles the City of Indianapolis is the 12 th most populous city in the United States (2010 census) Indianapolis is also the capital of the State of Indiana. The city is what "new regionalism" scholars would term a "consolidated" city and what "local public economy" scholars would term a "monocentric" city. The City of Indianapolis is a unified government with Marion County, Indiana. However, within the Marion County limits, there are three municipalities 31 that are independent from t he City of Indianapolis. These three municipalities make their own provisions for the delivery of trash collection and are not included within this case study Introduction to Indianapolis's Solid Waste Collection Interviews with officials from the City o f Indianapolis noted that the produc tion of trash collection services and recyclables collection services citywide for single family households (defined by the city as four or fewer units) within the city follow a franchise zone arrangement Citywide th er e are 270,000 units that meet the single family household definition (City Official Interviews, 2011 & 2012) and receive trash collection. Within the city, there are multiple services producers but each of these service producers is assigned a non overlapp ing franchise zone to be serviced. The city, on a periodic basis, bids out fo r services within the city's zones. Both the city itself (via the city's Department of Public Works) and private companies may compete to be awarded a !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 31 Lawrence, Spee dway, and Beech Grove

PAGE 158

! 138 service delivery zone. Col lection services are funded via property taxes at a flat rate ( City Official Interviews 2011 & 2012; Indianapolis Sustainability, 2011; www.indy.gov 2012 & 2014 ). Interviews revealed the specifics of this franchise zon e arrangement. T here are a total of twelve franchise zones within Indianapolis. The city self caps the Department of Public Works to a maximum of five service zones. As of 2012, t he City of Indianapolis, via their Department of Public Works, service d di stricts 5, 7, 8, 11, and 12. These five zones combined represent approximately 120,000 households serviced. The private firm, Republic Waste Services, service districts 2, 3, 4, 6, 9, and 10. The private firm, Waste Management, service d district 1. The se seven zones combined represent approximately 150,000 households serviced ( City Official Interviews 2011 & 2012; www.indy.gov 2012). Although waste collection is proprietary in nature for waste firms, the informatio n obtained in this dissertation is available due to bidding and contracting requirements imposed by the City of Indianapolis Indianapolis Trash Collection As noted above, t he City of Indianapolis makes provisions for trash collection services to single f amily households within the city via their franchise zone arrangement. Interviews revealed that t his service is citywide and collection is weekly. The two collection methods in use within the city are: (1) automated pick up using city or contractor issued barrels and (2) manual collection using household obtained barrels and / or trash bags. The collection of bulky / heavy items is also made available to single family households on a monthly basis curbside. The long term plan for the city is a full conve rsion to automated pick up from standardized barrels issued either by the city or by

PAGE 159

! 139 the contractor awarded a zone The city will complete this transition as zones come up for rebidding over time. City residents do not see a bill for this service and th e service is funded via the city's property taxes ( City Official Interviews 2011 & 2012; Indianapolis Sustainability, 2011; www.indy.gov 2012 & 2014 ). Interviewees noted that i n 2011 the actual spending for the seven zones contracted to private haulers was $19,446,896. For 2011, this translates to a cost of approximately $130 per year or $ 11 per month for trash collection services ( City Official Interviews 2011 & 2012; www.indy.gov 2012). Interviewees further noted that Indianapolis currently collects approximately 270,000 tons of disposal bound 32 waste annually (2011 tonnage was specifically 269,815 tons; first six months of 2012 tonnage was specifically 135,265 tons). This tran slates to 1.0 tons per household per year ( City Official Interviews 2011 & 2012; www.indy.gov 2012). Interviews revealed that a t this time m anual collection is provided on all but one of the city operated zones (i.e. o ne city operated zone utilizes automated collection ) T he zone operated by the firm Waste Management also utilizes manual collection All of the zones operated by the firm Republic Services have transitioned to an automated collection system. In the manu al collection zones households provide their own trash containers or trash bags and crews manually load waste. In the automated collection zones residents are provided a bin free of charge. A s of this writing both of the contractors and the city itself are all utilizing fleets of diesel powered vehicles (City Official Interviews, 2011 & 2012). !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 32 By contract, single family household waste in Indianapolis must be sent to the Covanta waste to energy facility located within the city, rather than directly to a landfill.

PAGE 160

! 140 The city calculates that the transition from manual to automated collection will improve both fiscal and environmental efficiency for the city. Fiscal efficienc y is improved, as each manual collection route requires two employees: a driver and a laborer. Whereas the automated collection route only requires one employee: a driver. Environmental efficiency is likewise improved as the collection via an automated s ystem is quicker and reduces idle time where fuel is being burnt to little or no benefit. In addition to budget and environmental related efficiency concerns, the city is likewise interested in moving away from manual collection because of worker injury is sues (City Official Interviews, 2011 & 2012). Indianapolis Recyclables Collection Interview ee s noted that t he City of Indianapolis makes provisions for the collection of recyclables. The city's recycling program completed its transition to its current fo rm in 2012. Indianapolis delivers citywide recyclables collection service. The recyclables collection program is tied to the trash collection franchise zones in terms of service producers however the program is funded and implemented separately from the trash collection program (City Official Interviews, 2011 & 2012) This service is city wide, collection is every two weeks, single stream in nature (all recyclables together), and collects a standard mix of recyclables 33 Prior to 2012, recyclables coll ection s ervice was previously provided via manual collection from 18 gallon containers, this service has transitioned to automated collection from 90 gallon containers. The cost for the collection of recyclables via curbside subscription is $6.50 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 33 Indianapolis's standard mix includes: #1 #7 plastic bottles and plastic containers (Styrofoam explicitly ex cluded); aluminum; corrugated, paperboard, and paper bags; glass bottles and jars; office paper; newspapers and magazines; and steel cans. !!

PAGE 161

! 141 per mont h and is available on a fee for service basis. The d istricts where Republic Services delivers trash collection services are likewise delivered recyclables collection services by Republic Services The d istricts where t he City of Indianapolis delivers tra sh collection are also serviced by Republic Services for recyclables collection via a contractual agreement between the City and Republic. The one district where Waste Management delivers trash collection has recyclables collection delivered via a private contractor employed by Waste Management ( City Official Interviews 2011 & 2012; www.indy.gov 2012). City Official interviews provided information on the adoption of recyclables collection by single family households. Adoption of the collection of recyclables has suffered historically from low participation rates in Indianapolis (as low as 10% in areas). It is both voluntary and provided only on a fee for service basis. Recent initiatives by the Mayor's office have increased these subscription rates (as high as 50% in areas) ( City Official Interviews 2011 & 2012; Indianapolis Sustainability, 2011 ) In 2011 Indianapolis collected 46,527 tons of recyclables 34 This translates to 0.17 tons per household per year citywi de. Because this recycling number includes both curbside recyclables collection and drop off recyclables collection a precise per ton per participating household number c ould not be provided as the number of single family households utilizing drop off loc ations c ould not be accurately established When the researcher pressed city officials on this issue they recommended simply utilizing the 46,527 number for two reasons: (1) The volume collected by the drop off sites is very small compared against the cur bside collection program (<10% of the total weight), and !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 34 This number includes both curbside recyclables collection and recyclables collected at the city's unstaffed recyc lables drop off locations.

PAGE 162

! 142 (2) Much of th is volume is coming from single family households who are not subscribing to the curbside collection thus it is still largely single family household volume ( City Official Interviews, 2011 & 2012). Lastly, it should be noted that the environmental efficiency measure is based on trash collection tonnages not recyclables collection tonnages and thus this disparity did not impact this calculation. Indianapolis Organics Collection There is limited service production of curbside organics collection for single family households in the City of Indianapolis. Each fall, the city offers leaf collection (limit 40 bags of leaves weekly during collection period) There are currently no plans to expand this program ( City Official Interviews 2011 & 2012; www.indy.gov 2012). Key Findings: Fiscal Efficiency, Environmental Efficiency, and Equity Fiscal Efficiency Indianapolis's operating costs for trash colle ction and recyclables collection in 2012 were $56,060,413 (City Official Interviews, 2011 & 2012) Given 270,000 single family household units are served by the program, this translates to a fiscal efficiency of $208 per Household per Year. Trash collect ion is funded out of the city's property tax base. Recyclables collection is funded on a fee for service basis. In terms of fiscal efficiency the city self identified the completed transition of their recyclables collection program to an automated colle ction system and the ir ongoing (as of 2014) transition of their trash collection to a fully automated cart collection service citywide as improvements to their program's fiscal efficiency (Public Official Interviews 2011 & 2012).

PAGE 163

! 143 Environmental Efficien cy Indianapolis's greenhouse gas footprint for the trash collection program across the twelve collection zones was calculated (ut ilizing the formula in Chapter III and the DOE GREET model) to be 64,800 tons CO 2e total (0.24 tons CO 2e per household per year ). In the case of Indianapolis t he greenhouse gas footprint for the disposal based footprint is actually calculated to be a negative number: 10,800 CO 2e per household per year ( 0.04 tons CO 2e per household per year). This is because Indianapolis utiliz ed a waste to energy facility for final disposal (the Covanta Energy facility) The EPA WARM model utilized to calculate greenhouse gas footprints (in combinati on with the formula in Chapter III ) takes into account the avoided energy production from higher carbon alternatives versus the carbon actually released by waste to energy facilities (WARM, 2013). This is how a negative number is generated in this case. Thus, the combined calculated total greenhouse gas footprint for Indianapolis's trash collection and trash disposal program is 54,000 tons CO 2e Given 270,000 single family households units are served by the program, this translates to an environmental efficiency of 0.20 tons CO 2e per household per year. As noted above, Indianapolis utilizes the Cov anta w aste to energy facility for their final disposal of solid waste. This facility is what is termed a "mass burn" facility. Waste is burnt to generate heat and this heat is used to generate steam. The generated steam is then fed into the downtown Ind ianapolis steam loop. This steam loop provides heating and cooling for buildings in downtown Indianapolis Thus the steam loop is an alternative to electric power providing this heating / cooling source and the Covanta Energy Facility is an alternative s ource of heat, reducing the need to utilize the coal

PAGE 164

! 144 burning facility that provides the remainder of the steam for Indianapolis's steam loop (Davis, 2009) Table 5.5 below summarizes the key findings from the Indianapolis case study in relation to this dis sertation's key research questions of fiscal efficiency, environmental efficiency, and equity in service delivery. Table 5.5 : Summary of Indianapolis Case Study Findings City Fiscal Efficiency ($ Per Household per Year) Environmental Efficiency (CO 2e per H ousehold per Year) Equity Indianapolis, IN $2 08 0.20 tons Equitable Average of All Monocentric Case Study Cities $142 1.22 tons Equitable Average of Both Franchise Zone Case Study Cities $2 65 1.25 tons Equitable Average of Both Polycentric Case Study Cities $175 $411 1.45 tons Equitable Equity In terms of this dissertation's equity hypothesis: (1) cost of service is equal citywide in that all residents all pay the same price for trash collection and recyclables collection is available at the same fee for service price citywide, (2) level of service is equal citywide weekly trash collection and periodic bulky item collection, and (3) breadth of service is equal citywide delivering trash and recyclables collection to al l single family households city wide. Based on th ese measures and based on interview comments, the Indianapolis program meets the equity hypothesis definition of e quitable in nature.

PAGE 165

! 145 City official interviews, however, noted that because not all trash collection in the city is delivered in the same manner (part of the city is manual collection and part of the city is automated collection) there are cases where residents perceived inequities in the delivery of their service. In particular, single family households serviced by automated co llection perceive that manual collection communities can dispose of more waste because the households in the automated areas have a set out limit of one 90 gallon cart and there is no set out limit in manual collection areas. City Officials counter that b ecause bulky item collection is provided on a monthly basis this is nothing more than a perception. Because the city is transitioning to fully automated collection this citizen perception will become moot within a few years. Lastly, city official noted t hat beyond this particular issue, city data are indicat ing an equal level of service delivery across service zones and that citizen satisfaction is equal in terms of whether the city deliver ed the service or one of the two private waste firms deliver ed the service (City Official Interviews, 2011 & 2012). The recyclables collection program in Indianapolis is offered citywide and thus the availability of collection meets the dissertation's equity hypothesis definition of equitable. However, because the servi ce is fee for service and su bscription based there are inequities in how this service is utilized by single family households. Like other case study cities 35 Indianapolis's program suffers from differentiated adoption by city residents. Also, because thi s service is only available via a fee for service delivery model there is inherently a greater financial impact on poorer city residents. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! $* Austin, TX; Denver, CO; and Fresno, CA all noted this problem of differential adoption.

PAGE 166

! 146 Discussion Beyond the Efficiency Hypotheses Each of the case study cities were asked a set of ope n ended questions (see Chapter III ) beyond the survey instrument questions (see Appendix B) asked of both case study cities and large N responding cities. This section highlights comments of note from these open ended questions. Fiscal Efficiency The franchise zone model is often touted as a model of fiscal efficiency because of the internalized competition created by this operating method (Savas, 2005). One interesting contrast between the two case study franchise zones cities that impacts this fiscal efficiency is that Indianapolis caps the number of service zones that may be serviced by the city and, by contrast, Phoenix caps the percentage of service zones that may be serviced by private contractors. Indianapolis has a self imposed cap of no more than five of the 12 s ervice zones may be serviced by the city at any given time. City official note that there have been numerous instances over the course of the bidding program where the city was the low bid in more than five zones, but because of this cap was in place the city could not service additional zones (City Official Interviews, 2011 & 2012). Phoenix diverges from this by capping the percentage of service zones that may be serviced by private contractors at 50% of the total service area across the city Because Ph oenix only has 3 franchise zones, this 50% limit is effectively a 66% limit where the City of Phoenix must serve at a minimum two of the three service zones. As of this writing (2014), the City of Phoenix is servicing all zones in house via city operation s.

PAGE 167

! 147 The franchise zone arrangement in Indianapolis was instituted in its current structural form during the Goldsmith administration in the 1990s as part of a broader "marketization" effort for city services (Savas, 2005) The Goldsmith administration sou ght to maximize fiscal efficiency in service delivery by incorporating a market based competition element to city services. Thus "privatization" was not necessarily the end goal but instead fiscal efficiency via competition was the goal (Savas, 2005). Th is included clear understanding of program costs and contract language built around clear performance standards and economic incentives toward favorable performance outcomes (Savas, 2005). "[Goldsmith] adopted a policy of managed competition, which meant that the competition would not be restricted to private companies and that work would not automatically be awarded to private firms" (Savas, 2005, pp.51 52). After Goldsmith took office he consolidated w hat was the twenty five small franchise zones (whic h at the time were served by a number of private waste firms) outside the city's core (which at the time was exclusively served by municipal employees). From the twenty five zones, eleven large consolidated zones were created; and from the one city core a rea the final zone was created. For all twelve of the new zones private firms and the city waste division both had an opportunity to compete on a recurring bid basis to be awarded a particular franchise zone area (Savas, 2005) with the exception being tha t the city could not service more than a total of five zones (City Official Interviews, 2011 & 2012)

PAGE 168

! 148 The City of Phoenix, Arizona Introduction to Phoenix With a population of 1,445,632 (2010 census) and a square mile city footprint of 517 square miles t he City of Phoenix is the 6 th most populous city in the United States (2010 census) Phoenix is the capital of the State of Arizona. The city is a council manager form of government where a city manager supervises city operations and executes and directs the policies from the c ity council a nine member council including the mayor. Introduction to Phoenix's Solid Waste Collection The City of Phoenix Department of Public Works Waste Management Program manages solid waste collection services for the city. T he department 's stated program goal is : "The Waste Management Program assists in providing a safe and aesthetically acceptable environment through effective, integrated management of the solid waste stream, including collection, disposal, source reduction and recycling activities" (phoenix.gov, 2013) While technically a franchise zone arrangement, city official interviews revealed that as of 2014 the City of Phoenix both makes provisions for and actually produces all solid waste collection services in the city for single family households The city itself, at this time, holds all three of the franchise zones in house. The City of Phoenix via the solid waste program produces trash collection services and recyclables collection services citywide for single family within the city. Citywide this represents approximately 394 ,000 units that receive trash collection. The City of Phoenix is the only case study city that provides trash collection twice a week. Collection of recyclables is provided on

PAGE 169

! 149 a weekly ba sis. While technically the citywide collection method falls under the franchise zone definition (because the city is broken into diff erent service areas) Phoenix certainly has the indications of a monocentric arrangement because: the city itself delivers the service, the service is delivered across the full service area of the city, and there is no overlap in the service production (City Official Interviews, 2012 ; phoenix.gov, 2013) With this caveat noted, for the purposes of th e two case study chapter s and thi s overall dissertation, Phoenix will remain categorized as a franchise zone city. Both trash collection and recyclables collection are funded via a city solid waste fee. The city notes: "The fee for solid waste services, including garbage and tras h collection and disposal, repair and replacement of containers, dead animal pickup, enforcement activities, and recycling and other environmental services, is on your monthly municipal bill. The solid waste fee begins when your water service is started a nd ends when water service is discontinued" (phoenix.gov, 2013). Phoenix Trash Collection Public Official interviews note that Phoenix delivers citywi de trash collection service, in house by public works, and is currently collecting approximately 500 ,000 t ons of trash annually from sing l e family households (actual collection for fiscal year 2011 2012: 565,529 tons) The trash collection method utilized in Phoenix for single family households is an automated collection system using city issued green 95 gall on barrels. The city utilizes side load collection vehicles with 30 cubic yards of capacity. The side load vehicles are equipped with a robotic arm to automate this collection. These routes require only one employee, a driver (City Official Interviews, 2 012 ; phoenix.gov, 2013).

PAGE 170

! 150 On a quarterly basis all single family households are provided a bulk and brush collection week. During this week households may set out up to 20 cubic yards of materials, or as the City of Phoenix describes it, "The size of a sma ll SUV" (phoenix.gov, 2013). Residents can set out bagg ed waste or neatly stacked waste. Materials allowed during these special collection weeks include: furniture, appliances, vegetation, water heaters, televisions or stereos, mattresses, carpeting, and large moving boxes. These trash collection events utilize manual collect operating rear load refuse vehicles. Crews of two are utilized for these bulky and brush collection events (City Official Interviews, 2012 ; phoenix.gov, 2013). A s of 2013 the major ity of Phoenix's trash and recycling collection trucks are diesel powered. However, the city is operating a sizable and growing pilot fleet of natural gas powered vehicles and the city has stated a goal of converting 30% of its trash collection fleet to n atural gas powered vehicles within the next five years The city has determined that natural gas vehicles are both more environmentally efficiency and fiscally efficient. Noting that they emit less pollution and are less expensive to operate than diesel powered vehicles (City Official Interviews, 201 2 ; phoenix.gov, 2013). Phoenix Recyclables Collection City Official interviews note that Phoenix delivers citywide recyclables collection service, in house by public works, and is currently collecting approxim ately 110,000 tons of recyclables annually from single family households (actual collection for fiscal year 2011 2012: 105,695 tons). This service is citywide, voluntary, single stream in nature

PAGE 171

! 151 (all recyclables together), and co llects a standard mix of r ecyclables 36 Trash and recyclable s are collected on the same day, with the one variation that trash collection is twice a week and recyclables collection is one a week. The collection is provided via automated side load collection vehicles utilizing city issued blue 95 gallon barrels. The automated trucks operate with one employee, a driver. Like trash, recyclables collection is also funded via the city's solid waste fee (City Official Interviews, 2012 ; phoenix.gov, 2013). Ph oe nix Organics Collection Th e city does not operate an organics collection program. Brush collection, however, is included in the city's quarterly bulky item collection program (City Official Interviews, 2012 ; phoenix.gov, 2013). Key Findings: Fiscal Efficiency, Environmental Effic iency, and Equity Fiscal Efficiency Phoenix's operating costs for trash collection and recyclables collection in 2012 37 were $126, 439 ,000 (City Official Interviews, 2012; phoenix.gov, 2013) Given approximately 394,000 single family household units 38 are se rved by the program, this translates to a fiscal efficiency of $322 per Household per Year. Trash and recyclables collection are supported out of the city's special solid waste fund that is paid for by the city's solid waste fee that is a line item on the city's monthly utility bill (City Official Interviews, 2012 ; phoenix.gov, 2013). !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 36 Phoenix's standard mix includes: #1 #7 plastic bottles; aluminum cans; corrugated and paperboard; glass bottles and jars; office paper, j unk mail, newspapers, magazines, catalogs, and phonebooks; and steel cans and empty aerosol cans. 37 Technically fiscal year 2011 2012. The City of Phoenix budget year begins on July 1. 38 City household counts based on billable solid waste fees: 392,825 for fiscal year 2009 2010, 394,000 for fiscal year 2010 2011, 395,785 for fiscal year 2011 2012.

PAGE 172

! 152 Environmental Efficiency Phoenix currently collects approximately 500,000 tons of landfill bound waste annually from single family households (565,529 tons in fiscal year 201 1 2012) (City Official Interviews, 2012; phoenix.gov, 2013) This translates to approximately 1.26 tons per household per year. Phoenix's overall greenhouse gas footprint for the trash collection program across the city was calculated (utilizing th e formu la in Chapter III and the DOE GREET model) to be 197,000 tons CO 2e total (0.5 tons CO 2e per household per year) and the greenhouse gas footprint for landfill based disposal was calculated (ut ilizing the formula in Chapter III and the EPA WARM model) to be 807,700 CO 2e per household per year (2.05 tons CO 2e per household per year) for a combined calculated greenhouse gas footprint of 1,004,700 tons CO 2e Given 394,000 single family households units are served by the program, this translates to an environmen tal efficiency of 2.55 tons CO 2e per household per year. Table 5.6 below summarizes the key findings from the Phoenix case study in relation to this dissertation's key research questions of fiscal efficiency, environmental efficiency, and equity in service delivery.

PAGE 173

! 153 Table 5.6 : Summary of Phoenix Case Study Findings City Fiscal Efficiency ($ Per Household per Year) Environmental Efficiency (CO 2e per Household per Year) Equity Phoenix, AZ $322 2. 55 tons Equitable Average of All Monocentric Case Study C ities $142 1.22 tons Equitable Average of Both Franchise Zone Case Study Cities $265 1.25 tons Equitable Average of Both Polycentric Case Study Cities $175 $411 1.45 tons Equitable It should be noted that Phoenix, AZ is the only case study city that provides twice a week trash collection. All other cities provide once a week collection. Equity In terms of this dissertation's equity hypothesis: (1) cost of service is equal citywide, (2) level of service is equal citywide twice a week trash collec tion and periodic bulky item collection, and (3) breadth of service is equal citywide delivering trash and recyclables collection to al l single family households city wide. Based on th ese three measures and based on interview comments, the Ph oe nix program meets the equity hypothesis definition of e quitable in nature. Discussion Beyond the Efficiency Hypotheses Each of the case study cities were asked a set of open ended questions (see Chapter III ) beyond the survey instrument questions (see Appendix B) aske d of both case study cities and large N responding cities. This section highlights comments of note from these open ended questions.

PAGE 174

! 154 Fiscal Efficiency The franchise zone model is often touted as a model of fiscal efficiency because of the internalized c ompetition created by this operating method (Savas, 2005). One interesting contrast between the two franchise zones cities that impact s this fiscal efficiency is that Indianapolis caps the number of service zones that may be operated by the city and, by c ontrast, Phoenix caps the percentage of total service area that may be operated by private contractors. Indianapolis has a self imposed cap of no more than five of the 12 service zones may be serviced by the city at any given time. Phoenix diverges from t his by capping the percentage of service zones that may be serviced by private contractors at 50% of the total citywide service area (Franciosi, 1998) Because Phoenix only has 3 franchise zones, this 50% limit is effectively a 66% limit where the City of Phoenix must serve at a minimum two of the three service zones As noted earlier, a s of this writing (2014), the City of Phoenix is servicing all zones in house via city operations. Ron Jensen, the Phoenix public works director until 1996 was the archit ect of the city's system of public private competition that he initiated in 1979. Jensen divided Phoenix into three sectors and put each sector out for competitive bid on a rotating basis (Savas, 2005). Proponents of the managed competition regime note tha t by demanding both high quality service and efficient service delivery, Phoenix forced its solid waste agency to reorganize, adopt more efficient practices, and compete with private companies for residential solid waste collection (Savas, 2005). Private firms had to present their most competitive bid possible as they always face d the city itself as one of the competitive

PAGE 175

! 155 bidders. In a study of solid waste collection services in Phoenix over a 20 year period researcher Ed Savas found inflation adjusted co sts of citywide trash collection had declined by 60 percent from the late 1970s to the late 1990s (Savas, 2005). Opponents of Phoenix's version of managed competition, however, note a number of issues: (1) the 50% city service minimum and (2) the require ments placed on private contractor bidders (Franciosi, 1988). The City of Phoenix has the following requirements on private contractors of residential solid waste collection: Previous experience (serving at least 50,000 households under exclusive contract ), surety bond ($2.5 million), Insurance ($2 million general liability; $5 million excess liability), medical benefits (equivalent to city employee benefits), displaced city employees (must be offered work by a winning bidder), and fleet restrictions (Flee t must be de signated exclusively for single family household collection service) (Franciosi, 1988). Some private contractors have complained that these requirements are onerous and have prevent ed them from submitt ing bids; other private contractors compla in these requirements inherently give the City of Phoenix an advantage in all open bids (Franciosi, 1988). Environmental Efficiency Interviews revealed that Phoenix is in the early stages of integrating natural gas vehicles into their fleet of refuse vehi cles. While their fleet is primarily diesel powered, the department has a small number of natural gas vehicles they are currently operating and they have plans over the next five years to continuing adding natural gas vehicles over time. Their current go al is a fleet that is approximately 30% natural gas powered. Phoenix c ites cost savings (and thus the fiscal efficiency benefits) T heir

PAGE 176

! 156 determination is that natural gas vehicles are less expensive to operate on a day to day bas is then diesel vehicles a nd the fuel source is likewise less expensive to purchase. However, Phoenix ci tes environmental benefits (and thus the environmental efficiency benefit) as their primary reason for moving a portion of their fleet in the direction of natural gas (City Offi cial Interviews, 2012). The City of Colorado Springs, Colorado Introduction to Colorado Springs With a population of 419,427 (2010 census) and a square mile city footprint of 195 square miles the City of Colorado Springs is the 41 st most populous city in the United States (2010 census) Colorado Spring's Trash Collection City Official interviews confirmed that t he City of Colorado Springs does not deliver trash collection services for single family households. The citywide collection method falls under the definition of a polycentric arrangement, private companies within the city deliver waste collection, there are multiple services producers within the city, and these service producers have overlapping service area s This open market, subscription base d collection scheme is funded via fee for service arrangements made between an individual household s and the company of their choice ( City Official Interview 2012; www.springsgov.com 2012). Most of these compani es also offer recyclables collection as a part of th e ir suite of service but at this time (as of 2014) none of these companies offer separate organics collection (bound for composting) and instead organics must simply be disposed of with the trash (Private Hauler Interviews, 2012).

PAGE 177

! 157 While there are many private waste haulers of various sizes and service areas serving Colorado Springs, there are four primary haulers (in terms of numbers of customers) : Waste Management (a national waste firm), Best Way (a loca l waste firm), Waste Connections (a national waste firm), and Springs Waste (a local waste firm). Trash collect is provided to single family households on a weekly basis ( Public Official Interview 2012; Private Haulers Interviews, 2012 www.springsgov.com 2012). Because of the proprietary nature of volumes, weights, and customers for waste firms in a polycentric environment, this will be the only point in this case study of Colorado Springs where the firms are di rectly named. Cost and customer data that were obtained are only presented in a range and do not include information identifying to the firm s Service is provided via a mix of automated and manual collection schemes. Private Hauler Interviews (2012) indi cated t rash collection service in Colorado Springs ranged from a low cost of $17 per month to a high cost of $24 per month. These prices are all based on automated cart based collection services, for carts ranging from 32 gallon up to 96 gallon. S ome hau lers only had a fee for trash collection services S ome haulers had a fee for trash service combin ed with a fuel service charge ; in these cases, t he cost numbers presented in this dissertation combine the two into one single monthly cost. This translates to a fee range of $204 $288 per household per year All of the four largest hauler firms tie recyclables collection service to a trash subscription one at no additional charge and three for an additional fee (Private Waste Hauler Interviews, 2012) Pr ivate Hauler Interviews (2012) noted that a utomated collection is the standard among the four largest waste collection firms in the city, the firms utilize company issued

PAGE 178

! 158 barrels of various size based on an individual household's specific subscription. Th e companies utilized various collection vehicle s of various size s but the stan dard among the firms was diesel powered vehicles (Private Waste Hauler Interviews, 2012) Colorado Springs Recyclables Collection Private Hauler interviews noted that r ecyclables collection service in Colorado Springs ranged from a low cost of $0 per month (that is, it was included in the cost of trash collection) to a high cost of $7 per month (Private Waste Hauler Interviews, 2012; Skumatz et al, 2012). All four of the firms i nterviewed in Colorado Springs offer recyclables collection to single family households. The result is that recyclables collection i s available to single family households in Colorado Springs on a fee for service basis For the four firms interviewed, t he service is provided every two weeks via single stream collection (all recyclables together). Among the four firms t here is a mix of automated and manual collection schemes however the most common (three of the four) was based on automated collection usi ng 64 gallon carts. The mix of materials accepted can best be described as a standard mix of recyclables 39 (Private Waste Hauler Interviews, 2012). The utilization of recyclables collection service is primarily driven by an individual household's desire t o recycle and the households translating their desire into a subscription for recyclables collect service often at an added cost to the household. Thus recycling in Colorado Springs is completely voluntary in nature. Ultimately, each waste hauler decides the s ervice option they offer, the four largest firms interviewed all deliver !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 39 A common mix of recyclable in Colorado Springs would include: #1 #7 plastic bottles and plastic containers; glass bottles and jars; aluminum cans; corrugated and paperboard; office paper; newspapers and magazines; and steel cans.

PAGE 179

! 159 recyclables collection ( City Official Interviews 2012; Private Hauler Interviews 2012; www.springsgov.com 2012). Obtaining a re cy cling rate for Colorado Springs proved challenging in that the collection tonnage data were not publically available. As the data available for this report could not provide a recycling rate, the author sought an outside source for an estimate. The est imated recycling rate for El Paso County (inclusive of Colorado Springs) is calculated to be between 8.6% (Colorado Department of Public Health and Environment calculation) and 11.4% (Skumatz Economic Research Associates calculation) both of these calculat ions were obtained from Skumatz et al (2012). Colorado Spring s Organics Collection As of 2014, none of the four private haulers in terviewed in Colorado Springs provide an organics collection service to single family households. One waste hauler interviewe d does provide organics collection, but at this time the service is only offered to commercial customers ( City Official Interview, 2012; Private Hauler Interviews, 2012 ). Key Findings: Fiscal Efficiency, Environmental Efficiency, and Equity Fiscal Efficien cy As noted above, Private Hauler Interviews (2012) revealed the cost of trash collection service among the firms interviewed ranged from $204 $288 per household per year and the cost of recyclables collection service (over and above the basic trash coll ection service) ranged from $0 $84 per household per year This translates to a fiscal efficiency of $204 $ 372 per h ousehold per y ear.

PAGE 180

! 160 Environmental Efficiency Colorado Springs's greenhouse gas footprint for trash collection programs across the city was calculated (utilizing the formula in Cha pter III and the DOE GREET model) to be 100,587 tons CO 2e total (0.78 tons CO 2e per household per year) and the greenhouse gas footprint for landfill based disposal was calculated (ut ilizing the formula in Chapt er III and the EPA WARM model) to be 145,723 CO 2e /household/year (1.13 tons CO 2e per household per year) for a combined calculated greenhouse gas footprint of 246,310 tons CO 2e Given 128,958 single family households units are served by the program, this translates to an environmental efficiency of 1.91 tons CO 2e per household per year. Table 5.7 below summarizes the key findings from the Colorado Springs case study in relation to this dissertation's key research questions of fiscal efficiency, environment al efficiency, and equity in service delivery. Table 5.7 : Summary of Colorado Springs Case Study Findings City Fiscal Efficiency ($ Per Househ old per Year) Environmental Efficiency (CO 2e per Household per Year) Equity Colorado Springs, CO $204 $ 372 1.91 tons Equitable Average of All Monocentric Case Study Cities $142 1.22 tons Equitable Average of Both Franchise Zone Case Study Cities $2 65 1.25 tons Equitable Average of Both Polycentric Case Study Cities $175 $411 1.45 tons Equitable

PAGE 181

! 161 Equity In t erms of this dissertation's equity hypothesis: (1) cost of service is "equal" citywide in that all residents are provided the same suite of service providers and the same suite of fee for service options 40 (2) level of service is equal citywide weekly tras h collection, and (3) breadth of service is equal citywide delivering trash with the option of a recyclables subscription to al l single family households city wide. Based on th ese measures, Colorado Springs's program meets the equity hypothesis definition of e quitable in nature. Discussion Beyond the Efficiency Hypotheses Each of the case study cities were asked a set of open ended questions (see Cha pter III ) beyond the survey instrument questions (see Appendix B) asked of both case study cities and large N responding cities. This section highlights comments of note from these open ended questions. Environmental Efficiency In terms of environmental efficiency this polycentric arrangement did not perform as well as the other two models. At a minimum, every household in Colorado Springs is services by four waste haulers on the same street. Given that there are a number of smaller waste haulers not interviewed as a part of this dissertation this number could in fact be significantly higher. At a minimum, th ere are four fold the number of trash collection trucks and four fold the number of recyclables collection trucks per household in Colorado Springs than in cities following a once a week collection scheme via a monocentric arrangement or a franchise zone a rrangement. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! ), With that said, there is certainly a wide range of prices in this market. Price vary greatly based on the firm hired and the fee package selected by a customer.

PAGE 182

! 162 City Official Interview (2012) indicated that the polycentric arrangement in Colorado Springs was unlikely to change Given this paradigm, the City Official interviewed offered a number of alternative and hybrid means by which the city could impact the provision of trash collection and recycling without becoming a service producer. Examples included (via city ordinance): a hauler fee used to provide other community and / or environmental services, reporting requirements for haulers, and the r equirement of a pay as you throw rate scheme (or at a minimum, mandating that pay as you throw is offered among a suite of options) ( City Official Interview 2012). T hese alternatives are already in place and implemented in Fort Collins, Colorado the fina l case study in this chapter. The City of Fort Collins, Colorado Introduction to Fort Collins With a population of 143,986 (2010 census) and a square mile city footprint of 54 square miles the City of Fort Collins is the 163 rd most populous city in the Uni ted States (2010 census) It is a midsized city that is known for being the home of Colorado State University. The city is a manager council system of municipal government with a seven member city council including the mayor. Introduction to Fort Collin s s Solid Waste Collection Among the case study cities, Fort Collins is the city where the distinction between the provision versus the production of trash collection services has the greatest significance (V. Ostrom, 1972, 2008). The City of Fort Collins does not participate in the production of single family household trash collection services and recyclables collection services, the production of these services is fully delivered by three private

PAGE 183

! 163 sector solid waste firms: Gallegos Sanitation Ram Waste S ystems and W aste Management (City Official Interviews, 2012) However, the City of Fort Collins is actively involved in the provision of these services. Via ordinance, the city council of Fort Collins activ ely guides the delivery of solid waste collecti on services. Some of their most significant requirements, as a condition of providing single family household solid waste collection services within the city, include: requiring pay as you throw billing options mandating recyclables collection be paired with trash collection service and requiring hauler registration and reporting (City Official Interviews, 2012) The city council has also set aggressive waste diversion goals and is moving in a "zero waste" direction for the city. In 1999 the city counci l of Fort Collins set a goal of diverting 50% of all municipal waste from landfills by 2010. In 2009 the city calculated a diversion rate of 38%. In March of 2014 the City C ouncil for Fort Collins set a goal of diverting all municipal waste from landfill s by 2030 (City Official Interviews, 2012 ; www.fcgov.com 2012, 2013, 2014; Trash Service Study, 2008; Zero Waste, 2013). Fort Collins Trash Collection City Official interviews (2012) and Private Hauler interviews (2012 ) confirmed the polycentric arrangement utilized by Fort Collins is an open (yet regulated) market, subscription based trash collection scheme that is funded via fee for service arrangements made between individual household s and the company of their choic e Mandated by city ordinance these fee for service arrangements must include pay as you throw payment option s This was first required by the Fort Collins City Council in 1995. Prior to this, i n 1992 a city ordinance mandated trash collection and rec yclables collection must b oth b e provided (City Official Interviews, 2012 ; www.fcgov.com 2012,

PAGE 184

! 164 2013, 2014). City officials noted the reasoning for utilizing a pay as you throw rate scheme in combination with mandatory recycling service production was to create a financial incentive for residents to recycle more and divert waste from the landfill (City Official Interviews, 2012) There are 60,816 single family households within Fort Collins serviced under this collectio n arrangeme nt, Fort Collins defines single family households as "residential properties" and defines these as any property where a communal system of trash collection is not employed. The ordinance further specifies this as multi unit house s with 8 units or less (City Official Interview s 2012 & 2013 ; www.fcgov.com 2012, 2013, 2014 ). Private Waste Hauler (2012) interviews confirmed that a ll three private waste haulers on a weekly basis provide collection of trash At this time all the private firms are utilizing diesel powered vehicles. The private waste haulers use a combination of manual and automated collection services based on their own corporate decision. Automated routes require only one employee; manual ro utes require two employees (Private Hauler Interviews, 2012) Fort Collins Recyclables Collection Both public official interviews (2012) and private firm interviews (2012) confirmed that in Fo rt Collins all trash three trash companies provid e citywide tras h and recyclables collection to single family households The se three private waste haulers combined currently collects over 60,000 tons of recyclables annually ; in 2010 the recycling program collection 61,634 tons (City Official Interviews, 2012). This service is single stream in nature (all recyclables together) and collects an extended mix of

PAGE 185

! 165 recyclables 41 By city ordinance recyclables collection must be provided at a minimum twice a month and must be collected on the same day as trash collection (Cit y Official Interview s 2012 ; www.fcgov.com 2012, 2013, 2014 ). In 2010 the city recycling ordinance was amended to require hauler s to provide single family household s a cart, free of charge for recycl ables collection and the cart must, at a minimum, be made available in the sizes of 64 gallons and 96 gallons. Collection is provided via automated side load collection vehicles utilizing the contractor issued containers While delivery of recycling services is mandatory for waste haulers actual recycling by residents is voluntary (City Official Interview, 2012 ; www.fcgov.com 2012, 2013, 2014 ). Fort Collins Organics Collection At this time F ort Collins government has no provision requ irements for organics collection; likewise, none of the three the singe family residential private haulers in Fort Collins deliver / produce an organics collection service. Instead, on the provision side, the city encourages residents to home compost via educational materials (City Official Interview, 2012 ; Private Hauler Interviews, 2012 ; www.fcgov.com 2014 ). One of the three waste haulers does provide organics collection for business customers and single family hous eholds via contractual agreements with homeowners associations (HOAs). For these customers : a cart is provided free of charge ; yard debris, vegetable food scraps, and compostable paper products are all collected together; and the service is provided on !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 41 Fort Collins ex tended mix includes: Aseptic cartons and waxed cartons; #1 #7 plastic bottles and plastic containers; aluminum cans, steel cans, and empty aerosol can; aluminum foil; glass bottles and jars; corrugated, paperboard, and paper bags; office paper, wrapping pa per, and junk mail; and newspapers, magazines, catalogs, and phonebooks.

PAGE 186

! 166 a weekly basis. The one private firm providing this service has not as yet (2014) extended this service to individual residential customers and / or single family household customers (Private Hauler Interviews, 2012). As part of Fort Collins plans to move t o "zero waste" in the future organics collection paired with composting has been identified as a priority. The report Zero Waste Fort Collins: Road to Zero Waste Plan published in December 2013 notes that a compostable materials collection system requirem ent should be added to the city's solid waste ordinance similar in nature to the recyclables collection requirement. The report recommends adding yard trimmings collection to single family household collection services by 2015 and the full phase out of la ndfill bound organics by 2018 (Zero Waste, 2013). Key Findings: Fiscal Efficiency, Environmental Efficiency, and Equity Fiscal Efficiency The cost of production of trash collection and recyclables collection by p rivate waste haulers operating in Fort Colli ns has a wide range. This is largely because of the wide range of pay as you throw options available to single family households. Based on 2012 costs (City Official Interviews, 2012; Private Hauler Interviews, 2012), the fiscal efficiency is $146 $450 per h ousehold per y ear. Environmental Efficiency Fort Collins greenhouse gas footprint for trash collection programs across the city was calculated (ut ilizing the formula in Chapter III and the DOE GREET model) to be 21,286 tons CO 2e total (0.35 tons CO 2e per household per year) and the greenhouse gas footprint for landfill based disposal was calculated (ut ilizing the formula in Chapter III

PAGE 187

! 167 and the EPA WARM model) to be 38,922 CO 2e per household per year (0.64 tons CO 2e per household per year) for a combine d calculated greenhouse gas footprint of 60,208 tons CO 2e Given 60,816 single family households units are served by the program, this translates to an environmental efficiency of 0.99 tons CO 2e per household per year. In addition to the calculations comp leted by this disser t ation, Fort Collins has analyzed the environmental impacts of solid waste collection vehicles in their city. They estimated residential trash trucks operating in Fort Collins generated between 200 300 tons of carbon dioxide (CO 2 ) pe r year. A report prepared for the city recommended eventual conversion within the city to new diesel vehicles meeting 2010 EPA air quality standards, natural gas powered vehicles, elect r ic hybrid vehicles, operate at idle and / or automatic engine shut of f technology, or a combination of these. The report also recommended the utilization of biodiesel over traditional diesel (Trash Service Study, 2008). Table 5.8 below summarizes the key findings from the Fort Collins case study in relation to this dissert ation's key research questions of fiscal efficiency, environmental efficiency, and equity in service delivery. Table 5.8 : Summary of Fort Collins Case Study Findings City Fiscal Efficiency ($ Per Househol d per Year) Environmental Efficiency (CO 2e per House hold per Year) Equity Fort Collins, CO $146 $450 0.99 tons Equitable Average of All Monocentric Case Study Cities $142 1.22 tons Equitable Average of Both Franchise Zone Case Study Cities $265 1.25 tons Equitable Average of Both Polycentric Case Stu dy Cities $175 $411 1.45 tons Equitable

PAGE 188

! 168 Equity In terms of this dissertation's equity hypothesis: (1) cost of service is "equal" citywide in that all residents are provided the same suite of service providers, the same suite of fee for service options, and all have the same pay as you throw payment option s it should be noted that this d oes generate a much wider range of potential pri ces to be paid by the customers ; (2) level of service is equal citywide weekly trash collection with periodic bulky item c ollection, and (3) breadth of service is equal citywide delivering trash and recyclables collection to all single family households citywide. Based on these three measures, the Fort Collins program meets the equity hypothesis definition of equitable in na ture. Interviews did not reveal any inequities of note (City Official Interviews, 2012; Private Hauler Interviews, 2012) however the researcher noted that the two polycentric case study cities were the only two cities to not offer any form of citywide orga nics collection. All four monocentric case study cities offer organics collection and composting as a base service to single family households and both franchise zone cities offer some form of limited organics collection in the case of Indianapolis fall l eaf collection and in the case of Phoenix brush and yard waste collection paired with their bulky item events. Discussion Beyond the Efficiency Hypotheses Each of the case study cities were asked a set of ope n ended questions (see Chapter III ) beyond the s urvey instrument questions (see Appendix B) asked of both case study cities and large N responding cities. This section highlights comments of note from these open ended questions.

PAGE 189

! 169 The Special Case of Fort Collins Compared to the other polycentric case st udy city, the City of Fort Collins exercises a far greater degree of control over the private waste haulers providing services within the ir city. In addition to the service delivery and pricing specific requirements discussed above, the program also requi res licensing fees and a highly detailed level of mandatory reporting. All solid waste haulers must obtain a Fort Collins specific vehicle license (at a fee of $100 per vehicle) and each vehicle must display a city identification stickers The city requi res extensive solid waste and recyclable reporting requirements in terms of weighs, volumes and types of materials collected. All waste firms agree to be subject to an audit by the city at any time at the city's discretion (City Official Interview, 2012 ; www.fcgov.com 2012, 2014 ). Fiscal Efficiency As a part of an evaluation of the Fort Collin's solid waste collection methods in 2008, the city looked at a number of alternatives designed to increase the fiscal efficien cy and effectiveness of collection programs. The alternative most seriously considered from the suite of recommendations was a recommendation for the city to consider a transition from the polycentric arrangement currently in place to a franchise zone bas ed arrangement (termed "districted collection" in city documents and reports to the city) (City Official Interviews, 2012; Trash Service Study, 2008). In 2009 the c ity c ouncil directed city staff to investigate the creation of a pilot program in one area of the city utilizing a "districted collection" (franchise zone) arrangement. As a part of this investigation the city when as far as completing a competitive bid process for this zone. The bid results indicated there was the potential of

PAGE 190

! 170 significant cos t savings for residents within the zone with no change in level of service or breadth of service, however residents would lose the ability to choose which hauler they desired to receive service from. After considerable public input and debate the city cou ncil decided to maintain the current system in place and not execute the pilot initiative. F urther discussions about franchising trash and / or recyclables collection services were discontinued at that time (City Official Interviews, 2012). Environmenta l Efficiency In addition to the fiscal and environmental efficiencies this dissertation has already identified, the Trash Service Study (2008) report added the following additional considerations when looking at moving from a polycentric arrangement to a franchise zone arrangement : reduced roadway maintenance due to reduced truck traffic, improved neighborhood aesthetics and safety due to reduced truck traffic, and reduced neighborhood noise due to reduced truck traffic. The estimate, determined by R3 Cons ulting Group, was that a citywide franchise zone arrangement would result in CO 2 emission r eductions on the order 140 tons annually ( Trash Service Study 2008).

PAGE 191

! 171 C HAPTER V I RESULTS FROM LARGE N DATA ANALYSIS Research Question and Hypotheses This dissert ation's overarching research question is: How does the organization of solid waste collection at the city level relate to: (1) fiscal efficiency, (2) environmental efficiency, and (3) equity of service production? This question was translated into four measurable and testable hypotheses Hypothesis 1 .1 : Fiscally Efficient Hypothesis: The monocentric model will be most fiscally efficient. Justification: Lack of service production overlap and economies of scale for the monocentric model will produce the m ost fiscally efficient outcomes. Literature Grounding: Prediction based on Savas (1977, 1987) solid waste studies, where Savas found that a monocentric service delivery model was most efficient from a cost perspective. Findings from Case Studies: Lend s sup port for H ypothesis 1 .1 The four m onocentric cities had an average service production cost of $142 per household per year. The two franchise zone cities had an average annual cost of $265. The two polycentric cities had an average annual cost of $293. Hypothesis 1 .2 : Fiscally Efficient Hypothesis: The monocentric model with private sector production will be most fiscally efficient.

PAGE 192

! 172 Justification: Lack of service production overlap, competitive bidding by private sector service producers, and economie s of scale for the monocentric model will produce the most fiscally efficient outcomes. Literature Grounding: Prediction based on Savas (1977, 1987) solid waste studies, where Savas found that a monocentric service delivery model with private sector servic e provision was most efficient from a cost perspective 42 Findings from Case Studies: Lend s support for H ypothesis 1 .2 The annual service production cost for the one monocentric private city (Knoxville) was $124 per household per year. Among the three mo nocentric public cities the annual cost was: $142 for Austin $145 for Denver and $156 for Fresno Hypothesis 2: Environmentally Efficien t Hypothesis: Monocentric models will be most environmentally efficient. Justification: Lack of service production ov erlap will lead to a more environmentally efficient outcome because only a single trash truck, rather than multiple trash trucks, will service any one street in non overlapping service production situations. This overlap minimization reduces the multiplie d environmental footprint (in terms of fuel consumption) from multiple trash trucks servicing the same streets. Literature Grounding: Findings from urban area greenhouse gas (GHG) footprint studies by Ramaswami et al. (E.g. Hillman & Ramaswami, 2009; Rama swami et !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! )# Savas (1987) full findings were: monocentric model with private sector production was most efficient, followed closely by monocentric model with public sector production, and poly centric model was least fiscally efficient.

PAGE 193

! 173 al, 2011; Chavez & Ramaswami, 2011) suggest that minimizing service overlap may als o minimize GHG footprints. Findings from Case Studies: Partial support for H ypothesis 2. For environmental efficiency, two of the four monocentric case study citi es were within the "top four" most environmentally efficient cities (Austin and Fresno). All four of th e "top four" cities ha d a CO 2e of less than 1 ton per household per year. The other two "top four" cities were: one was Polycentric (Fort Collins) and one wa s Franchise Zone (Indianapolis). The next two most efficiency cities are the remaining two monocentric cities (Denver and Knoxville), both of these cities ha d a CO 2e of greater than 1 but less than 1.5 tons per household per year. The last two citi es were the least environmentally efficient, both with CO 2e of approximately 2 tons per household per year, one wa s Polycentric (Colorado Springs) and one wa s Franchise Zone (Phoenix). Hypothesis 3: Equitable Hypothesis: Monocentric models will be most eq uitable. Justification: Citywide service production should lead to consistent service delivery across an entire city and eliminate service disparities from different zones of service and different service producers, allowing the monocentric model to delive r the most equitable service production. Literature Grounding: DeHoog, Lowery, and Lyons (1991) found support for the advantages of consolidated government based on equity grounds in service delivery and in a different study Lowery, Lyons, and DeHoog (1995 ) found

PAGE 194

! 174 "citizen satisfaction" with government service delivery was higher in monocentric models than polycentric models. Findings from cas e studies: Partial support for H ypothesis 3. On a cost of service basis the monocentric case studies cities have the lowest cost on a per household basis. In terms of breadth of service, the monocentric case study cities each present a more varied suite of services than franchise zone case study cities and polycentric case study cities. All four monocentric cities off er trash collection, recyclables collection, and organics collection -! Roadmap for Chapter VI Th e following section is a description of the sample dataset utilized and a number of summary tables and descriptive statistics related to the sample dataset the independent variable s the dependent variables, and the control variables ut ilized in the regression models N ext there is a section present ing statistical and regression analyses completed in relation to the dissertations hypotheses for fiscal efficien cy (Hypotheses 1.1 and 1.2) and environmental efficiency (Hypothesis 2). For these thre e hypotheses a number of multiple linear regressions were utilized toward analyzing the three efficiency hypotheses. This section also includes a n examination of the c haracteristics of the polycentric cities in relation to fiscal efficiency a test of the fiscal efficiency regression model fit compared against case study cities, and a sensitivity analysis specific to the environmental efficiency hypothesis This sensit ivity analysis compares the environmental efficiency when all waste is disposed of as trash versus when waste is partially disposed of as trash, partially as recyclables, and partially as organics bound for

PAGE 195

! 175 composting. This analysis utilizes a subset of 4 0 of the 102 large N cities. These 40 cities have trash disposal, recyclables collection, and organics bound for composting collection and all 40 cities reported via the survey the ir tonnage for each of these three waste streams. Finally th is chapter pre sent s a discussion for the dissertation's equity hypothesis (Hypothesis 3) looking at variations in level of service variations in breadth of service, and variations in cost of service as indicators of equity / inequity in service delivery Last there is a conclusion section that summarizes the major findings Sample Data and Descriptive Statistics Sample Dataset and Sample Population This dataset includes 102 cities from a sample population of 429 United States cities (23.8% response rate) These data we re from a nationwide survey of solid waste professions conducted in the fall of 2013 specific to single family household curbside solid waste collection programs. The sample population for this dataset include d cities that ha d populations greater than 50, 000, cities that were a "central core city" (as defined by this dissertation, see Chapter III for further discussion), and cities among the "lower 48" states (i.e. Alaska, Hawaii, Washington D.C., and U.S. territories were excluded). (Appendix C is a full list of the sample population of central core cit ies ) Both individual cities and the num ber of "single famil y households" serviced by solid waste collection programs with in each city served as a key unit s of analysis Th e number of single family househo ld s served was a number defined by and counted by individual cities and the number was provided in the city survey responses. For the 102 cities that responded to the survey, the number of single family households served per

PAGE 196

! 176 city ran ged from 10,590 househ olds to 394,000 households One way the response sample (N=102) varied from the sample population (N= 429) is the average size of cities (in terms of single family households) responding to the survey tended to be smaller, on average, than the mean of the s ample po pulation cities. Where the mean number of households per city in the response sample was 60,924, the mean number of households per city in the sample population was 92,804. Likewise, where the median number of households per city in the response sample was 29,400 the median number of households per city in the sample population was 42,662. This outcome, however, is not surprising. Given the median is a significantly smaller number than then mean, this reflects a larger body of cities tending to the smaller end of the number of households scale with a small number of very large cities moving the mean in a larger direction thus the survey response bias in the smaller city direction is not surprising. Responses to the surveys were completed by solid waste professional. Often these same individuals have a professional interest in the results of the survey. Given the professional nature of the survey recipients, the researcher contends the response versus non response does not represent a bias to the sample. Geographically 40 of the cities responding to the survey were from the Western United States 43 28 of the cities responding to the survey were from the Southern United States 44 23 of the cities responding to the survey were from the Midwest ern Un ited !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 43 U.S. Census West Region includes: Arizona, California, Colorado, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, and Wyoming. 44 U.S. Census South Region includes: Alabama, Arkansas, Delaware, Flo rida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, and West Virginia.

PAGE 197

! 177 States 45 and 11 of the cities responding to the survey were from the Northeast ern United States 46 Table 6.1 below breaks down these survey response cities by geographic region and service production method and as a comparison includes the total numbe r of cities from each geographic region in the survey's total sample population. In general, in terms of geographic region representation, the response cities represent a satisfactory match in relation to the sample population's geographic distribution. With the exception of franchise zones and polycentric in the Northeast, each region of the United States has a t least one city that represents each service production model. Table 6.1: Survey Cities by Geographic Region and Service Production Model Geogra phic Region And Service Production Model Survey Response Counts Total Regional Survey Response Counts Total Regional Sample Population Western United States Monocentric Public 8 40 130 Monocentric Private 16 Franchise Zones 14 Polycentric 2 S outhern United States Monocentric Public 12 28 144 Monocentric Private 12 Franchise Zones 1 Polycentric 3 Midwestern United States Monocentric Public 7 23 96 Monocentric Private 9 Franchise Zones 2 Polycentric 5 Northeastern Unite d States Monocentric Public 5 11 59 Monocentric Private 6 Franchise Zones 0 Polycentric 0 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 45 U.S. Census Midwest Region includes: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebra ska, North Dakota, Ohio, South Dakota, and Wisconsin. 46 U.S. Census Northeast Region includes: Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, and Vermont.

PAGE 198

! 178 The geographic region breakdown indicates a slight bias in terms of number of cities represented by the Western Region of the United States. The Wes tern Region has approximately a 30% response rate. By contrast, the other three regions each have approximately a 20% response rate. Given the study's major finding that polycentric service production is least efficient, and given the N ortheast ern United States and the P acific states of the Western Region have the highest solid waste service production costs in the United States, it is of note that all ten polycentric cities are in the S outh Midwest or Rocky Mountain West the regions of the United State s associated with lowest solid waste service production costs With this noted, however, the author acknowledges the polycentric sample is simply too small and this (along with the overall N) should be increased via additional surveying to increase the ge neralizability of this study. Descriptive Statistics This section presents descriptive statistics in relation to the dissertations independent variable of solid waste service production scheme and dependent variables of fiscal efficiency and environmenta l efficiency (the third dependent variable of equity is discussed in its own stand along sec tion toward the end of the chapter) This section also includes a discussion for some of the control variables utilized, specifically: level of service production, breadth of service production, level of automation for collection fleet fuel source for collection fleet, funding mechanism for collection programs work force unionization, and citywide solid waste related goal. Independent Variabl es The research hypot heses for this disse rtation are built around testing the efficiency and equity of three solid waste collection schemes: polycentric service

PAGE 199

! 179 production, franchise zone service production, and monocentric service production. From the 102 survey responses: 1 0 cities followed a polycentric service arrangement and 17 cities utilized the franchise zone service arrangement. The remaining 75 cities utilized the monocentric service delivery arrangement 42 of these were service produced by a single contractor (a pr ivate waste hauler) (these are referred to hereafter as monocentric private) and 33 of these were city operated arrangements (these are referred to hereafter as monocentric public) For fiscal efficiency H ypothesis 1.1 and the environmental efficiency hyp othesis polycentric, franchise zones, and monocentric are the independent variable s For fiscal efficiency H ypothesis 1.2 polycentric, franchise zones, monocentric public, and monocentric private are the independent variable s Dependent Variables The depe ndent variables for the regression analyses are f iscal efficiency and environmental efficiency This section will first present fiscal efficiency and will then present environmental efficiency. Across all survey cities, the fiscal efficiency had a wide ra nge. Fiscal efficiency was defined as the cost of trash collection per household per year. The mean cost of trash collection among the cities surveyed was $17 7.53 per household per year (median $172.96). The franchise zone model was slightly below this mean with an average cost of $165.76 per household per year (median $172.00). The monocentric was even further below the mean with an average cost of $162.75 per household per year (median $165 .00 ). The polycentric model was the one model above the mean with an average cost of $308.38 (median $274.83). Table 6. 2 below presents this cost of trash collection service production in more detail broken down by service delivery schemes Table 6. 3

PAGE 200

! 180 then presents the one sample t test results for the mean across the service production categories of polycentric, franchise zone and monocentric Table 6. 4 then presents the one sample t test results for the mean across the service production categories of polycentric, franchise zone monocentric public, and monocen tric private The one sample t test presents strong evidence that the polycentric model results in a higher overall cost per household (t=2.993, df=9, p<.05, 2 tailed) the monocentric model results in a lower overall cost per household (t= 2.215, df=74, p<.05, 2 tailed) and the monocentric public model results in a lower overall cost per household (t= 3.581, df=31, p<.001, 2 tailed) Franchise zone and monocentric private did not test as statistically significantly different than the mean fiscal efficie ncy value Table 6. 2 : Cost of Trash Collection Service Production Service Delivery Scheme Cost of Service Production ($ per Household per Year) Highest Cost Lowest Cost Mean Cost Standard Deviation Median Cost Polycentri c (N = 10) $583.33 $182.00 $308. 38 $138.24 $274.83 Franchise Zones (N = 17) $314.16 $95.67 $165.76 $79.07 $172.00 Monocentric (N = 75) $436.59 $69.58 $162.75 $57.79 $165.00 Monocentric P ublic (N = 32) $320.91 $69.58 $142.19 ** $55.81 $145.61 Monocentric P rivate (N = 43) $436.59 $8 0.00 $178.04 $54.99 $178.99 All Schemes (N = 102) $583.33 $69.58 $17 7.53 $83.79 $172.96 = Significant at the .05 level, ** = Significant at the .01 level, *** = Significant at the .001 level

PAGE 201

! 181 T able 6.3: One sample t test Results for Cost of Trash Col lection Service Production: Polycentric, Franchise Zones, and Monocentric Service Delivery Scheme Test Value = $177.53 t df Significance Mean Difference 95% Confidence Interval Polycentric (N = 10) 2.993 9 .015 130.85 31.96 to 229.74 Franchise Zones (N = 17) .614 16 .548 11.77 52.43 to 28.89 Monocentric (N = 75) 2.215 74 .030 14.78 28.08 to 1.49 Table 6.4: One sample t test Results for Cost of Trash Collection Service Production: Polycentric, Franchise Zones, Monocentric Public, Monocentric Private Service Delivery Scheme Test Value = $177.53 t df Significance Mean Difference 95% Confidence Interval Polycentric (N = 10) 2.993 9 .015 130.85 31.96 to 229.74 Franchise Zones (N = 17) .614 16 .548 11.77 52.43 to 28.89 Monocentric Public (N = 32) 3.581 31 .001 35.34 55.46 to 15.21 Monocentric Private (N = 43) .061 42 .952 .51 16.41 to 17.43 Figure 6.1 incorporates these findings into the fragmentation overlap 2x2 diagram.

PAGE 202

! 182 Figure 6.1: Hypothesis 1.1 Fiscal Efficie ncy Findings Across all survey cities, the environmental efficiency likewise had a wide range. Environmental efficiency was defined as the CO 2e per household per year generated by the combination of CO 2e emitted in the transportation of trash and the CO 2 e emitted by the disposal of the trash. The mean CO 2e of trash collection among the cities surveyed was 1.514 CO 2e per household per year (median 1.58). The franchise zone model was ! ! ! Polycentric Franchise Zones Monocentric Overlap No Overlap Consolidated Fragmented Number Cities: 10 Range: $182 $583 Mean: $308 (Std. Dev. $138) Median: $275 (All units: $ / Household / Year) Number Cities: 17 Range: $96 $314 Mean: $166 (Std. Dev. $79) Median: $172 (All units: $ / Household / Year) Number Cities: 75 Range: $70 $437 Mean: $163 (Std. Dev. $58) Median: $165 (All units: $ / Household / Year)

PAGE 203

! 183 below this mean with an average of 1.320 CO 2e per household per year (me dian 1.58 CO 2e ). The monocentric was also b elow the mean with an average of 1.493 CO 2e per household per year (median 1.58 CO 2e ). The polycentric model was the one model above the mean with an average of 2.036 CO 2e (median 1.91 CO 2e ). Table 6. 5 below pr esents this environmental efficiency of trash collection service production in more detail. Table 6. 6 then presents the one sample t test results for the mean across the service production categories of polycentric, franchise zone, and monocentric for the dependent variable of envi ronmental efficiency. Table 6.7 then presents the one sample t test results for the mean across the service production categories of polycentric, franchise zone, and monocentric for the dependent variable of route only environme nta l efficiency. Finally, Table 6.8 presents the one sample t test results for the mean across the service production categories of polycentric, franchise zone, and monocentric for the dependent variable of disposal only environmental efficiency. The one sample t test for environmental efficiency presents strong evidence that the polycentric model results in a less environmentally efficient service delivery than the monocentric and franchise zone service production models (t=2.9 57 df= 8 p<.05, 2 tailed) Table 6. 5 : Environmental Efficiency of Trash Collection Service Production Service Delivery Scheme Environmental Efficiency of Service Production (CO 2e per Household per Year) Largest Footprint Smallest Footprint Mean CO 2e Standard Deviation Median CO 2e Polycentric 2.88 0.99 2.036 0.5331 1.91 Franchise Zones 2.01 0.20 1.320 0.4994 1.58 Monocentric 2.55 0.56 1.493 0.4213 1.49 All Schemes 2.88 0.20 1.514 0.4766 1.58 = Significant at the .05 level, ** = Significant at the .01 level, *** = Significant at the .001 level

PAGE 204

! 184 Table 6. 6: One sample t test Results for Environmental Efficiency of Trash Collection Service Production Service Delivery Scheme Test Value = 1.5 1 t df Significance Mean Difference 95% Confidence Interval Polycentric (N = 9 ) 2.9 57 8 01 8 .526 .116 to .935 Franchise Zones (N = 17) 1.569 16 136 .190 .447 to .067 Monocentric (N = 71 ) .321 70 .749 .016 .116 to 0.84 Table 6. 7: One sample t test Results for Route only Environmental Efficiency of Trash Collection Service Produ ction Service Delivery Scheme Test Value = 0.39 t df Significance Mean Difference 95% Confidence Interval Polycentric (N = 9 ) 2.881 8 .020 0.481 .096 to .866 Franchise Zones (N = 17) .727 16 .478 0.038 .147 to .072 Monocentric (N = 71) 1.837 70 .070 .054 .113 to .005 Table 6. 8: One sample t test Results for Disposal only Environmental Efficiency of Trash Collection Service Production Service Delivery Scheme Test Value = 1.13 t df Significance Mean Difference 95% Confidence Interval Polyce ntric (N = 10) .290 8 779 .034 .239 to .308 Franchise Zones (N = 17) 1.569 16 .136 .190 .067 to .447 Monocentric (N = 71) .740 70 .462 .028 .047 to .103

PAGE 205

! 185 Figure 6.2 incorporates these findings into the fragmentation overlap 2x2 diagram. Figure 6.2: Hypothesis 2 Environmental Efficiency Findings Control Variables This section includes counts and / or descriptive statistics for some of the control variables utilized in the efficiency regression models and the eq uity analysis Some of these control variables were eventually removed from the models, as the controls did not ! ! ! Polycentric Franchise Zones Monocentric Overlap No Overlap Con solidated Fragmented Environmental Efficiency : Route: 0.31 1.75 (M: 0.87 SD: 0.50) Disp osal : 0.58 1.37 (M: 1.16 SD: 0.36) Total: 0.89 3.12 (M: 2.04 SD: 0.53) (Units are CO 2e / HH / Yr) Environmental Efficiency : Route: 0.24 1.01 (M: 0.35 SD: 0 .21 ) Disp osal : 0.04 2.05 (M: 0.97 SD: 0.38 ) Total: 0.20 2.55 (M: 1.32 SD: 0.50 ) (Units are CO 2e / HH / Yr) Environmental Efficiency : Route: 0.10 0.70 (M: 0.34 SD: 0.24 ) Disposal: 0.50 1.30 (M: 1.16 SD: 0.32 ) Total: 0.60 1.92 (M: 1.49 SD: 0.42 ) (Units are CO 2e / HH / Yr) No Overlap Overlap

PAGE 206

! 186 present statistically significant relationships with the dependent variables The control variables are presented in the following order: leve l of service, level of automation fuel source, work force unionization, program funding, breadth of service and citywide solid was te goal The section closes with a discussion of hybrid drivetrains this was not ultimately utilized as a control variable a nd this section includes a discussion as to why. Level of Service: The majority of cities surveyed, 93 in total delivered trash collection at a standard level of service. Among the ten polycentric cities, eight cities had standard collection and two citie s had greater than standard collection (for both of these two the extra service was twice a week trash collection). Among the 17 franchise zone cities, 16 cities delivered standard collection and one delivered greater than standard collection (again, twic e a week trash collection). Among the 75 monocentric cities, 69 cities delivered standard collection, two delivered greater than standard collection (again, twice a week trash collection), and four delivered a less than standard level of collection (th re e collected trash every two weeks; one did not offer a regular bulky item collection). Level of Automation : The survey as administered included three categories for level of automation : fully automated service, partially automated service, and manual coll ection. A collection worker never having to exit the truck most completely characterizes a fully automated service. All trash can be collected with a single employee and single truck. Partially automated service is characterized by an occasional employe e assist to the automation process (i.e. occasionally exiting of truck to position a cart for collection); partially automated service is generally characterized by crew sizes of 1 2 employees. Employees rather than machinery execute the trash collection process in a

PAGE 207

! 187 fully manual collection situation. Fully manual collection is generally characterized by crew sizes of 2 3 employees. E xamination of the dissertation's case stud y cities indicated most cities utilized a combination of fully and partially aut omated collection. These cities would have fully automated routes where access r oadways and alleyways w ere sufficiently wide but would utilize partially automated collection where a neighborhood's geometry simply did not allow for full automation. Beca use of th is observation the measure for automation was simplified to a two point scale: (1) automated collection (encompassing fully automated systems, partially automated systems, and all combinations in between) and (2) fully manual collection systems. The findings were that a total of 89 cities (87.3%) utilized some form of automated collection and 13 cities (12.7%) utilized fully manual collection. Across all three service delivery models automated collection was dominant. Seven of the 10 polycentri c cities utilized automated collection; 15 of the 17 franchise zone cities utilized automated collect; 67 of 75 monocentric cities utilized automated collection Fuel Source: The survey was used to identify a range of fuels being utilized by collection fle ets including: diesel, natural gas (both compressed and liquid) propane, and other alternative fuels" The survey data indicated that there were only two categories in use among cities that responded to the survey : fleets that were fully diesel powered and fleets that were fueled by a combination of natural gas 47 and diesel powered vehicles. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! )& There are two types of natural gas vehicles in common use among trash fleets today : compressed natural gas vehicles and liquid natural gas vehicles. Both, however, are still rare enough that this category is simply combine d i nto a single "natural gas" category.

PAGE 208

! 188 The findings were 87 cities (85.3%) utilized diesel fueled fleets and 15 cities (14.7%) utilized mixed fleets of both diesel and natural gas powered trucks. Among the cities surveyed: none of the polycentric cities had adopted mixed fleets, two of the franchise zone cities had adopted mixed fleets, and 13 of the monocentric cities had adopted mixed fleets. Work Force Unionization: In terms of work force unionizati on, 26.4% of the total work forced was reported to be unionized among cities responding to the survey For pol ycentric cities, one city was reported nonunionized and nine cities did not report. For franchise zone cities, eight cities reported as unionize d and nine cities reported as nonunionized. For monocentric cities, 19 cities reported as unionized, 50 cities reported as nonunionized, and six cities did not report. Table 6 9 below presents these findings in counts and percentages. Table 6 9 : Unionizat ion of Trash Collection Workforce Unionized or Non Unionized? Polycentric Franchise Zone Monocentric Totals Unionized 0 Cities 0% Polycentric 0% All Cities ** 8 Cities 47% Franchise 7.8% All Cities 19 Cities 25.3% Mono 18.6% All Cities 27 Cities 26.4% Ci ties Nonunionized 1 City 10% Polycentric 1% All Cities 9 Cities 53% Franchise 8.8% All Cities 50 Cities 66.7% Mono 49% All Cities 60 Cities 58.8% Cities Not Reported 9 Cities 90% Polycentric 8.8% All Cities 0 Cities 0% Franchise 0% All Cities 6 Cities 8 % Monocentric 5.9% All Cities 15 Cities 14.7% Cities This is a column specific percentage only (Polycentric, Franchise Zone, and Monocentric). That is each column totaled equals 100%. ** "All Cities" percentage total across the three columns and the th ree rows. That is all nine cells totaled equals 100%. Program Funding: In terms of the program funding mechanism, among cities responding to the survey, 65 cities ( 63.8% ) had a flat rate system to fund their trash

PAGE 209

! 189 collection services and 37 cities ( 36 .3% ) had a variable rate system (pay as you throw) to fund their trash collection services. Among polycentric cities, four cities had variable rate structures and six cities utilized a flat rate system. Among franchise zone cities, six cities had variabl e rates and 11 cities had a flat rate Among monocentric cities, 27 cities had variable rates and 48 cities had a flat rate Table 6 10 below presents these findings in counts and percentages. Table 6 10 : City Funding Mechanism for Solid Waste Collection F unding Mechanism Polycentric Franchise Zone Monocentric Totals Variable Rates (PAYT) 4 Cities 40% Polycentric 3.9% All Cities ** 6 Cities 35.3% Franchise 5.9% All Cities 27 Cities 36% Mono 26.5% All Cities 3 7 Cities 36.3 % Cities Flat Rate 6 Cit ies 60% P olycentric 5.9% All Cities 11 Cities 64.7 % Franchise 10 .8% All Cities 48 Cities 64% Mono 47.1% All Cities 65 Cities 63 .8% Cities This is a column specific percentage only (Polycentric, Franchise Zone, and Monocentric). That is each column totaled equal s 100%. ** "All Cities" percentage total across the three columns and the two rows. That is all six cells totaled equals 100%. Breadth of Service: T he majority of cities surveyed, 92 cities (90.2%) delivered trash collection and recyclable collection service s likewise 68 cities (66.7%) in total delivered trash collection recyclables collection, and organics collection as a suite of service s Only ten cities (9.8%) delivered only trash collection service. No cities (0%) delivered the combination of trash collection service and organics collection service without recyclables collection service. Among the ten polycentric cities, all ten offered recyclables collection along with trash collection in their suite of services, however, none of the ten offer ed organics collection along with trash collection and recyclables collection

PAGE 210

! 190 Among the 17 franchise zone cities, 15 offered recyclables collection along with trash collection and 12 offered organics collection along with both trash and recyclables collec tions as their suite of services. Thus two cities did not have recyclables or organics collection programs (only trash collection) and t hree cities did not have organics collection program but did have recyclables collection with their trash collection pr ogram Among the 75 monocentric cities, 67 offered recyclables collection along with trash collection and 5 7 offered organics collection along with both trash and recyclables collections as their suite of services. Thus eight cities did not have recyclabl es or organics collection program s and 1 8 cities did not have organics collection program but did have a recyclables collection program and a trash collection program G oal: The majority of cities surveyed, 73 cities had some kind of solid waste related go al for their city. Of these 73 cities, 33 had already achieved their goal. Table 6. 1 1 compares city level solid waste management goals against breadth of services offered.

PAGE 211

! 191 Table 6 1 1 : Comparing Number of Cities with City level Solid Waste Goals Agains t Service Production Models and Breadth of Service s Offered City Goal? Breadth of Service Goal? Goal Achieved? Poly Fran Mono Poly Fran Mono Yes Trash 0 0 6 N/A N/A 5 Trash + Recycle 4 6 9 2 1 5 Trash + Recycle + Organics 0 8 40 N/A 1 19 All Cities 73 33 No Trash 0 1 3 N/A Trash + Recycle 6 1 7 Trash + Recycle + Organics 0 1 9 All Cities 28 * One city did not report Hybrid Drivetrains: The survey was used to identify how many fleets utilized electric assisted hybrid driv etrains combined with electric assisted braking technology in their fleet vehicles over traditional drivetrain vehicles. Amongst cities responding to the survey, however, this option was not indicated. Among case study cities it was noted that fleets had none of these vehicles or a few of these vehicles on an experimental basis. Thus it would appear this technology is still too new to have any level of fleet wide penetration among cities that responded to the survey Because of this finding the electric assisted hybrid drivetrain was ultimately dropped as a control variable from the

PAGE 212

! 192 environmental efficiency analysis and this paragraph is the extent of the discussion of this technology Efficiency Hypothese s Analyse s Regression Analyses Three series of l inear regression analyses were conducted. Each of the th r ee hypotheses were tested using Ordinary Least Squares Regression : fiscal efficiency (Hypotheses 1.1 and 1.2) and environmental efficiency (Hypothesis 2) This section will first discuss the findin gs for the two fiscal efficiency hypotheses and will then discuss the findings for the environmental efficiency hypothesis. Fiscal Efficiency Model s A series of linear regression models were run to test the two research hypotheses. This section will fir st discuss the model runs related to testing Hypothesis 1.1 for monocentric, franchise zone, and polycentric models. The section will then turn to model runs related to testing Hypothesis 1.2 for monocentric with public production, monocentric with privat e production, franchise zone, and polycentric model. Table 6. 1 2 below includes the variables utilized for both fiscal efficiency hypotheses (and the environmental efficiency hypothesis) Table 6. 1 2 : Dependent, Independent, and Control Variables for Fiscal Efficiency Models and Environmental Efficiency Model Label Signifying Type of Variable Utilized with Model(s) Fiscal Fiscal Efficiency ($ per Household per Year) Dependent Fiscal Env Environmental Efficiency (CO 2e per Household per Year) Dependent Environmental Poly Polycentric Service Production Independent Fiscal and Environmental

PAGE 213

! 193 Label Signifying Type of Variable Utilized with Model(s) Fran Franchise Zone Service Production Independent Fiscal and Environmental Mono Monocentric Service Production Independent Fiscal (1.1) and Environmental MonoPub Mo nocentric with Public Service Production Independent Fiscal (1.2) MonoPri Monocentric with Private Service Production Independent Fiscal (1.2) LSP 1 = High Level of Service Production 0 = Standard Level of Service Production Control Fiscal and Environme ntal BSP 1 = Low Breadth of Service Production 0 = Standard Breadth of Service Production Control Fiscal and Environmental AUTO 1 = Manual Collection 0 = Automated Collection Control Fiscal and Environmental FUND 1 = Flat Rate Funding Mechanism 0 = Var iable Rate Funding Mechanism Control Fiscal and Environmental UNIO 1 = Unionized Work Force 0 = Non unionized Work Force Control Fiscal FUEL 1 = Diesel only Fleet 0 = Diesel and Natural Gas Fleet Control Environmental Less50K 1 = Less than 50,000 Househ olds 0 = Greater than 50,000 Households Control Fiscal and Environmental OneIsOne 1 = City Definition of a Single family Household is Truly a Single family Household 0 = City Definition of a Single family Household is more inclusive than just Single famil y Households (such as Duplexes, Triplexes, Etc.) Control Fiscal and Environmental GOAL 1 = City has a Solid Waste Specific Goal 0 = City does not have a Solid Waste Specific Goal Control Fiscal and Environmental ConDep 1 = City is within a State with a C ontainer Deposit Law 0 = City is not within a State with a Container Deposit Law Control Fiscal and Environmental YWBan 1 = City is within a State with a Landfill Ban on Yard Waste Disposal 0 = City is within a State without a Landfill Ban on Yard Waste D isposal Control Fiscal and Environmental

PAGE 214

! 194 Label Signifying Type of Variable Utilized with Model(s) HHDens Household Density = (# Households) per (# Square Miles Citywide) Control Fiscal and Environmental West Western United States Geographic Region (U.S. Census) : Arizona, California, Colorado, Idaho, Montana, N evada, New Mexico, Oregon, Utah, Washington, and Wyoming Geographic Control Fiscal and Environmental Midwest Midwestern United States Geographic Region (U.S. Census) : Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, and South Dakota Geographic Control Fiscal and Environmental NorEast Northeastern United States Geographic Region (U.S. Census) : Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont Geographic Control Fiscal and Environmental South Southern United States Geographic Region (U.S. Census) : Alabama, Arkansas, Delaware, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, and West Virginia Geographic Control Fiscal and Environmental Fiscal Efficiency Model for Hypothesis 1.1 Equation 1.1 below is the initial regression equation that was utilized to test Hypothesis 1.1 Multiple linear regressions were run to test the significance of the variables thus equation 1.1 represents the linear regression equation prior to any of the control variables being eliminated. FISCAL 1.1 ) = c + b 1 (D fran ) + b 2 (D mono ) + b 3 (D LSP ) + b 4 (D BSP ) + b 5 (D AUTO ) + b 6 (D FUND ) + b 7 (D UNIO ) + b 8 (D Less50K ) + b 9 (D OneIsOne ) + b 10 (D GOAL ) + b 11 (D ConDep ) + b 12 (D YWBan ) + b 13 (D HHDens ) + b 14 (D WEST ) + b 15 (D MIDWEST ) + b 16 (D SOUTH ) + e Equation 1.1: Initial Linear Regression Equation for T esting Hypothesis 1.1

PAGE 215

! 195 A total of t welve control variables were initially inclu ded in the initial regression model. The se controls were tested for significance. Table 6. 1 3 below summarizes the most notable model runs and the p values for both the independent and the control variables included in each model run In each case, the m odel itself had explanatory value. Ultimately the two control variables found to be significant across various model runs were high level of service production and low breadth of service production (trash only collection) Table 6. 1 3 : Linear Regression Ana lysis Runs for Fiscal Efficiency Hypothesis 1.1 p values for Independent and Control Variables MODEL RUN Monocentric Service Production Franchise Zone Service Production High Level of Service Production Low Breadth of Service Production Automation Fundin g Mechanism Unionization Container Deposit Law Landfill Yard Waste Ban Solid Waste Goal Household Density West (Geographic) Midwest (Geographic) South (Geographic) 1 .006 .006 .001 .006 .347 .474 .355 .962 .824 .971 .146 .290 .214 .563 2 .005 .005 .000 005 .335 .481 .364 .957 .139 .237 .200 .556 3 .005 .007 .001 .009 .372 .469 .386 .837 .966 .981 .192 4 .005 .008 .001 .008 .594 .316 .420 5 .000 .000 .000 .041 Model R un 1 included all control variables. The geographic control w as split into fo u r dummy categories. The model included the West, Midwest, and South dummies and omitted the Northeast dummy. This model tested as having explanatory value. ANOVA s ignificance for model was .004 ( .01 significance level). Model Run 2 test ed the geographic controls while removing the two state law controls of container deposit law and landfill yard waste ban. Model Run 3 did the reverse of Model Run 2 and tested the two state law controls of container deposit law and landfill yard waste ba n while removing the geographic controls. Ultimately neither the

PAGE 216

! 196 state law controls nor the geographic controls tested as significant and all were ultimately removed from the final model. However, both models tested as having explanatory value. ANOVA s ig nificance for Model Run 2 w as .001 (.001 significance level). ANOVA s ignificance for Model Run 3 was .001 (.001 significance level). Model Run 4 is an example utilizing a leaner set of control variables in this specific run high level of service product ion, low breath of service production, level of automation, funding mechanism, and workforce unionization were included. Ultimately only high level of service production and low brea d th of service production tested as significant. ANOVA significance for Model Run 4 was .000 (.001 significance level). Model Run 5 is the final run with the four significant variables. Both independent monocentric and franchise zone test as significant and the two controls of high level of service production and low breadth of service production test as significant. ANOVA significance for Model Run 5 was .000 (.001 significance level). The following more in depth analysis for Hypothesis 1.1 is based on the findings from Model Run 5. Table 6.1 4 below presents the regression analysis results for fiscal efficiency for the independent variable service production categories of monocentric and franchise zones with polycentric serving as the omitted category.

PAGE 217

! 197 Table 6. 1 4 : Regression Analysis for Fiscal Efficiency Hypothesis 1.1 Independent and Control Variable B (Unstan darized) 95% Confidence Interval Standard Error t p value (Constant) $282.55 $239.34 to $325.75 21.769 Monocentric (IV) $119.20 $164.67 to $73.72 22.913 5.202 .000 Franchise Zone (IV) $118.43 $171.59 to $65.27 26.783 4.422 .000 High Level of Service Production (CV) $129.18 $67.32 to $191.05 31.171 4.144 .000 Low Breadth of Service Production ( CV) $50.62 $99.18 to $2.06 24.466 2.069 .041 Note: Sum of squares for regression 287162.257 (R 2 = 405; Adjusted R 2 = .380) F score = 16.503; model significance = .000 (.001 significance level) The multiple regression analysis was conducted to evaluate how well service production (polycentric, franchise zones, and monocentric) predicted fiscal efficie ncy. The linear relationship between service production and fiscal efficiency was significant (F score = 16.503, p<.00 1 ) The adjusted R 2 indicates this model predicts approximately 40% of the variation observed in this model. T wo control variables were s ignificant: high level of service production (p<.001) and low breadth of service production (p<.05). Holding high level of service production (HLSP) and low breadth of service production (LBSP) constant, the average annual cost per household under polycent ric is $282.55. Again holding HLSP and LBSP constant, the average annual cost per household under monocentric is $163.35 ($119.20 less than polycentric) and the average annual cost per household under franchise zone is $164.12 ($118.43 less than polycentr ic). A HLSP (most often twice a week trash collection instead of once a week trash collection) on average added $129.18 to the annual cost of service across the service

PAGE 218

! 198 production categories and a LBSP (trash only collection, no recyclables collection and no organics collection) on average decreased the annual cost of service production across the service production categories by $50.62. Table 6. 15 below presents the zero order and partial correlations for the independent and control variables with the de pendent variable of fiscal efficiency. Table 6. 15 : Zero order and Partial Correlations for Fiscal Efficiency Variable Correlation between IVs and CVs and Fiscal Efficiency Correlations between IVs and CVs and Fiscal Efficiency Controlling for all Other Var iables Monocentric (IV) .30 .47 Franchise Zones (IV) .06 .41 High Level of Service (CV) .44 .39 Low Breadth of Service (CV) .23 .21 As predicted by Hypothesis 1.1 the mean value for the mon ocentric model is considerably more fiscally efficien t than the polycentric model holding high level of service production and low breadth of service production constant However, the model does not find any difference of statistical significance between monocentric and franchise zone service production in re lation to fiscal efficiency. ( B ecause these numbers represent dollar values, a smaller number represents a higher efficiency ) Regression diagnostics indicated linearity in the relationship between the independent variables and the dependent variable. Residuals and errors presented homogeneity and as normally distributed, thus the model and residuals analysis presented as homoscedastic and consequently negative for heteroskedasticity. Variables operated largely independent of each other. Diagnostics indicated very low collinearity between the variables and thus did not indicate problems of

PAGE 219

! 199 multicollinearity between the variables all VIFs were less than five across the different model runs and across the different independent and control variables. Li kewise interaction effects between variables were low and did not present multicollinearity concerns. While the model did have four outliers the model was stronger with the outliers included in that the outliers were logically aligned along the line whethe r included or excluded from the model indicating that the outliers were not exercising excessive leverage / influence over the model. Given this finding and the small overall sample size it was decided the model was better with the outliers left within it than excluded. Fiscal Efficiency Model for Hypothesis 1.2 Equation 1.2 below is the initial regression equation that was utilized to test Hypothesis 1.2. Multiple linear regressions were run to test the significance of the variables thus equation 1.2 rep resents the linear regression equation prior to any of the control variables being eliminated. This second fiscal efficiency model had monocentric service production split into two categories: monocentric service production by the city (MonoPub) and monoc entric service production by a single private sector service producer (MonoPri). (FISCAL 1.2 ) = c + b 1 (D fran ) + b 2 (D MonoPub ) b 3 (D MonoPri ) + b 4 (D LSP ) + b 5 (D BSP ) + b 6 (D AUTO ) + b 7 (D FUND ) + b 8 (D UNIO ) + b 9 (D Less50K ) + b 10 (D OneIsOne ) + b 11 (D GOAL ) + b 12 (D ConDep ) + b 13 (D YWBan ) + b 14 (D HHDens ) + b 15 (D WEST ) + b 16 (D MIDWEST ) + b 17 (D SOUTH ) + e Equation 1.2: Initial Linear Regression Equation for T esting Hypothesis 1.2 A total of t welve control variables were initially included in the regression model. These controls w ere tested for significance. Table 6. 1 6 below summarizes the most notable model runs and the p values for both the independent and the control variables

PAGE 220

! 200 included in each model run In each case, the model itself had explanatory value. Ultimately the two control variables found to be significant across various model runs were high level of service production and low breadth of service production (trash only collection). Table 6. 1 6 : Linear Regression Analysis Runs for Fiscal Efficiency Hypothesis 1.2 p val ues for Independent and Control Variables MODEL RUN Monocentric Public Service Production Monocentric Private Service Production Franchise Zone Service Production High Level of Service Production Low Breadth of Service Production Automation Funding Mechan ism Unionization Container Deposit Law Landfill Yard Waste Ban Solid Waste Goal Household Density West (Geographic) Midwest (Geographic) South (Geographic) 1 .002 .009 .006 .001 .029 .403 .300 .208 .943 .691 .921 .144 .388 .240 .429 2 .002 .007 .004 .001 .022 .379 .308 .220 .942 .138 .314 .247 .398 3 .002 .008 .005 .001 .036 .454 .293 .243 .843 .863 .904 .216 4 .001 .007 .004 .000 .029 .651 .183 .271 5 .000 .000 .000 .000 .018 Model Run 1 included all control variables. The geo graphic control was split into four dummy categories. The model included the West, Midwest and South dummies and omitted the Northeast dummy. This model tested as having explanatory value. ANOVA significance for model was .001 (.001 significance level) Model Run 2 tested the geographic controls while removing the two state law controls of container deposit law and landfill yard waste ban. Model Run 3 did the reverse of Model Run 2 and tested the two state law controls of container deposit law and land fill yard waste ban while removing the geographic controls. Ultimately neither the state law controls nor the geographic controls tested as significant and all were ultimately removed from the final model. However, both models tested as having explanatory value.

PAGE 221

! 201 ANOVA significance for Model Run 2 was .000 (.001 significance level). ANOVA significance for Model Run 3 was .000 (.001 significance level). Model Run 4 is an example utilizing a leaner set of control variables, in this specific run high level of service production, low breath of service production, level of automation, funding mechanism, and workforce unionization were included. Ultimately only high level of service production and low breadth of service production tested as significant. ANOVA significance for Model Run 4 was .000 (.001 significance level). Model Run 5 is the final run with the five significant variables. All three independent s, monocentric public, monocentric private, and franchise zone test as significant and the two contro ls of high level of service production and low breadth of service production test as significant. ANOVA significance for Model Run 5 was .000 (.001 significance level). The following more in depth analysis for Hypothesis 1.2 is based on the findings from Model Run 5. Table 6.1 7 below presents the regression analysis results for fiscal efficiency for the independent variable service production categories of monocentric public, moncentric private, and franchise zones with polycentric serving as the omitted category.

PAGE 222

! 2 02 Table 6.1 7 : Regression Analysis for Fiscal Efficiency Hypothesis 1.2 Independent and Control Variable B (Unstan dardized) 95% Confidence Interval Standard Error t p value (Constant) $282.21 $239.68 to $324.74 21.425 Monocentric Public (IV ) $137.86 $186.18 to $89.54 24.343 5.663 .000 Monocentric Private (IV) $106.28 $152.78 to $59.78 23.426 4.537 .000 Franchise Zone (IV) $119.44 $171.59 to $65.27 26.364 4.531 .000 High Level of Service Production (CV) $130.87 $69.95 to $191.79 30.689 4.264 .000 Low Breadth of Service Production (CV) $39.98 $88.89 to $8.93 24.640 1.623 .018 Note: Sum of squares for regression 304627.247 (R 2 = .430; Adjusted R 2 = .400) F score = 14.459; model significance = .000 (.001 significance level) T he multiple regression analysis was conducted to evaluate how well service production (polycentric, franchise zones, monocentric public, and monocentric private) predicted fiscal efficiency. The linear relationship between service production and fiscal ef ficiency was significant (F score = 1 4.459 p<.00 1 ) The adjusted R 2 indicates this model predicts approximately 40% of the variation observed in this model. Two control variables were significant: high level of service production (p<.001) and low breadt h of service production (p<.05). Holding high level of service production (HLSP) and low breadth of service production (LBSP) constant, the average annual cost per household under polycentric is $282. 21. Again holding HLSP and LBSP constant, the average a nnual cost per household under monocentric public is $1 44.35 ($1 37.86 less than polycentric) the average annual cost per household under monocentric private is $175.93 ($106.28 less

PAGE 223

! 203 than polycentric), and the average annual cost per household under franch ise zone is $1 62.77 ($11 9.44 less than polycentric). A HLSP (most often twice a week trash collection instead of once a week trash collection) on average added $1 30.87 to the annual cost of service across the service production categories and a LBSP (tras h only collection, no recyclables collection and no organics collection) on average decreased the annual cost of service production across the service production categories by $ 39.98 Table 6. 18 below presents the zero order and partial correlations for t he independent and control variables with the dependent variable of fiscal efficiency. Table 6. 18 : Zero order and Partial Correlations for Fiscal Efficiency Variable Correlation between IVs and CVs and Fiscal Efficiency Correlations between IVs and CVs and Fiscal Efficiency Controlling for all Other Variables Monocentric Public (IV) .29 .50 Monocentric Private (IV) .01 .42 Franchise Zones (IV) .06 .42 High Level of Service (CV) .44 .40 Low Breadth of Service (CV) .23 .16 As predicted by Hypoth esis 1.2 the mean value for monocentric private is considerably more fiscally efficient than polycentric holding high level of service production and low breadth of service production constant. (Because these numbers represent dollar values, a smaller nu mber represents a higher efficiency.) However, the model does not find any difference of statistical significance between the monocentric private, monocentric public, and fr anchise z one service production models in relation to fiscal efficien cy. As monoc entric private was predicted to be most fiscally efficient by H ypothesis 1.2 this finding only lends part ial support for the hypothesis.

PAGE 224

! 204 Regression diagnostics indicated linearity in the relationship between the independent variables and the dependent vari able. Residuals and errors presented homogeneity and as normally distributed, thus the model and residuals analysis presented as homoscedastic and consequently negative for heteroskedasticity. Variables operated largely independent of each other. Diagnos tics indicated very low collinearity between the variables and thus did not indicate problems of multicollinearity between the variables all VIFs were less than five across the different model runs and across the different independent and control variables Likewise interaction effects between variables were low and did not present multicollinearity concerns. While the model did have four outliers the model was stronger with the outliers included in that the outliers were logically aligned along the line w hether included or excluded from the model indicating that the outliers were not exercising excessive leverage / influence over the model. Given this finding and the small overall sample size it was decided the model was better with the outliers left with in it than excluded. Fiscal Efficiency Discussion Beyond the Research Hypothesis for Polycentric Cities This section specifically looks at the ten polycentric cities from the large N and explores if any patterns can be garnered for these cities in terms o f fiscal efficiency compared against environmental efficiency, level of service production, geographic region, number of waste haulers serving each city, and the number of households being served within each city. Table 6. 19 below presents these findings for all ten of the polycentric cities. The table is sorted from least fiscally efficient to most fiscally efficient polycentric city.

PAGE 225

! 205 Table 6. 19 : Comparing Fiscal Efficiency Against Other Factors for Polycentric Cities Fiscal Efficiency ($ Per Household Per Year)* Environmental Efficiency (CO 2e Per Household Per Year)** Level of Service Production Geographic Region of United States Number of Waste Haulers with >15% Market Share Total Number of Waste Haulers Total # Households $583.33 2.58 High South 6 9 12,000 $533.03 1.83 High Midwest 6 7 13,320 $298.00 0.99 Standard West 3 3 60,816 $287.32 2.11 Standard Midwest 4 4 22,487 $285.42 1.85 Standard South 6 6 22,487 $264.24 2.27 Standard South 6 13 92,670 $249.50 1.91 Standard West 4 9 128,958 $210.00 Not Reported Standard Midwest 6 23 61,707 $191.00 2.88 Standard Midwest 5 5 16,010 $182.00 1.90 Standard Midwest 3 3 29,000 Because Fiscal Efficiency is reported in $ per household per year, the higher the number the lower the efficiency. ** Because E nvironmental Efficiency is reported in CO 2e per household per year, the higher the number the lower the efficiency. Three observations surface from this table: the two cities with substantially lower fiscal efficiency than the other eight polycentric ci ties have three characteristics in common: both deliver a high level of service production (twice a week trash collection) 48 the two are the smallest cities among the ten, and the two tend to be on the higher end of the number of waste haulers producing se rvice within the cities served. There does not appear to be any particular association between fiscal efficiency and geographic region of the United States where the cities are located. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! )' The two cities are, in fact, the only two polycentric cities delivering a high level of service production among the survey response cities.

PAGE 226

! 206 Testing Regression Model Fit Against Case Study Cities As a final test of the regression models, the models predicted fit was tested against the actual findings for the eight case study cities. Because the omitted category from the regression was polycentric service production, the constant represents the mean cost of t his service production category with a standard level of service production and a high breadth of service production The control variable of "level of service production" was split into two categories: high level of service production and standard level of service production. A "yes" for high level had a coefficient adjustment of $129.18 for the 1.1 test and $130.87 for the 1.2 test. Among the case study cities, seven had a standard level of service production. Only Phoenix had a high level of service production, thus Phoenix is the one case study city with this adjustment. The control variable of "breadth of service" was split into two categories: low breadth of service ( trash only collection) and standard breadth of service (trash and recyclables col lection or trash, recyclables, and organics collection). A "yes" for low breadth had a coefficient of adjustment of $50.62 for the 1.1 test and $39.98 for the 1.2 test. However, all eight of the case study cities had a standard breadth of service and t hus none have this adjustment. Table 6. 20 summarizes the findings from the test of regression model fit against the eight case study citie s and Table 6. 2 1 compares the differences between the predicted versus the actual outcomes against the standard devia tion. For seven of the eight case studies the fit of the model is better than observed standard deviation for each of the service production models. Only the case study city of Fort Collins actual versus predicted fell outside of the standard deviation.

PAGE 227

! 207 Table 6. 20 : Testing Regression Model Fit Against Case Study Actual Findings City Service Productio n Model Constant Service Productio n Model Level of Service Productio n Breadth of Service Productio n Predicted Outcome Actual Outcome Testing with Equation 1.1: Monocentric, Franchise Zone, and Polycentric Austin Mono $282.55 $119.20 0 0 $163.35 $142 Denver Mono $282.55 $119.20 0 0 $163.35 $145 Fresno Mono $282.55 $119.20 0 0 $163.35 $156 Knoxville Mono $282.55 $119.20 0 0 $163.35 $124 Indianapolis Fran $282.55 $118.43 0 0 $164.12 $130 Phoenix Fran $282.55 $118.43 $129.18 0 $293.30 $322 CO Springs Poly $282.55 0 0 0 $282.55 $204 $372 Ft Collins Poly $282.55 0 0 0 $282.55 $146 $450 Testing with Equation 1.2: Mono Public, Mono Private, Franch ise Zone, and Polycentric Austin MonoPub $282.21 $137.86 0 0 $144.35 $142 Denver MonoPub $282.21 $137.86 0 0 $144.35 $145 Fresno MonoPub $282.21 $137.86 0 0 $144.35 $156 Knoxville MonoPri $282.21 $106.28 0 0 $175.93 $124 Indianapolis Fran $282.21 $119.44 0 0 $162.77 $130 Phoenix Fran $282.21 $119.44 $130.87 0 $293.64 $322 CO Springs Poly $282.21 0 0 0 $282.21 $204 $372 Ft Collins Poly $282.21 0 0 0 $282.21 $146 $450 In reviewing Table 6. 20 the regression model fit with the case study cities, Phoenix moves from being an outlier among the case studies to a logical fiscal efficiency based on the control variable of high level of service production coming into play and impacting the city's fiscal efficiency.

PAGE 228

! 208 Table 6. 21 : Testing Regressio n Model Fit Against Case Study Actual Findings : Differences Between Predicted and Actual Versus Standard Deviation City Service Productio n Model Predicted Outcome Actual Outcome Difference between Predicted and Actual Standard Deviation Outcome Testing with Equation 1.1: Monocentric, Franchise Zone, and Polycentric Aust in Mono $163.35 $142 21.35 57.79 Difference between predicted and actual < Standard Deviation Denver Mono $163.35 $145 18.35 57.79 Difference between predicted and actual < Standard Deviation Fresno Mono $163.35 $156 7.35 57.79 Difference between predi cted and actual < Standard Deviation Knoxville Mono $163.35 $124 39.35 57.79 Difference between predicted and actual < Standard Deviation Indianapolis Fran $164.12 $130 34.12 79.07 Difference between predicted and actual < Standard Deviation Phoenix Fran $293.30 $322 28.70 79.07 Difference between predicted and actual < Standard Deviation CO Springs Poly $282.55 $204 $372 65.76 89.45 138.24 Difference between predicted and actual < Standard Deviation Ft Collins Poly $282.55 $146 $450 136.55 167.45 138.24 Difference between predicted and actual > Standard Deviation Testing with Equation 1.2: Mono Public, Mono Private, Franchise Zone, and Polycentric Austin MonoPub $144.35 $142 2.35 55.81 Difference between predicted and actual < Standard Deviation Denver MonoPub $144.35 $145 0.65 55.81 Difference between predicted and actual < Standard Deviation Fresno MonoPub $144.35 $156 11.65 55.81 Difference between predicted and actual < Standard Deviation Knoxville MonoPri $175.93 $124 51.93 54 .99 Difference between predicted and actual < Standard Deviation Indianapolis Fran $162.77 $130 32.77 79.07 Difference between predicted and actual < Standard Deviation Phoenix Fran $293.64 $322 28.36 79.07 Difference between predicted and actual < St andard Deviation CO Springs Poly $282.21 $204 $372 78.21 89.79 138.24 Difference between predicted and actual < Standard Deviation Ft Collins Poly $282.21 $146 $450 136.21 167.79 138.24 Difference between predicted and actual > Standard Deviatio n

PAGE 229

! 209 When the difference between predicted and actual is compared against the standard deviation for each service production category, the model's predicted outcome (both for Hypotheses 1.1 and 1.2) produces a strong fit between predicted and actual outco mes and a tighter variance than that from the standard deviation alone. This is a good indication that the models are indeed predictive of fiscal efficiency when service production model, level of service, and breadth of service are all considered. In pa rticular, the model for monocentric public arrangements generated prediction eerily close to the actual values. Environmental Efficiency Mode ls A series of linear regression models were run to test the environmental efficiency research hypothes is. This se ction will discuss the model runs related to testing Hypothesis 2 for monocentric, franchise zone, and polycentric models. See Table 6. 12 in the fiscal efficiency section above for a full list of the variables utilized for this hypothes i s. Equation 2 bel ow is the initial regression equation that was utilized to test Hypothesis 2. Multiple linear regressions were run to test the significance of the variables thus equation 2 represents the linear regression equation prior to any of the control variables be ing eliminated. (ENV) = c + b 1 (D fran ) + b 2 (D mono ) + b 3 (D LSP ) + b 4 (D BSP ) + b 5 (D AUTO ) + b 6 (D FUND ) + b 7 (D FUEL ) + b 8 (D Less50K ) + b 9 (D OneIsOne ) + b 10 (D GOAL ) + b 11 (D ConDep ) + b 12 (D YWBan ) + b 13 (D HHDens ) + b 14 (D WEST ) + b 15 (D MIDWEST ) + b 16 (D SOUTH ) + e Equation 2: Initial Linear Regression Equation for Testing Hypothesis 2 A total of t welve control variables were initially included in the regression model. These controls were test ed for significance. Table 6.2 2 below summarizes the most

PAGE 230

! 210 notable model runs and th e p values for both the independent and the control variables included in each model run. Ultimately the two control variables found to be significant across various model runs were high level o f service production and one is one (that is a single family household definition is truly only single family households (no multi unit housing such as duplexes, triplexes, etc.). Table 6. 2 2 : Linear Regression Analysis Runs for Environmental Efficiency Hypothesis 2 Significance Levels for Model, Independent and C ontrol Variables MODEL RUN Model Significance Lev Monocentric Franchise Zone Level of Service Prod. Breadth of Service Prod. Automation Funding Mechanism Fuel Household Density 1 Is 1 Sing. Fam. Container Deposit Law Landfill Yard Waste Ban Goal West (Geo graphic) Midwest (Geographic) South (Geographic Model R uns for Environmental Efficiency as the Dependent Variable 1 .007 .074 .008 .007 .908 .864 .184 .361 .595 .878 .661 .126 .803 .977 .869 2 .002 .162 .042 .009 .694 .781 .352 .281 .034 .893 .663 .07 9 .742 .969 .773 3 .000 .146 .047 .006 .802 .794 .309 .298 .030 .834 .085 4 .000 .153 .039 .003 .705 .815 .310 .271 .032 .078 .796 .914 .722 5 .000 .000 .004 .049 .208 .021 .287 6 .000 .033 .009 .005 .026 Model R uns for Rou te Environmental Efficiency as the Dependent Variable 1 .001 .000 .003 .339 .521 .308 .392 .531 .347 .577 .303 .410 .489 .919 .401 2 .000 .001 .015 .385 .262 .536 .636 .642 .043 .573 .308 .318 .502 .915 .522 3 .000 .001 .011 .151 .237 .545 .599 .640 .049 .280 .295 4 .000 .001 .012 .179 .290 .534 .523 .727 .045 .325 .720 .614 .319 5 .000 .000 .004 .049 .208 .021 .287 6 .000 .000 .001 .043 .050 Model R uns for Dispos al Environmental Efficiency as the Dependent Variable 1 117 .515 .309 .005 .488 .529 .288 .083 .936 .501 .780 .174 .805 .963 .356 2 .083 .368 .518 .007 .676 .381 .406 .074 .263 .520 .782 .136 .900 .971 .364 3 .046 .341 .646 .016 .516 .405 .377 .086 .229 .537 .170 4 .045 .323 .554 .005 .697 .409 .425 085 .250 .135 .969 .779 .196 5 .044 .351 .648 .034 .475 .221 .321 6 .021 .375 .371 .892 .852 Each model run was conducted three times: once for environmental efficiency, once for the route only portion of environmental efficiency and once for the disposal only portion of environmental efficiency. Model Run 1 included all control variables. The geographic control was split into four dummy categories. The model included the West, Midwest, and South dummies and omitted the Northea st dummy. The models for environmental efficiency and route only environmental efficiency both tested as having explanatory value however the disposal

PAGE 231

! 211 only environmental efficiency model did not ANOVA significance for these three models were: .00 7 (.01 significance level) for environmental efficiency; .001 ( 001 significance level) for route only; and .117 (not significant) for disposal only. Model Run 2 replaced the household density control variable (that did not test as significant) with the one is o ne control variable. This alternate control variable tested as significant in all model runs it was included in. The models for environmental efficiency and route only environmental efficiency again both tested as having explanatory value, however the di sposal only environmental efficiency model again did not. ANOVA significance for these three models were: .002 (.01 significance level) for environmental efficiency; .000 (.001 significance level) for route only; and .083 (not significant) for disposal on ly. Model Run 3 tested the landfill yard waste ban controls while removing the state law control for container deposit law and geographic controls. Model Run 4 instead tested the geographic controls and removed the two state law controls of container de posit law and landfill yard waste ban Ultimately neither the yard waste ban control nor the geographic controls tested as significant and all were ultimately removed from the final model. However, all three models tested as having explanatory value. ANO VA significance for these three models were: .000 (.001 significance level) for environmental efficiency for Model Run 3 and Model Run 4;

PAGE 232

! 212 .00 0 (.001 significance level) for route only for Model Run 3 and Model Run 4 ; and 046 (.05 significance level) for M odel Run 3 and .045 (.05 significance level) for Model Run 4 for disposal only. Model Run 5 is an example utilizing a leaner set of control variables, in this specific run high level of service production, low breath of service production, level of automat ion, funding mechanism, and fuel were included. Ultimately only high level of service production and one is one tested as significant. ANOVA significance for these three models were: .000 (.01 significance level) for environmental efficiency; .000 (.001 significance level) for route only; and .044 (.05 significance level) for disposal only. Model Run 6 is the final run with the four significant variables. Both independents : monocentric and franchise zone test as significant and the two controls of high l evel of service production and one is one test as significant. ANOVA significance for these three models were: .000 (.01 significance level) for environmental efficiency; .000 (.001 significance level) for route only; and .021 (.05 significance level) for disposal only. The following more in depth analysis for Hypothesis 2 is based on the findings from Model Run 6 Table 6. 2 3 below pr esents the regression analysis results for environmental efficiency for the independent variable service production categor ies of monocentric and franchise zones with polycentric serving as the omitted category.

PAGE 233

! 213 Table 6. 2 3 : Regression Analysis for Environmental Efficiency Hypothesis Independent and Control Variable B (Unstan darized) 95% Confidence Interval Standard Error t p Dependent Variable: Environmental Efficiency (Constant) 1.722 1.401 to 2.043 .161 Monocentric (IV) 0.335 0.642 to 0.027 .155 2.162 .033 Franchise Zone (IV) 0.485 0.845 to 0.125 .181 2.673 .009 Level of Service (CV) 0.584 0.185 to 0.984 201 2.904 .005 One Is One (CV) 0.207 0.025 to 0.389 .092 2.258 .026 Note: Sum of squares for regression 3.220 (R 2 = .345; Adjusted R 2 = .301) F score = 7.896; model significance = .000 (.001 significance level) The multiple regression analysis was co nducted to evaluate how well service production (polycentric, franchise zones, and monocentric) predicted environmental efficiency. The linear relationship between service production and environmental efficiency was significant (F score = 7.896, p<.001) T he adjusted R 2 indicates this model predicts approximately 30% of the variation observed in this model. Two control variables were significant: high level of service production (p<.0 1 ) and city definition of single family household is indeed one single fa mily household, termed "One Is One" (p<.05). Holding high level of service production (HLSP) and "One Is One" constant, the average annual environmental footprint on a per household basis (environmental efficiency) under polycentric is 1.722 tons of CO 2e Again holding HLSP and "One Is One" constant, the average annual environmental efficiency under monocentric is 1.387 tons of CO 2e (0.335 less than polycentric) and the average annual environmental efficiency under franchise zones is 1.237 tons of CO 2e ( 0. 485 less than polycentric). A

PAGE 234

! 214 HLSP (most often twice a week trash collection instead of once a week trash collection) on average added 0.584 tons of CO 2e to the annual environmental efficiency across the service production categories and "One is One" on a verage added 0.207 tons of CO 2e to the annual environmental efficiency across the service production categories. Table 6. 2 4 below presents the zero order and partial correlations for the independent and control variables with the dependent variable of env ironmental efficiency. Table 6. 2 4 : Zero order and Partial Correlations for Environmental Efficiency Variable Correlation between IVs and CVs and Fiscal Efficiency Correlations between IVs and CVs and Fiscal Efficiency Controlling for all Other Variables M onocentric (IV) .07 .22 Franchise Zones (IV) .19 .27 High Level of Service (CV) .38 .29 "One is One" (CV) .37 .23 As predicted by Hypothesis 2 the mean value for monocentric is more environmentally efficient than polycentric, holding high level of service production and "One is One" constant. (Because these numbers represent carbon dioxide emission equivalencies a smaller number represents a higher efficiency.) However, the model does not find any statistical difference of significance between m onocentric and franchise zone service production in relation to environmental efficienc y. As monocentric was predicted to be most environmentally efficient by H ypothesis 2 this finding only lends partial support for the hypothesis. Regression diagnostics indicated linearity in the relationship between the independent variables and the dependent variable. Residuals and errors presented

PAGE 235

! 215 homogeneity and as normally distributed, thus the model and residuals analysis presented as homoscedastic and consequently negative for heteroskedasticity. Variables operated largely independent of each other. Diagnostics indicated very low collinearity between the variables and thus did not indicate problems of multicollinearity between the variables all VIFs were less than five across the different model runs and across the different independent and control variables. Likewise interaction effects between variables were low and did not present multicollinearity concerns. While the model did have three outliers the model was stronger with the outliers included in that the outliers were logically aligned along the line whether included or excluded from the model indicating that the outliers were not exercising excessive leverage / influence over the model. Given this finding and the small overall sample size it was decided the model was better with the outliers left within it than excluded This section now continues the environmental efficiency analysis, first completing the same regression runs but substituting route only en vironmental efficiency as the dependent variable and then again repeating the same regression runs but this time substituting disposal only environmental efficiency as the dependent variable

PAGE 236

! 216 Table 6. 2 5 : Regression Analysis for Environmental Efficiency H ypothesis Utilizing Route only Environmental Efficiency as the Dependent Variable Independent and Control Variable B (Unstan darized) 95% Confidence Interval Standard Error t p Dependent Variable: Route only Environmental Efficiency (Constant) 0.711 0.5 08 to 0.914 .102 Monocentric (IV) 0.098 0.627 to 0.238 .098 4.413 .000 Franchise Zone (IV) 0.115 0.629 to 0.173 .115 3.494 .001 High Level of Service (CV) 0.262 0.009 to 0.514 .127 2.054 .043 One Is One (CV) 0.114 0.001 to 0.230 .058 1.97 4 .050 Note: Sum of squares for regression 3.025 (R 2 = .324; Adjusted R 2 = .295) F score = 11.021; model significance = .000 (.001 significance level) The multiple regression analysis was conducted to evaluate how well service production (polycentric, franchise zones, and monocentric) predicted the route only portion of environmental efficiency. The linear relationship between service production and route only environmental efficiency was significant (F score = 11.021, p<.001). The adjusted R 2 indicate s this model predicts approximately 30% of the variation observed in this model. Two control variables were significant: high level of service production (p<.05) and city definition of single family household is indeed one single family household, termed "One Is One" (p<.05). Holding high level of service production (HLSP) and "One Is One" constant, the average annual route only environmental efficiency under polycentric is 0.711 tons of CO 2e Again holding HLSP and "One Is One" constant, the average annu al environmental efficiency under monocentric is 0.613 tons of CO 2e (0.098 less than polycentric) and the average annual route only environmental efficiency under franchise

PAGE 237

! 217 zones is 0.596 tons of CO 2e (0.115 less than polycentric). A HLSP (most often twic e a week trash collection instead of once a week trash collection) on average added 0.262 tons of CO 2e to the annual route only environmental efficiency across the service production categories and "One is One" on average added 0.114 tons of CO 2e to the an nual route only environmental efficiency across the service production categories. Table 6. 2 6 below presents the zero order and partial correlations for the independent and control variables with the dependent variable of route only environmental efficien cy. Table 6. 2 6 : Zero order and Partial Correlations for Route only Environmental Efficiency Variable Correlation between IVs and CVs and Fiscal Efficiency Correlations between IVs and CVs and Fiscal Efficiency Controlling for all Other Variables Monocen tric (IV) .28 .42 Franchise Zones (IV) .05 .34 High Level of Service (CV) .34 .21 "One is One" (CV) .34 .20 Although Hypothesis 2's prediction was in relation to total environmental efficiency, the mean value for monocentric's route only environme ntal efficien cy is indeed more efficien t than polycentric, holding high level of service production and "One is One" constant. (Because these numbers represent carbon dioxide emission equivalencies, a smaller number represents a higher efficiency.) Howev er, the model does not find any difference of statistical significance between monocentric and franchise zone service production in relation to route only environmental efficiency. Regression diagnostics indicated linearity in the relationship between the independent variables and the dependent variable. Residuals and errors presented

PAGE 238

! 218 homogeneity and as normally distributed, thus the model and residuals analysis presented as homoscedastic and consequently negative for heteroskedasticity. Variables operated largely independent of each other. Diagnostics indicated very low collinearity between the variables and thus did not indicate problems of multicollinearity between the variables all VIFs were less than five across the different model runs and across the different independent and control variables. Likewise interaction effects between variables were low and did not present multicollinearity concerns. While the model did have three outliers the model was stronger with the outliers included in that the out liers were logically aligned along the line whether included or excluded from the model indicating that the outliers were not exercising excessive leverage / influence over the model. Given this finding and the small overall sample size it was decided the model was better with the outliers left within it than excluded. Table 6. 2 7 : Regression Analysis for Environmental Efficiency Hypothesis Utilizing Disposal only Environmental Efficiency as the Dependent Variable Independent and Control Variable B (Unstan darized) 95% Confidence Interval Standard Error t p Dependent Variable: Disposal only Environmental Efficiency (Constant) 1.011 0.760 to 1.262 .126 Monocentric (IV) 0.098 0.143 to 0.338 .121 .807 .422 Franchise Zone (IV) 0.084 0.366 to 0.198 142 .590 .557 High Level of Service (CV) 0.323 0.010 to 0.635 .157 2.049 .043 One Is One (CV) 0.092 0.050 to 0.235 .072 1.288 .201 Note: Sum of squares for regression 1.282 (R 2 = .117; Adjusted R 2 = .079) F score = 3.058; model significance = .021 ( .05 significance level)

PAGE 239

! 219 The multiple regression analysis was conducted to evaluate how well service production (polycentric, franchise zones, and monocentric) predicted the disposal only portion of environmental efficiency. While the linear relationship between service production and disposal only environmental efficiency was significant (F score = 3.058, p<.05) neither the independent variables of monocentric or franchise zone nor the control variables of "One is One" tested as significant and the adjus ted R 2 indicates this model predicts only about 8% of the variation observed in this model Only the control variable of high level of service production tested as significant. Because of the marginal significance of this model further analysis of the di sposal only environmental efficiency model was not conducted. Environmental Efficiency Discussion Beyond Research Hypothesis While the regression analysis completed lends partial support for th is dissertation's research hypothesis that the monocentric serv ice production model will be the most environmentally efficient th at analysis wa s limited only to the trash collection portion of a city's environmental footprint. One common point of discussion in the solid waste literature is the relative environmental footprint of landfill disposal of municipal solid waste versus the related footprint for the same waste if it is recycled and/or composted. Among the 102 large N cities, a subset of 30 cities provided sufficient data to complete an analysis that look ed at the relative environmental footprint and relative environmental efficiency across these three disposal options. These 30 cities all offered a full breadth of collection services: trash collection, recyclables collection, and organics collection. Additio nally, these 30 cities provided comprehensive disposal tonnage data

PAGE 240

! 220 across the three collection services. The EPA WARM model was used to translate this tonnage data into a total greenhouse gas footprint and a CO 2e per household per year environmental effi ciency Table 6. 28 presents the statistical summary data for the tonnages from these same cities Table 6. 2 9 then presents the results of the greenhouse gas footprint analysis utilizing the EPA WARM model and the related statistical summary data The bas eline measure in this table utilizes the assumption that all of the waste is being disposed of in a landfill. The alternate waste management scenario in this table then presents the disposal environmental efficiency based on the "current mix" of disposal within the city which for each city includes a mix of landfill disposal, recycling, and organics bound for composting T able 6. 28 : Summary Data for Tonnage Per Household Data for 30 Cities: Landfill Disposal, Recycling, and Organics Bound for Composting L andfill Disposal (Tons Per Household) Recycling (Tons Per Household) Organics Composted (Tons Per Household) Highest Landfill Tons 1.52 Lowest Landfill Tons 0.39 Highest Recycling Tons 0.56 Lowest Recycling Tons 0.06 Highest Organics Tons .8 8 Lowest Organics Tons .09 Median 1.18 0.18 0.25 Mean 1.08 0.22 0.31 Standard Deviation 0.334 0.133 0.224

PAGE 241

! 221 T able 6. 2 9 : Disposal Environmental Efficiency of All Landfill Disposal Versus Landfill Disposal, Recycling, and Organics Composting Base line Analysis: All Landfill Disposal (CO 2e Per Household Per Year) Alternate Waste Management Scenario: Current City's Mix of Landfill Disposal, Recycling, and Organics Bound for Composting (CO 2e Per Household Per Year) Largest Footprint: 1.463 Largest F ootprint: 1.192 Smallest Footprint: 0.252 Smallest Footprint: 0.731 Median: 0.98 Median: 0.45 Mean: 1.08 Mean: 0.51 Standard Deviation: 0.349 Standard Deviation: 0.586 Across the 30 cities the average disposal environmental efficiency was 1.08 tons CO 2e per household per year if all waste collected went to the landfill for disposal. By contrast, across the 30 cities the average disposal environmental efficiency was 0.51 tons CO 2e per household per year utilizing each of the city's current mix of tonnage bound for the landfill, tonnage bound for recycling, and tonnage bound for composting. The finding from this analysis is that combining recycling and composting to reduce the waste bound for landfill disposal does indeed reduce the environ mental footprint and increase the environmental efficiency for a city in terms of disposal based environmental efficiency measures. Equity Hypothesis Analysis Equity Hypothesis Discussion This dissertation's equity hypothesis measured indicators of equity and / or inequity in service delivery in three ways : (1) cost of service production on a within city

PAGE 242

! 222 basis (measured on a cost per household per year basis) ; (2) level of service on a within city basis (measured in terms of frequency of trash collection a nd frequency of periodic bulky item collection); and (3) breadth of service on a within city basis (measured in terms of delivering trash and / or recyclables collection and / or organics collection). In addition to this hypothesis based equity analysis, this section will complete a cross city multiple city equity analysis from the dataset, also based on the three equity measures noted above. Across all three measures of equity, and across all three service delivery schemes, there was a universal finding of equity in service production on a with in city basis. P olycentric, franchise zones and monocentric collection schemes were equally equitable in the delivery of services on a within city basis. No cities reported variations on cost of service on a with in city basis, with the exception of pay as you throw, and in all cases of pay as you throw the same variable rate options were offered to all customers. Likewise level of service had a collective finding of equity across the cities responding to the surv ey Eight survey cities reported inequities in breadth of service 49 on a within city basis, however follow up with each of these cities indicated that the inequity in breadth of service was tied to piloting of new collection programs (either for recyclable s or for organics). Prior to data collection the researcher for this dissertation decided that programs that were designated as pilot in nature would not count as an inequity in service delivery. Thus breath of service, across the surveyed cities, also i ndicated equitable service delivery. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! )( An indicator of inequity related to breadth of service would be part of a city particular neighborhoods, or particular households receiving recyclabl es collection services and/or organics collection service while other areas neighborhoods, or households did not have access to this same service within the same city.

PAGE 243

! 223 Equity Discussion Beyond Research Hypothesis In the end the equity measure as defined in this dissertation was too blunt a measure to detect nuances on the equity front. The following is an across multiple cities anal ysis still utilizing the three indicators of inequity from the original equity hypothesis: cost of service, level of service, and breadth of service. This analysis compares across the three service production schemes of polycentric service delivery, franc hise zone service delivery, and monocentric service delivery. Table 6 30 provides a visual of this comparison. Table 6 30 : Equity Across Multiple Cities and Across Service Production Models Service Production Scheme Cost of Service ($ per Household per Ye ar) Level of Service Breadth of Service Polycentric Service Mean Cost of Service: $308.38 Median Cost of Service: $274.83 2 cities had high level of service 8 cities had standard level of service 0 cities had trash only 10 cities had trash and recycla bles 0 cities had trash, recyclables, and organics Franchise Zones Mean Cost of Service: $173.54 Median Cost of Service: $172.00 1 city had high level of service 16 cities had standard level of service 2 cities had trash only 3 cities had trash and r ecyclables 12 cities had trash, recyclables, and organics Monocentric Service Mean Cost of Service: $166.75 Median Cost of Service: $165.00 2 cities had high level of service 69 cities had standard level of service 4 cities had low level of service 8 cities had trash only 11 cities had trash and recyclables 56 cities had trash, recyclables, and organics

PAGE 244

! 224 Data presented in Table 6. 30 support that the a nnual cost of service, on average, is least expensive under the monocentric model and most expensiv e under the polycentric model. In terms of level of service with 91.2% of all cities surveyed (93 cities total) delivering a standard level of trash collection service, this service is largely delivered in an equitable manner across the three service pro duction models. In terms of breadth of service, it is of note that n one of the polycentric cities surveyed delivered organics collection as a service. By contrast 70.5% of franchise zone cities and 74.7% of monocentric cities delivered organics collectio n services Thus when analyzed on an across multiple city basis the equity findings do inde ed become more noteworthy L evel of service is still largely delivered in an equitable manner across all three of the service production schemes However, in terms of both cost of service and breadth of service the polycentric scheme presents clear inequities when compared against both the franchise zone and monocentric service production schemes. Cost of service under a polycentr ic arrangement is more expensive, o n average, than cost of service under both franchise zone and monocentric arrangements. Organics collection service was not available in any polycentric cit ies s urveyed B y contrast, organics collection service was available in 70%+ of the franchise zone and monocentric cities responding to the survey. Figure 6. 3 below is a visual representation of this finding across service production models.

PAGE 245

! 225 Figure 6. 3 : Large N Equity Compared on an Across Multiple Cities Basis ! ! ! Polycentric Franchise Zones Monocentric Overlap No Overlap Consolidated Fragmented Least equitable in terms of cost of service and breadth of service. Equitable in terms of level of service. Most equitable in terms of cost of service and breadth of service. Equitable in terms of level of service. Most equitable in terms of cost of service and breadth of service. Equitable in terms of level of service.

PAGE 246

! 226 Conclusions : Zoomi ng Out from the Hypothesis Level The hypothesis level of analysis primarily utilized throughout this chapter, focuses on variations in the depend variable s between the three independent variable service production models of monocentric, franchise zones, and polycentric (for Hypothesis 1.1 and Hypothesis 2) or focuses on variations in the depend variables between the four independent variable service production models of monocentric private, monocentric public, franchise zones, and polycentric (for Hypothe sis 1.2). T he analysis of the Large N dataset lent support for fiscal efficiency hypothesis 1.1, partial support for fiscal efficiency H ypothesis 1.2, partial support for the environmental efficiency hypothesis, and partial support for the equity hypothes is. When one zooms out from this hypothesis based level of analysis however the Large N dataset clearly highlights a demarcation between cities utilizing the polycentric service production model and cities utilizing the franchise zone and the monocentric service production models. Among the cities surveyed, in terms of fiscal efficiency and environmental efficiency, the franchise zone and the monocentric cities performed at statistically significant levels of higher efficiency than the polycentric cities Th e large N findings in relation to equity when compared on an across multiple cities bases, also indicated superior service production by the franchise zone and monocentric service production models for two of equity's three measures. In terms of leve l of service, all three service production models performed largely equally and thus equitably. However, in terms of breadth of service and in terms of cost of service, the franchise zone and the monocentric cities outperformed the polycentric cities surv eyed and were more equitable in their service delivery than the polycentric cities.

PAGE 247

! 227 CHAPTER V I I CONCLUSIONS Key Findings Both the case study research and the large N research lend support to the dissertation's fiscal efficiency, environmental efficiency and equity research hypotheses Using the performance measures defined within this dissertation and the dataset gathered for this dissertation, polycentric service production delivers solid waste collection services at a lower level of fiscal and enviro nmental efficiency then the monocentric and the franchise zone service production models. While the mean cost of service is lowest for monocentric service production (when measured as a single category) and lowest for monocent ric public service productio n (w hen this category is split into monocentric private and monocentric public), the monocentric (including both monocentric public and monocentric private) and franchise zone models fiscal efficiency means are within statistical confidence intervals to a ll be considered uniform in terms of fiscal efficiency. Likewise, the mean environmental efficiency is lowest for polycentric and highest for franchise zone service production however again both the franchise zone and the monocentric models environmental e fficiency means are within statistical confidence intervals for both to be considered uniform in terms of environmental efficiency. In relation to equity, the polycentric model is found to present inequities for two of the three equity measures when compar ed against the monocentric and the franchise zone models. For level of service all three service production models deliver equitable service. However, for breadth of service and cost of service, the polycentric model is less

PAGE 248

! 228 equitable in its service deli very than the monocentric and the franchise zone service production model s The following section presents this information in more detail and specifically in relation to the dissertation's research hypotheses. Following this section the results are again presented, in this case the results are presented from a practical significance perspective. Key Findings in Relation to Research Hypotheses Hypothesis 1.1 predict ed that the monocentric service production model would be the most fiscally efficient. Case study cities lent support to this hypothesis. For large N cities, t he mean price of service on an annual basis among monocentric cities responding to the survey was $163 (SD $58; CI: $149 to $176 ) 50 (All units are dollars per household per year.) By co ntrast, the franchise zone model was $166 (SD: $79; CI: $125 to $206 ) and the polycentric model was $308 (SD: $138; CI: $209 to $407 ) These data indicate that among the cities responding to the survey, a citizen living in a city with monoc entric service production or with franchi se zone service production for trash collection will pay an annual rate for service that is much less expensive than a citizen living in a city with polycentric trash collection. These findings were found to be statistically sign ificant across multiple model runs. This finding lends further validation to solid waste studies conducted by Savas (1977, 1987) with similar findings. It must, however, be noted that the differences in means between the monocentric and the franchise zon e models did not represent a statistically significance difference from one another in terms of fiscal efficiency. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 50 SD = Standard deviation; CI = 95% Confidence interval.

PAGE 249

! 229 While case study cities len d support 51 for Hypothesis 1.2 the finding among large N cities lend only partial support. Hypothesis 1.2 predict ed that the monocentric service provision with private sector production (monocentric private) would be the most fiscally efficient. With a mean annual cost of $178 (SD: $55; CI: $161 to $195) monocentric private was indeed substantially more fiscally ef ficient than the polycentric service production model with a mean annual cost of $308 (SD: $138; CI: $209 to $407) However, among the sample both the franchise zone arrangement and the monocentric service provision with public sector production (monocent ric public) had means lower than monocentric private. Franchise zones had a mean annual cost of $166 (SD: $79; CI: $ 125 to $206) and monocentric public had a mean annual cost of $142 ( SD: $56; CI: $122 to $193) Differences in means between polycentric a nd the other three models all tested at statistically significant levels in relation to fiscal efficiency. However, the models did not find any difference of statistical significance between the monocentric private, monocentric public, and franchise zone service production models in relation to fiscal efficiency. This finding however, is potentially of note as it differs from solid waste studies conducted by Savas (1977, 1987). Sava's earlier research had lent support for Hypothesis 1.2 as his studies f ound the monocentric private arrangement to be the most fiscally efficient. For the fiscal efficiency hypotheses two control variables were statistically significant: high level of service production (most commonly twice a week trash collection) and low br eadth of service production (trash only collection no recyclables !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! *" Knoxville, the one monocentr ic private case study city, had an average annual cost for trash collection of $124. The other three monocentric case study cities were monocentric public arrangements. Austin's annual cost was $142; Denver's annual cost was $145; and Fresno's annual cos t was $156.

PAGE 250

! 230 and no organics collection). This is of note because it lends further support to the finding that polycentric service production is least fiscally efficient in that by holding level and bre adth of service constant in the regression s there is not an indication that polycentric cities are receiv ing more services for their higher cost. Both case study cities and large N cities lend partial support for Hypothesis 2. This environmental efficie ncy hypothesis predicted that the monocentric service production model would be the most environmentally efficient. With a mean annual environmental efficiency of 1.49 (SD: 0.42; CI: 1.40 to 2.35) tons CO 2e per household per year the monocentric cities s urveyed were indeed more environmentally efficient than the polycentric cities surveyed. By contrast, the polycentric cities had a mean annual environmental efficiency of 2.04 (SD: 0.53; CI: 1.63 to 2.45) tons CO 2e per household per year. However, among cities responding to the survey it was the franchise zone cities with the smallest average annual environmental efficiency. Franchise zone cities had a mean annual environmental efficiency of 1.32 (SD: 0.50; CI: 1.07 to 1.58) tons CO 2e per household per y ear. Again d ifferences in means between polycentric and the other two models each t ested at statistically significant levels in relation to environmental efficiency. However, the models did not find any difference of statistical significance between the monocentric and the franchise zone service production models in relation to environmental efficiency. This dissertations third hypothesis deal t with equity and proved to be the most challenging to measure. The nuances gleaned from this dissertation's case study research were most helpful in teasing out findings of substance for the equity hypothesis. Hypothesis 3 predicted the monocentric model would be the most equitable. Equity was

PAGE 251

! 231 measured in three ways: variations in cost of service, variations in le vel of service, and variations in breadth of service. The initial hypothesis considered equity solely on a within city basis. However, the final analysis completed for this dissertation considered equity both on a within city basis and on an across multi ple cities and across multiple service delivery models basis. Looking at equity solely on the within city basis, both case studies and large N surveys presented a finding that there was equity in cost of service, level of service, and breadth of service de livery across all case study cities, across all cities surveyed and across all three service production models. Thus, on a within city basis, the data did not lend support for Hypothesis 3. Instead, the finding was that service on a within city basis wa s delivered on an equitable basis in terms of cost of service, level of service, and breadth of service. Further evaluation of the case study cities, however, offered a more nuanced picture to this finding. While all eight case study cities delivered a br eadth of services and a level of service on an equitable basis, four cities noted inequities in the adoption of the services available. Among these four cities, there was an observation that different neighborhoods within their city often had different le vels of adoption of recyclables collection and organics collection. In terms of breadth of service, a further observation was noted that was then applied to the large N cities. Among the eight case study cities, all four of the monocentric cities offere d organics collection. By contrast, the two franchise zone cities offered only very limited organics collection programs (for one it was fall leaf collection

PAGE 252

! 232 only and for the other it was only in conjunction with bulky item collection events). Furthermor e, neither of the two polycentric cities offered any organics collection at all. This breadth of service observation was then applied to the large N surveys with the finding that none of the ten polycentric cities had any form of organics collection. By c ontrast, 12 of the 17 franchise zone cities offered some form of organics collection and 56 of the 75 monocentric cities offered some form of organics collection. Based on this finding, all three of the equity measures were re evaluated from an across mult iple cities perspective rather than from a within city perspective The new finding was that among surveyed cities the level of service production is indeed delivered on an equitable basis by all three service delivery models. However, inequities arise b oth in terms of cost of service and in terms of breadth of service. In both of these cases, the monocentric model is more equitable in service delivery than the polycentric model among surveyed cities. Thus this revised method of looking at equity lends partial support to Hypothesis 3. Practical Significance The practical significance for specialists in the municipal solid waste management field is that this dissertation finds that among the cities surveyed, there is a clear demarcation between cities ut ilizing the polycentric service production model and cities utilizing the franchise zone and the monocentric service production models. This dissertation lends support to the assertion that among surveyed cities, monocentric service production (including both monocentric private and monocentric public ) and franchise zone service production are substantially less expensive on an annual per household basis than the equivalent service offered to citizens in cities with a polycentric

PAGE 253

! 233 service production arrange ment. Likewise, monocentric service production and franchise zone service production have substantially smaller environmental footprints than the equivalent service offered to citizens in cities with a polycentric service production arrangement. Addition ally, cities providing a high level of service (such as twice a week trash collection) that are looking to reduce costs and improve their fiscal efficiency should look to a standard level of service (specifically once a week trash collection). Further, t he breadth of services for solid waste collection offered in cities served by monocentric and fra nchise zone service production are more likely to include organics collection amongst their suite of services than in cities served by poly centric service prod uction. Key Findings from Case Study Thick Description The "thick description" research specific to the case study research offers four key findings: (1) Across all service production models programs are actively making changes to increase efficiency, (2 ) Sometimes programs must make trade offs between maximizing efficiency and maximizing equity, (3) Some monocentric programs are actively making changes to increase equity, and (4) Across a l l service production models programs are actively making changes t o increase citizen satisfaction. All eight case study cities had undertaken some kind of program change specifically to increase efficiency. Austin, Denver, Indianapolis, and Fort Collins are all in the process of transitioning to automated collection pr ograms to increase fiscal efficiency. Austin, Fresno, and Phoenix are all in the process of transitioning their collection fleets to include natural gas powered vehicles to increase both fiscal and

PAGE 254

! 234 environmental efficiency. All eight of the case study ci ties have transitioned to automated single stream recycling for fiscal efficiency reasons. The cities of Denver and Austin present an interesting case pair because faced with the same decision one city opted for an option that maximized efficiency and one city opted for an option that maximized equity. Both cities had noted differential adoption of their recyclables collection and / or their organics collection programs across neighborhoods within their cities. In both cases the recyclables collection se rvice was provided free of charge to residents. To maximize fiscal efficiency, Denver makes the default for their recyclables and organics collection programs "opt out". The logic is that if a citizen wants these services they will take the time to call and request the service. To maximize equity, Austin makes the default for their recyclables and organics collection programs "opt in". The logic is that automatically providing bins for the service will maximize usage and maximize equity and if a citizen truly does not want the service they may still call to have the service canceled. As noted in the last paragraph, both Austin and Denver have noted problems with differential adoption of their recyclables collection and / or their organics collection. L ikewise, Fresno ha s also noted this problem of differential adoption. To address this problem all three of these monocentric service production cities have targeted education programs specifically to low adoption neighborhoods and have geared these educat ion campaigns to, in particular, encourage recycling in these neighborhoods. Five of the eight case study cities explicitly mentioned efforts to improved citizen satisfaction:

PAGE 255

! 235 Austin noted that the residents of their city "demand" progressive integrated solid waste management service production; Via a city commissioned survey, Denver found 78% of residents rated their refuse collection program "good to excellent"; Fort Collins explored a transition from polycentric to franchise zone service production bu t ultimately opted to remain with polycentric service production when citizens expressed a high level of satisfaction with the current system ; Indianapolis transitioned their recyclables collection program from a manual collection, dual stream program with 18 gallon bins to a fully automated, single stream program with 90 gallon bins largely driven by citizen dissatisfaction with the older method; and Knoxville, as the one case study city sticking with the utilization of a manual collection program for tras h, explicitly cited citizen satisfaction with the current system as a key reason for remaining with their "tried and true" program. Contributions to the Literature Expanded Theoretical Paradigm: New Service Production Arrangements This dissertation has loo k ed at the organization of city level service delivery primarily via the lens of overlap and fragmentation in service production, and how fragmentation and / or overlap influence the efficiency and equity of service delivery. There are two literatures tha t frame city level services in terms of this fragmentation versus overlap viewpoint: (1) t he "local public economy" literature (McGinnis, 2011; V.

PAGE 256

! 236 Ostrom, 1972; V. Ostrom, Tiebout, & Warren 1961 ); and (2) the "new regionalism" literature (Lyon & Lowery, 19 89a 1989b ; Rusk, 1993; Pierce, Johnson, & Hall, 1993; and Downs, 1994 ). Both literatures speak to the question of overlap and fragmentation. However, a contention of this dissertation is that both literatures tie these two independent considerations tog ether. Thus, to date, research is primarily along a spectrum from non overlapping and consolidated to the other extreme of overlapping and fragmented By decoupling fragmentation from overlap this dissertation is able to presents an expanded picture that could be applied to future research. While more fully defined in the "local public economy" literature, both literatures also explore this overlap and fragmentation question from the perspectives of both service provision and service production ( Musgrav e 1959 ; V. Ostrom, 2008 ; Oakerson, 1999) It is important to note again that this dissertation solely explores discusses, and researches this overlap and fragmentation question from the service production side of the equation. Future research utilizing the decoupled version of the "four quadrants" 2x2 for overlap versus fragmentation and researching the service provision side of the equation is a logical follow up to this production specific project. Both the local public economy literature and this dis sertation define a service that is non overlapping and consolidated as a m onocentric arrangement (McGinnis, 2011). Both the local public economy literature and this dissertation define a service that is both overlapping and fragmented as a polycentric arra ngement (McGinnis, 2011).

PAGE 257

! 237 As noted throughout, w hile this dissertation utilizes th e polycentric monocentric construct, it also expands upon it. This dissertation operates from the assumption that overlap and fragmentation can and do operate independent of one another. When this assumption is explored, the two point spectrum of monocentricity to polycentricity is transformed into a two by two arrangement with two additional fields to be considered. Figure 7.1 : Independent Overlap and Fragmentation Two by Two ! ! ! Polycentric Franchise Zones Functional Consolidation Monocentric Overlap No Overlap Consolidated Fragmented ! !

PAGE 258

! 238 Figure 7.1 above is a visual illustration of this two by two. In decoupling fragmentation f rom overlap this dissertation conceptually added two additional categories to the already researched monocentric and polycentric categori es. A third arrangement that is non overlapping yet fragmented is defined as the franchise zone arrangement. In a franchise zone scheme, citywide there will be multiple service producers and multiple non overlapping geographic service delivery zones, howe ver within any one given zone there will only be one single service producer. A fourth arrangement that is overlapping yet consolidated is defined as the functional consolidation arrangement. In a functional consolidation arrangement, multiple government units, each having jurisdiction for a particular service provision and / or service production, will all collectively and voluntarily consolidate under a single service delivery mechanism for a specific service. Thus while this quadrant has the potential of overlap the mutually agreed upon consolidation has precedence in this quadrant. Also, in relation to the overlap, it is important to note these same government entities maintain their individual autonomy in other service provision and / or production areas and, likewise, their defined geographic boundaries will remain intact. This decoupling of fragmentation from overlap and identification of two additional service production categories represents a contribution to the local public economy and new re gionalism literatures. This dissertation, while opening this expanded theoretical perspective, however only explores a limited component of this new theory development. First, this dissertation only explores the service production side (not the service p rovision side). Second, this dissertation is from only one service delivery perspective, that of trash collection, leaving many additional service delivery areas as yet

PAGE 259

! 239 unexplored. Third, because of the spe cifics of trash collection, only the polycentric monocentric, and franchise zone service delivery schemes are applicable, leaving the functional consolidation service delivery scheme also as yet unexplored. New Service Production Performance Measures As noted above, t o apply this new theoretical paradi gm t his dissertatio n researche d the fragmentation versus overlap debate within the context of city level solid waste (tr ash) service delivery to single family households and utilize d three performance measure s to illustrate differences in service delivery performance. The dissertation 's overarching research question wa s : How does the organization of solid waste collection at the city level relate to: (1) fiscal efficiency, (2) environmental efficiency, and (3) equity of service production? Fiscal efficienc y wa s defined as providing the most service for the least cost (E. Ostrom, 2011; Oakerson, 1999; V. Ostrom, Bish, E. Ostrom, 1988; Savas, 1977, 1987, 2005) The literature generally simply refers to this performance measure as "efficiency" however this di ssertation applied a more precise definition as it also developed a new measure of efficiency This dissertation offers a further contribution to the literature by envisioning and executing a new method of measuring efficiency in environmental rather tha n monetary terms. The new measure of Environmental efficiency wa s defined as providing the most service with the smallest environmental footprint (Hillman & Ramaswami, 20 10 ; Chavez & Ramaswami, 2011). Equity wa s defined as fairness, impartiality, or equal ity in service (Oakerson, 1999) In this solid waste collection co ntext this translate d to how evenly a service is

PAGE 260

! 240 provided across multiple service areas in terms of cost of service, level of service, and breadth of service The original equity hypothesi s tied to this research question looked at this question on a within city basis. However, the final analysis within this dissertation also looked at this question from an across multiple cit ies and across multiple service production arrangements basis as well. Equity while often discussed within the literature is much less frequently measured. ( Two notable exceptions that explicitly measure equity are DeHoog, Lowery, & Lyons 1991 and Lowery, Lyons, & DeHoog, 1995.) This dissertation o ffers a further con tribution to the literature by envisioning and executing a new method for measuring and evaluating equity New Insights to Old Questions Lastly, w hile both "new regionalism scholars and "local public economy" scholars offer quantitative analyses supp ortive of their arguments, a closer look at these studies illuminates an interesting insight. Oakerson (1999) found that studies favorable to consolidated service arrangements tended to be capital intensive service deliveries, such as water and wastewater utilities. Oakerson (1999) likewise found that studies favorable to polycentric service arrangements tended to be people intensive service deliveries, such as policing and education. Th is dissertation offered an excellent study frame for furthering prev ious research as solid waste collection exhibits aspects of both capital intensive and people intensive service delivery. As noted above, o ne limitation to this dissertation's new application of decoupled fragmentation overlap continuum was that by utiliz ing solid waste collection as the research frame the "functional consolidation" quadrant was not applicable in the solid

PAGE 261

! 241 waste context. Future research specifically explor ing this quadrant and utilizing a different service provision and / or service produ ction (perhaps regional mass transit districts) would be of benefit to the literature. Figure 7.2 below is an illustration of the 2x2 specifically in the solid waste collection context Figure 7 2 : Solid Waste Collection Specific Two by Two Example ! ! ! Polycentric Franchise Zones Monocentric Overlap No Overlap Consolidated Fragmented Private Sector Service Production Example: Colorado Springs, CO Public and Private Sector Service Production Example: Indianapolis, IN "Exclusive Franchise" Public and Private Sector Service Production Example: Denver, CO

PAGE 262

! 242 The original research completed for this dissertation ultimately included eight interview driven case studies and 102 internet survey driven large N United States cities. All of these cities fell within the three quadrants of monocentric, polycentric, or franchise zone service delivery. Theoretical Limitations While this dissertation certainly offers contributions to the literature, particularly in relation to the revised polycentric monocentric continuum, the study certainly does have theoretical limitations. Because this dissertation is focused on the service production rather than service provision some of the nuances of the provision side are lost in the newl y refined definitions of the de coupled overlap and fragmentation 2x2 diagra m. For example, McGinnis and E. Ostrom (2012, p.15) note, "a polycentric system . includes crosscutting jurisdictions . community based organizations play critical supporting roles in a polycentric system of governance". In a similar vein, while this dissertation has identified from a theoretical point of view the "functional c onsolidation" category and identified some regional governance examples, such as mass transit districts, as fitting within this category, this is obviously the limits of res earch into this quadrant for this dissertation. However, this is certainly an area ripe for further research. As noted by McGinnis and E. Ostrom (2012, p.17), "Special districts and other hard to categorize jurisdictions are critical components of federa lism, even though they remain a controversial subject among scholars of federalism". A further theoretical limitation for this dissertation's research design is the factors of fragmentation versus overlap are explored solely from a "yes" or "no" perspecti ve. That is to say, in determining polycentricity, franchise zones, and monocentricity, the

PAGE 263

! 243 measurement for these is simply whether there is or is not fragmentation or whether there is or is not consolidation and likewise whether there is overlap or there is not overlap. In reviewing the literature, in particular related to discussions of service provision rather than service production, there are often discussions of degrees of polycentricity and / or degrees of monocentricity (V. Ostrom, 1972; Oakerson, 1999; McGinnis & E. Ostrom, 2012) This nuance of "degrees" is lost within this dissertation as from the service production side the analysis follow ed more of a "yes" or "no" rather than a "degrees" approach. Research Limitations This section will di scuss four key limitations to this dissertations research. The first is a set of limitations tied to the final size of the N from survey responses. The second two limitations are tied to the new performance measures of environmental efficiency and equity 52 that were utilized within this dissertation both of these performance measures were untested prior to this dissertation and both could now be refined and improved based on observations related to their application within this dissertation. The final lim itation is the lack of inclusion of effectiveness as a performance measure within this dissertation. One key limitation for this dissertation was the N of 102 cities for the statistical study. This N is small enough that it is reasonable to question t he transferability and generalizability of the results from the study particularly for polycentric cities where the N is only ten cities. The size of this N could also raise concerns of sample bias. Expanding this N via either re surveying the current sample population or surveying a !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! *# While equity has been measured in other research studies the method by which equity was measured within this dissertation was new and specific to this research design.

PAGE 264

! 244 larger sample population of United States cities, would greatly benefit this study. Likewise, with a large N additional techniques to minimize concerns of sample bias could be utilized, such as propensity score matching, a s an example. On a related note, a deliberate delimitation imposed on this dissertation's research design was the use of only central core cities. One logical expansion of this research would be to remove this restriction and look at either all cities in the United States with populations greater than 50,000 or simply all cities in the United States. Either of these two options would help to expand the overall size of the N for this project, increasing generalizability and transferability of the results. Anecdotal evidence also suggests that cities that are suburb rather than central core cities are most likely to adopt the polycentric rather than the monocentric service production model. Thus this expanded sample population would likely have the added b enefit of expanding the N for cities utilizing polycentric service production. A second key limitation of the large N portion of this dissertation is that the data are a "one year" snapshot rather than findings over a series of multiple years. This concer n was partially addressed via the case study cities research. This research offer ed a multi year look at programs However, with that said, repeated surveys over multiple years would, over time, develop a more complete and dynamic picture o f the solid wa ste management programs among survey response cities. This over time element would also allow for the research of cities that transition from one service production scheme to another over time. While fiscal efficiency is a well established performance mea sure in the literature, this dissertation was a first attempt at adding environmental efficiency as a new

PAGE 265

! 245 performance measure. Refinement of this new measure is certainly appropriate after this first attempt. In particular, the disposal only portion of t he measure had difficulty testing at a statistically significant level. Special attention should likely be paid to this part of the measure. The environmental efficiency measure is also very specific to this dissertation's solid waste management context. A more generalizable version of this measure that can be utilized for other city level service deliveries would certainly be of both practical and academic benefit. Eq uity is often discussed in the literature More rarely is equity actually quantified a nd measured in the literature. Thus this dissertation wa s a further attempt in the nascent literature that works to quantify and measure equity at the city level. Given the findings of this dissertation that with in city service delivery was largely equit able it will likely be most fruitful to continue to pursue the across multiple cit ies and across multiple service production schemes method of measuring equity and inequities in service delivery. Lastly, a third performance measure commonly measured along side efficiency and equity is the performance measure of effectiveness. Thus not including effectiveness within this dissertation creates a deficit in the findings. Effectiveness is generally defined as the successful delivery of a service. Basically a n effective program would mean the production of a service meets the provision goals for that service. Cost is not a factor for effectiveness, it is simply a measure of was the job well done or not (Oakerson, 1999; Dietz, Dolsak, E. Ostrom & Stern, 2002). With the existing dataset or with a future expanded dataset, an additional hypothesis specifically designed to measure pro gram effectiveness could be developed and added as a future research effort.

PAGE 266

! 246 Future Research This section will first discuss future research that was directly pointed to within the research limitations section discussed above and will then turn to some additional potential future research options for both this specific dataset and this overall line of research. An expanded N would be a top priority in terms of future research. While this dissertation's finding s are promising, an expanded N would greatly enhance them particularly in terms of generalizability Thus a top future research priority would be additional surveying designed t o expand the N of the study. It is the researcher's opinion that this should include eliminating the deliberate delimitation of only central core cities yet maintaining the deliberate delimitation of only looking at cities with population greater than 50, 000 Future research should include refinement of all three of the performance measures (fiscal efficiency, environmental efficiency, and equity) and should include the addition of an appropriate effectiveness performance measure One approach to measur e effectiveness is pointed to by Folz (1996, 1999, 2004) where he utilized objective measures of quality to assess program effectiveness across multiple cities. Another approach to measuring effectiveness is pointed to by Ramaswami et al. (2008, 2011, 201 2) where city level goals are quantified and then each city's completion, partial completion, or non completion of the goals are assessed. Another deliberate delimitation to this dissertation was to restrict th is dissertation's resear ch to the United State s. W hile many countries deliver trash collection services in very different ways than the United States, a research design with

PAGE 267

! 247 minimal alterations could be applied in a Canadian or an Australian context. Other countries would require increasingly altere d research designs. As previously noted, w hile on a theory level th is dissertation has identified the functional consolidation service delivery scheme, the trash specific context of this dissertation's research precluded the inclusion of this fragmentation overlap quadrant within the research design. Just a s this dissertation h as successfully added franchise zones to the previously established categories of monocentric and polycentric service production a future research effort in a different frame of ser vice delivery should venture to formally add via quantifiable research the functional consolidation service production model Additionally, this dissertation is specifically on the service production side of municipal services. Research on the service pr ovision side of municipal services utilizing the decoupled fragmentation overlap frame of research would likewise be of great benefit to the literature. Lastly, in the broadest terms, what would this researcher have done differently if he were to start ove r ? First, have the environmental efficiency measure more fully developed earlier in the research process. It was developed on an iterative process during the research project and could still benefit from further refinement. Earlier development would lik ely have benefited this iterative process. Second, have a shorter and more focused survey instrument. Ultimately not all questions from the survey were utilized in the results. It is likely the length of the original survey instrument led directly to so me cities not responding. A shorter survey and more focused survey could still answer the questions within this dissertation's research design, could still answer a question related to effectiveness, and would likely improve response rate by being shorter Another

PAGE 268

! 248 observation in relation to the survey is an acknowledgement of how important the cost and budget related reporting portion of the survey is to this research frame. While other areas of the survey could likely be shortened the survey would likel y merit an expansion of questions in relation to costs and budgets to provide additional crosschecks and assurances of the accuracy of cost related data. Finally throughout th e process of writing th is dissertation informal discussions about the disserta tion have often led to a hint of what should likely be a fourth performance measure: responsiveness to citizen s This fourth measure is discussed in the literature but is not as cohesively articulated as "the 3Es" of effectiveness, efficiency, and equity. Salamon (2002) describes it, in part, when he defines "legitimacy and political feasibility" as an evaluation tool. Likewise, Denhardt and Denhardt (2011) describe it in part when they discuss the inherent transparency of a "new public service" approach This responsiveness to citizens measure, however, is not new. V. Ostrom (1972) in discussing polycentricity notes the importance of voice in articulating demands, being responsive to citizen requests, and citizen self determination. "The criterion of self determination implies that the government of a public enterprise will be controlled by the decisions of its constituents" (V. Ostrom, 1972, p.72) This quote harkens the author of this dissertation back to the case study city of Fort Collins. The ci ty had completed extensive research determining that from both a fiscal efficiency and an environmental efficiency point of view a transition from a polycentric service arrangement to a franchise zone arrangement would be merited. However, the city then l istened to their citizens, and their citizens clearly said they liked and wanted to keep the polycentric service production system that was already in place. And thus, in the case of Fort

PAGE 269

! 249 Collins, the as yet not fully defined performance measure of respon siveness to citizens trumped both fiscal efficiency and environmental efficiency. Conclusions This dissertation finds that among the cities surveyed, there is a clear demarcation between cities utilizing the polycentric service production model and cit ies utilizing the franchise zone and the monocentric service production models. Among the cities surveyed, in terms of fiscal efficiency and environmental efficiency, the franchise zone and the monocentric cities performed at statistically significant lev els of higher efficiency than the polycentric cities. This dissertation's findings in relation to equity also indicated superior service production by the franchise zone and monocentric service production models for two of equity's three measures. In ter ms of level of service, all three service production models performed largely equally and thus equitably However, in terms of breadth of service and in terms of cost of service, the franchise zone and the monocentric cities outperformed the polycentric c ities surveyed and were more equitable in their service delivery than the polycentric cities Figure 7. 3 below visually presents a summar y of these findings across all three of the service production models.

PAGE 270

! 250 Table 7. 3 : Fiscal Efficiency, Environmental Efficiency, and Equity Findings Across All Three of the Service Production Models The practical significance of these findings are that cities looking to improve the efficiency and equity of the trash collection service delivery, including recyclables collection and organics collection among their suite of services, would most likely best be ! ! ! Polycentric Franchise Zones Monocentric Overlap No Overlap Consolidated Fragmented Number Cities: 10 Fiscal Efficiency: $182 $583 / Household / Year (Mean: $308) Environmental Efficiency: 0.89 3.12 CO 2e / H ousehold / Y ea r (Mean: 2.04) Equity: Least equitable in terms of cost of service and breadth of service. Number Citie s: 1 7 Fiscal Efficiency: $ 96 $ 314 / Household / Year (Mean: $ 166 ) Environmental Efficiency: 0.20 2.55 CO 2e / H ousehold / Y ear (Mean: 1.32 ) Equity: Most equitable in cost of service and breadth of service Number Cities: 75 Fiscal Efficiency: $ 70 $ 4 37 / Household / Year (Mean: $ 163 ) Environmental Efficiency: 0.60 1.92 CO 2e / H ousehold / Y ea r (Mean: 1.49 ) Equity: Most equitable in cost of service and breadth of service

PAGE 271

! 251 served by looking to either the monocentric (including both monocentric public and monocentric private) or the franchise zone service production models rather than the polycentric service production model.

PAGE 272

! 252 R EFERENCES Aligica, P. D. & P. Boettke (2011). The two social philosophies of Ostroms' institutionalism. Policy Studies Journal : 39(1): 29 49. Aligica, P. D. & V. Tark (2013). Co production, po lycentricity, and value heterogeneity: The Ostroms' public choice institutionalism revisited. American Political Science Review : 107(4): 726 741 Baer, S. E. & V. L. Marando (2000). The sub districting of cities: Applying the polycentric model. Polycen tricity colloquium presentation. Benton, J. E. & D. Gamble (1984). City/county consolidation and economies of scale: Evidence from a time series analysis of Jacksonville, Florida. Social Science Quarterly : 65: 190 198. Bish, R. L. (1971). The public economy of metropolitan areas Chicago, IL, USA: Markham Press. Bish, R. L. & V. Ostrom (1974). Understanding urban government: Metropolitan reform reconsidered Washington, DC, USA: American Enterprise Institute for Public Policy Research. Bohm, R. A., D. H. Folz, T. C. Kinnaman, & M. J. Podolsky. (2010). The costs of municipal waste and recycling programs. Resource, Conservation and Recycling 54(2010): 864 871. Cannon, J.S. (2006). Greening garbage trucks: Trends in alternative fuel use, 2002 2005 New York, NY, USA: INFORM, Inc. Report. Chavez, A. & A. Ramaswami (2011). Progress toward low carbon cities: Approaches for transboundary GHG emissions' footprinting. Carbon Management 2(4): 471 482. City of Colorado Springs Website (2012). www.springsgov.com Last accessed August 1, 2012. City of Denver Climate Action Plan (2007). http://www.greenprintdenver.org/docs/Den verClimateActionPlan.pdf Last accessed April 27, 2012. Collier D., H. E. Brady, & J. Seawright (2004). Sources if leverage in causal inference: Toward an alternative view of methodology. In H. E. Brady & D. Collier Eds., Rethinking social inquiry: Dive rse tools, shared standards. Lanham, MD, USA: Rowman & Littlefield.

PAGE 273

! 253 Copeland, D. (2013). NIMBY Notes: Unwelcome Oversight. Waste Age : May, 2013. Cowan, R. & D. Century (2003). Takedown: The fall of the last mafia empire New York, NY, USA: Berkley. D avis, M.W. (2009). Covanta Energy Indianapolis: A case study of integrated waste management. Unpublished manuscript. Davis, M. W. & C. M. Weible (2011). Linking social actors by linking social theories: Toward improved GHG mitigation. Carbon Managemen t 2(4): 483 491. DeHoog, R. H., D. Lowery, & W. E. Lyons (1991). Metropolitan fragmentation and suburban ghettos: Some empirical observations on institutional racism. Journal of Urban Affairs 12(November): 479 493. Denhardt, J. V. & R. B. Denhardt (201 1). The new public service: Service, not steering Third edition. Armonk, New York, USA: M.E. Sharpe. Dietz, T., N. Dolsak, E. Ostrom, & P. C. Stern (2002). The drama of the commons. In eds. E. Ostrom, T. Dietz, N. Dolsak, P. C. Stern, S. Stonich, and E. U. Weber The drama of the commons Washington, DC, USA: National Academy Press. Dillman, D. A. (2007). Mail and internet surveys: The tailored design method, second edition Hoboken, NJ, USA: John Wiley & Sons, Inc. Downs, A. (1994). New visions for metropolitan America Washington, DC, USA: Brookings Institution Press. Durning, D. (1989). The impacts of city county government consolidation Athens: Carl Vinson Institute of Government. Elazar, D. J. (1987). Exploring Federalism Tuscaloos a, AL, USA: University of Alabama Press. Erie, S., J. Kirlin, & F. Rabinovitz (1972). Can something be done? Propositions on the performance of metropolitan institutions. In L. Wingo, ed. Reform of metropolitan governments Washington, DC, USA: Resour ces for the Future. Feiock, R. C. & J. B. Carr (1997). A reassessment of city/county consolidation: Economic development impacts. State and Local Government Review Folz, D. H. (1996). Survey research for public administration Thousand Oaks, CA, USA : SAGE. Folz, D. H. (1999). Municipal recycling performance: A public sector environmental success story. Public Administration Review 59(4): 336 345.

PAGE 274

! 254 Folz, D. H. (2004). Service quality and benchmarking the performance of municipal services. Public Administration Review 64(2): 209 220. Folz, D. H. & J. N. Giles (2002). Municipal experience with "Pay as you throw" policies: Findings from a national survey. State and Local Government Review 34(2): 105 115. Franciosi, R. (1998). Garbage in, garb age out: An examination of private/public competition by the City of Phoenix. Goldwater Institute Arizona Issue Analysis 148 Frisken, F. & D. F. Norris (2001). Regionalism reconsidered. Journal of Urban Affairs : Volume 23(5): 467 478. Gerlat, A. (201 3). Citizen group files suit against California city's waste privatization. Waste Age: January 2013. George, A. L. & A. Bennett (2005). Case studies and theory development in the social the social sciences Cambridge, MA, USA: MIT Press. Gerring, J. (2007). Case study research: Principles and practices New York, NY, USA: Cambridge University Press. Goertz, G. (2006). Social science concepts: A User's Guide Princeton, NJ, USA: Princeton University Press. Hawley, A. & B. G. Zimmer (1970). The Metropolitan Community Beverly Hills, CA: Sage Publications. Healey, J. F. (1999). Statistics: A tool for social research fifth edition Belmont, CA, USA: Wadsworth Publishing Company. Heikkila, T. (2006). The contribution of public choice analysis to decision making theories in public administration: Past, present, and future. In Editor G. Morcol Handbook of decision making (2006), pp. 2 1 3 7. CRC Press Taylor & Francis Group. Heikkila, T. (2004). Institutional boundaries and common pool resour ce management: A comparative analysis of water management in California. Journal of Policy Analysis and Management 23(1): 97 117. Heikkila, T. (2001). Managing common pool resources in a public se rvice industry: The case conjunctive water management (Do ctoral dissertation). University of Arizona, Tucson, AZ, USA. Hillman, T., B. Janson, & A. Ramaswami (2011). Spatial allocation of transportation greenhouse gas emissions at the city scale. Journal of Transportation Engineering 137: 416 425.

PAGE 275

! 255 Hillm an, T. & A. Ramaswami (2010). Greenhouse gas emission footprints and energy use benchmarks for eight U.S. cities. Environmental Science Technology 44: 1902 1910. Judd, D. R. & T. Swanstrom (2006). City politics: The political economy of urban Americ a (fifth edition) New York, NY, USA: Pearson Education, Inc. Keating, M. (1995). Size, efficiency and democracy: Consolidation, fragmentation and public choice. In D. Judge, G. Stoker, & H. Wolman (Eds.) Theories of urban politics (117 134). London, UK: SAGE. Kennedy, C., J. Steinberger, B. Gasson, Y. Hansen, T. Hillman, M. Havranek, D. Pataki, A. Phdungsilp, A. Ramaswami, G. V. Mendez (2010). Methodology for inventorying greenhouse gas emissions from global cities. Energy Policy 38(2010): 4828 48 37. Kiser, L. L. & E. Ostrom (1982). The three worlds of action: A metatheoretical synthesis of institutional approaches. In Editor E. Ostrom Strategies of political inquiry 179 222. Beverly Hills, CA: Sage. Lijphart, A. (1971). Comparing Politics and the Comparative Method. American Political Science Review : 65(3): 682 693. Lind, M. (1997). A Horde of Lilliputian Governments. The New Leader : 5(May): 6 7. Lowery, D., W. E. Lyons, & R. H. DeHoog (1995). The empirical evidence for citizen inform ation and a local market for public goods. The American Political Science Review 89(3): 705 707. Lyons, W. E. & D. Lowery (1989a). Citizen responses to dissatisfaction in urban communities: A partial test of a general model. Journal of Politics 51(4): 842 868. Lyons, W. E. & D. Lowery (1989b). Governmental fragmentation v ersus c onsolidation: Five public choice myths about creating informed, involved, and happy citizens. Public Administration Review 49(6): 533 543. McGinnis, M. (1999). Introduction. In Editor M. D. McGinnis Polycentricity and local public economies (1999), pp. 1 27. Ann Arbor, MI, USA: University of Michigan Press. McGinnis, M. (2011). An introduction to IAD and the language of the Ostrom workshop: A simple guide to a complex fra mework. Policy Studies Journal : 39(1): 169 183. McGinnis, M. & E. Ostrom (2012). Reflections on Vincent Ostrom, public administration, and polycentricity. Public Administration Review 72(1): 15 25. Miller, C. (2013). Circular File: A Tale of Four Cit ies. Waste Age : August 2013.

PAGE 276

! 256 Musgrave, R. A. (1959). The theory of public finance: A study in public economy. New York, NY, USA: McGraw Hill Book Company. Oakerson, R. J. (1999). Governing local public economies: Creating the civic metropolis Oakla nd, CA, USA: Institute for Contemporary Studies Press. Oakerson, R. J. & R. B. Parks (1988). Citizen voice and public entrepreneurship: The organizational dynamic of a complex metropolitan county. In Editor M. D. McGinnis Polycentricity and local public economies (1999), pp. 306 328. Ann Arbor, MI, USA: University of Michigan Press. Oakerson, R. J. & R. B. Parks (2011). The study of local public economies: Multi organizational, multi level institutional analysis and development. Policy Studies Jour nal 39(1): 147 167. O'Brien, J. K. (2010). The benchmarking of residential recycling and yardwaste collection services. MSW Management July/August: 10 13. O'Brien, J. K. (2012). Important findings of the SWANA applied research foundation's research groups. MSW Management September/October: 12 18. O'Brien, J. K. & V. O Johnson (201 2 ). Lowering greenhouse gas emissions by reducing the frequency of curbside recycling MSW Management June: 12 16. Orfield, M. (1997). Metropolitics: A regional ag enda for community and stability Washington DC, USA: Brookings Institution Press. Ostrom, E. (2011). Background on the institutional analysis and development framework. Policy Studies Journal 39(1) 7 28. Ostrom, E. (2010). Beyond markets and s tates: Polycentric governance of complex economic systems. American Economic Review 100: June 2010: 641 672. Ostrom, E. (2007). Institutional Rational Choice: An Assessment of the Institutional Analysis and Development Framework. In Sabatier, P., ed. ( 2007) Theories of the Policy Process Boulder, CO, USA: Westview. Ostrom, E. (2005). Understanding institutional diversity Princeton, NJ: Princeton University Press. Ostrom, E. (2000). The danger of self evident truths. Political Science and Politic s 33(1): 33 44.

PAGE 277

! 257 Ostrom, E. (1972). Metropolitan reform: Propositions derived from two traditions. In Editor M. D. McGinnis Polycentricity and local public economies (1999), pp. 139 160. Ann Arbor, MI, USA: University of Michigan Press. Ostrom, E., R. B. Parks, & G. P. Whitaker (1974). Defining and measuring structural variations in interorganizational arrangements. In Editor M. D. McGinnis Polycentricity and local public economies (1999), pp. 265 283. Ann Arbor, MI, USA: University of Michigan Pr ess. Ostrom, E., R. B. Parks, & G. P. Whitaker (1978). Patterns of metropolitan policing Cambridge, MA, USA: Ballinger Publishing Company. Ostrom, V. (2008). The intellectual crisis in American public administration (Third Edition). The Universit y of Alabama Press, Tuscaloosa, AL, USA. Ostrom, V. (1972). Polycentricity (Part 1). In Editor M. D. McGinnis Polycentricity and local public economies (1999), pp. 52 74. Ann Arbor, MI, USA: University of Michigan Press. Ostrom, V., R. Bish, & E. Ost rom (1988). Local government in the United States San Francisco, CA, USA: ICS Press. Ostrom, V. & E. Ostrom (1977). Public goods and public choice. In Editor M. D. McGinnis Polycentricity and local public economies (1999), pp. 75 106. Ann Arbor, MI, USA: University of Michigan Press. Ostrom, V. & E. Ostrom (1991). Public goods and public choices: The emergence of public economies and industry structures. In V. Ostrom (Ed.) The meaning of American federalism (163 197). San Francisco: Institute for Contemporary Studies Press. Ostrom, V., C. M. Tiebout, & R. Warren (1961). The organization of government in metropolitan areas: A theoretical inquiry. In Editor M. D. McGinnis Polycentricity and local public economies (1999), pp. 31 51. Ann Arbor, MI, USA: University of Michigan Press. Owen, J. C. (1992). Indianapolis unigov: A focus on restructured executive authority. Paper presented at the Southern Political Science Association annual meeting. Peretz, J. H, B. E. Tonn, D. H Folz (2005). Explain ing the performance of mature municipal solid waste recycling programs. Journal of Environmental Planning and Management 48(5): 627 650. Pierce, N. R., with C. W. Johnson & J. S. Hall (1993). Citistates: How urban America can prosper in a competitive w orld Washington, DC, USA: Seven Locks Press.

PAGE 278

! 258 Poteete, A. R., M. A. Janssen, E. Ostrom (2010). Working together Princeton, NJ, USA: Princeton University Press. Ramaswami, A., M. Bernard, A. Chavez, T. Hillman, M. Whitaker, G. Thomas, & M. Marshall ( 2012). Quantifying carbon mitigation wedges in U.S. cities: Near term strategy analysis and critical review. Environmental Science & Technology 46: 3629 3642. Ramaswami, A., T. Hillman, B. Janson, M. Reiner, & G. Thomas. (2008). A demand centered, h ybrid life cycle methodology for city scale greenhouse gas inventories. Environmental Science & Technology 42(17): 6455 6461. Ramaswami, A., D. Main, M. Bernard, A. Chavez, A. Davis, G. Thomas, & K. Schnoor (2011). Planning for low carbon communities i n US cities: A participatory process model between academic institutions, local government and communities in Colorado. Carbon Management 2(4): 397 411. R3 Consulting Group ( 2008 ) Trash service study final report A report prepared for the City of Fort Collins. Rusk, D. (1995). Cities without suburbs second edition Washington, DC, USA: Woodrow Wilson Center. Saha, D. & R. G. Paterson (2008). Local government efforts to promote the "three Es" of sustainable development: Survey in medium to large c ities in the United States. Journal of Planning and Education Research 28(21): 21 37. DOR: 10.1177/0739456X08321803. Salamon (2002 ) The tools of government: A guide to the new governance New York, NY, USA: Oxford University Press Savas, E. S. (197 7). An empirical study of competition in municipal service delivery. Public Administration Review 37(November December): 717 724. Savas, E. S. (1987). Privatization: The key to better government Chatham, NJ: Chatham House Publishers. Savas, E. S. ( 2005). Privatization in the city: Successes, Failures, Lessons Washington, DC: CQ Press. Savitch, H. V. & R. K. Vogel (2000). Paths to new regionalism. In J. S. Davies & D. L. Imbroscio (Eds.) Theories of urban politics, second edition (106 124). Lo s Angeles, CA, USA: SAGE. Savitch, H. V. & R. K. Vogel (2008). Regionalism and urban politics. Paths to new regionalism. State and Local Government Review.

PAGE 279

! 259 Seamon, F. & R. C. Feiock (1995). The political implications of consolidated government: The c ase of Jacksonville. International Journal of Public Administration Shadish, W. R., T. D. Cook, D. T. Campbell (2002). Experimental and quasi experimental designs for generalized causal inference Boston, MA, USA: Houghton Mifflin. Simon, H (1965). Administrative Behavior: A study of decision making processes in administrative organization New York: Free Press. Singleton, R. A. & B. C. Straits (2010). Approaches to social science research, fifth edition New York, NY, USA: Oxford University Pr ess. Schlager, E. (2007). A comparison of frameworks, theories, and models of policy process. In Sabatier, P. Editor. Theories of the Policy Process, 293 320. Boulder, CO: Westview Press. Skumatz, L. A. & D. J. Freeman (2006). Pay as you throw (PA YT) in the US: 2006 update and analysis Skumatz Economic Research Associates Report for EPA Office of Solid Waste, Washington, DC. Skumatz, L. A., D J Freeman, D D'Souza, & D Bement (2012). Rethinking recycling in El Paso County Skumatz Economic Research Associates Report, Superior, CO. Skumatz, L. A., C. Vanderloop, & D. DeLillio (2013). 2013 Roadmap to Commercial Waste Reduction Prepared for the Urban Sustainability Director's Network. Prepared by Skumatz Economic Research Associates Super ior, CO and the City and County of Denver CO SWANA (2000). Manager of landfill operations training and certification course manual Prepared for Solid Waste Association of North America (SWANA) Revisions prepared by Gannett Fleming. Swanstrom, T. ( 2001). What we argue about when we argue about regionalism. Journal of Urban Affairs : 23(5): 479 496. Teske, P., M. Schneider, M. Mintrom, & S. Best (1993). Establishing the micro foundations of a macro theory: Information movers, and the competitive l ocal market for public goods. The American Political Science Review, 87(3): 702 713. Tiebout, C. (1956). A pure theory of local expenditures. The Journal of Political Economy 64(5): 416 424. Tooten, T. (2010). Resilience in public administration: T he work of Elinor and Vincent Ostrom from a public administration perspective. Public Administration Review 70(2): 193 202.

PAGE 280

! 260 U.S. Census Bureau. Metropolitan and micropolitan statistical areas. http://www.census.gov/population/www/metroareas/aboutmetro.html Last accessed April 1 201 4 U.S. Department of Energy. The Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation Model (GREET). http://greet.es.anl.gov/ Last accessed April 1 201 4 U.S. Environmental Protection Agency. Greenhouse Gas Equivalencies Calculator. http://www.epa.gov/cl eanenergy/energy resources/calculator.html Last accessed: April 1, 201 4 U.S. Environmental Protection Agency. Waste Reduction Model (WARM). http://www.epa.gov/clima techange/wycd/waste/calculators/Warm_home.html Last accessed April 1 201 4 Van De Ven, A. H. (2007). Engaged scholarship: A guide for organizational and social research New York, NY, USA: Oxford University Press. Warner, M. & A. Hefetz (2002). The uneven distribution of market solutions for public goods. Journal of Urban Affairs : 24(4): 445 459. Weimer, D. L. & A. R. Vining (2005). Policy analysis: Concepts and practice fourth edition Upper Saddle River, NJ, USA: Pearson Prentice Hall. Y ates, D. (1977). The ungovernable city: The politics of urban problems and policymaking Cambridge, MA, USA: The MIT Press. Young, Oran R. (1999). Hitting the mark. Environment 41(8): 20 29. Zero Waste Associates ( 2013 ) Zero waste Fort Collins: Ro ad to zero waste plan A report prepared for the City of Fort Collins.

PAGE 281

! 261 A PPENDIX A : K EY DEFINITIONS IN ALPHABETICAL ORDER Cartel (Specific to solid waste management): A service delivery system characterized by overlapping yet consolidated service. Whil e there is an appearance of multiple service providers, behind this faade there is in fact a club of collaborators who pre determine who will serve what property at what cost. Thus it is consolidated in that the independent actors act in collusion to the ir exclusive benefit (Davis, 2014) Consolidation: Consolidated is defined as a singular integrated unit. For consolidated provision, this unit is one single governmental unit that has jurisdiction over some form of service provision. For consolidated service production, this unit must be one singular producer delivering the defined service (Aligica & Boettke, 2011; V. Ostrom, 2008; Oakerson, 1999). It is important to note that within the literature the terms "monocentric" and "consolidated" are often considered to be interchangeable, however within this dissertation the two have distinct definitions (Davis, 2014) Economies of Scale: Characterized by a situation where the average cost of a good or service decreases as the number of goods that are pr oduced increases or the number of consumers delivered a service increases (V. Ostrom, Bish, & E. Ostrom, 1988). Effectiveness: Effectiveness is defined as the successful delivery of a service. Basically an effective program would mean the production of a service meets the provision goals for that service. Cost is not a factor for effectiveness, it is simply a measure of was the job well done or not (Oakerson, 1999; Dietz, Dolsak, E. Ostrom & Stern, 2002). Efficiency: Efficiency takes into account comp arative measures. For this study efficiency will be been divided into two measures: fiscal efficiency and environmental efficiency Fiscal efficiency is specific to providing the most service for the least cost. Similarly environmental efficiency is spe cific to providing the most service with the smallest environmental footprint (Oakerson, 1999; McGinnis, 2011). Equity: Equity is a measure of how evenly a level of service is provided across a service area or across multiple service areas. Does everyone receive an equal level of service or are there pockets of uneven service delivery? Do some communities receive better or worse service than other communities? While equity was not a measure in early fragmentation consolidation research, it has been inco rporated as a relevant variable by a number of scholars who have continued research in this vein (Oakerson, 1999; Baer & Marando, 2000; McGinnis, 2011). Fragmentation : Fragmentation is defined as multiple differentiated units. For fragmented provision, t hese units must be multiple governmental units having jurisdiction over some form of service provision. (As defined by this dissertation, fragmentation does not have a geographic element. The geographic element is key to defining overlap and non overlap. ) For fragmented service production, these units must be multiple producers and these multiple producers each need to be delivering the service (Oakerson, 1999; Frisken & Norris, 2001; Oakerson & Parks, 1988, 2011 ; Davis, 2014 )

PAGE 282

! 262 Franchise Zones: This ar rangement is non overlapping yet fragmented. In a franchise zone scheme, citywide there will be multiple service producers and multiple non overlapping geographic service delivery zones, however within any one given zone there will only be one single serv ice producer (Davis, 2014) Functional Consolidation: This arrangement is overlapping yet consolidated. In a functional consolidation arrangement, multiple government units, each having jurisdiction for a particular service provision and / or service pro duction, will all collectively and voluntarily consolidate under a single service delivery mechanism for a specific service. Thus while this arrangement has the potential of overlap the mutually agreed upon consolidation has precedence (Davis, 2014) I nstitutional Analysis and Development Framework: A public policy framework that focuses on institutions and acknowledges: multiple levels of analysis (including theories underlying the framework); the centrality of an action situation; the importance of ru les, their configurations, and their level of rule making (constitution, collective choice, day to day); and the significance of community attributes, such as trust, reciprocity, common understanding, social capital, and cultural repertoire (Ostrom, 2007; McGinnis, 2011). To further understand this framework we shall further define each of its three key words: "Institutions are human constructed constraints or opportunities within which individual choices take place and which shape the consequences of thei r choices" (McGinnis, 2011, p.170); "Analysis involved decomposition of institutional contexts into their component parts as a prelude to understanding how these parts affect each other and how institutions shape outcomes" (McGinnis, 2011, p.170); and "Dev elopment is interpreted broadly as referring to the processes of dynamic changes of institutions as well as changes in their effects over time" (McGinnis, 2011, p.170). Monocentric: Both the local public economy literature and this dissertation define a service that is non overlapping and consolidated as a m onocentric arrangement (McGinnis, 2011). No Overlap (Non overlap): Is defined as the complete lack of overlying, compounded, co occurrence of service provision or production within the same geograp hic area. Within any defined geographic area there is a single service provider and / or single service producer. For non overlapping service provision there must be only one governmental unit exercising its right to execute service provision within a de fined geographic region. For non overlapping service production there must only be one service producer delivering a defined service within a defined geographic region (Davis, 2014) Overlap: Overlap is defined as the real or potential overlying, compoun ded, co occurrence of service provision or production by various entities within the same

PAGE 283

! 263 geographic area. For real overlapping provision there must be multiple governmental units having jurisdiction over some form of service provision and (at a minimum) at least two of these multiple governmental units are exercising their right to execute this service provision in the same geographic region as another unit. For potential overlapping provision there must be multiple governmental units having jurisdiction over some form of service provision however one or none of these multiple governmental units are exercising their right to execute this service provision in the same geographic region as other units. For real overlapping service production there must be multiple producers of the service and (at a minimum) at least two of these producers are exercising their right to deliver this service in the same geographic region as another unit. For potential overlapping service production there must be multiple prod ucers of the service however one or none of these producers are exercising their right to deliver this service in the same geographic region as other units. (While the local public economy literature states that "overlap" can incorporate both a spatial el ement and a function element (Oakerson, 1999) this dissertation, in clarifying definitions of overlap / non overlap and fragmentation / consolidation, defines overlap and non overlap first and primarily as a spatial element and secondarily in terms of ther e needing to be multiple governmental units provision side or multiple producers. This dissertation takes the stand that the "functional element" is not defined via overlap and non overlap but instead is defined via fragmentation and consolidation.) (Davi s, 2014). Polycentric: Both the local public economy literature and this dissertation define a service that is both overlapping and fragmented as a polycentric arrangement (McGinnis, 2011). Production: T he act of producing a good or physically delive ring a service. Production may be delivered by government or by private entities acting as government's agent. (Oakerson, 1999; McGinnis, 2011). Provision: I s buil t around decisions of what public goods and services will be provided, what private activit ies require regulation, how and how much revenue to be raised, and it arranges for how the production of a public good or service will be conducted Provision nearly always is solely the purview of government (Oakerson, 1999; McGinnis, 2011). Solid Waste Collection: Solid waste collection can be colloquially called trash collection, garbage collection, or rubbish removal. Principally, the service must include the collection of trash/waste. Often the service is in conjunction with the further collection or recyclables (bound for recycling rather than disposal) and organics (bound for composting rather than disposal) (Davis, 2014)

PAGE 284

! 264 A PPENDIX B : S URVEY INSTRUMENTS 1: Case Study Phone Interview Questionnaire (Not Annotated) Interview Questio n naire GENER AL QUESTIONS 1.1 In terms of solid waste collection services, how does your city define a "single family" household? (Many cities including free standing multiple unit houses within their definition up to some ordinance designated cut off point.) 1.2 Do es your city provide trash collection service to single family households in the city (either directly or via a contractor)? [ If no skip to question 1.6 / If yes continue. ] 53 1.3 Is this trash service citywide? And if not citywide please explain. 1.4 Is this service provided directly by city employees or by a contracted private company or multiple private companies? 1.5 What is the name of the division with in the city o r the name of the private contractor(s) currently delivering this trash collection ser vice? [ After answering skip to question 1.8] 1.6 Are city residents required to individually subscribe to trash collection service from a company of their choice? 1.7 In your best professional assessment, what are the names of companies that provide ser vice on a citywide basis and, at a minimum, hold at least 20% of the market share for single family household trash colle ction services within your city? 1.8 Does your city have a specific waste reduction goal or a sustainability goal tied to waste reduct ion? 1.9 If yes, w hat is this goal? 1.10 Has your city met the goal? If yes, when? !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 53 For phone interview questionnaire, interviewer will move to next appropr iate question based on directions in brackets. For internet survey instrument, survey software will automatically move to next appropriate question based on directions in brackets.

PAGE 285

! 265 1.11 Are solid waste collection crews: unionized, non unionized, or a mix of unionized / non unionized? 1.12 If a mix, what percentage do you estimate is unionized and what percentage do you estimate is non unionized? TRASH COLLECTION QUESTIONS (Recycl ables and organics questions follow in another section below) 2.1 In 2012, h ow many households were eligible to be served by the single family household t rash collectio n program? 2.2 In 2012, h ow many households subscribed to / were actively served by the single family household trash collection program? 2.3 What is the frequency of the single family household trash collection service? 2.4 What type of collection ve hicle is utilized? Side load, rear load, front load? Cubic yard capacity of truck? Diesel, natural gas, propane, or other? Standard engine or electric assisted "hybrid" engine? 2.5 Would you describe your single family household trash collection as: autom ated, semi automated, or manual? 2.6 What is the typical number of members of a collection crew? 2.7 Are collection containers (e.g. bins, carts, containers, dumpsters, etc.) for trash collection provided to households ? [ If no, skip to 2.10] 2.8 Are these containers provided to households at no direct cost or is there a fee to residents for the containers ? 2.9 What is the size in (gallons or cubic yards) of the container provided? 2.10 Does your city provide periodic "bulky item" collection in conj unction with the single family household trash collection service? If yes, what is the frequency of this "bulky item" collection program? 2.11 What were the total tons of trash disposed specifically from single family households in 201 2 ?

PAGE 286

! 266 2.12 What was the total cost of the city's single family household trash collection program in 201 2 ? (Total cost includes all direct & indirect costs and any payments to contractors ) 2.13 How is the single family household trash collection program funded? 2.14 D oes your community have a variable fee or unit pricing system (pay as you throw) for single family household trash collection? 2.15 How many years has this trash collection program operated in its current form? RECYCLABLES COLLECTION QUESTIONS (O rganic s questions follow in another section below) 3.1 Does your city provide recyclables collection service to single family households in the city (eithe r directly or via a contractor)? (NOTE: Drop off programs are not being assessed by this survey; if your program is a drop off program indicate no.) [ If no skip to next section.] 3.2 Is this service provided directly by city employees or by a contracted private company or multiple private companies? 3.3 What is the name of the division within the city o r the name of the private contractor(s) currently delivering this single family household recyclables collection service? 3.4 In 2012, h ow many households were eligible to be served by the single family household recyclables collection program? 3.5 Is this recyclables collection program mandatory or voluntary? 3.6 In 2012, h ow many households subscribed to / were actively served by the single family household recyclables collection program? 3.7 Is this recyclables collection service citywide? And i f not citywide please explain. 3.8 What is the frequency of the recyclables collection service? 3.9 Is collection of recyclables scheduled for the same day as the collection of trash? 3.10 What type of collection vehicle is utilized? Side load, rear load, front load sectioned curb sort truck ?

PAGE 287

! 267 Cubic yard capacity of truck? Diesel, natural gas, propane, or other? Standard engine or electric assisted "hybrid" engine? 3.11 Would you describe your single family household recyclables collection as: automa ted, semi automated, or manual? 3.12 Would you describe your re cyclables collection as: single stream, dual stream, or curb sort? Single stream (Example: All Recyclables Together) Dual stream (Example: Paper/Fiber in one container / All else in another container) Curb Sort (Example: Glass, Plastics, Paper, Metals all separated from one another) 3.13 What is the typical number of members of a collection crew? 3.14 Are collection containers (e.g. bins, carts, containers, dumpsters, etc.) for recyclabl es collection provided to households ? [If no, skip to question 3.17.] 3.15 Are these containers provided to households at no direct cost or is there a fee to residents for the containers ? 3.16 What is the size in (gallons or cubic yards) of the contain er provided? 3.17 What were the total tons of recyclables collected specifically from single family households in 2012 3.18 What was the total cost of the city's single family household recyclables collection program in 201 2 ? (Total cost includes all d irect & indirect costs and any payments to contractors ) 3.19 How is the recyclables collection program funded? 3.20 How much total revenue, if any, was obtained from the sale of recyclables collection program via the collection program in 201 2 ? 3.2 1 Below please check all applicable recyclables collected as a part of the collection program: ___ #1 PET Plastic Bottles and Jugs ___ #1 Other non bottle non jug PET Plastics ___ #2 HDPE Plastic Bottles and Jugs ___ #2 Other non bottle non jug PET Plastic s ___ #3 V Plastic Bottles and Jugs

PAGE 288

! 268 ___ #3 Other non bottle non jug V Plastics ___ #4 LDPE Plastic Bottles and Jugs ___ #4 Other non bottle non jug LDPE Plastics ___ #5 PP Plastic Bottles and Jugs ___ #5 Other non bottle non jug PP Plastics ___ #6 PS Plast ic Bottles and Jugs ___ #6 Other non bottle non jug PS Plastics ___ #7 Other Plastic Bottles and Jugs ___ #7 Other non bottle non jug Plastics ___ Residential Aluminum Cans ___ Residential Aluminum Foil ___ Residential Steel "Tin" Cans ___ Non can Ferrous Metal ___ Non can Non ferrous Metal ___ White Goods ___ Mixed Office Paper (Color and White) ___ Motor Vehicle / Lead Acid Batteries ___ Alkaline Batteries ___ Non alkaline non rechargable batteries ___ Non alkaline non lead acid rechargeable batteries ___ White Office Paper (High Grade) ___ Mixed Office Paper ___ Newspaper ___ Magazines ___ Phonebooks ___ Cardboard / Corrugated ___ Paperboard ___ Clear Glass Bottles and Jars ___ Brown Glass Bottles and Jars ___ Green Glass Bottles and Jars ___ Aseptic Pack aging (Ex ample : Juice Box) ___ Waxed Paper Packaging (Example: Milk Cartons) ___ Textiles (Write in / Specify) ___ Electronics (Write in / Specify) ___ Other (Write in / Specify) 3.22 How many years has this recyclables collection program operated in its current form? ORGANICS COLLECTION QUESTIONS (Organics collection is defined as the collection of organic matter such as yard waste or food waste bound for composting rather than disposal.) 4.1 Does your city provide organics collection service to sing le family households in the city (eithe r directly or via a contractor)? (NOTE: Drop off programs are not being assessed by this survey; if your program is a

PAGE 289

! 269 drop off program indicate no.) [ If no skip to next section.] 4.2 Is the program year round, ye ar round except for some winter months, or limited to only fall leaf collection? ___ Year round collection [If yes skip to question 4.4] ___ Year round collection except some winter months [If yes continue to question 4.3] ___ Fall leaf collection only [ If yes skip to next section.] 4.3 What months of the year is the service is provided? What months of the year is the service not provided? 4.4 Is this service provided directly by city employees or by a contracted private company or multiple private c ompanies? 4.5 What is the name of the division within the city or the name of the private contractor(s) currently delivering this organics collection service? 4.6 In 2012, h ow many single family households were eligible to be served by the organics coll ection program? 4.7 Is this organics collection program mandatory or voluntary? 4.8 In 2012, h ow many households subscribed to / were actively served by the organics collection program? 4.9 Is this single family household organics collection service citywide? And if not citywide please explain. 4.10 What is the frequency of the organics collection service? 4.11 Is collection of organics scheduled for the same day as the collection of trash or recyclables ? ___ Yes, same day as trash ___ Yes, same day as recyclables ___ Yes, same day as both trash and recyclables ___ No, different day 4.12 What type of collection vehicle is utilized? Side load, rear load, front load? Cubic yard capacity of truck? Diesel, natural gas, propane, or other? Standard en gine or electric assisted "hybrid" engine? 4.13 Would you describe your single family household organics collection as: automated, semi automated, or manual?

PAGE 290

! 270 4.14 What is the typical number of members of a collection crew? 4.15 Are collection containe rs (e.g. bins, carts, containers, dumpsters, etc.) for organics collection provided to households ? [If no, skip to question 4.18.] 4.16 Are these containers provided to households at no direct cost or is there a fee to residents for the containers ? 4.1 7 What is the size in (gallons or cubic yards) of the container provided? 4.18 What were the total tons of organics collected specifically from single family households in 201 2 ? 4.19 What was the total cost of the city's single family household trash co llection program in 201 2 ? (Total cost includes all direct & indirect costs and any payments to contractors ) 4.20 How is the organics collection program funded? 4.21 How much total revenue, if any, was obtained from the sale of finished compost colle ction program via the collection program in 201 2 ? 4.22 Below please check all applicable organics for composting collected as a part of the single family household collection program: ___ Yard Trimmings (Grass, brush, wood) ___ Fruit and Vegetable Food S crap s ___ Compostable Paper Products ___ Other (Write in / Specify) 4.23 How many years has this organics collection program operated in its current form? OPEN ENDED QUESTIONS SPECIFIC TO INTERVIEW 5.1 D escribe recent (within three years) changes to any of these collection programs and perceptions of how the change (s) has (have) impacted the programs efficiency and equity 5.2 D escribe near future (within three years) pending changes to any of these collection programs. Was the decision to implement changes related to issues of efficiency and/or equity ? 5.3 For any of these collection programs, describe a program element that you view as particularly efficient or inefficient (fiscally or environmentally)

PAGE 291

! 271 5.4 For any of these collection programs, describe a program element that citizens receiving the service view as particularly efficient or inefficient (fisca lly or environmentally) 5.5 For any of these collection programs, describe a program element that you view as particularly equitable or i nequitable. 5.6 For any of these collection programs, describe a program element that citizens receiving the service view as particularly equitable or inequitable. 5.7 How would you describe the level of citizen satisfaction within your city in relati on to your single family household trash collection program? 5.8 How would you describe the level of citizen satisfaction within your city in relation to your single family recyclables trash collection program? 5.9 How would you describe the level of ci tizen satisfaction within your city in relation to your single family household organics collection program? FINAL QUESTIONS 6.1 Which of these best describes your position? Check all that apply: ___ I work for the city's public works department ___ I work for the city's trash program ___ I work for the city's recycling program ___ I work for the city's organics program ___ I work for the city's sustainability program ___ I work for a private waste hauler 6.2 Which of these best describes your job ti tle? Check the one that is most applicable: ___ City Program Executive ___ City Program Director ___ City Program Manager ___ City Program Coordinator ___ City Program Assistant ___ Private Waste Hauler ___ Other (Write in / Specify) 6.3 Years in this p osition? 6.4 Years of experience in solid waste management?

PAGE 292

! 272 6.5 May we contact you if we have further questions? ___ Yes ___ No 6.6 Would you like to receive a copy of the final report generated from this research? ___ Yes ___ No 6.7 If yes to eithe r of the above, please provide contact information here: Name: Phone: E mail: Mailing Address: 6.8 Would you prefer an electronic or a paper copy of the final report? ___ Electronic ___ Paper 2: City Official Survey Instrument ( Annotated ) City Off icial Survey GENERAL QUESTIONS 1.1 In terms of solid waste collection services, how does your city define a "single family" household? (Many cities including free standing multiple unit houses within their definition up to some ordinance designated cut off point.) Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: C ity level definition of "single family household" 1.2 Does your city provide trash collection service to single family

PAGE 293

! 273 households in the city (either directly or via a contractor)? [ If no skip to question 1.6 / If yes continue. ] 54 Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables : Breadth o f service production 1.3 Is this trash service citywide? And if not citywide please explain. Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 1.4 Is this service provided directly by city employees or by a contracted private company or multiple private companies? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 1.5 What is the name of the division with in the city o r the name of the private contractor( s) currently delivering this trash collection service? [ After answering skip to question 1.8] Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 1.6 Are city residents required to individually subscribe t o trash collection service from a company of their choice? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 1.7 In your best professional assessment, what are the names of companies that provide service on a citywide basis and, at a minimum, hold at least 20% of the market share for single family household trash colle ction services within your city? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 1.8 D oes your city have a specific waste reduction goal or a sustainability goal tied to waste reduction? !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 54 For phone interview questionnaire, interviewer will move to next appropr iate question based on directions in brackets. For Internet survey instrument, survey software will automatically move to next appropriate question based on directions in brackets.

PAGE 294

! 274 Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 1.9 If yes, w hat is this goal? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 1.10 Has your city met the goal? If yes, when? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 1.11 Are solid waste colle ction crews: unionized, non unionized, or a mix of unionized / non unionized? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: W ork force unionization 1.12 If a mix, what do y ou estimate is the percentage of unionized labor and what is the percentage of non unionized labor? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: W ork force unionization TR ASH COLLECTION QUESTIONS (Recycl ables and organics questions follow in another section below) 2.1 In 2012, h ow many households were eligible to be served by the single family household t rash collection program? Relevant Propositions / Dependent Variabl es: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: C ity level definition of "single family household" 2.2 In 2012, h ow many households subscribed to / were actively served by the single family household trash collecti on program? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity

PAGE 295

! 275 2.3 What is the frequency of the single family household trash collection service? Relevant Propositions / Dependent Variables: Fiscal Effici ency, Environmental Efficiency, and Equity Relevant Control Variables: Level of service production 2.4 What type of collection vehicle is utilized? Side load, rear load, front load? Cubic yard capacity of truck? Diesel, natural gas, propane, or other? S tandard engine or electric assisted "hybrid" engine? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: L evel of automation utilized in service production and truck fuel source. 2.5 Would you describe your single family household trash collection as: automated, semi automated, or manual? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of auto mation utilized in service production 2.6 What is the typical number of members of a collection crew? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of automation utilized in service production. 2.7 Are collection containers (e.g. bins, carts, containers, dumpsters, etc.) for trash collection provided to households ? [ If no, skip to 2.10] Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environment al Efficiency, and Equity Relevant Control Variable s: L evel of automation utilized in service productio n

PAGE 296

! 276 2.8 Are these containers provided to households at no direct cost or is there a fee to residents for the containers ? Relevant Propositions / Depend ent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variable s: L evel of automation ut ilized in service production 2.9 What is the size in (gallons or cubic yards) of the container provided? Relevant Propositions / De pendent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 2.10 Does your city provide periodic "bulky item" collection in conjunction with the single family household trash collection service? If yes, what is the frequency of this "bulky item" collection program? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of service production 2.10 What were the total tons of trash disposed specifically from si ngle family households in 201 2 ? Relevant Propositions / Dependent Variables: Environmental Efficiency 2.11 What was the total cost of the city's single family household trash collection program in 201 2 ? (Total cost includes all direct & indirect costs a nd any payments to contractors ) Relevant Propositions / Dependent Variables: Fiscal Effici ency and Equity 2.12 How is the single family household trash collection program funded? Relevant Propositions / Dependen t Variables: Fiscal Efficiency and Equit y Relevant Control Variable s: F unding mechanism 2.13 Does your community have a variable fee or unit pricing system (pay as you throw) for single family household trash collection?

PAGE 297

! 277 Relevant Propositions / Dependent Variables: Fiscal Efficiency and Equ ity Relevant Control Variables: F unding mechanism 2.14 How many years has this trash collection program operated in its current form? Relevant Propositions / Dependent Variables: Fiscal Efficiency and Equity RECYCLABLES COLLECTION QUESTIONS (O rganics questions follow in another section below) 3.1 Does your city provide recyclables collection service to single family households in the city (eithe r directly or via a contractor)? (NOTE: Drop off programs are not being assessed by this survey; if your p rogram is a drop off program indicate no.) [ If no skip to next section.] Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables : B readth of service production 3.2 Is this servic e provided directly by city employees or by a contracted private company or multiple private companies? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 3.3 What is the name of the division within the c ity or the name of the private contractor(s) currently delivering this single family household recyclables collection service? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 3.4 In 2012, h ow many hous eholds were eligible to be served by the single family household recyclables collection program? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: C ity level definition of "sin gle family household"

PAGE 298

! 278 3.5 Is this recyclables collection program mandatory or voluntary? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 3.6 In 2012, h ow many households subscribed to / were actively served by the single family household recyclables collection program? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: C ity level definition of "single family household". 3 .7 Is this recyclables collection service citywide? And if not citywide please explain. Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables : B readth of service production 3.8 W hat is the frequency of the recyclables collection service? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of service production and breadth of service production 3 .9 Is collection of recyclables scheduled for the same day as the collection of trash? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of service production and breadth of service production 3.10 What type of collection vehicle is utilized? Side load, rear load, front load sectioned curb sort truck ? Cubic yard capacity of truck? Diesel, natural gas, propane, or other? Standard engine or electric assisted "hybrid" eng ine? Relevant Propositions / Dependent Variables: Fiscal Efficiency,

PAGE 299

! 279 Environmental Efficiency, and Equity Relevant Control Variables: Level of automation utilized in service production and truck fuel source 3.11 Would you describe your single family house hold recyclables collection as: automated, semi automated, or manual? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: L evel of automation utilized in service production 3 .12 Would you describe your recyclables collection as: single stream, dual stream, or curb sort? Single stream (Example: All Recyclables Together) Dual stream (Example: Paper/Fiber in one container / All else in another container) Curb Sort (Example: Glas s, Plastics, Paper, Metals all separated from one another) Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of service production, b readth of service production, and le vel of automation utilized in service production 3.13 What is the typical number of members of a collection crew? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of automation utilized in service production 3.14 Are collection containers (e.g. bins, carts, containers, dumpsters, etc.) for recyclables collection provided to households ? [If no, skip to question 3.17.] Relevant Propositions / Dependent Variables: Fis cal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of automation utilized in service production

PAGE 300

! 280 3.15 Are these containers provided to households at no direct cost or is there a fee to residents for the containers ? Re levant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 3.16 What is the size in (gallons or cubic yards) of the container provided? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 3.17 What were the total tons of recyclables collected specifically from single family households in 20 1 2? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 3.18 What was the tota l cost of the city's single family household recyclables collection program in 201 2 ? (Total cost includes all direct & indirect costs and any payments to contractors ) Relevant Propositions / Dependen t Variables: Fiscal Efficiency and Equity Relevant Co ntrol Variables: F unding mechanism 3.19 How is the recyclables collection program funded? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: F unding mechanism 3.20 How muc h total revenue, if any, was obtained from the sale of recyclables collection program via the collection program in 201 2 ? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: F und ing mechanism 3.21 Below please check all applicable recyclables collected as a part of the collection program: ___ #1 PET Plastic Bottles and Jugs ___ #1 Other non bottle non jug PET Plastics ___ #2 HDPE Plastic Bottles and Jugs ___ #2 Other non bottl e non jug PET Plastics ___ #3 V Plastic Bottles and Jugs

PAGE 301

! 281 ___ #3 Other non bottle non jug V Plastics ___ #4 LDPE Plastic Bottles and Jugs ___ #4 Other non bottle non jug LDPE Plastics ___ #5 PP Plastic Bottles and Jugs ___ #5 Other non bottle non jug PP Pla stics ___ #6 PS Plastic Bottles and Jugs ___ #6 Other non bottle non jug PS Plastics ___ #7 Other Plastic Bottles and Jugs ___ #7 Other non bottle non jug Plastics ___ Residential Aluminum Cans ___ Residential Aluminum Foil ___ Residential Steel "Tin" Cans ___ Non can Ferrous Metal ___ Non can Non ferrous Metal ___ White Goods ___ Mixed Office Paper (Color and White) ___ Motor Vehicle / Lead Acid Batteries ___ Alkaline Batteries ___ Non alkaline non rechargable batteries ___ Non alkaline non lead acid recha rgeable batteries ___ White Office Paper (High Grade) ___ Mixed Office Paper ___ Newspaper ___ Magazines ___ Phonebooks ___ Cardboard / Corrugated ___ Paperboard ___ Clear Glass Bottles and Jars ___ Brown Glass Bottles and Jars ___ Green Glass Bottles and Jars ___ Aseptic Packaging (Ex ample : Juice Box) ___ Waxed Paper Packaging (Example: Milk Cartons) ___ Textiles (Write in / Specify) ___ Electronics (Write in / Specify) ___ Other (Write in / Specify) Relevant Propositions / Dependent Variables: Fiscal Eff iciency, Environmental Efficiency, and Equity Relevant Control Variables : B readth of service production and level of automation utilized in service production 3.22 How many years has this recyclables collection program operated in its current form? Re levant Propositions / Dependent Variables: Fiscal Efficiency and

PAGE 302

! 282 Equity ORGANICS COLLECTION QUESTIONS (Organics collection is defined as the collection of organic matter such as yard waste or food waste bound for composting rather than disposal.) 4.1 Do es your city provide organics collection service to single family households in the city (eithe r directly or via a contractor)? (NOTE: Drop off programs are not being assessed by this survey; if your program is a drop off program indicate no.) [ If no sk ip to next section.] Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of service production and breadth of service production 4.2 Is the program year round, year roun d except for some winter months, or limited to only fall leaf collection? ___ Year round collection [If yes skip to question 4.4] ___ Year round collection except some winter months [If yes continue to question 4.3] ___ Fall leaf collection only [If yes skip to next section.] Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of service production and breadth of service production 4.3 What months of the year is the se rvice is provided? What months of the year is the service not provided? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of service production and breadth of service pro duction 4.4 Is this service provided directly by city employees or by a contracted private company or multiple private companies? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Envi ronmental Efficiency, and Equity 4.5 What is the name of the division within the city or the name of the private contractor(s) currently delivering this organics collection service?

PAGE 303

! 283 Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 4.6 In 2012, h ow many si ngle family households were eligible to be served by the organics collection program? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: C ity level definition of single family household". 4.7 Is this organics collection program mandatory or voluntary? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 4.8 In 2012, h ow many households subscribed to / were actively served by th e organics collection program? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: C ity level definiti on of "single family household" 4.9 Is this single family household organi cs collection service citywide? And if not citywide please explain. Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: C ity level definiti on of "single family household" 4.10 What is the frequency of the organics collection service? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables : Level of service production and breadth of service production 4 .11 Is collection of organics scheduled for the same day as the collection of trash or recyclables ? ___ Yes, same day as trash

PAGE 304

! 284 ___ Yes, same day as recyclables ___ Yes, same day as both trash and recyclables ___ No, different day Relevant Propositions / D ependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables : Level of service production and breadth of service production 4.12 What type of collection vehicle is utilized? Side load, rear load, front load? Cu bic yard capacity of truck? Diesel, natural gas, propane, or other? Standard engine or electric assisted "hybrid" engine? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level o f automation utilized in service production a nd truck fuel source 4.13 Would you describe your single family household organics collection as: automated, semi automated, or manual? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Enviro nmental Efficiency, and Equity Relevant Control Variables: Level of automation utilized in service production 4.14 What is the typical number of members of a collection crew? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of automation utilized in service production 4.15 Are collection containers (e.g. bins, carts, containers, dumpsters, etc.) for trash collection provided to households ? [If no, skip to question 4 .18.] Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of automation utilized in service

PAGE 305

! 285 production 4.16 Are these containers provided to households at no direct cost or is there a fee to residents for the containers ? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 4.17 What is the size in (gallons or cubic yards) of the container provided? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity 4.18 What were the total tons of organics collected specifically from single family households in 201 2 ? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environm ental Efficiency, and Equity 4.19 What was the total cost of the city's single family household trash collection program in 201 2 ? (Total cost includes all direct & indirect costs and any payments to contractors ) Relevant Propositions / Dependent Varia bles: Fiscal Efficiency and Equity Relevant Control Variables: F unding mechanism 4.20 How is the organics collection program funded? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Va riables: F unding mechanism 4.21 How much total revenue, if any, was obtained from the sale of finished compost collection program via the collection program in 201 2 ? Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficienc y, and Equity Relevant Control Variables: F unding mechanism 4.22 Below please check all applicable organics for composting collected as a part of the single family household collection program: ___ Yard Trimmings ( Such as: Grass, brush, wood)

PAGE 306

! 286 ___ Fruit a nd Vegetable Food Scrap s ___ Compostable Paper Products ___ Other (Write in / Specify) Relevant Propositions / Dependent Variables: Fiscal Efficiency, Environmental Efficiency, and Equity Relevant Control Variables: Level of service production and breadth of service production 4.23 How many years has this organics collection program operated in its current form? Relevant Propositions / Dependent Variables: Fiscal Efficiency and Equity FINAL QUESTIONS 5.1 Which of these best describes your position? Check all that apply: ___ I work for the city's public works department ___ I work for the city's trash program ___ I work for the city's recycling program ___ I work for the city's organics program ___ I work for the city's sustainability program ___ I wo rk for a private waste hauler 5.2 Which of these best describes your job title? Check the one that is most applicable: ___ City Program Executive ___ City Program Director ___ City Program Manager ___ City Program Coordinator ___ City Program Assistant ___ Private Waste Hauler ___ Other (Write in / Specify) 5.3 Years in this position? 5.4 Years of experience in solid waste management? 5.5 May we contact you if we have further questions? ___ Yes ___ No

PAGE 307

! 287 5.6 Would you like to receive a copy of the fin al report generated from this research? ___ Yes ___ No 5.7 If yes to either of the above, please provide contact information here: Name: Phone: E mail: Mailing Address: 5.8 Would you prefer an electronic or a paper copy of the final report? ___ Elec tronic ___ Paper 3: Private Waste Company Survey Instrument ( Not Annotated) Private Waste Hauler Survey GENERAL QUESTIONS 1.1 In terms of solid waste collection services, how does your c ompany define a "single family" household? (Many cities includ ing free standing multiple unit houses within their definition up to some ordinance designated cut off point.) 1.2 Does your c ompany provide trash collection service to single family households in the city? [ If no skip to question 1. 5 / If yes continue. ] 55 1.3 Does your company provide citywide trash collection service s ? And if not citywide please explain. 1.4 Is this service provided directly by company employees or by a subcontractor or multiple subcontractors ? 1.5 Are solid waste collection crews: unionized, non unionized, or a mix of !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! 55 For phone interview questionnaire, interviewer will move to next appropr iate question based on directions in brackets. For internet survey instrument, survey software will automatically move to next appropriate question based on directions in brackets.

PAGE 308

! 288 unionized / non unionized? 1.6 If a mix, what do you estimate is the percentage of unionized labor and what is the percentage of non unionized labor? TRASH COLLECTION QUESTIONS (Recycl ables and organics questions fo llow in another section below) 2.1 In 2012, h ow many households subscribed to / were actively served by your company's single family household trash collection program? 2.2 What is the frequency of the single family household trash collection service? 2.3 What type of collection vehicle is utilized? Side load, rear load, front load? Cubic yard capacity of truck? Diesel, natural gas, propane, or other? Standard engine or electric assisted "hybrid" engine? 2.4 Would you describe your single family house hold trash collection as: automated, semi automated, or manual? 2.5 What is the typical number of members of a collection crew? 2.6 Are collection containers (e.g. bins, carts, containers, dumpsters, etc.) for trash collection provided to households ? [ If no, skip to 2.9] 2.7 Are these containers provided to households at no direct cost or is there a fee to residents for the containers ? 2.8 What is the size in (gallons or cubic yards) of the container provided? 2.9 What were the total tons of trash disposed specifically from your company's single family households in 201 2 ? 2.10 What is the per household cost for single family household trash collection provided by your company in 201 2 ? ( A range, based on level of service or in city location varia tions is acceptable.) 2.11 Does your com pany have a variable fee or unit pricing system (pay as you throw) for single family household trash collection? If yes, is this required by the city in which you are providing collection services? RECYCLABLES COLLECTION QUESTIONS

PAGE 309

! 289 (O rganics questions follow in another section below) 3.1 Does your company provide recyclables collection service to single family households in the city? (NOTE: Drop off programs are not being assessed by this survey; if your progra m is a drop off program indicate no.) [ If no skip to next section.] 3.2 Is this service provided directly by company employees or by a subcontractor or multiple subcontractors ? 3.3 Is this recyclables collection program mandated by the city? 3.4 In 2012, h ow many households subscribed to / were actively served by your company's single family household recyclables collection program? 3.7 Is this recyclables collection service citywide? And if not citywide please explain. 3.5 What is the frequency of the recyclables collection service? 3.6 Is collection of recyclables scheduled for the same day as the collection of trash? 3.7 What type of collection vehicle is utilized? Side load, rear load, front load sectioned curb sort truck ? Cubic yard cap acity of truck? Diesel, natural gas, propane, or other? Standard engine or electric assisted "hybrid" engine? 3.8 Would you describe your single family household recyclables collection as: automated, semi automated, or manual? 3.9 Would you describe you r re cyclables collection as: single stream, dual stream, or curb sort? Single stream (Example: All Recyclables Together) Dual stream (Example: Paper/Fiber in one container / All else in another container) Curb Sort (Example: Glass, Plastics, Paper, Metals all separated from one another) 3.10 What is the typical number of members of a collection crew? 3.11 Are collection containers (e.g. bins, carts, containers, dumpsters, etc.) for recyclables collection provided to households? [If no, skip to questio n

PAGE 310

! 290 3.14.] 3.12 Are these containers provided to households at no direct cost or is there a fee to residents for the containers ? 3.13 What is the size in (gallons or cubic yards) of the container provided? 3.14 What were the total tons of recyclables c ollected by your company specifically from single family households in 20 1 2? 3.15 What is the per household cost for single family household recyclables collection provided by your company in 201 2 ? ( A range, based on level of service or in city location variations is acceptable.) 3.16 Below please check all applicable recyclables collected as a part of the collection program: ___ #1 PET Plastic Bottles and Jugs ___ #1 Other non bottle non jug PET Plastics ___ #2 HDPE Plastic Bottles and Jugs ___ #2 Ot her non bottle non jug PET Plastics ___ #3 V Plastic Bottles and Jugs ___ #3 Other non bottle non jug V Plastics ___ #4 LDPE Plastic Bottles and Jugs ___ #4 Other non bottle non jug LDPE Plastics ___ #5 PP Plastic Bottles and Jugs ___ #5 Other non bottle n on jug PP Plastics ___ #6 PS Plastic Bottles and Jugs ___ #6 Other non bottle non jug PS Plastics ___ #7 Other Plastic Bottles and Jugs ___ #7 Other non bottle non jug Plastics ___ Residential Aluminum Cans ___ Residential Aluminum Foil ___ Residential Ste el "Tin" Cans ___ Non can Ferrous Metal ___ Non can Non ferrous Metal ___ White Goods ___ Mixed Office Paper (Color and White) ___ Motor Vehicle / Lead Acid Batteries ___ Alkaline Batteries ___ Non alkaline non rechargable batteries ___ Non alkaline non le ad acid rechargeable batteries ___ White Office Paper (High Grade) ___ Mixed Office Paper ___ Newspaper ___ Magazines ___ Phonebooks

PAGE 311

! 291 ___ Cardboard / Corrugated ___ Paperboard ___ Clear Glass Bottles and Jars ___ Brown Glass Bottles and Jars ___ Green Glass Bottles and Jars ___ Aseptic Packaging (Ex ample : Juice Box) ___ Waxed Paper Packaging (Example: Milk Cartons) ___ Textiles (Write in / Specify) ___ Electronics (Write in / Specify) ___ Other (Write in / Specify) ORGANICS COLLECTION QUESTIONS (Organics c ollection is defined as the collection of organic matter such as yard waste or food waste bound for composting rather than disposal.) 4.1 Does your company provide organics collection service to single family households in the city? (NOTE: Drop off prog rams are not being assessed by this survey; if your program is a drop off program indicate no.) [ If no skip to next section.] 4.2 Is the program year round, year round except for some winter months, or limited to only fall leaf collection? ___ Year rou nd collection [If yes skip to question 4.4] ___ Year round collection except some winter months [If yes continue to question 4.3] ___ Fall leaf collection only [If yes skip to next section.] 4.3 What months of the year is the service is provided? What months of the year is the service not provided? 4.4 Is this service provided directly by company employees or by a subcontractor or multiple subcontractors ? 4.5 Is this organics collection program mandated by the city? 4.6 In 2012, h ow many household s subscribed to / were actively served by the organics collection program? 4.7 Is this single family household organics collection service citywide? And if not citywide please explain. 4.8 What is the frequency of the organics collection service? 4. 9 Is collection of organics scheduled for the same day as the collection of trash or recyclables ? ___ Yes, same day as trash

PAGE 312

! 292 ___ Yes, same day as recyclables ___ Yes, same day as both trash and recyclables ___ No, different day 4.10 What type of collecti on vehicle is utilized? Side load, rear load, front load? Cubic yard capacity of truck? Diesel, natural gas, propane, or other? Standard engine or electric assisted "hybrid" engine? 4.11 Would you describe your single family household organics collection as: automated, semi automated, or manual? 4.12 What is the typical number of members of a collection crew? 4.13 Are collection containers (e.g. bins, carts, containers, dumpsters, etc.) for organics collection provided to households ? [If no, skip to qu estion 4.16.] 4.14 Are these containers provided to households at no direct cost or is there a fee to residents for the containers ? 4.15 What is the size in (gallons or cubic yards) of the container provided? 4.16 What were the total tons of organics collected by your company specifically from single family households in 201 2 ? 4.17 What is the per household cost for single family household organics collection provided by your company in 201 2 ? ( A range, based on level of service or in city location variations is acceptable.) 4.18 Below please check all applicable organics for composting collected as a part of the single family household collection program: ___ Yard Trimmings ( Such as: Grass, brush, wood) ___ Fruit and Vegetable Food Scrap s ___ Com postable Paper Products ___ Other (Write in / Specify) FINAL QUESTIONS 5.1 Which of these best describes your position? Check all that apply: ___ I work for the city's public works department ___ I work for the city's trash program ___ I work for the city's recycling program

PAGE 313

! 293 ___ I work for the city's organics program ___ I work for the city's sustainability program ___ I work for a private waste hauler 5.2 Which of these best describes your job title? Check the one that is most applicable: ___ City Program Executive ___ City Program Director ___ City Program Manager ___ City Program Coordinator ___ City Program Assistant ___ Private Waste Hauler ___ Other (Write in / Specify) 5.3 Years in this position? 5.4 Years of experience in solid waste manag ement? 5.5 May we contact you if we have further questions? ___ Yes ___ No 5.6 Would you like to receive a copy of the final report generated from this research? ___ Yes ___ No 5.7 If yes to either of the above, please provide contact information her e: Name: Phone: E mail: Mailing Address: 5.8 Would you prefer an electronic or a paper copy of the final report? ___ Electronic ___ Paper

PAGE 314

! 294 A PPENDIX C: S AMPLE POPULATION OF CENTRAL CORE CITIES Table C.1: Sample Population of Central Core Cities Ov er 50,000 in Population Rank in this Study Rank in 2010 Census Geographic Area April 1, 2010 Population Estimate (as of July 1) Place State Census Estimates Base 2010 2011 1 1 New York City New York 8,175,133 8,175,133 8,186,443 8,244,910 2 2 Los A ngeles California 3,792,621 3,792,625 3,795,761 3,819,702 3 3 Chicago Illinois 2,695,598 2,695,598 2,698,283 2,707,120 4 4 Houston Texas 2,099,451 2,099,430 2,108,278 2,145,146 5 5 Philadelphia Pennsylvania 1,526,006 1,526,006 1,528,074 1,536,471 6 6 P hoenix Arizona 1,445,632 1,445,656 1,448,531 1,469,471 7 7 San Antonio Texas 1,327,407 1,327,606 1,334,431 1,359,758 8 8 San Diego California 1,307,402 1,307,406 1,311,516 1,326,179 9 9 Dallas Texas 1,197,816 1,197,816 1,201,715 1,223,229 10 10 San Jose California 945,942 952,612 955,091 967,487 11 11 Jacksonville Florida 821,784 821,784 822,883 827,908 12 12 Indianapolis Indiana 820,445 820,442 821,708 827,609 13 13 Austin Texas 790,390 790,390 795,378 820,611 14 14 San Francisco California 805,235 805,235 805,340 812,826 15 15 Columbus Ohio 787,033 787,073 788,696 797,434 16 16 Fort Worth Texas 741,206 742,030 745,231 758,738 17 17 Charlotte North Carolina 731,424 731,424 734,216 751,087 18 18 Detroit Michigan 713,777 713,777 711,70 0 706,585 19 19 El Paso Texas 649,121 649,152 651,881 665,568 20 20 Memphis Tennessee 646,889 646,889 647,780 652,050 21 21 Boston Massachusetts 617,594 617,594 618,147 625,087 22 22 Seattle Washington 608,660 608,660 610,480 620,778 23 23 Denver Colorado 600,158 600,008 603,440 619,968 24 24 Baltimore Maryland 620,961 620,961 620,560 619,493 25 26 Nashville Tennessee 601,222 601,222 602,537 609,644 26 27 Louisville Kentucky 597,337 597,337 598,207 602,011 27 28 Milwaukee Wisconsin 594,833 5 94,832 595,407 597,867 28 29 Portland Oregon 583,776 583,776 585,474 593,820 29 30 Oklahoma City Oklahoma 579,999 580,001 582,352 591,967 30 31 Las Vegas Nevada 583,756 583,748 584,539 589,317 31 32 Albuquerque New Mexico 545,852 545,852 547,392 55 2,804 32 33 Tucson Arizona 520,116 520,097 521,180 525,796 33 34 Fresno California 494,665 494,735 496,181 501,362

PAGE 315

! 295 34 35 Sacramento California 466,488 466,488 467,575 472,178 35 36 Long Beach California 462,257 462,257 462,645 465,576 36 37 Kansas City Missouri 459,787 459,787 460,724 463,202 37 39 Virginia Beach Virginia 437,994 437,994 439,122 442,707 38 40 Atlanta Georgia 420,003 420,005 422,387 432,427 39 41 Colorado Springs Colorado 416,427 416,427 419,745 426,388 40 42 Raleigh North Ca rolina 403,892 403,903 406,432 416,468 41 43 Omaha Nebraska 408,958 408,962 410,170 415,068 42 44 Miami Florida 399,457 399,457 400,509 408,750 43 45 Tulsa Oklahoma 391,906 391,901 393,166 396,466 44 46 Oakland California 390,724 390,724 391,445 39 5,817 45 47 Cleveland Ohio 396,815 396,815 396,166 393,806 46 48 Minneapolis Minnesota 382,578 382,578 383,108 387,753 47 49 Wichita Kansas 382,368 382,368 383,085 384,445 48 50 Arlington Texas 365,438 365,438 367,021 373,698 49 51 New Orleans Lo uisiana 343,829 343,829 347,907 360,740 50 52 Bakersfield California 347,483 347,462 348,541 352,428 51 53 Tampa Florida 335,709 335,709 336,820 346,037 52 58 St. Louis Missouri 319,294 319,294 319,008 318,069 53 59 Riverside California 303,871 303 ,871 305,679 310,651 54 60 Corpus Christi Texas 305,215 305,215 305,349 307,953 55 61 Pittsburgh Pennsylvania 305,704 305,704 306,956 307,484 56 62 Lexington Kentucky 295,803 295,803 296,792 301,569 57 63 Stockton California 291,707 291,707 292,711 296,357 58 64 Cincinnati Ohio 296,943 296,950 296,797 296,223 59 66 St. Paul Minnesota 285,068 285,068 285,414 288,448 60 67 Toledo Ohio 287,208 287,208 287,031 286,038 61 68 Newark New Jersey 277,140 277,140 277,185 277,540 62 69 Greensboro Nort h Carolina 269,666 269,668 270,364 273,425 63 70 Plano Texas 259,841 259,841 261,888 269,776 64 71 Lincoln Nebraska 258,379 258,381 259,044 262,341 65 72 Buffalo New York 261,310 261,310 261,229 261,025 66 74 Fort Wayne Indiana 253,691 253,691 254, 015 255,824 67 75 Jersey City New Jersey 247,597 247,597 247,876 250,323 68 77 St. Petersburg Florida 244,769 244,769 244,683 244,997 69 78 Orlando Florida 238,300 238,304 238,902 243,195 70 79 Norfolk Virginia 242,803 242,803 242,915 242,628 71 8 0 Laredo Texas 236,091 236,100 237,252 241,935 72 82 Madison Wisconsin 233,209 233,209 233,732 236,901 73 83 Lubbock Texas 229,573 229,573 230,713 233,740 74 84 Durham North Carolina 228,330 228,329 229,027 233,252 75 85 Winston Salem North Carolin a 229,617 229,617 230,077 232,385 76 88 Baton Rouge Louisiana 229,493 229,493 229,852 230,139 77 90 Reno Nevada 225,221 225,229 225,717 227,511 78 91 Chesapeake Virginia 222,209 222,209 223,052 225,050 79 93 Irving Texas 216,290 216,290 216,942 220 ,702 80 95 Fremont California 214,089 214,089 214,490 216,916 81 98 Birmingham Alabama 212,237 212,244 212,225 212,413 82 100 Rochester New York 210,565 210,565 210,578 210,855 83 101 Boise City Idaho 205,671 205,671 206,252 210,145 84 102 Spokane Washington 208,916 208,916 209,296 210,103

PAGE 316

! 296 85 103 Montgomery Alabama 205,764 205,771 206,083 208,182 86 104 Des Moines Iowa 203,433 203,433 204,191 206,599 87 105 Richmond Virginia 204,214 204,214 204,159 205,533 88 106 Fayetteville North Carolina 200,564 200,564 201,022 203,945 89 107 Modesto California 201,165 201,165 201,541 202,751 90 108 Shreveport Louisiana 199,311 199,311 199,819 200,975 91 109 Tacoma Washington 198,397 198,397 198,434 200,678 92 111 Aurora Illinois 197,899 197,897 19 8,288 199,672 93 112 Fontana California 196,069 196,069 196,737 199,028 94 113 Akron Ohio 199,110 199,110 199,005 198,402 95 116 Augusta Georgia 195,844 195,844 196,175 196,494 96 117 Little Rock Arkansas 193,524 193,524 193,944 195,314 97 118 Mobi le Alabama 195,111 195,107 195,166 194,914 98 119 Columbus Georgia 189,885 189,885 190,371 194,107 99 120 Amarillo Texas 190,695 190,695 191,398 193,675 100 123 Salt Lake City Utah 186,440 186,443 187,082 189,899 101 124 Grand Rapids Michigan 188, 040 188,041 188,166 189,815 102 125 Tallahassee Florida 181,376 181,376 181,626 182,965 103 126 Huntsville Alabama 180,105 180,120 180,829 182,956 104 127 Worcester Massachusetts 181,045 181,045 181,208 181,631 105 128 Knoxville Tennessee 178,874 1 78,874 179,225 180,761 106 129 Newport News Virginia 180,719 180,719 180,618 179,611 107 131 Brownsville Texas 175,023 175,027 175,764 178,430 108 132 Providence Rhode Island 178,042 178,042 178,077 178,053 109 133 Santa Clarita California 176,320 176,320 176,469 177,601 110 134 Overland Park Kansas 173,372 173,372 173,870 176,185 111 135 Jackson Mississippi 173,514 173,516 173,707 175,561 112 136 Garden Grove California 170,883 170,883 171,302 173,470 113 137 Chattanooga Tennessee 167,674 1 67,978 168,393 170,136 114 138 Oceanside California 167,086 167,086 167,630 169,569 115 140 Fort Lauderdale Florida 165,521 165,521 165,900 168,528 116 143 Port St. Lucie Florida 164,603 164,603 165,166 166,149 117 144 Vancouver Washington 161,791 1 61,807 162,432 164,759 118 146 Springfield Missouri 159,498 159,509 159,634 160,660 119 147 Lancaster California 156,633 156,633 156,743 157,693 120 149 Cape Coral Florida 154,305 154,305 154,734 157,476 121 150 Eugene Oregon 156,185 156,191 156,26 0 156,929 122 151 Peoria Arizona 154,065 154,067 154,378 156,637 123 152 Sioux Falls South Dakota 153,888 153,890 154,443 156,592 124 153 Salem Oregon 154,637 154,637 155,029 156,244 125 156 Palmdale California 152,750 152,750 152,880 153,867 126 157 Springfield Massachusetts 153,060 153,060 153,134 153,155 127 158 Salinas California 150,441 150,441 150,929 152,994 128 160 Rockford Illinois 152,871 152,871 152,807 152,222 129 162 Joliet Illinois 147,433 147,440 147,745 148,402 130 163 Fort Collins Colorado 143,986 143,991 144,425 146,762 131 165 Kansas City Kansas 145,786 145,786 146,070 146,453 132 166 Paterson New Jersey 146,199 146,199 146,309 146,427

PAGE 317

! 297 133 168 Escondido California 143,911 143,911 144,375 146,032 134 169 Bridgeport Connecticut 144,229 144,229 144,463 145,638 135 170 Syracuse New York 145,170 145,170 145,237 145,151 136 172 Alexandria Virginia 139,966 139,966 140,894 144,301 137 173 Hollywood Florida 140,768 140,768 141,095 143,357 138 175 Mesquite Texas 139, 824 139,824 140,245 142,674 139 176 Sunnyvale California 140,081 140,081 140,449 142,287 140 177 Dayton Ohio 141,527 141,527 141,696 142,148 141 179 Savannah Georgia 136,286 136,280 136,565 139,491 142 183 Hampton Virginia 137,436 137,436 137,328 1 36,401 143 187 McAllen Texas 129,877 129,877 130,759 133,742 144 189 Columbia South Carolina 129,272 129,276 129,612 130,591 145 190 Killeen Texas 127,921 127,921 129,030 130,018 146 192 New Haven Connecticut 129,779 129,779 129,774 129,585 147 19 3 Topeka Kansas 127,473 127,473 127,717 128,188 148 196 Cedar Rapids Iowa 126,326 126,326 126,528 127,905 149 197 Waco Texas 124,805 124,805 125,351 126,697 150 198 Visalia California 124,442 124,442 124,816 126,432 151 199 Elizabeth New Jersey 12 4,969 124,969 125,195 125,660 152 201 Gainesville Florida 124,354 124,354 124,394 125,326 153 202 Hartford Connecticut 124,775 124,775 124,789 124,867 154 206 Stamford Connecticut 122,643 122,643 122,848 123,868 155 207 Coral Springs Florida 121,09 6 121,098 121,382 123,338 156 208 Charleston South Carolina 120,083 120,083 120,479 122,689 157 209 Carrollton Texas 119,097 119,097 119,729 122,640 158 210 Lafayette Louisiana 120,623 120,621 120,905 122,130 159 216 Allentown Pennsylvania 118,032 118,032 118,232 119,141 160 217 Beaumont Texas 118,296 118,296 118,347 118,548 161 219 Abilene Texas 117,063 117,063 117,404 118,117 162 220 Evansville Indiana 117,429 117,429 117,483 117,825 163 221 Victorville California 115,903 115,921 116,299 1 17,597 164 222 Independence Missouri 116,830 116,830 116,969 117,213 165 224 Springfield Illinois 116,250 116,272 116,479 117,076 166 226 Athens Georgia 115,452 115,453 115,450 116,084 167 227 Provo Utah 112,488 112,488 113,153 115,321 168 228 Peor ia Illinois 115,007 115,025 114,895 115,234 169 229 Ann Arbor Michigan 113,934 113,934 114,076 114,925 170 230 Lansing Michigan 114,297 114,297 114,247 114,605 171 232 Midland Texas 111,147 111,147 111,255 113,931 172 233 Berkeley California 112,5 80 112,578 112,765 113,905 173 234 Norman Oklahoma 110,925 110,925 111,426 113,273 174 239 Columbia Missouri 108,500 108,500 108,857 110,438 175 240 Waterbury Connecticut 110,366 110,366 110,360 110,189 176 241 Manchester New Hampshire 109,565 109, 565 109,634 109,830 177 242 Miami Gardens Florida 107,167 107,167 107,452 109,680 178 243 Elgin Illinois 108,188 108,188 108,395 109,104 179 244 Wilmington North Carolina 106,476 106,476 106,785 108,297 180 246 Rochester Minnesota 106,769 106,769 1 06,943 107,890 181 248 Lowell Massachusetts 106,519 106,519 106,706 107,584

PAGE 318

! 298 182 249 Pueblo Colorado 106,595 106,595 106,864 107,577 183 251 San Buenaventura (Ventura) California 106,433 106,433 106,706 107,514 184 252 Gresham Oregon 105,594 105,594 105,905 107,439 185 253 Fargo North Dakota 105,549 105,549 105,884 107,349 186 254 Carlsbad California 105,328 105,328 105,671 106,888 187 257 Fairfield California 105,321 105,323 105,524 106,126 188 260 Green Bay Wisconsin 104,057 104,057 104,567 105,809 189 263 Billings Montana 104,170 104,170 104,505 105,636 190 269 Wichita Falls Texas 104,553 104,553 104,679 103,931 191 270 Palm Bay Florida 103,190 103,190 103,278 103,227 192 272 Temecula California 100,097 100,146 100,766 102,464 193 273 Daly City California 101,123 101,123 101,286 102,362 194 274 Odessa Texas 99,940 99,940 99,962 102,106 195 275 Erie Pennsylvania 101,786 101,786 101,826 101,807 196 277 Pompano Beach Florida 99,845 99,845 100,070 101,617 197 278 Flint Michigan 102,434 102,434 102,271 101,558 198 279 South Bend Indiana 101,168 101,170 101,139 101,081 199 280 West Palm Beach c Florida 99,919 99,920 100,128 101,043 200 282 Davenport Iowa 99,685 99,685 100,003 100,802 201 283 Rialto California 99,171 99,170 9 9,508 100,662 202 284 Santa Maria California 99,553 99,553 99,730 100,277 203 285 Broken Arrow Oklahoma 98,850 98,847 99,195 100,073 204 286 Kenosha Wisconsin 99,218 99,226 99,379 99,738 205 287 North Charleston South Carolina 97,471 97,471 97,868 99,727 206 288 Las Cruces New Mexico 97,618 97,618 98,200 99,665 207 289 Boulder c Colorado 97,385 97,385 97,574 98,889 208 291 Lakeland Florida 97,422 97,419 97,551 98,589 209 292 Tyler Texas 96,900 96,900 97,250 98,564 210 294 Lawton c Oklahoma 96 ,867 96,867 97,873 98,177 211 295 Albany New York 97,856 97,856 97,760 97,660 212 300 Roanoke Virginia 97,032 97,032 96,856 96,714 213 302 Portsmouth Virginia 95,535 95,535 95,627 95,684 214 304 New Bedford Massachusetts 95,072 95,072 95,115 95,183 215 305 College Station Texas 93,857 93,862 94,232 95,142 216 307 Greeley Colorado 92,889 92,889 93,395 94,962 217 309 San Angelo Texas 93,200 93,200 93,562 94,544 218 310 Yuma Arizona 93,064 93,066 93,436 94,536 219 311 Brockton Massachusetts 93 ,810 93,810 93,965 94,316 220 315 Vacaville California 92,428 92,428 92,593 93,088 221 318 Yakima Washington 91,067 91,067 91,509 92,512 222 320 Macon Georgia 91,351 91,346 91,416 91,856 223 321 Lee's Summit Missouri 91,364 91,364 91,477 91,668 22 4 323 Tuscaloosa Alabama 90,468 90,483 90,633 91,605 225 324 Hesperia California 90,173 90,173 90,481 91,534 226 325 Sparks Nevada 90,264 90,264 90,463 91,195 227 326 Roswell Georgia 88,346 88,347 88,846 91,168 228 333 Redding California 89,861 89 ,861 89,904 90,138 229 334 Miami Beach Florida 87,779 87,779 88,012 89,840

PAGE 319

! 299 230 335 Waukegan Illinois 89,078 89,078 89,210 89,426 231 336 Rio Rancho New Mexico 87,521 87,521 88,071 89,320 232 338 Santa Barbara California 88,410 88,409 88,565 89,045 233 339 Fall River Massachusetts 88,857 88,857 88,897 88,962 234 340 Lawrence Kansas 87,643 87,643 87,955 88,727 235 341 Reading Pennsylvania 88,082 88,082 88,351 88,414 236 342 Longmont Colorado 86,270 86,270 86,450 87,712 237 344 Fort Smith Ark ansas 86,209 86,209 86,284 87,152 238 345 Nashua New Hampshire 86,494 86,494 86,549 86,704 239 348 Norwalk Connecticut 85,603 85,603 85,746 86,460 240 349 Chico California 86,187 86,187 86,187 86,290 241 350 Duluth Minnesota 86,265 86,265 86,230 86 ,277 242 351 San Leandro California 84,950 84,950 85,108 86,071 243 352 Greenville North Carolina 84,554 84,562 84,856 86,017 244 356 Boca Raton Florida 84,392 84,392 84,566 85,329 245 357 Deltona Florida 85,182 85,182 85,191 85,219 246 359 Suffol k Virginia 84,585 84,592 84,872 84,930 247 361 Trenton New Jersey 84,913 84,913 85,009 84,899 248 363 Asheville North Carolina 83,393 83,393 83,574 84,458 249 365 Clifton New Jersey 84,136 84,136 84,201 84,269 250 368 Bloomington Minnesota 82,893 82,893 83,013 84,057 251 369 Ogden Utah 82,825 82,825 83,170 83,949 252 372 Sioux City Iowa 82,684 82,695 82,826 82,967 253 374 Nampa Idaho 81,557 81,565 81,774 82,755 254 375 Warwick Rhode Island 82,672 82,672 82,628 82,361 255 376 Livermore Cal ifornia 80,968 80,968 81,121 82,039 256 381 Danbury Connecticut 80,893 80,893 81,023 81,671 257 384 Bloomington Indiana 80,405 80,407 80,675 81,381 258 385 Longview Texas 80,455 80,455 80,572 81,336 259 386 Champaign Illinois 81,055 81,055 81,168 8 1,291 260 390 Concord North Carolina 79,066 79,066 79,308 80,597 261 391 O'Fallon Missouri 79,329 79,479 79,755 80,519 262 392 Hemet California 78,657 78,657 79,140 80,467 263 393 Cranston Rhode Island 80,387 80,387 80,404 80,392 264 395 Merced C alifornia 78,958 78,958 79,257 80,232 265 396 Gary Indiana 80,294 80,294 80,314 80,221 266 397 Mission Texas 77,058 77,061 77,588 79,368 267 402 Chino California 77,983 77,983 78,227 79,059 268 403 Racine Wisconsin 78,860 78,860 78,898 78,853 269 404 Lake Forest California 77,264 77,264 77,454 78,439 270 405 Bend Oregon 76,639 76,639 76,735 77,905 271 406 Napa California 76,915 76,945 77,150 77,867 272 407 Indio California 76,036 76,038 76,504 77,780 273 408 Redwood City California 76,815 76,815 76,936 77,745 274 410 Albany Georgia 77,434 77,434 77,595 77,683 275 411 New Rochelle New York 77,062 77,062 77,156 77,606 276 412 Avondale Arizona 76,238 76,238 76,392 77,518 277 413 Bryan Texas 76,201 76,201 76,525 77,321 278 415 St. Jose ph Missouri 76,780 76,803 76,658 77,185 279 417 Bloomington Illinois 76,610 76,610 76,735 77,071 280 418 Lawrence Massachusetts 76,377 76,377 76,514 76,976 281 420 Meridian Idaho 75,092 75,092 75,308 76,750

PAGE 320

! 300 282 423 Lynchburg Virginia 75,568 75,568 75,716 76,504 283 424 Palm Coast Florida 75,180 75,176 75,430 76,499 284 425 Deerfield Beach Florida 75,018 75,018 75,191 76,389 285 426 Kennewick Washington 73,917 73,917 74,461 76,224 286 427 Decatur Illinois 76,122 76,122 76,115 76,096 287 428 Melbourne Florida 76,068 76,068 76,131 76,095 288 429 Scranton Pennsylvania 76,089 76,089 76,084 75,995 289 430 Baldwin Park California 75,390 75,390 75,453 75,940 290 431 Chino Hills California 74,799 74,799 75,053 75,928 291 434 Bethlehem Pennsy lvania 74,982 74,982 75,068 75,266 292 435 Mountain View California 74,066 74,066 74,262 75,235 293 436 Fayetteville Arkansas 73,580 73,580 73,921 75,102 294 440 St. George Utah 72,897 72,897 73,107 74,770 295 441 Kalamazoo Michigan 74,262 74,262 7 4,352 74,743 296 446 New Britain Connecticut 73,206 73,206 73,215 73,261 297 447 Appleton Wisconsin 72,623 72,623 72,768 73,243 298 448 Folsom California 72,203 72,203 72,346 73,001 299 449 Canton Ohio 73,007 73,014 72,978 72,919 300 454 Gastonia North Carolina 71,741 71,741 71,780 72,068 301 455 Plymouth Minnesota 70,576 70,576 70,678 71,561 302 456 Auburn Washington 70,180 70,180 70,376 71,517 303 457 Rochester Hills Michigan 70,995 70,995 71,013 71,452 304 458 Springdale Arkansas 69,797 69,792 70,157 71,397 305 459 Wilmington Delaware 70,851 70,852 70,920 71,305 306 461 Pawtucket Rhode Island 71,148 71,148 71,163 71,153 307 462 Waukesha Wisconsin 70,718 70,718 70,740 70,867 308 463 Jacksonville North Carolina 70,145 70,145 70,719 70,801 309 464 Union City California 69,516 69,516 69,645 70,436 310 467 Muncie Indiana 70,085 70,085 70,091 70,080 311 469 Passaic New Jersey 69,781 69,781 69,835 69,893 312 470 Missouri City Texas 67,358 67,358 67,964 69,774 313 472 Mount Pleas ant South Carolina 67,843 67,874 68,097 69,357 314 473 Gulfport Mississippi 67,793 67,793 68,060 69,220 315 474 Rapid City South Dakota 67,956 67,956 68,236 69,200 316 475 Turlock California 68,549 68,549 68,677 69,089 317 477 Iowa City Iowa 67,862 67,873 68,051 68,947 318 479 Waterloo Iowa 68,406 68,406 68,462 68,653 319 480 Santa Fe New Mexico 67,947 67,943 68,098 68,642 320 481 Jonesboro Arkansas 67,263 67,261 67,442 68,547 321 482 Warner Robins Georgia 66,588 66,588 67,013 68,500 322 484 Manteca California 67,096 67,171 67,404 68,254 323 485 Loveland Colorado 66,859 66,859 67,070 68,203 324 487 Lafayette Indiana 67,140 67,140 67,190 67,947 325 491 Missoula Montana 66,788 66,788 66,873 67,290 326 493 Temple Texas 66,102 66,130 66, 691 67,188 327 494 Union City New Jersey 66,455 66,455 66,529 67,187 328 497 Eau Claire Wisconsin 65,883 65,888 65,978 66,623 329 499 Youngstown Ohio 66,982 66,982 66,846 66,571 330 501 Portland Maine 66,194 66,194 66,125 66,363

PAGE 321

! 301 331 502 Oshkosh W isconsin 66,083 66,083 66,119 66,344 332 504 Schenectady New York 66,135 66,135 66,221 66,273 333 505 St. Cloud Minnesota 65,842 65,842 65,902 66,169 334 507 Harlingen Texas 64,849 64,869 65,140 66,122 335 508 Davis California 65,622 65,622 65,705 66,016 336 509 Dothan Alabama 65,496 65,496 65,637 66,010 337 510 Flagstaff Arizona 65,870 65,870 65,956 65,914 338 513 Palo Alto California 64,403 64,403 64,571 65,412 339 516 Jackson Tennessee 65,211 65,211 65,227 65,187 340 518 Yuba City Calif ornia 64,925 64,925 65,009 65,050 341 520 Eagan Minnesota 64,206 64,206 64,303 64,765 342 526 Lorain Ohio 64,097 64,097 64,135 64,152 343 528 Johnson City Tennessee 63,152 63,150 63,333 63,815 344 530 Fort Myers Florida 62,298 62,296 62,460 63,512 345 531 Janesville Wisconsin 63,575 63,575 63,558 63,479 346 534 Pasco Washington 59,781 59,781 60,600 63,186 347 535 Lodi California 62,134 62,134 62,350 63,133 348 536 Victoria Texas 62,592 62,592 62,629 63,131 349 542 Bismarck North Dakota 61,2 72 61,274 61,575 62,665 350 544 Council Bluffs Iowa 62,230 62,228 62,377 62,466 351 547 Madera California 61,416 61,416 61,650 62,219 352 548 Utica New York 62,235 62,235 62,233 62,110 353 549 Homestead Florida 60,512 60,512 60,673 61,940 354 550 Coon Rapids Minnesota 61,476 61,476 61,593 61,904 355 551 Eden Prairie Minnesota 60,797 60,797 60,885 61,657 356 553 Haverhill Massachusetts 60,879 60,879 60,987 61,351 357 556 Waltham Massachusetts 60,632 60,632 60,717 61,181 358 558 Gaithersburg Maryland 59,933 59,933 60,160 61,045 359 559 Daytona Beach Florida 61,005 61,005 61,011 61,028 360 560 Terre Haute Indiana 60,785 60,785 60,807 60,961 361 561 Vineland New Jersey 60,724 60,724 60,810 60,952 362 563 Marysville Washington 60,020 60, 020 60,191 60,785 363 569 Encinitas California 59,518 59,518 59,711 60,400 364 570 Greenville South Carolina 58,409 59,139 59,390 60,379 365 571 Santa Cruz California 59,946 59,948 60,049 60,342 366 572 Springfield Ohio 60,608 60,608 60,554 60,333 367 575 Cheyenne Wyoming 59,466 59,483 59,739 60,096 368 576 Malden Massachusetts 59,450 59,450 59,563 60,071 369 577 Lancaster Pennsylvania 59,322 59,322 59,693 60,058 370 581 Springfield Oregon 59,403 59,403 59,431 59,695 371 586 Cupertino Cali fornia 58,302 58,302 58,455 59,220 372 587 Ames Iowa 58,965 58,965 58,981 59,042 373 589 Great Falls Montana 58,505 58,579 58,711 58,950 374 590 Bowling Green Kentucky 58,067 58,062 58,242 58,894 375 592 Grand Junction Colorado 58,566 58,564 58,402 5 8,704 376 593 Des Plaines Illinois 58,364 58,364 58,425 58,617 377 595 Petaluma California 57,941 57,941 58,041 58,453 378 598 Dubuque Iowa 57,637 57,637 57,778 58,234 379 603 Idaho Falls Idaho 56,813 56,813 57,022 57,646 380 606 Owensboro Kentuc ky 57,265 57,267 57,323 57,605 381 610 Rocky Mount North Carolina 57,477 57,475 57,482 57,433

PAGE 322

! 302 382 613 White Plains New York 56,853 56,853 56,923 57,258 383 619 Berwyn Illinois 56,657 56,657 56,718 56,906 384 623 Ocala Florida 56,315 56,315 56,333 5 6,517 385 628 Valdosta Georgia 54,518 54,763 54,978 56,019 386 629 Casper Wyoming 55,316 55,316 55,333 55,988 387 630 Anderson Indiana 56,129 56,129 56,136 55,957 388 632 Taunton Massachusetts 55,874 55,874 55,899 55,940 389 634 Decatur Alabama 5 5,683 55,683 55,738 55,902 390 635 Woodland California 55,468 55,468 55,539 55,806 391 636 Novi Michigan 55,224 55,224 55,238 55,583 392 638 New Brunswick New Jersey 55,181 55,181 55,234 55,444 393 639 Carson City Nevada 55,274 55,274 55,212 55,439 394 645 Porterville California 54,165 54,165 54,326 55,023 395 647 Pocatello Idaho 54,255 54,255 54,373 54,810 396 648 Corvallis Oregon 54,462 54,461 54,428 54,674 397 649 Elyria Ohio 54,533 54,533 54,564 54,581 398 650 Auburn Alabama 53,380 53,3 91 53,587 54,566 399 653 Hanford California 53,967 53,967 54,050 54,284 400 660 Port Arthur Texas 53,818 53,818 53,840 53,937 401 663 Highland California 53,104 53,104 53,286 53,903 402 664 Delano California 53,041 53,041 53,402 53,819 403 665 Manh attan Kansas 52,281 52,281 52,551 53,678 404 666 Noblesville Indiana 51,969 51,968 52,321 53,515 405 670 Colton California 52,154 52,155 52,331 52,940 406 671 Lake Havasu City Arizona 52,527 52,532 52,575 52,935 407 673 Blue Springs Missouri 52,57 5 52,575 52,637 52,749 408 674 Grand Forks North Dakota 52,838 52,838 52,903 52,631 409 680 Sarasota Florida 51,917 51,983 52,060 52,341 410 682 Pensacola Florida 51,923 51,923 51,986 52,197 411 683 Florissant Missouri 52,158 52,158 52,149 52,145 412 684 Yucaipa California 51,367 51,371 51,546 52,139 413 687 Battle Creek Michigan 52,347 52,347 52,322 52,093 414 688 Smyrna Georgia 51,271 51,271 51,396 51,982 415 690 La Crosse Wisconsin 51,320 51,320 51,419 51,719 416 691 Peabody Massachuset ts 51,251 51,251 51,344 51,653 417 693 Watsonville California 51,199 51,199 51,300 51,586 418 696 Elkhart Indiana 50,949 50,964 50,962 51,320 419 697 Bellevue Nebraska 50,137 50,137 50,405 51,319 420 700 Saginaw Michigan 51,508 51,508 51,455 51,230 421 701 Milford c Connecticut 51,271 51,271 51,268 51,189 422 702 Charleston West Virginia 51,400 51,371 51,351 51,177 423 703 Perth Amboy New Jersey 50,814 50,814 50,870 51,093 424 704 Burlington North Carolina 49,963 50,201 50,336 50,925 425 705 Albany Oregon 50,158 50,158 50,222 50,724 426 706 Joplin Missouri 50,150 50,150 50,277 50,559 427 711 Bradenton Florida 49,546 49,546 49,639 50,193 428 712 Troy New York 50,129 50,129 50,140 50,120 429 713 Niagara Falls New York 50,193 50,193 50, 211 50,086