Evaluation of local economic development incentives

Material Information

Evaluation of local economic development incentives the case of Colorado Enterprise Zone program
Sonmez, Zafer
Publication Date:
Physical Description:
ix, 176 leaves : ; 28 cm


Subjects / Keywords:
Enterprise zones -- Evaluation -- Colorado ( lcsh )
Industrial promotion -- Evaluation -- Colorado ( lcsh )
Economic policy -- Evaluation ( fast )
Enterprise zones -- Evaluation ( fast )
Industrial promotion -- Evaluation ( fast )
Economic policy -- Evaluation -- Colorado ( lcsh )
Colorado ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 156-164).
General Note:
Department of Urban and Regional Planning
Statement of Responsibility:
by Zafer Sonmez.

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:
255559577 ( OCLC )
LD1193.A78 2008m S66 ( lcc )

Full Text
Zafer Sonmez
B.A., Gazi University, 2005
A thesis submitted to the
University of Colorado Denver
in partial fulfillment
of the requirements for the degree of
Master of Urban and Regional Planning

This thesis for the Master of Urban and Regional Planning
degree by
Zafer Sonmez
has been approved by
Thomas A. Clark
Edward Kamp

Sonmez, Zafer (Master of Urban and Regional Planning)
Evaluation of Local Economic Development Incentives: The Case of Colorado
Enterprise Zone Program
Thesis directed by Professor Thomas A. Clark
This thesis examines the evaluation of local economic development policies and
business development incentives. Because program evaluation remains one of the
important questions in local economic development, a significant part of the study
has been devoted to review of discussions regarding evaluation of local economic
development in general and business incentives in particular. Enterprise zone
programs are among the most commonly used economic development tools by
state and local governments in the United States since their emergence in the early
1980s. Enterprise zone programs and their impacts on local economies constitute
the core of this study because even after a large number of enterprise zone
evaluations, the conclusions are still generally not definitive and often do not
directly inform policy. Previous research, for instance, has focused greatly on
urban enterprise zones yet few studies have examined rural enterprise zones
extensively. Furthermore, long term-assessment of impacts of enterprise zones is
also lacking in the literature. Consequently, this study critically examines
previous evaluation studies of enterprise zones and, in order to fill the gaps in the
literature, it analyzes long term impacts of Colorados enterprise zone program on
economic development of rural areas. Results of longitudinal analysis suggest that
the enterprise zone program improves economic development in the short term, 5-
10 years, but this impact is not evident in the long term, after 10 years. Cross-
sectional analysis of rural zone regions and survey research of zone directors
imply a positive relationship between aggressive zone management and
effectiveness of the program. Longitudinal analysis gives some clues about
progressive stages of economic development in small towns as well.
This abstract accurately represents the conten^^fjJhe_^andidaufs
recommend its publication.
lomas A. Clarl

1. INTRODUCTION.............................................1
General Background....................................1
Purpose of the Study..................................3
Scope of the Study....................................4
Research Questions and Significance of the Study......4
Overview of the Study.................................4
Introduction......................................... 6
Justifications for Evaluation of Local Economic Development
Arguments against Evaluation of Local Economic Development
Basic Concepts.......................................11
Critical Understandings in Evaluation................15
Growth versus Development......................16

Fiscal Health of City versus Economic Health of City.19
Efficiency versus Effectiveness......................20
Competitive Advantage versus Comparative Advantage..21
The Anatomy of Success...............................21
Approaches to Evaluation of Local Economic Development
A Critical Review of Economic Development Evaluations......30
Measurement of Success of LED through Evaluative
Limitations of Past Evaluation Studies and New Approaches..43
A Critical Evaluation of Local Economic Development
A New Evaluation Framework...........................49
Increasing the Use of Results of Evaluation Studies..54
Economic Development Incentives............................57
History of Economic Development Incentives...........58
Arguments for Economic Development Incentives........61
Arguments against Economic Development Incentives...64
Methods to Evaluate Effects of Economic Development
Future of Economic Development Incentives..........75

Review of the Enterprise Evaluation Literature.......87
Critique of the Current Literature...................96
Methodology and Study Design........................100
Case Study: Colorado Enterprise Zone Program........104
Longitudinal Analysis........................117
Quasi-Experimental Control Group Method......134
Cross-Sectional Analysis and Survey Research.138
4. CONCLUSIONS...........................................148

2.1: Role progression in capacity building..................................18
2.2: Payoffs from competitive subsidization.................................69
2.3: Why local governments pursue offering incentives.......................70
3.1: Colorado enterprise zones.............................................107
3.2: Relationship between economic development score and degree of zone

2.1: Traditional indicators versus indicators for critical evaluation..46
3.1: Colorado enterprise zones legislative history.....................106
3.2: Enterprise zone tax credits and incentives........................Ill
3.3: Population by selected zone counties, 1980-2000...................119
3.4: Annual per capita personal income by selected zone counties, 1980-2000..123
3.5: Employment by selected zone counties, 1980-2000...................125
3.6: Occupational wages by selected zone counties, 1980-2000...........129
3.7: Selected economic and social indicators by selected zone counties, 1990-
3.8: Employment by Industry in Chaffee and Teller Counties, 1987.......135
3.9: Population, employment, income and occupational wages in Chaffee and
Teller Counties, 1980-2000.............................................137
3.10: Population and employment by selected zone counties and regions, 1990-
3.11: Income and wage by selected zone counties and regions, 1990-2000.141
3.12: Economic development scores of regions...........................142
3.13: Results of survey of enterprise zone region directors............144

I thank my supervisor Thomas A. Clark for his invaluable suggestions and
comments throughout the study. Without his guidance, this study could not have
been completed. In addition, for his commitment to science and education, he
deserves to be remembered as a symbol for generations to come. I thank Edward
Kamp too for his comments on my study. I also must mention my thanks to the
Fulbright which gave me an opportunity to study in the U.S. Without its support
neither could I pursue a graduate study nor complete this thesis. After finishing
this study, I discovered that I needed to know far more about economic
development planning than I did before. This realization will help me to fill the
many gaps in my knowledge in the future.

General Background
Program evaluation has been labeled one of the big questions in planning
and public management (Hill, 2000 in Reese and Fasenfest, 2004). Although
the crusade for economic development endures as the central concern of local
governments in the United States and a vast amount of political and economic
resources continue to be directed towards achieving this goal, the careful and
systematic evaluation of economic development policy has been largely
eschewed (Imbroscio, 2004).
Jackson (2001) points to the lack of theoretical thinking in evaluation and
explains how they have led to two major problems in economic development.
First, policy managers have given insufficient attention to what they wish to
achieve. Second, they have given insufficient attention to how this effect will
be achieved and how their proposed policy would deliver this effect. Bartik
(1994 in Tao and Feiock, 1999) shares similar concerns with Jackson by
arguing that the capacity for sophisticated evaluation at the local level is still
underdeveloped in many places and there has been little guidance for those
seeking to link inputs to outcomes. In a very practical way, the lack of
measurable outcomes can feed a growing sense that governments are reneging
on promises made. Thus, evaluations that link policy to outcomes may be
viewed as a long-term political necessity.

Increasing the importance of evaluation in local economic development
(LED) and its urgency as a key research requirement started to be discussed in
the literature during the last several decades. The increased awareness of the
problems in effective targeting as well as the sensitivity of government
departments to charges of efficiency have created an imperative to evaluate
existing and emerging policies (Haughton, 1990). Today, rigorous evaluation
is both justified and necessary to demonstrate the value of expenditures to the
tax payer at local and national levels (Hughes, 1991).
Being one of the most used economic development tools by state and local
governments enterprise zone programs constitute an important place in the
academic discussions. As in the case of many other economic development
policies, there is no consensus in the literature whether or not enterprise zones
are effective tools of economic development. According to advocates of the
concept, enterprise zones help reduce economic development disparities
among cities and are generators of economic activity, rather than just
distributors of that activity (Rubin and Richards, 1992). For example, recently
Peters and Fisher (2002) in their case study of 75 cities in 1994 found that
enterprise zone incentives reduced the economic development disparity
among cities.
Opponents of the concept argue that implementation of enterprise zones in the
United States has not been a success. They assert that U.S. economic-
development policy, when evaluated from a national perspective, is a zero-
sum game because economic development policy has not generated net new
American investment; it merely moves investment around. Moreover, it has a
negative net impact on the national economy (economic development policy
distorts the workings of private location decisions). Indeed, there is a

consensus in the literature that a business chooses a location when primary
factors such as proximity to markets, labor, infrastructure and accessibility
exist. Moreover, it is argued that state and local taxes account for such a small
share of spatially variable business costs that their influence is swamped by
other larger costs, such as labor and transportation and federal taxes, which
massively reduce the actual variability of state and local taxes (Peters and
Fisher, 2002). In addition, economists question government policies designed
to encourage large corporations to increase employment rolls (Wolf, 1990).
Critics further argue that enterprise zones do not have magical qualities that
can overcome all physical, social, and economic barriers to revitalization. The
myriad social and physical problems plaguing many urban neighborhoods
(e.g. decaying infrastructure, high crime rates, and inadequate school systems)
are suggested not responsive to targeted development incentives. Enterprise
zone areas that suffered severe economic blight require much more assistance
than an enterprise zone can provide (Wilder and Rubin, 1996).
Discussions above indicate that enterprise zone evaluation studies are not
conclusive yet and continue as to whether area-base incentives are an effective
tool of economic development. Findings from the existing enterprise zone
literature are difficult to translate into useful policies, and conflicting results
limit the ability to generalize them (Peters and Fisher, 2007; Bondonio and
Greenbaum, 2007). In short, analysis of the enterprise zone programs is still
seen essential for improved public policy.
Purpose of the Study
The purpose of this study is to discuss issues in evaluation of local economic
development, critically review previous evaluation studies of enterprise zones

from a larger economic development perspective, and finally, to evaluate
Colorados enterprise zone program and its impacts on economic development
of rural areas.
Scope of the Study
This study reviews the literature on local economic development policies and
business development incentives in the United States. It reviews previous
enterprise zone evaluation studies. As a case study, it evaluates Colorados
enterprise zone program and its impacts on economic development of rural
areas. Analysis includes 19 counties of Colorado that are completely
designated as an EZ and having population of less than 35,000 or more than
Research Questions and Significance of the Study
This study focuses on long term economic development impacts of rural
enterprise zone programs in small towns in order to fill the existing gap in this
area of the literature. It tries to answer two questions: (1) How do enterprise
zone programs affect economic development of rural areas? (2) Does
aggressive management enhance impact of rural enterprise zones? Answers of
these questions not only will help understand reasons of policy success and
failures in different areas but also will help local and state governments make
more effective economic development policies in the future.
Overview of the Study
This chapter has provided the background to this study and its objectives and
scope. Chapter 2 provides a theoretical and empirical framework for the study
by reviewing current literature on evaluation of local economic development
policies and effectiveness of business incentives. In this chapter, some basic
and critical concepts are identified briefly and then evaluation methodologies

are discussed in detail. First part of this chapter ends with developing a new
evaluation framework and several suggestions to increase the use and
effectiveness of evaluation studies by practitioners. Second part of chapter 2
looks at the historical development of economic development incentives in the
U.S. and reviews methods of evaluating effects of economic development
incentives. Last part of this chapter discusses problems of economic
development incentives in the context of market failure and changing context
of economic development and concludes with some policy recommendations.
Chapter 3 focuses on enterprise zone programs and their evaluations. First of
all, a brief background is provided and then previous studies of enterprise
zones are reviewed. This chapter explains the methods used for conducting the
research and for the analysis of the data in this study. It describes scope of
analysis, data types for analysis, and its sources. Finally, it evaluates the
Colorados enterprise zone program and discusses results of longitudinal
analysis, quasi-experimental control group method, and cross-sectional
analysis and survey research.
Chapter 4 summarizes the research findings, draws conclusions from those
findings, and indicates policy implications of these findings. Limitations of
the study and suggestions for further research are considered.

The purpose of evaluation is to provide continuous feedback in ever fine detail
about the conditions under which policy succeeds or fails (Jackson, 2001).
Rossi and Freeman (1985 in Bartik, 1995: 1) define evaluation as the
systematic application of social research procedures in assessing the
conceptualization and design, implementation, and utility of social
intervention programs. Turok (1990, in Foley, 1992: 557) has defined the
same concept more broadly as monitoring and assessment of the impact and
effectiveness of policies and programs. On the other hand, Reese and
Fasenfest (1997: 282) describe it as the efficiency of development efforts as
the relationships between inputs-the resources such as budget, staff time,
personnel that are allocated toward economic development and outputs- jobs,
businesses, tax-bases. A broader definition than above was developed by
Cave et al. (1990 in Jackson 2001: 7). They explain evaluation as focus on
the three Es: economy, (the extent to which the inputs are minimized),
efficiency (the relationship of the output of an activity or organization to the
associated inputs), and effectiveness (the extent to which outputs contribute
final objectives).

Justifications for Evaluation of Local Economic Development Policies
The importance of evaluations in LED is justified by Bovaird (1994: 599) as
the following: The development of a richer understanding of both policy
success and failure in different LED programs is needed to enable a more
balanced judgment of the role and limitations of LED, both at national and
local levels. Bovaird (1993) argues that since important failures have
occurred, both in LED policies and in service delivery, they need to be
understood and lessons should be learned from them. In another study Bovaird
(1992) argues that the diverse paradigms from which LED is currently being
studied mean that there is an oversupply of hypotheses relative to available
evidence which can be used to test them. There is a clear need for more
operational formulations of many of the conceptual frameworks. For example,
Rhodes et al. (2005) discuss inadequate understanding of the theory of change
behind the underlying policy intervention, insufficient development of
evaluation theory, and method constrained to assess the effect of area based
initiatives on local area outcomes in England.
Foley (1992) approaches the necessity of evaluation from a different
perspective by arguing that resources for future generations are restricted and
it is therefore important to target assistance through schemes which have the
greatest impact on particular disadvantaged groups in a locality. Restricted
resources can only be used in the most cost-effective manner if there is some
measure of the relative efficiency of different schemes. Since the relative
efficiency of different schemes can only be examined through evaluation, it is
likely the number of evaluation studies will continue to rise in the future.
Evaluation studies also can provide a useful input to the policy-making and
day-to-day management process. Evaluation may not always provide the

results policy makers would like to hear, but by learning from results it should
be possible to reduce the replication of mistakes and increase the benefits
(Foley, 1992). Bartik (1995) suggests that the idea of high-quality evaluation
as an essential part of carrying out high-quality economic development
programs should be encouraged so that development professionals really
know what works and what doesnt work in economic development.
Van der Knaap (1995, in Jackson, 2001) shows how evaluation can contribute
to organizational learning. He breaks learning into three separate components
each of which benefits from evaluation.
Corrective learning. Evaluation can provide feedback showing
whether implementation is in line with plans.
Cognitive development through a process of refining schemata.
Evaluation can provide stimulus and inspiration.
Social learning through dialogue and debate. Evaluation can provide a
forum for communication and the development of alliances.
As discussions above demonstrate, although policy evaluation in LED is
necessary, evaluation studies are limited. Several reasons have been suggested
for the paucity of empirical evaluations of economic development polices:
Robust evaluations are extremely difficult to design and conduct;
Evaluations are time and resource intensive and are often the first
activities to go in periods of public-sector fiscal retrenchment.
Randomized control groups in particular raise political and ethical

Extant evaluations typically do not suggest how programs can be
improved so are of little practical use to policy makers;
Public-sector auditing agencies that often conduct evaluations do not
tend to use rigorous methodologies; and, probably most important,
Policy makers fear the political repercussions of possible negative
evaluation (Bartik and Bingham, 1997 in Reese and Fasenfest, 2004).
In spite of these difficulties, the number of evaluation studies has been
increasing. According to Foley (1992), the growing level of recent interest in
economic development policy evaluation has been founded on a number of
factors. Primary of these are accountability, understanding the impact of
policy, cost-efficiency and cost-effectiveness. To Willis (1983 in Foley,
1992), major reason for the increase in evaluation studies is a realization
amongst policy-makers that they do not always understand the effects of
policies. The range of initiatives, both formal and informal, is well known to
policy-makers, but much less is known about the impact of these policies on
economic development.
Arguments against Evaluation of Local Economic Development Policies
Despite the seemingly strong arguments in favor of evaluation, it is worth
considering briefly the challenges facing evaluation and the reasons why it
may be unwise to consider evaluation of economic development initiatives.
Storey (1990 in Foley, 1992) proposed three arguments against evaluation.
First, the diversity of initiatives is so great that no evaluation system could
take them into account. This argument is countered by the fact that if clear
policy objectives are defined it should be possible to measure the
effectiveness of policy in achieving these goals. Second, the problem of
unemployment is so serious that efforts should be directed exclusively to

solving the problem. The third argument is that many policy initiatives are so
new that evaluation would be premature.
Bartik (1995) argues that so many external variables exist that it is virtually
impossible to determine the real impact on social problems resulting from
policy intervention. Likewise, Keating (1995) argues that it is often
impossible to disentangle cause and effect in economic development. For
example, do cities with a higher education level attract industry, or does the
presence of industry attract a more highly educated population and, therefore,
provide the resources for higher spending levels on education? Is education a
development input or a social output? Do high tax incentives deter industry, or
do cities with a poor industrial base have to have higher tax rates because of a
low tax base? It is also argued that cities cannot be measured against each
other because there are too many uncontrolled variables. Equally important
are the social and political effects of policy, which must be taken into account
on appraisal. Indeed, these considerations may be the most important. Nor
can we see cities as driven inexorably by the logic of global competition to
adopt identical policies. On the contrary, there is ample evidence that the
politics of cities is becoming more complex and that non-market factors may
impinge on the definition of development path which it takes (Swanstorm,
1993 in Keating, 1995: 19). Keating (1995) argues that rather than view
development purely as policy, we must also see it as politics.
These arguments are supported by a recent study of Reese and Rosenfeld
(2002). After analyzing several cities in Michigan, Ohio, and Ontario these
researchers concluded that economic development decisions and the results of
these decisions are part and parcel of the larger community fabric they are
inherently socially embedded. To understand development decision making

through an economic or institutional lens is very difficult. Disparate aspects of
politics and process may be accurately reflected, but the overall picture is
fragmented and ultimately misleading.
Level of analysis posses another challenge for evaluation studies. Identifying
and choosing the level of analysis requires awareness that area, citywide, or
gross policy outcomes may be very different than targeted, neighborhood, or
net outcomes. Policies that may appear to fail at the macro level may well
be successful at the micro level. In this sense a decision about level of analysis
reflects a value choice requiring specification of exactly whose benefits serve
as the evaluative criteria (Reese and Fasenfest, 2004). As it will be discussed
later in this chapter, there are methods developed to assess impacts of
economic development programs at the micro (firm level) and macro level
(community or city level).
Finally, when the outcomes of interest shift from the relatively easily
quantifiable measuring of job and tax-base growth to variables such as
income distribution, changes in poverty, job security, employment change by
sector, and political empowerment, the challenges multiply exponentially
(Reese and Fasenfest, 1999: 3-8).
Basic Concepts
Before discussing evaluation issues, approaches, and methodologies, it is
necessary to briefly discuss some of the essential concepts in evaluation
studies to be able to understand easily discussions that follow. These are
deadweight, displacement, additionality, supplier multiplier, and income and
employment multiplier.

Deadweight: This phenomenon occurs when some or all of the intended
impact from a project would have occurred anyway for instance, a
partnership would have formed or a community business would have
employed the same numbers of persons despite the help given (Armstrong et
al., 2001).
Displacement: The occurring of displacement is explained by Turok and
Wannop (1990 in Foley, 1992: 580) as displacement is very visible when
firms relocate, but it arises more commonly when increased sales, output and
employment in assisted firms are achieved at the direct expense of sales,
output and employment in their unassisted competitors. The most common
way to present displacement values is to consider the proportion of local trade
lost (displaced) by competitors as a result of the activities of the assisted firm
(Foley, 1992). Similarly Armstrong et al. (2001: 686) state that displacement
occurs when support to establish a business in one area displaces activity from
competitor business. In these situations support will simply redistribute jobs
from one place to another. The issue of how to measure displacement is one of
the thorniest of all community economic development (CED). Implications of
policy from this result could be that economic policies should support
growing or basic/export oriented industries to minimize displacement so as to
increase the success of economic development programs.
Additionality: This concept is defined as; the amount of output from a
policy as compared with what would have occurred without intervention
(HM Treasury, 1988 in Jackson, 2001). This definition goes on to note that
additionality can apply to inputs as well as to outputs. Inputs in this context
refer to financial additionality. If financial assistance is provided to someone
to do something they would have done anyway, there is zero additionality

(also called deadweight spending) (Foley, 1992). It is obviously unrealistic to
assert that every job at a program or an initiative was created due to policy.
Jackson (2001) points the failure to isolate additionality in the past economic
development evaluations. The aggregate picture presented by indicators has
been accepted uncritically, without separating out effects which can rightfully
be attributed to policy.
Storey (1990 in Foley, 1992) notes three main approaches which have been
used to measure additionality. First is a comparison of the performance of
assisted companies with unassisted firms. However, this approach is difficult
to apply since obtaining a control sample to provide matched pairs of
companies with any degree of legitimacy is challenging. Second is to use
regression techniques to isolate the effects of assistance on financially-assisted
and non-assisted firms (Wren, 1989 in Foley, 1992). It is noted that this
technique is rarely possible due to paucity of data. The final approach
identified by Storey (1990 in Foley, 1992) involves the use of interviews with
firms or individuals receiving assistance. This form of analysis has been
widely used in many evaluation studies. Whilst some commentators have
correctly asserted that additionality depends in part upon the subjective
viewpoint (and honesty) of a respondent, efforts have been made in evaluation
studies to integrate the interviewers judgment of additionality to temper
unrealistic views (Foley, 1992).
Supplier Multiplier: Supplier multipliers are used in evaluation studies to
investigate the job-creation impact of additional purchases by companies from
other local companies. The impact of supplier multipliers is found by asking
companies what changes in their purchasing activity (of goods and services)
might arise from the assistance provided to the company (Foley, 1992). This is

an additional beneficial effect of CED. An intervention which helps the
community business to grow leads it to increase its demand for inputs the
suppliers of some of which will be located in the same area. With respect to
this concept it is important to note that smaller companies are likely to possess
low levels of external control and higher degrees of local purchasing, and as a
result they create high supplier ratios (Watts, 1981 in Foley, 1992). However,
purchasing of inputs from firms elsewhere can also be of great long term
benefit if it allows lower costs to be enjoyed and community businesses to
develop greater competitiveness in whatever markets they choose to compete
(Armstrong et al., 2001).
Income and Employment Multiplier: These are an additional effect. They
arise through CED intervention via increasing spending by households in the
local economy. These then generate additional income and jobs through the
various rounds of the Keynesian multiplier process (Haughton, 1998;
McCormick and West, 1998 in Armstrong et al., 2001). The basic theory of
the income multiplier is defined by Brownrigg (1974 in Foley, 1992: 582) as
the following: A money injection into an economic system, whether national
or regional, will cause an increase in the level of income in that system by
some multiplier of the original injection. Any money spent in an economy (or
area) will generate additional income in that area. A portion of this additional
income will be spent again in that area in a second round of spending. This
process is continuous, with many later rounds of spending. It only ceases to
have an impact when all subsequent income has leaked out of the economy,
through spending in other areas. One of the main influences on the size of a
multiplier is the level or proportion of the subsequent income flows (in later
rounds) which leak out of the economy after each round of expenditure. It
should be noted that leakages from the circular flow of income are larger the

smaller is the geographical area considered. This is potentially a serious
problem for those types of CED policies which rely on income and spending
being retained within the local community which is known as grassroots
approach to CED (Haughton, 1998; McCormick and West, 1998 in Armstrong
et al., 2001). If the local firms sales help to prevent local spending from
leaking away from the area (a process known as internal blocking of the
multiplier), the local orientation of their activities may be not bad thing
(Armstrong et ah, 2001).
Multiplier effects are usually calculated using input-output matrices (Willis,
1983 in Foley, 1992) and there is a need to possess accurate details about
social characteristics, consumption patterns, and the industrial structure of an
area. For example, franchise and corporate retail stores such as McDonalds
restaurants, were examined to demonstrate how these enterprises drained
income away from local communities; and community reinvestment advocates
began studying bank landing practices to determine if banks lent as much to
inner-city neighborhoods as the neighborhoods own deposits warranted
(Reese and Fasenfest, 2004). In general, both the economic democracy and
ghetto economic development literature developed an understanding of how
endogenous development strengthening internal community linkages could
rebuild and restructure community economies (Yin, 2004).
Critical Understandings in Evaluation
In this section, some concepts that are widely used in local economic
development discussions are defined and subtle differences between them,
which are believed to be crucial to fully understand evaluations of local
economic development policies in general and business incentives in
particular, are explained.

Growth versus Development
Growth-development distinction is of fundamental importance to economic
developers (Malizia and Feser, 1999). However, what is meant by local
growth and development is unclear and not well-specified. Is it job
growth, investment growth, output growth, productivity growth, or escalating
rates of new firm formation (Plummer and Taylor, 2001)?
It is argued that growth and development are not the same. Although
economic growth and economic development are related processes,
development is more fundamental. Economic development leads to and
sustains competitiveness; economic growth results from competitiveness
(Malizia and Feser, 1999). Growth involves changes in overall aggregates
such as height or weight, while development includes changes in capacities -
physical coordination, learning capacity, or ability to adapt to changing
circumstances (Herrick and Kindleberger, 1983 in Reese and Fasenfest, 2004).
Development supports the sequence from invention to innovation and then to
commercialization. It leads to higher levels of welfare in terms of (1) income
level, (2) income distribution, and (3) income stability (Malizia and Feser,
1999). Development increases the quality of our overall life. Economies are a
factor in that development, but not the only one. Growth brings an increase in
our economic resources, but not always strengthening of our community
resources, or even of our economic structures. Growth can occur with
negative effects of the quality of personal and community life. Witness the
decline of the human environment with growth; and witness also, in the
growth of some metropolitan areas, the deterioration of the personal and
communal structures of life in the central and inner cities (Jeep, 1993).
Eocalities experiencing economic development over time should be able to

sustain economic growth that leads to improvements in per capita income
levels and to less unemployment and stability.
In the local development context, the goal of economic growth traditionally
has meant getting the local community back to work and building the local tax
base at the lowest cost to the public. Such goals have driven communities and
evaluations for years because job creation and tax-base growth usually means
votes. Moreover, citizens expect governmental officials to attract business
(Rubin and Zorn, 1985 in Reese and Fasenfest, 2004).
The practice of economic development in the U.S. is concerned primarily with
achieving economic growth at particular locations. As a result, economic
development, as a process, is often reduced to and equated with economic
growth (Malizia and Feser, 1999). However, focus of economic development
has changed in recent years. The development concept has broadened in
scope, with attention shifting to broad community economic development,
rather than being limited to jobs created and businesses started. Quality of life,
public services, and education opportunities are often included in community
development efforts (Walzer, 1995). In this context LED goals are defined as
the following: (a) increase the stability as well as the gross level of income for
the population; (b) increase local control over both market and government
operations, particularly in those aspects that affect citizens in poverty; and (c)
increase economic and political empowerment of all sectors of a community
(See Fig. 1), including individual citizens (Brown & Warner, 1991 in Reese
and Fasenfest, 1997). Reese and Fasenfest (1997) add to these the following:
long-range improvements in attractiveness and, ultimately, the quality of life
in the community.

Group as beneficiary/user of service
Group as a consultant and representative of local opinion
Group as a source of project ideas/focus for activity
Group as a deliverer of service and generator of economic
Group as a long-term (strategic) service partner
Proactive Supplier
Figure 2.1: Role progression in capacity building (Ekos in Armstrong et al.,
Ray (2000) suggests that local development initiatives should be seen as, in
essence, entities that are interpreting, learning and changing over time.
Development, according to this view is an on-going dialogue, experiment or
multiple explorations; the logic inherent is that the way in which development
is understood and the way in which development action is designed must
recognize and use this locally specific dynamics. Finally, it is important to
note that economic development is a long-term process of change with the

goals of increased employment, income, and community wealth. Yet, many
communities continue to look for quick results (Daniels, 1989).
Fiscal Health of City versus Economic Health of City
Enhancing the tax base is a common objective as most local taxes are based
on property values and this suggests policies to increase land and property
values and to develop unused land. It is not always necessary to increase the
level of economic activity, employment, or revenue flow in order to achieve
this. Operations which displace small enterprises with low values by high-
valued development may even destroy employment while raising the tax base.
Therefore a distinction must be made between the fiscal health of a city and its
economic health (Ladd and Yunger, 1988 in Keating, 1995). Further, unless
the enhanced assessment base is tilted towards externally owned businesses,
the effect of raising assessed values is merely to extract more revenue from
the income generated within the city; in other words, it is equivalent to an
increase in tax rates (Walzer, 1995). Economic health of a city is important,
because it is argued that projects in some cities may not achieve their stated
objectives simply because of the poor economic health of the city; whereas in
other cities, a robust economy carries even the riskiest projects (Pagano and
Bowman, 1992).
It is also important to briefly state the difference between city economic
health and health of city residents. Evaluation research has often
confused city economic health with the health of city residents. The
former, measured by the number of private jobs per resident, but is not the
same as the economic health of city residents which is measured by their per
capita income (Ladd and Yinger, 1989 in Reese and Fasenfest, 2004: 5).

Efficiency versus Effectiveness
It is very necessary to establish the critical difference between efficiency and
effectiveness since they are very commonly used in evaluation studies.
Efficiency is to maximize output (per given unit of input) or maintain the
same level of output with reduced input costs (Bamekov et al., 1990 in Foley,
1992: 561). On the other hand, effectiveness is the extent to which policy
objectives are achieved (HM Treasury, 1988 in Foley, 1992: 561). In local
economic initiatives, this generally means the overall effectiveness of the
initiative in relation to economic objectives such as the number of jobs created
or retained. At the firm level, efficiency increases with increased productivity
and the realization of internal economies of size and scale. At the industry
level, efficient production is reflected in productivity gains, usually through an
industrial complex or district that results in localization economies. For the
entire local economy, efficient production depends upon competitive
advantage which, at any given time, is a factor of the functional specialization
(s) of the areas industrial structure. Efficient production can lead to sustained
economic growth and, possibly, to greater stability (Malizia and Feser, 1999).
The difference between the two is subtle, but important since it leads to the
final and the most important concept in current evaluation literature: cost-
effectiveness. The difference between effectiveness and efficiency is clearly
demonstrated using two examples (HM Treasury, 1988 in Foley, 1992). An
initiative may be very efficient in promptly paying grants to eligible
individuals, but the grants may not have the desired affect on their behavior.
Conversely, an initiative may be highly effective in providing jobs to
everyone who enters a scheme, but if the cost of the initiative was very high
per person it would not be good value for money and not efficient. In short,
effectiveness is defined without reference to costs while efficiency is defined

with reference to costs, and the concept of efficiency and effectiveness can
both be applied to the same final objective.
Competitive Advantage versus Comparative Advantage
Developed in the context of international trade, comparative advantage, stress
the relative strength of an industry in its local economy1. Recently, it has been
argued that relative local advantages may not be sufficient for the vitality of a
local industry. Michael Porter stresses the concept of competitive
advantage, a focus upon a local industrys competitors throughout the larger
economy (1990 in Rosentraub and Przybylski, 1996: 315). Local industry has
a competitive advantage only if it has an advantage over others in its industry,
whereas comparative advantage is a relative concept focused upon local
industries. Competitive advantage encourages firms and communities to look
at all competitors throughout the world.
The Anatomy of Success
In order to gauge success in managing LED policy and initiatives, it is
necessary to have an agreed definition of what constitutes a success or a
high performance. This is not as unproblematic as many other commentators
on LED and urban policy have assumed. There is first the question of how to
define organizational performance; second, the issue of performance for
whom?, and, third, the need to position analysis of the overall service system
(to which the organization contributes) or within the context of the
contribution made by the local and wider social environment (Bovaird, 1994:
580). The judgment of success in policy or in implementation of programs and
initiatives need to address the issue of success for whom?. In the U.S., for
1 Specifically, an industry has a comparative advantage when the goods it produces have a
lower price with respect to other goods in its local economy versus its price on the
international market. The goods local relative price is lower than the goods world
relative price.

example, it is clear that most LED initiatives are not aimed at disadvantaged
groups. The U.S. surveys cited by Hambleton show that only a small
proportion of the cities (less than 10 per cent) view it as important to channel
economic benefits to the poorer areas of their city or to focus efforts on the
needs of economically disadvantaged residents (1990 in Bovaird, 1994: 581-
Approaches to Evaluation of Local Economic Development Policies
Effectiveness studies in LED have evolved more from systems analysis and
management accounting than from welfare economics. There have, for
example, been no influential attempts to apply a full social cost-benefit
analysis to LED. Rather, evaluation has focused on the actual achievement of
stated high-level LED objectives. This has led to some potentially serious
biases, especially the inadequate treatment of longer-term benefits and costs
and the neglect of wider social costs and benefits of programs. Few studies
have used discounted cash-flow techniques to deal with a future stream of
LED benefits and costs (Bovaird, 1992). This is especially likely to be
important in those urban regeneration initiatives which involve short-term
losses for longer term gain (e.g. employment losses through clearing inner-
city industrial areas in order to develop them for firms with a longer term
Local economic development evaluation studies have been undertaken with
the aid of either macro or micro-economic theory or techniques (Willis, 1985
in Foley, 1992). Ten years ago (and earlier, in fact), macro techniques
exemplified by shift-share, input-output and econometric studies
predominated. Shift-share studies were criticized for problems related to time-
scale and difficulties in determining the precise effects of multiple policies.

Input-output models were used to estimate direct and indirect job creation
through recreational expenditure (Archer, 1973; Henderson and Cousins,
1975; Vaughan, 1977 in Foley, 1992) and advanced factory building
(Hubbard, 1982 in Foley, 1992: 560). Econometric techniques were first used
in the U.S. on an experimental basis by the General Accounting Office
between 1974 and 1978 (Bamekov et al., 1990 in Foley, 1992: 560). Their
first study using this input-output method evaluated more than 50 federal
programs. This econometric model has been criticized on several grounds.
Critiques primarily focus its inability to distinguish differences between direct
and indirect jobs and the concentration of most input-output models on job
creation when other objectives also exist (Foley, 1992).
Although micro-economic theory and techniques have been used for some
time in the field of human-capital theory (Becker, 1964; Pissardes, 1981 in
Foley, 1992), Willis (1985 in Foley, 1992) argues that they are not so well-
known and used amongst planners, and those concerned with urban and
regional planning. However, intervening period has witnessed the adoption of
micro-economic techniques as the dominant method of economic policy
evaluation. These studies have all adopted a micro-economic approach based
on interviews with companies or other individuals involved with an initiative
(Foley, 1992).
According to Bartik (1995), evaluation methodologies can be divided into two
categories: process or formative evaluations and outcome, impact, or
summative evaluations. Process evaluations focus on how a program is
delivered. Impact evaluations focus on the programs results. Bartik suggests
that process and outcome evaluations should not be thought of as being
mutually exclusive alternatives, but as being complementary. The process

evaluation or a survey evaluation is subjective enough that it may be easier to
manipulate the evaluation process or reinterpret the results to make the
program look better; if study shows however, that firms participating in the
economic development program, compared to a truly comparable group, show
no difference in performance, then it is difficult to argue that the program
Rossi and Freeman (1993 in Reese and Fasenfest, 2004) distinguished three
types of evaluations: Process or formative evaluation; outcome, impact, or
summative evaluation; and efficiency evaluation or measurement of net
effects. Process or formative evaluation looks at program implementation,
monitor program operations, and often provides data to improve program
administration. For example, a formative evaluation of the empowerment
zones would look at which programs are operating and have expended funds,
and the extent of meetings involving community participation. Summative
evaluation assesses the utility of a program in achieving goals. Outcome and
impact evaluations look at the extent to which a program causes change in the
ways desired. An empowerment zone evaluation could look, for instance, at
the numbers of zone residents employed in zone businesses. Impact or
outcome evaluation also assesses whether the changes in outcome measures
are due to zone program activities; at whether, for instance, businesses hired
more residents of the empowerment zone because of zone programs.
Efficiency evaluation assesses whether the program is superior to obvious
alternatives in achieving the desired goals.
Another extensively used tool is the cost-benefit analysis (CBA), but it is
argued that failures to include opportunity costs make CBA practice most
hurtful to poor people. It is usually in benefits foregone that policy makers

exclude poor people (Buss and Yancer, 1999). Casson (1994 in Buss and
Yancer, 1999) argues that multilateral international donor organizations, by
using this method, often fail to help poor people and, in many cases, leave
poor people worse off.
Evaluations on the basis of achievement of objectives are faced with the
difficulty of LED: employment creation, reduction of unemployment levels
and employment effects on target social groups (typically women, the young
unemployed, inner city residents or ethnic-minority groups). Few studies have
looked at the more general economic growth affects of LED, in terms if
output growth or income generation. Hambleton (1990 in Bovaird, 1992: 361)
reports the findings of two major surveys of U.S. mayors and LED
professionals which showed that the aims of LED were most commonly cited
as: creating jobs, expanding or maintaining the tax base or tax revenues,
promoting downtown development, developing, and recruiting or attracting
new business.
Some other evaluative studies have focused upon perceived critical success
factors in the LED process, especially new firm formation, growth in
confidence in indigenous firms and improved image of the area in the eyes of
potential inward investors (Tricker and Bozeat, 1983; BiC, 1987; Kennet and
Bamford, 1988; Turok and Richardson, 1989 in Bovaird, 1992)
Although most evaluations assess programs or projects, local economic
development policy in the U.S. has always included assessments of the
aggregate performance of local economies. These macro evaluations are quite
common and usually focus on the extent to which local economy is doing

better than it did in the past or in comparison to the economy of a similar city
or metropolitan region (Beauregard, 1999).
Bartik (1995: 21) suggests 3 approaches to economic development evaluation.
First community evaluations, evaluations that directly estimate how some
economic development initiatives affect an overall community; second, firm
evaluations: evaluations that directly estimate how an economic development
program affects individual firms; third community impact evaluations:
evaluations that calculate the impact of some assumed change in firm on a
To Bartik, types of economic development evaluations include process
evaluation, before and after evaluation, survey evaluation, evaluation based on
models of firm behavior, comparison group evaluation, randomized control
group evaluation and evaluation that use models of community impact.
Another way to measure the effects of economic development programs is to
compare the performance of program participants with a comparison group of
non-participants. By definition, this comparison group will not be randomly
chosen, but there will be some effort to make the group comparable to the
program participants.
Comparative evaluations studies compare actual against alternative actual,
with against without, before against after, actual against expected or
actual against ideal (Berk and Rossi, 1990 in Jackson, 2001: 15). Pawson and
Tilley (1997 in Jackson, 2001: 16) argue that most powerful and practical
comparisons are between different versions of the same program. Authors
such as Chen (1990 in Jackson, 2001: 21), Pawson and Tilley (1997 in
Jackson, 2001: 21) argue that evaluation should move away from broad

questions about whether the program produced the impact intended, toward
more sensitive contextual analysis looking at where and why programs are
most effective. The where would look at environmental factors, types, and
motivations of participants, the why demands probing of program
mechanisms. As Pawson and Tilley, and Chen (1990 in Jackson, 2001) argue,
that refinement and analysis of the conceptual foundations of programs has
been neglected because methodologists have taken a black box approach
focusing on impact, looking at the overall relationships between the inputs
and outputs of a program without concern for the transformation processes in
the middle.
Another type of approach was developed by Reese and Fasenfest (2004).
Job chains approach to evaluating job creation provides a more explicit
focus on the extent to which economic development policies improve the
welfare of low and medium-skilled workers. It is not enough that policies
simply increase jobs indeed economic development programs do more than
just generate discrete jobs. Each job creates chain reactions in the market,
where residents move up and into jobs vacated by others. Thus the question of
policy success relates to the quality of the jobs, and who is getting them,
and how are the jobs impacting the local labor market more generally. Job
chain evaluations incorporate a more stringent definition of success because
not all jobs are equal; in terms of their contribution to local welfare, jobs at the
bottom of the chain. On the other hand, jobs going to nonresidents or
immigrants do not contribute local benefits and truncate the overall job
Turok (1990 in Foley, 1992) identifies four different evaluation categories that
encapsulate all the approaches and methodologies discussed above. The first

and basic style of evaluation is the conventional approach used internally
many organizations to monitor the effectiveness of programs in allocating
resources and achieving targets set internally. This is called Internal Review:
administrative effectiveness. The primary concern of this type of evaluation
is to measure performance in delivering outputs, rather than their eventual
economic effects or efficiency. This type of analysis provides no indication of
the impact of the initiative on local economic conditions (Foley, 1992).
The second type of evaluation study is External review: financial efficiency
(Turok, 1990 in Foley, 1992). This type of approach develops measures of
performance by relating policy outputs more directly to resource inputs. It has
been argued that this approach places a great emphasis on economy in the use
of public resources and efficiency in the provision of public services, rather
than their effectiveness in generating desired economic outcomes (Foley,
1992). The approach has its roots in private-sector methods which emphasize
the control of costs and inputs rather than outputs or outcomes. It has been
suggested that such an approach discourages the long-term process of
regeneration through investment in training, physical infrastructure and other
longer term policies (Turok, 1989 in Foley, 1992).
The third type of evaluation is newer and less easily defined. This approach is
called social accounting (Turok, 1990 in Foley, 1992). It is seen as a
reaction against the narrow scope of the preceding accounting methods (Harte,
1986; Geddes, 1988 in Foley, 1992). The scope of social accounting is broad,
with a primary concern about effectiveness in terms of the economic and
social outcomes of policy, not the immediate physical outcomes or direct
financial costs. This type of approach involves investigating the secondary
effects of policy and the variety of distributional consequences for both

individuals and organizations. Such studies also investigate quality issues such
as wages, skills and conditions (Foley, 1992).
Given the pervasive ramifications of distributional consequences of economic
development, their importance warrants a brief discussion. According to Sen
(1992 in Hopkins, 2001), the most important distributional question is the
equality of capabilities, not the equality of opportunities or the equality of
outcomes directly. It is argued that equality of basic goods is not sufficient
because individuals are not equal a priori in physical capacity, mental
capacity, wealth, cultural norms, social status, and many other attributes.
Equality of capabilities is distinct from and a more appropriate measure than
equality of outcomes because equality of capabilities holds that individuals an
intrinsically valued determiner of their own status rather than an instrument
for achievement of equal outcomes (Hopkins, 2001). Actually much earlier
John Friedman (1981: 5 in Boothroyd and Davis, 1993: 236) mentioned that:
The challenge is not only to create a more equal distribution of
income, but also to ensure every household has access to knowledge
and pertinent skills, social and political organizations, instruments of
production (including access to good health), relevant information,
social networks, and financial means to enhance its ability to pursue its
own objectives in cooperation with others.
The final type of evaluation described by Turok (1990 in Foley, 1992) is
Understanding and Explanation. It is argued that policy implementation and
program evaluation are now into their third generation of research, having
moved well beyond program outcome monitoring and toward explanation of
implementation (Goggin et al., 1990; Palumba and Calsita 1990 in Talen,

1996). This approach is concerned with obtaining a deeper understanding of
how policies work. They go beyond measurement of simple policy impacts
and explore how and why they are produced. It is argued that this type of
evaluation should go beyond simple before-and-after snapshot comparisons
and look at longitudinal analysis in the processes of economic growth and
change and the forces and mechanisms which induce it. The longitudinal
approach certainly offers a deeper understanding of the process of change
within the economy and of the extent to which policy has any longer-term
effects (Foley, 1992).
A Critical Review of Economic Development Evaluations
As evaluation studies increase, many evaluations of evaluation studies are
completed to review these studies. In this section, some critiques developed
by these review studies and some essential issues and concerns raised are
Wolman et al. (1994) concluded in their study evaluating the success of urban
success stories that cities which have been perceived as urban success stories
have not, in fact, been successful, at least in so far as improving the economic
well-being of their residents. The changes in the economic well-being of
residents of cities that are typified as urban success stories between 1980 and
1990 did not differ from-and in some cases was inferior to-change in the
economic well-being of residents of other cities that were like the urban
success stories example of distressed in 1980.
Likewise, it is argued that many commonly perceived success stories of local
economic development programs in developing countries have not been
successful by all standards. In one recent case in Africa, for example, a

particular community stood as successful in a national program that funded
community groups for housing, infrastructure, and small-business finance.
The people in this community had a very positive view of the local economic
development program and the way it benefited them, and this image of good
performance had spread to the outside. In reality, the communitys
performance in terms of housing unit completions, housing loan repayments,
completion and adequate operation of basic infrastructure systems, gains in
the level and quality of employment, and increases in household income was
generally no better than in communities in which the program was perceived
as a failure. Clearly, this successful community felt empowered, and enough
residents believed that things had improved sufficiently to judge the
development program positively. The actual contribution to the local
economy, however, was well below what the technical designers had
expected. According to Smoke (1997), this is a common scenario in
developing countries.
The normative nature of economic development policy leads to questions
beyond evaluation process and techniques. Tao and Feiock (2004 in Reese and
Fasenfest, 2004) raise more fundamental questions of ends versus means and
highlight that explicit and systematic comparisons of the latter is as important
as evaluations of the former. In this way they remind academics and policy
makers that evaluation can be more than a retrospective exercise in policy
outcome assessment. Truly critical evaluations of economic development
strategies require a prospective analysis of the means the policies
Two different means, the same end, but in implementation very different
values emerge. The clear lesson is that critical assessment must consider all

aspects of the policy process: means or policies, process or implementation,
and ends or outcomes. Reese and Fasenfest (1999) emphasizing on both
means and ends provides a new way to think about economic development
policy choice and new definitions of good policy.
Foley (1992) argues that levels of economic and employment growth within
an economy will be affected by two main factors, time and local
characteristics. These two factors are important because they will affect the
ability of any economic initiative to create jobs. He notes that, surprisingly,
these are two factors which many evaluation studies have failed to consider.
First is national trends in economic growth and the second is local economic
and social-political structure. Growth in national and local economies is
generally cyclical in nature. Differences in economic factors and the industrial
structure of a locality can have a significant effect on employment and
recruitment patterns. For example, Doeringer et al. (1987) found that
industrial mix interacting with national trends is the most significant influence
on a states short- and long-term economic performance in the U.S. (in Wong,
2002). Some evaluation reports note influence of these factors but many of the
studies have not attempted to compare the effectiveness of policies between
different geographical areas (Foley, 1992).
Another criticism is developed by Howe (1987), who contends that social
analysis is used particularly at the evaluation stage, after projects were
completed. Although this use may affect the continuation or reorientation of a
project, the problem with greater responsiveness at this stage is that it is
irretrievably late, and can give the impression that the primary role of social
analysis is to complain about project failures.

Bovaird (1994) suggests that the evaluation of policy failure needs to be set in
a much more convincing contextual framework than has hitherto been the
case. Differentiation between the effectiveness of services, the success of the
communities which they serve and the excellence of the organizations which
provide them is an essential first step to unraveling some of the different
meanings of performance assessment in LED. Mechanistic evaluations of
intermediate output measures have become symbolic rituals which disguise
rather than reveal the effectiveness of LED approaches and therefore engender
the credibility of evaluation as a concept. It is urgently necessary that this
ground is reclaimed by evaluation theorists before it is too late.
According to Bartik (1995), evaluations with a good comparison group or a
control group are likely to do the best job of measuring the true quantitative
effects of economic development programs. Surveys of client firms provide
some qualitative evidence on programs effectiveness. This qualitative
evidence may be particularly valuable when comparing similar programs. But
surveys of clients alone are not a reliable means of quantifying exactly what
difference the economic development program has made to a firms
performance. Such surveys cannot really address the crucial issue of whether
the quantitative effects of economic development programs are worth their
costs. He further argues that in order to effectively prove that a program
accomplishes its goals, the evaluation must be at the highest two levels of
evaluation-measuring effectiveness or, assessing impact. Simply enumerating
outcomes is not sufficient. A good economic and fiscal impact study should
be sensitive to at least the following five issues:
1. Export-base versus non-export base companies. Economic
development assistance to non-export base companies, companies that
primarily sell locally, is likely to have some adverse effects on the

market share and performance of other locally-oriented firms. These
adverse competitive effects on unassisted local firms are less likely for
economic development assistance to export base firms.
Increases in local employment caused by economic development
programs are likely to lead to in-migration. For instance, research
reviewed by Bartik (1993) indicates that for every 5 jobs created in a
metropolitan economy, 4 jobs are likely to go in-migrants in the long-
run (five years or so). This increase in population reduces the
employment benefits of economic development programs to the
original local residents, and is likely to raise the fiscal costs of
economic development programs. Policy makers should understand
that job growth, particularly when quite rapid, can produce undesirable
outcomes (Beauregard, 1999).
2. Adverse affects due to increased costs and other congestion effects. As
an area grows, land and housing prices increase, and thus force up
wages. In addition, growth may cause greater traffic congestion and
environmental problems. All these greater costs are likely to
discourage employment growth in some firms. Hence, the initial
increase in local economic employment growth due to some economic
development program is likely to be offset to some extent in the long-
run due to these higher costs.
3. Public service costs. Additional business activity and population will
create the need for additional public expenditure. Too many analyses
of economic development seem to pretend that growth only increases
the tax base without increasing spending needs.

4. Public capital costs. Some of the most expensive public expenditures
required by growth are likely to be the capital costs associated with
building new infrastructure, or retrofitting old infrastructure to
accommodate growth. These increased costs for additional population
and business will often be more expensive, on per person or per
employee basis, than the capital costs for the current population and
employment base. Retrofitting infrastructure is often expensive. In
addition, the existing infrastructure may have already been paid for by
past generations, and may have been partially funded by federal grants.
5. Good quantitative evaluations of economic development programs
require the use of a comparable group of unassisted communities or
firms. Such quantitative studies can be usefully supplemented by well-
designed qualitative surveys of assisted firms, done so that results can
be compared across similar programs.
Bartik (1995) suggests a set of questions and says that to answer these
questions a cause and effect relationship must be established between the
program and the outcome. What is the result of the activities described in the
process of evaluation? What happened to target population? How is it
different from before? Have unanticipated outcomes occurred and are they
desirable? Have program objectives been achieved? To ascertain what the
results of the program are, some overtime measures are needed. How are the
program recipients different from the way they were before? What would
have happened in the absence of the program? Does the program work? What
are the other factors that may have contributed to changes in the recipients?

Finally, Smoke (1997) argues that the main concern is that the evaluation of
local economic development policies is often narrowly framed and based on
largely implicit values. The major supposition is the need to go beyond
conventional measures of economic development policy output, given the
inadequacy of such measures to incorporate some of the broader quality of
life, distributional, and local capacity-building concerns that many analysts
feel are as important as to local economy as narrowly defined economic gains.
The next section will discuss these issues.
Measurement of Success of LED through Evaluative Criteria
Rolnick (1992 in Tao and Feiock, 1999) points to insufficiency of aggregate
economic indicators in explaining real progress and argues that although
development programs may be successful in terms of increases in aggregate
economic indicators such as jobs or income, there may be simultaneous
decline in certain neighborhoods or sectors of the community. In this section,
evaluation criteria used in evaluation studies are discussed in terms of their
appropriateness and comprehensiveness.
Evaluative indicators such as income distribution, changes in poverty rate, job
security, employment change by sector, and political empowerment act as a
starting point to shift the focus of evaluation from primarily market affects of
economic development policies toward a social economy perspective, which
encompasses the extent to which public sector expenditures actually bring
public or community benefits. The challenge of incorporating alternative or
social economy indicators further complicates the number of extant evaluation
techniques. Indeed, the issue what appropriate and reasonable outcomes are to
be used in economic development evaluations has come under some debate.
For example, Bartik and Bingham (1997 in Reese and Fasenfest, 1999) have

suggested that ultimate outcomes might be replaced by proximate
outcomes in the face of methodological and design challenges, arguing that,
even if ultimate outcomes such as poverty reduction cannot be measured or
are not achieved, programs may still be beneficial if the proximate goal (say,
job growth) is. On the other hand, Reese and Fasenfest (1999) argue that
although this point is well taken, what if the assumption that proximate goals
lead to subsequent benefits breaks down. What if additional jobs do not go to
local residents or create negative environmental affects or public-service
pressures? What if new jobs occur in economic sectors that benefit only
higher income individuals? What if alternative programs might have produced
greater community benefits?
Tao and Feiock (2004 in Reese and Fasenfest, 1999) in evaluating two
targeted state-level economic development programs in Florida, provide
several important lessons about different outcome scenarios. First, they point
to an important methodological issue in most economic development
evaluations: There is a critical difference between aggregate city-level and
area-level impacts. This is particularly important when targeted programs are
being evaluated. For example, targeted economic development programs
might have no effect on aggregate income growth (apparently a failure) but
can improve disparities in income distribution at both the area and citywide
level (a successful outcome). If income growth, as usual, is the sole or primary
focus of the evaluative narrative, important positive outcomes are missed.
Molotch (1991 in Reese and Fasenfest, 1997) calls for more social criteria in
evaluating economic development policies highlighting the shortcomings of
approaches that focus solely on job creation, he suggests that additional
outcomes be examined, including the quality of the work experience that will

be used in constructing the project and in the labor it would house (e.g. craft
work v. tedious assembly, safe v. dangerous work, well-paid v. poorly paid);
the impact of the project on the external environment, social as well as
physical, including the capacity of the project to contribute to a self-sustaining
local region; the use value of the product that would result from the project
(e.g. production of bread v. production of cigarettes).
Wiewel, Teitz, & Giloth (1994 in Reese and Fasenfest, 1997) place emphasis
on another evaluation criterion and assert that mobilizing political power is
also an integral part of what community development means. Assessment of
local economic development outcomes must include not only indicators of
employment and economic growth but also empowerment and systematic
change. The former represent whether populations in need have a role in
collective action; the latter indicates whether power and resource patterns are
altered in a sustainable manner (Me Kee, 1989 in Reese and Fasenfest, 1997).
In discussing appropriate social indicators of development, Herrick and
Kindleberger (1983 in Reese and Fasenfest, 1997) suggest a focus on life
expectancy, health and nutritional status, educational attainment, housing
quality, and social insurance program coverage.
Reese and Fasenfest (1999) raise the following questions and posit
evaluations that do not consider these issues do not tell the full economic
development story.
What kinds of jobs are created in which economic sector?
Who gets the job?
How do new jobs affect the distribution of income in a community?

How do policies contribute to political empowerment of
Do more jobs actually lead to other benefits, and are there concomitant
unintended costs?
How do policies affect specific areas, neighborhoods, or groups,
particularly the poor?
Are the evaluative goals really important to the community and
residents targeted?
In addition those, Hudson and Sadler (1987 in Foley, 1992) and Moore
(Cameron, 1990 in Foley, 1992) ask similar questions: How many jobs are
being created? Who gets them? At what cost? How many jobs might have
been created without assistance? We also need to ask, how many jobs might
have been lost without assistance?
Cameron and Turok (1990 in Foley, 1992) point out the importance of private
sector investment and call for an evaluation criterion that assesses the extent
to which economic policy stimulate private sector investment. They note the
current governments desire to increase the amount of private-sector projects
(so increasing the net benefit of public spending on financial assistance) since
it is usually difficult to attract private-sector funding for some projects.
Robinson et al. (1987 in Foley, 1992) suggest that deadweight concept needs
to be considered in evaluation studies as well; given the growing interest in
cost and expenditure factors and a recent increase in the number of studies
investigating deadweight spending, which has been defined as the extent to
which projects would have gone ahead even without public assistance.

To Foley (1992), the desegregation of employment impact and policy effects
on employment are required and they can be investigated in two ways. First,
there is a relatively well-established methodology to examine the number of
jobs which might have been created without the initiative. These
investigations usually try to find additional employment created by an
initiative. Secondly, investigations of the type of job or individual who obtains
the jobs created by an initiative are starting to become more common in
evaluation studies. Finally, as well as noting other benefits of projects, several
studies also note the additional costs which many studies fail to consider
should be included. For example, Turok (1990) and others (Blake, 1973;
Hodge and Whitby, 1979; Pearce, 1983 in Foley, 1992) state that full
exchequer costs should be considered by the project since the immediate costs
used in calculations usually only include the costs of funding provided by an
initiative and they fail to consider the personal and management costs
involved in administrating an initiative.
Critique of Job Creation as a Main Indicator of Success of LED
Despite this diversity of factors, most evaluation studies have focused on job
creation and the cost-effectiveness of job creation because jobs are regarded
as the principal policy objective. The majority of evaluation studies have
therefore used job creation as their basic measure of success (Foley, 1992). As
a result, state and local economic development strategies evolve
incrementally, without underlying economic theory except that more jobs are
good and less jobs are bad (Beaumont and Hovey, 1985 in Reese and
Fasenfest, 1997).
Job creation as a main indicator of successful LED has been criticized on
various grounds. The location of the links in the job chain, and the role of in-

migrants and local disturbances to chain formation are all likely to be
determinants of the success of job creation in local economic development.
The occupational distributions of new jobs strongly affect the degree to which
jobs translate into local welfare improvements (Felsenstein and Persky, 1999).
A research of National Association of State Development Agencies (Poole et
al., 1999) shows that there are difficulties in estimating the cost impact of new
development because the costs depend upon who gets the new jobs. If the new
jobs are taken by in-migrants, the project will have much greater cost impacts
on schools and other population-based services than if the new job-takers are
existing residents. Since these individuals and families already live in the
community, their employment would not strain existing capacity levels.
Similarly, Turok and Wannop (1990 in Foley, 1992) note that comparisons
between different projects solely on the basis of cost per job created can be
misleading because they overlook additional economic and social benefits,
including the re-allocation of jobs to disadvantaged groups, jobs created
temporarily, and the kinds of jobs created. McGowan and Ottensmeyer
(1993) argue that local economic development is a wealth creation process. It
is not in and of it self, job creation. This subtle but important distinction
(Vaughan, 1985 in McGowan and Ottensmeyer, 1993: 72) is one not always
understood by those charged with making public policy. Moreover, because
policymakers are driven primarily by political expediency, they have a
tendency to confuse the creation of jobs with economic development. Giloth
(1992) argues that jobs have become ingrained in the minds of politicians and
the public as the appropriate sign of success and as a result, the wrong debates
occur (e.g., whether firms have created jobs), and government officials
attempt to keep the discrepancies between jobs and performance the public

Another criticism is that the welfare gain to the individual worker from a new
job will not in general be equal to that workers wage. Rather it will be a
smaller amount, equal to the difference between wages on the new job and the
workers wage in his or her next best alternative. In the extreme case of
smooth and perfectly functioning labor markets, these alternative wages will
be close to the wage level on a new job. Indeed, this is why, in a fully
employed market, wages are not only a private cost to the business but a
social cost as well. In a fully employed labor market, wages indicate the value
of alternative production given up when a worker shifts a new enterprise. In
such a world, simply counting additional employment may not teach us very
much about the efficacy of economic development programs (Courant, 1994
in Persky, Felsenstein & Carlson, 2004).
Much of this criticism points not to the objective of short-term job creation,
but the increasing of long-term competitiveness of the area in a changing
national and world economy (Hughes, 1991). It is argued that although the
jobs indicator may be difficult to dislodge, economic development
practitioners should give attention to alternatives such as business
competitiveness or wealth creation that may more appropriately measure firm
health and satisfy the public need for accountability. Likewise, many
initiatives may not have a direct impact on employment even when they are
targeted at promoting company growth. For example, Robinson and Wren
(1987 in Foley, 1992) note that assistance can increase productivity, output
and profitability, lead to the introduction of new products and processes, and
increase research and development, exports and market share.

Limitations of Past Evaluation Studies and New Approaches
Although evaluation literature has reached a certain point, there are still some
challenges that future studies have to meet. Limitations of evaluation
methodologies and challenges are stemming from complexity of economies
themselves, the multidimensional nature of economic development, diversity
of interest groups, and gathering of reliable data about impacts of economic
development policies, programs and projects for a given time frame.
All LED evaluation studies face four major modeling challenges: the
estimation of the gross direct effects of initiatives, the incorporation of
multiplier effects in the economy, the separation of the agencys net affects
from those caused by other agencies, and the estimation of the additionality
of the public-sector contribution (Bovaird, 1992).
Estimation of the gross direct effects of LED initiatives requires data for the
before and after periods and a way of controlling for the effects of external
changes not connected to the initiatives (e.g. recession or interest rates). The
most common approaches involve interview surveys of firms, sometimes in a
longitudinal panel to establish changes which have occurred (since they
received the LED initiative) in employment, turnover, profit, investment, and
so on. Control groups of firms not benefiting from the initiatives are
sometimes used; when this is not the case, contingent evaluations are used
(Willis, 1990 in Bovaird, 1992); firms are asked how they would have
responded in the absence of the initiatives.
The main alternative to this approach is components-of-change analysis. This
analysis attempts to separate local or regional economic changes during the
period of the LED policy or initiative which were not attributable to national

trends, to the industrial structure of the area or to the local effects of the
business cycle (Moore et al., 1986 in Bovaird, 1992). This approach requires
clear-cut start and end dates for LED initiatives or clear time periods between
which the strength of the LED initiative varies (Bovaird, 1992).
Another limitation of the past evaluation studies is their incapability of
including the multiplier effects. Although their importance is well known
theoretically, in practice most evaluation studies have not sought to provide
fresh empirical evidence of multiplier effects in the local economy, and only
few studies have partially investigated local industrial linkages. They have
either ignored the multiplier effects (implicitly or explicitly assuming that they
are equal for all initiatives), or have used the values generated in previous
In practice, LED evaluations have most commonly sought to explore
additionality in the case of those public agency initiatives where surveys
could be undertaken of the firms benefiting from the initiatives. These surveys
have attempted to establish whether the public intervention influenced the
likelihood of the project taking place, its final size or the date at which it was
undertaken. The most common way of comparing the results of LED
evaluations has been to focus upon estimates of the cost-per-job ratio
(Bovaird, 1992). In this case, decisions have to be made on how to estimate
the aggregate net effect on jobs, taking into account those:
Permanent and temporary jobs created
New jobs and jobs preserved (which would have otherwise

Entirely new jobs, jobs relocating from outside the study area, and jobs
displacing workers in firms already in the study area.
Moore and Townroe (1990; the Polytechnic of Central London, School of
Planning 1990 in Foley, 1992) indicates that the practical, methodological and
conceptual problems facing those undertaking evaluation exercises are
considerable by quoted Britain experience. Those are:
Unclear or vague objectives for policy;
Objectives poorly quantified or defined in a qualitative manner;
Difficulties in determining full exchequer or resource-cost
Difficulty of measuring secondary effects;
Reliance on the accuracy, perception and honesty of those involved
with an initiative, particularly with respect to sensitive issues.
To Foley (1992), a major conceptual issue is the geographical parameters
which should be used for any evaluation study. The geographical (and
accounting) parameters chosen for an evaluation have a considerable
influence on additionality, displacement and multiplier values. A wider
geographical boundary will have a number of effects over a small boundary.
These are, primarily, that with a wider boundary more transfers are likely
within the (defined) local area and therefore additionality will be lower.
Multiplier values are higher with wider boundaries, but displacement values
are also likely to be higher. Ultimately, if one adopts national boundaries, the
benefits of a local policy may be zero or negligible. But as Turok (1990 in
Foley, 1992) points out, the impact (and additionality) of policy at the local

level may be far more important in overcoming local disadvantages and
spatial disparities. Therefore most studies adopt parameters such as local
authority boundaries or arbitrary distances (Foley, 1992).
Conflicts among goals constitute another challenge for future evaluation
studies. Identifying the distinctions between economic and non-economic
outcomes seems a logical step toward linking goals to policy. Although some
of these goals may be compatible with promoting economic growth, others
may conflict with or undercut the economic goals of development; for
example, supporting environmental or growth management goals by seeking
specific types of industries. Targeting development benefits to specific
categories of recipients or geographical areas based on need might also
conflict with aggregate economic goals in certain contexts. This suggests that
economic development programs serve multiple and perhaps conflicting
purposes including development, distribution, and redistribution. Even when
officials seek aggregate economic growth, short-term economic gains may
conflict with long-term economic performance (Tao and Feiock, 1999). For
example, Beauregard (1999) explains inconsistencies among economic
development outcomes using New York City. He shows how the three most
desired outcomes of economic development intervention-job growth, local tax
revenue growth, and income growth interrelate over time.
A Critical Evaluation of Local Economic Development Policies
A recent study by Reese and Fasenfest provides one of the broadest
discussions about evaluation of LED. There is a story in development circles
that perhaps best illustrates the tensions and dilemmas raised between
traditional and critical evaluations.

Consider a village at the edge of a cliff with a hospital at its base. Villagers
fall from the cliff and break bones, quickly overwhelming the hospitals
capabilities to serve the village. The traditional (in this case modem
medical) assessment results in efforts to expand the hospital and increase
its capacity to serve the community. A more critical evaluation may
determine that the real beneficiaries of such a solution are those in the
medical professions making more for services provided, when what is
really needed is a fence at the top of the cliff to prevent more accidents.
From whose perspective are the solutions posed and benefits accrued to be
assessed (Reese and Fasenfest, 2004: 49-50).
Critical evaluations make explicit an inquiry into the reality of economic
development policies they entail the allocation of public revenues and
resources to the private sector, a public subsidization of private interests.
Critical evaluations make that discriminating judgment and hence ask
questions about reasonable public expectations for program outcome (Reese
and Fasenfest, 2004). It is argued that critical evaluations of local economic
development policies go beyond the simple and narrow consideration of
immediate outcomes and explicitly address broader and more fundamental
To what extent do economic development policies, as public
expenditures, serve public as opposed to private goals and needs?
To what extent do economic development policies benefit those most
in need within communities?
To what extent do economic development policies build and enhance
community stability as opposed to reinforcing market imperatives?
How should conflicting interests within cities be balanced in creating
and implementing development policies?

To what extent is policy means as opposed to policy ends an important
consideration in evaluating economic development policy?
How do policymakers assess the public costs of private development
How might policymakers systematically assess and choose among
policies using social as opposed to purely economic criteria?
Are economic development policies that are targeted toward public
rather than private interests inherently less rational, or might such
policies actually address both social and economic considerations? In
other words, are social and economic values inherently at odds?
How does the selection and implementation of particular evaluation
methodologies affect the results of outcome assessments? How do the
methods used color the findings?
How can qualitative considerations be taken into account in
quantitative evaluations?
What is the role of community stakeholders in evaluating economic
development policies, and how might they most effectively be
These researchers further suggest the following indicators and argue that a
shift in indicators or measures of success leads to very different conclusions
about the effects of policy (See Table 2.1.)-

Table 2.1 Traditional indicators versus indicators for critical evaluation
Traditional Indicators Indicators for Critical Evaluations
Employment growth Employment change by sector
Change in per capita income Income distribution
Job generation Job generation by sector
Private jobs per resident Wage and non-wage employment
Change in GNP/GDP Change in percent in poverty
New business attraction Diversity of economic base
New business starts Extent of citizen participation
Sales increases Job security
Increased tax base Pay levels of new jobs
Business expansion Safety of new jobs
Dollars of investment capital Quality / challenge of new jobs Recession-proof economy Value use of new products
Source: Reese and Fasenfest, 2004: 9.
It is argued that when the goal of economic development programs is
community development, success is no longer merely the increase in jobs
and business; instead, policies must foster structural and institutional changes
which promote a more equitable distribution of new jobs and income
generated by growth (Reese and Fasenfest, 1997 in Tao and Feiock, 2004).
Tendler (1987 in Tao and Feiock, 2004) suggests that measurement of
economic development must measure the well-being of residents including the
redirection of resources to the poorest segments of the community. Thus it
would seem that policies that call for redistribution of resources (e.g., jobs,
business opportunities, education) would be the logical starting point for an
examination of the effectiveness of both economic and community
development efforts (Tao and Feiock, 2004).
A New Evaluation Framework
Review of the literature shows that there is a clear evolution in evaluation
studies. Early studies rarely considered secondary factors or estimates of cost
per job. These have now become commonplace in most evaluation studies.

Also pressure to achieve higher levels of value for money has increased and
this has led many evaluation studies to place greater emphasis on financial
issues such as economy and efficiency. In addition, greater attention is now
being focused on the beneficiaries of economic policy (Foley, 1992). And
recently it is observed there is a shift from quantitative
methodologies/indicators to qualitative methodologies/indicators.
Most evaluation studies have followed the guidelines offered by Robinson and
Wren (1987 in Foley, 1992), who noted that the need to narrow down the
complex overlay of policy measures to manageable proportions, using data
which can be reliably collected, while at the same time allowing the
possibility to draw comparisons to give meaning to figures of cost-
effectiveness. Robinson and Wren correctly assert that spurious sophistication
should be avoided. Such an approach should provide the sort of results policy-
makers require without the sophistication which developed in the late 1970s,
when evaluation became too complex and as a result largely unusable. Foley
(1992) supports this view by saying that it is questionable whether increased
sophistication would really add any value to results obtained for comparative
purposes as claimed by some.
Although the review of the literature shows that there is not a single identified
indicator that can provide an adequate measure of LED performance,
following are the mostly discussed indicators or issues:
Productivity measures such as gross domestic product per head and
gross value added in manufacturing per employee
Employment level and change

Unemployment level, change and duration demographic variables such
as population change and net in-migration level
Property-market-related indicators such as planning applications
(Wong, 2002: 1854)
Three key issues related to jobs created: Job quality, job allocation and
duration of job (Storey, 1990 in Foley, 1992: 564).
Tao and Feiock (1999) argue that if the expectation at both the state and local
levels is that disadvantaged communities should be the primary beneficiaries
of programs, local assessments should prove far more meaningful than
aggregate measures. However, evaluation models and measures must
incorporate attention to the distributional consequences of policy action;
among the residents within a community and between communities within an
urban area because valued outcomes cannot be defined solely in terms of a
single community (Reese and Fasenfest, 1997). On the other hand,
Beauregard (1999) suggests that more attention needs to be paid socio-spatial
variations in economic development outcomes. The aggregate performance of
a local economy is important, but much more must be said about internal
Turner (1999 in Reese and Fasenfest, 1999) addresses more political aspects
of economic development stories. She suggests considering political
empowerment as an outcome of local economic development and for
distinguishing community development from aggregate economic
development. She explained importance of this indicator by explaining a
Florida study where she found that more successful communities had residents
with more entrepreneurial skills, and visions. Ray (1999a in Ray, 2000)

suggests that evaluation must go beyond the specifics of the initiative in
question and focus on the general principles of endogeneity.
Myers (1987) and Castells and Hall (1994 in Wong, 2002) point to the lack of
literature in the relatively newly developed and qualitative indicators by
arguing that the nature of the relationship between LED and factors like
infrastructure, industrial structure and knowledge and technology are not that
well understood. Porter (1990, 1991 in Wong, 2002) shares similar concerns
as well and suggests that in order fully to explain local economic
performance, the inclusion of the demand-side conditions, performance of
private business sectors, and of supporting industry are required.
The theories and empirical research of these soft factors are less well-
established because the casual relationships between them and the LED
process are usually not straightforward. There is also an inherent difficulty in
grasping the concept of these factors because their contribution is not directly
measurable. All these findings naturally lead to the conclusion that the
contribution of different factors to LED is circumstantial and diffused, and
much more empirical work is required to explore their logistical relationships.
The causal relationship between LED and quality of life is a difficult and
controversial topic as well. It is important to note that quality of life is a result
of local economic development as well as reason a contributor to local
economic development as argued by Castells (1989 in Wong, 2002) who
regards quality of life as a result of the characteristics of the high-tech
industry in Silicon Valley (its newness and highly educated workforce) rather
than the determinant of its location pattern.

It is a well established fact that investment and technical change are key
requirements of continued economic growth and development in competitive
economies. The progressive increases in productivity that it helps to bring
about improve competitiveness, enhance profitability and generate resources
for further investment, higher incomes and greater consumption (Turok,
1989). The capacity of technological absorption and diffusion is firmly
believed to be a critical basis for the creation of local economic growth in the
long run (Castells and Hall, 1994; European Commission, 1994; HM
Government, 1993 in Wong, 2002). Accordingly, these emerging concepts
need to be considered by future evaluation studies as well.
The policy-intervening capacity of organizations such as local authorities,
chambers of commerce and other agencies is found influential to the process
of LED (Blakely, 1994; Bennett and Krebs, 1991 in Wong, 2002). Clarke
(1995 in Wong, 2002), challenges the presumption that more capacity tends to
be associated with better and more progressive decisions as there are no
thresholds to determine when these conditions are met. Although, due to the
qualitative nature of institutional capacity, it seems not possible to devise any
reliable and valid measure for this factor, new studies must find ways to
measure and include those indicators as well.
Finally, it is worth explaining briefly an ideal evaluators perspective. Jackson
(2001: 19) explains what literature discusses concerning how the evaluator
should be.
Objectivity (Scriven, 1976). Evaluation findings should not be
contaminated by the personal preferences or ideology of the evaluator.

Reproducibility (Cronbach, 1982). Another evaluator using the same
methodology should obtain the same results.
Autonomy (Newman and Brown, 1996). Evaluators should be free
from the outside influence.
Emphatic Neutrality (Patton, 1990). Evaluators should be perceived as
sympathetic to program participants but open about findings.
In conclusion, the new forms of local economic development involve multiple
measures of success, a broader range of economic development impacts,
increased organizational and economic complexity, and immense local
variation (Giloth, 1992). Many authors in Jackson (2001: 20) have argued that
power needs re-balancing through giving a greater role to stakeholders.
Variants on this view include responsive evaluation (Stake, 1975),
democratic evaluation (McDonald, 1976), naturalistic evaluation (House,
1980), participatory evaluation (Choudhary and Tandon, 1988), fourth
generation evaluation (Guba and Lincoln, 1989), empowerment evaluation
(Fetterman et al., 1996), and utilization-focused evaluation (Patton, 1996).
All these theories advocate qualitative methodologies and participation in
various degrees.
Increasing the Use of Results of Evaluation Studies
Since one of the most important points of academic research is to bridge the
gap between theory and practice to ensure the utilization of scientific studies
by practitioners, this section devoted to a discussion of reasons for
underutilization of evaluation studies and problems associated with it.
Further, the increased use of evaluation results is explored.

As previous sections show, it is clear that evaluative studies of LED have
become fashionable and widely practiced during the several decades.
However, it is still far from certain that they have a real place in the policy-
making process (as opposed to their value as symbolic activities and ex-post
policy rationalization). For example, Bovaird (1992) argues that there is a
clear need for evaluation methodologies to be codified by LED agencies, in
order that really significant comparisons can be made and lessons of LED
success and failure disseminated effectively. After investment, few
municipalities conduct ongoing evaluations of firms that receive incentives;
most city agencies superficially monitor business performance and have little
or no recourse when firms leave town or when firms fail to produce their
promised benefits (Giloth, 1991).
Research in the U.S. has consistently revealed that evaluation does not have a
major effect on programs. Several possible explanations for this weak link are;
that policy is the product of political as well as technical factors; that
programs change slowly because of resistance from vested interests; that
evaluation reports fail to speak directly to the concerns of policy makers and
implementers (Jackson, 2001).
Theorists have identified several ways of increasing the use of evaluation
results (Patton, 1996; Shadish et al., 1995; Georghiou, 1995 in Jackson, 2001).
These can be related to four principles: First, evaluators can increase the
relevance of their work through concentrating on the factors which users can
control, and giving priority to questions relevant to pending decisions.
Second, evaluators can increase receptivity through identifying users early on,
involving users in the evaluation, and providing interim findings. Third, the
absorbability of evaluation can be increased through providing

recommendations, including non-technical executive summaries, matching
summaries to the interests of different stakeholders, and disseminating
findings through informal meetings as well as reports. Fourth, credibility of
evaluation studies which can be increased through using evaluators with
appropriate status who are demonstrably fair and competent.
Turok (1989) suggests that evaluative research would be more useful and
convincing if the focus is broadened and if greater attention is paid to
understanding causal mechanisms and processes. According to Foley (1992)
evaluative research should go beyond simply trying to isolate and measure
policy effects: it should examine how changes are produced or why they are
not. More effort should be made to analyze the mechanisms linking policies
and outcomes, and the conditions that facilitate or obstruct their effectiveness.
This requires some exploration of the underlying forces and processes of
economic change. Moreover, Felsenstein and Persky (1999) argue that
understanding is necessary to improve the reliability of research findings and
to make future policy interventions more effective. They suggest that serious
evaluation of local economic development efforts must go beyond simple
impact analysis. In particular, a local welfare perspective requires us to
distinguish between total new payroll and real improvements in residents
standard of living. To follow up on this distinction requires an appreciation of
local labor markets-and particularly of job chains. It must be recognized that
welfare gains are spread more broadly across the population than are the new
jobs that create them.

Economic Development Incentives
States and localities use economic development incentives to influence the
location of business investments. Poole et al., (1999: 10-11) define incentives
as programs with budgeted and allocated public dollars that are directly or
indirectly invested in activities of businesses. Incentives also may be defined
as economic development programs that assist businesses without providing
direct financial assistance. For instance, tax policies of states, property
valuation, accelerated deprecation, and interest rate subsidies are among these
types of programs. Other forms of incentive assistance for businesses in this
category include technical assistance, subsidized higher education, and public
infrastructure. Below three types of incentives are discussed briefly.
Direct Financial Incentives: These incentives are programs that provide
direct monetary assistance to a business from the state or through a state-
funded organization. The assistance is provided through grants, loans, equity
investments, loan insurance and guarantees. These programs generally address
business financing needs but also may be invested in workforce training,
market development, modernization, and technology commercialization.
Indirect Financial Incentives: These incentives include grants and loans to
local governments and community organizations to support business
investment or development. The recipients include communities, financial
institutions, universities, community colleges, training providers, venture
capital investors, and childcare providers. In many cases, the funds are tied to
one or more specific business location or expansion projects. Other programs
are targeted toward addressing the general needs of the business community,

including infrastructure, technical training, new and improved highway
access, airport expansions and other facilities.
Tax Incentives: These incentives are widely used as a strategy for leveraging
business investments. Policy makers and practitioners view them as a means
of reducing the cost of doing business and leveling the playing field in the
competition with other states. Generally tax incentives can be classified into
five subcategories: (1) credits, (2) abatements/reductions, (3) exemptions, (4)
refunds and (5) other special tax treatment to encourage business investment
in the state (Poole et al., 1999).
History of Economic Development Incentives
Review of incentives offered by state and local governments to influence the
location of business investments shows that there are three different waves.
The first wave was dominated by programs designed specifically to attract
footloose firms from old industrial areas to growing regions, such as the South
or West. The typical tools of the this first wave were subsidized loans or direct
payments to firms for relocation expenses, tax reductions, subsidies applied to
the cost of plant facilities or utilities, and competitive and expensive industrial
recruitment programs (Bradshaw and Blakely, 1999). The same period, before
1985, was characterized by Grant et al., (1995) as a supply side approach,
premised on increasing the supply of capital or labor to business without
altering the basic industrial composition of the states economy. Such supply-
side strategies are sometimes characterized as relocation incentives (or, more
disparagingly, as smokestack chasing), since they are aimed, first, at
dissuading in-state industrial firms from moving and, second, at enticing out-
of-state firms to relocate within the state.

Second-wave strategies offered indirect types of firm-level assistance, such as
creating new businesses, increasing investment capital, developing incubators,
or providing assistance to help local businesses grow or expand. Eisinger
(1988 in Grant et al., 1995) identifies this major shift in state economic
development strategy from older supply-side approaches to new demand-
side policies, in which states take on innovative entrepreneurial functions
such as venture capital funding, high technology development, and export-
promotion projects. By the early 1980s, states began operating and expanding
existing firms.
The second wave is giving a way to the third wave in early 1990s. In general,
third-wave strategies shift focus on local development by creating the context
for economic growth through public-private partnerships, networks that
leverage capital and human resources to increase the global competitiveness
of a group of strategically linked firms. Third-wave program do not eliminate
first- and second-wave strategies, but they give specific purpose and focus to
the use of these techniques (Bradshaw and Blakely, 1999). Key to third-wave
economic development programs is the identification of key industry clusters
specific to each region (Waits and Howard, 1996 in Bradshaw and Blakely,
Third-wave economic development efforts focus multiple existing resources
of state government on a continuing broadening of the foundation for
economic development. From the first waves focus on attracting firms
external to the state and the second waves focus on both internal and external
firms, the third wave gives emphasis to the community and institutional
resources that allow firms to succeed in what is an increasingly global context.
Indeed, the ultimate success of third-wave efforts is not measured by short

term gains, but what has been described as a positive-sum industrial policy, in
which all firms benefit from state efforts and the industry expands due to its
global advantages rather than local incentives (Bradshaw and Blakely, 1999).
As it will be discussed later, many argue that global competitive advantages
are crucial for long term economic development.
The role of networking is the classic low-cost economic development tool.
Third-wave strategy in general and networking in particular defines new roles
for state and local governments. The new-generation strategy seeks to reduce
the role of state government as a direct supplier of programs in favor of a role
a wholesaler, in which the emphasis is on seeding, leveraging, and general
capacity building functions. The notion is that governments prove most
effective when they enable individuals, institutions, firms, and communities to
develop the means to help economic development. Third wave policies
include investment in job training and education, industrial modernization
initiatives, support for the community-level economic development planning,
and encouragement of industrial clusters of firms for the purpose of pooling
resources to achieve higher levels of international competitiveness than each
firm could manage on its own (Eisinger, 1995). Similarly, Hatch (1991 in
Bradshaw and Blakely, 1999) points out that the best role for the state is to
concentrate its resources on coordinating services to groups of firms, proving
strategic information on growing markets, and catalyzing manufacturing
networks. These efforts require that the major economic development role is
to be a broker who mobilizes firm into cooperative systems in which
common needs are addressed and specialized services are collectively
acquired as a regional hub around which an industrial structure can be formed.
This is the future of economic development.

Arguments for Economic Development Incentives
In general, the academic literature on effects of economic development
incentives is divided between studies finding major effects associated with
growth and those finding negative or inconclusive results. Thus far, negative
and inconclusive study findings are more numerous than those finding
positive correlations. Findings showing an association between taxes and
economic growth tend to be disparate and, in general, give little guidance to
policy makers (Poole et al., 1999). In this section, how business development
incentives are justified and why they are seen necessary for economic
development will be discussed.
Wassmer (1992) discusses four determinants of local economic development.
First, exogenous structural change in the macro economy and second the
national business cycle which influences local economic development
intertemporally, third, local economic and demographic factors, which are
likely to be endogenously determined with local development. The last and
fourth determinants of a citys economic development over time are the
incentive package offered by the city.
Indeed, there are some empirical as well as theoretical studies that support
Wassmers argument. For instance, Marlin (1990), who surveyed
effectiveness of economic development subsidies across states in the U.S.,
found that those states that emphasized economic development in the use of
their tax-exempt debt did indeed experience accelerated economic growth,
while those that did not do so experienced no discernible impact on their
growth rates. Similarly, as recently as 2002, Bradshaw analyzed the effects of
small business loans in California, 1991-1995 and found that firms getting
loan guarantees generated more than double the number of jobs generated by

comparable firms statewide. However, neither of these two studies questioned
the social and distributional impact of incentives and sustainability of these
developments over the long run.
Many economists argue that the U.S. is a highly mobile society; as many as
14% of the U.S. metropolitan population moves between metro areas in any 4-
year period. In this context, subsiding new investment at a particular place
merely makes that place more attractive to in-migrants (Marston, 1985 in
Peters and Fisher, 2004)
Most of the documented competition among areas appears to be of an
intraregional nature, suggesting that these incentives encourage firms to move
from one community to another within the region. This coincides with the
finding in several research studies that taxes matter most in a business location
decision when the locational choices have been narrowed down to the substate
regional level. The research also indicates that economically disadvantaged
communities tend to offer more generous incentives than more prosperous
communities. It is not clear, however, that distressed communities offering the
biggest incentives receive the greatest economic boost in return. Yet it appears
that incentives are particularly valuable for distressed areas facing challenges
related to high unemployment (Bartik, 1991 in Poole et al., 1999)
It is also argued that for state and local incentive competition to benefit the
nation as a whole, the benefits to communities gaining jobs must exceed the
losses the communities that would otherwise have had those jobs. This will
occur only to the extent that incentives are targeted at poorer populations, who
stand to gain more from job growth than residents of low-employment or
middle-class communities. Thus the alleviation of unemployment or poverty

becomes an additional rationale for incentives, and poorer targeting of
incentives becomes an additional criterion for assessing their effectiveness
(Peters and Fisher, 2004).
Luger and Bae (2005) examined the effectiveness of state business tax
incentive programs in North Carolina and found that programs were generally
successful in creating jobs in economically distressed areas. The tax incentive
program, however, was not cost-effective in inducing new employment in
North Carolina. Their study also suggests that the targeting of incentives to
distressed areas might help those areas overcome competitive disadvantages
relative to other places (e.g., agglomeration economies and quality of life).
Finally, in a recent theoretical study Bartik (2005) concluded that some
incentives are beneficial for two reasons: First, because corporations are
becoming more footloose, and second, increased local employment rates yield
social benefits. However, he mentions that incentives are often wasteful for
two reasons: local policymakers often overestimate the benefits of incentives,
and the local debate over incentives is dominated by business interests.
According to this researcher, who questions the rationale for state and local
governments to offer economic development incentives to attract or retain
large businesses and whether there is a good rationale for why such incentive
use appears to be rising over time, one plausible rationale is that incentives are
increasingly perceived as a necessary cost incurred to produce social benefits.
He argues that it is increasingly plausible that incentives might help attract or
retain business, and thereby produce benefits such as greater employment
rates and a stronger state and local fiscal situation. However, he states that
incentives costs are often large while incentives benefits are often modest.
The next section examines costs, benefits and effectiveness of incentives.

Arguments against Economic Development Incentives
As long being one of the mainstays of economic development policy, the
provision of business incentives has been the target of the most criticism at the
same time. At the heart of the controversy over economic development
incentives is the fundamental yet unresolved question: are they cost-effective
strategies for achieving economic growth? There are many critics who believe
that virtually all incentives should be banned (Burstein and Rolnick, 1995 in
Peters and Fisher, 2004). Furthermore, for some it seems a game in which
there are substantially more players than prizes, as a result there are inevitably
sizable numbers of losers (Boothroyd and Davis, 1993).
Incentives have been criticized from the equity point of view by some while
others criticize them on the ground of their contribution to revenue and citys
fiscal health. Most critics generally question the effectiveness of incentives in
terms of sustaining economic growth and development in the long term. For
example, Krumholz (1991) argues that the effectiveness of local economic
development in terms of net new jobs or taxes -its essential public purpose -
is largely unknown. He points equity of local economic development and
states that economic development practitioners have been criticized for freely
using various inducements with little idea of or interest in their effectiveness
in achieving the overall objectives -new jobs and net increases. A study that
Krumholz cites examined 21 subsidized projects in five cities (Birmingham,
Hartford, Milwaukee, St. Louis, and Toledo) found that the projects had
produced far fewer jobs and other benefits than that were promised.
According to Redmond and Goldsmith, this should not be surprising given
that large corporations get most of the subsidies and small businesses produce
most of jobs (1986 in Krumholz, 1991). Krumholz concludes that local
economic development as currently practiced provides tangible benefits to

developers, land owners, lawyers, construction union members, suppliers,
politicians, and development officials. Tangible job benefits for the poor and
unemployed residents of central cities cannot be demonstrated and net benefits
to the overall fiscal condition of these cities is questionable.
Wolkoff s (1990) argument is that subsidies paid by communities increase the
returns to stock holders, many of whom may live in other jurisdictions,
supports Krumholzs. This result is troublesome for many political leaders
who care about distribution and would prefer to capture more of the surplus
for their constituents. Some critics of industrial subsidy policy have argued
that efforts should be made to restrict the degree of bidding for businesses so
that more benefits can be captured by local tax payers and fewer by holders of
capital. However, the incentives in a world of severe competition, what
Wolkoff calls zero-sum game, make it unlikely that the participants will be
able to police themselves. In another study, it is discussed that although the
public rhetoric of incentives invariably focuses on the job gains for
unemployed and underemployed locals, the real beneficiaries of incentive
induced growth are the owners of immobile and scarce resource: land (Peters
and Fisher, 1997). Thus, economic development policy is likely to a
regressive effect on the local distribution of income.
In another study evaluating the economic development incentives in the
Detroit Metropolitan Area, Wassmer (1992) concludes that in only 5 instances
(31%) incentives caused the desired additive effect on local economic
development. His studys results indicate that a more reasonable premise is
that city incentives exert no effect, or are only positively associated with
further economic decline.

Bumier (1992) argues that tax abatement can be costly. With the addition of
new firms come the costs of providing additional public services such as
police, water, and sewer. Local schools do not receive as much tax revenue as
they would if abatements had not been used. Further, it is not unusual for
firms to request, in addition to tax abatement, job-training funds, reduced
water and sewer rates, building and infrastructure improvements, and low
interest loans. Finally, there is the possibility, however remote, that a firm
would have located in the community without incentives.
Another criticism is developed around business location decision. Surveys of
corporate executives show that policy inducements enter business calculations
only when proximity to raw materials and markets, cost of energy, availability
of skilled workers, and other important investment criteria are equally well-
satisfied by more than one location. Since sites are seldom equivalent on these
data, incentives are generally unimportant; investors will simply choose the
site that is economically soundest (Kieschnick, 1981 in Hanson, 1993).
Moreover, it is asserted that even if a plant can be attracted to a community by
subsidies, it is generally a plant so marginal that it will be closed in the first
economic downturn (Thurow 1989 in Boothroyd and Davis, 1993).
Leicht and Jenkins (1994) examined three strategies of state economic
development. They evaluated (1) entrepreneurial approach forcing on new
firm and technology development, (2) industrial recruitment strategy
emphasizing financial incentives for the relocation or expansion of existing
enterprises and (3) a deregulation approach that minimizes governmental
control over private enterprise. They found that the entrepreneurial policies
encouraged the creation of new firms, but these new firms did not
significantly trickle down to workers in terms of new jobs or to the states

economy in terms of expanded production. Similarly, industrial recruitment
strategy prevented businesses from declaring bankruptcy but had little effect
on manufacturing employment and total state product. Finally, deregulation
appears to have credited higher business failure rates without significantly
altering the other growth dimensions. None of the strategies created
manufacturing jobs or growth in state product, which are meaningful
indicators of growth. They concluded that creating new enterprises and
reducing business failures will not quickly guarantee economic growth. It is
indicated that these policies need more detailed and specific evaluation over a
longer period of time before we can gauge their economic impact.
Peters and Fisher (2004) found whether or not incentives induce significant
new investment is not clear. Since these programs probably cost state and
local governments about $40-50 billion a year, one would expect some clear
and undisputed evidence of their success. This is not the case. In fact, there
are very good reasons theoretical, empirical, and practical- to believe that
economic development incentives have little or no impact on firm location
and investment decisions.
Bartik (1994 in Peters and Fisher, 2004) has argued that it is highly likely that
incentives are always revenue negative. His arguments assumption is that
economic activity is not very sensitive to taxes; in fact, he assumes that the
elasticity is around -.3, so that a 10% cut in taxes would produce just 3%
increase in investment or jobs. He then demonstrates that the net change in tax
revenue is approximately equal to the percentage increase in jobs minus the
percentage cut in taxes. Obviously, if the elasticity is -.3, the net result is
negative (3% minus 10%). For instance location decisions, the elasticity is
almost certainly much less than 1, so the percentage change in taxes, and the

net effect is a loss of revenue. Furthermore, the bigger the tax cut, the larger
the revenue loss.
Similarly Peters and Fisher (2004) argue that effects of incentives on the
location choices of businesses at the interstate level are so weak that, state
governments and their local governments in the aggregate probably lose far
more revenue by cutting taxes to firms that would have located in that state
anyway than they gain from the firms induced to change location. Some
localities may gain revenue by offering locally funded incentives, because
taxes may provide more leverage over location decisions at the intra-
metropolitan level. But this will come at the expense of other localities,
generally within the same state, so the state gains nothing and local fiscal
effects cancel out. And if the state funds locally targeted incentives, the state
is merely spending money to move tax-paying firms from one place to
another; once again the local fiscal effects cancel out, but now the entire
incentive cost is a state loss. And these fiscal loses are not trivial; the cost per
job could be massive.
In addition, many critics argue that economic development incentives create a
zero-sum game. Zero-sum games describe the class of games where gains to
one party are strictly at the expense of the other. In the economic development
context, this characterization has been used to describe the outcome of the
bidding wars between jurisdictions. Critics of subsidy policies argue that the
competitive use of industrial subsidies, for the purpose of luring footloose
firms, leads to development at one location at the expense of development
elsewhere. Figure 2.2 depicts the zero-some aspect of this game in a simple
game matrix. The zero-sum aspect of this analysis implies, for example, that

Indianas effort to lure an electronics plant comes at the expense of the
location in Ohio that would otherwise have captured the investment.
Competition among Localities and States
State B
Subsidize Do not Subsidize
State A
Do not
Figure 2.2: Payoffs from competitive subsidization Note: a < c < b; a > b/2
(Wolkoff, 1990: 338).
To see the zero-sum aspect of this game, examine the payoffs that result when
one jurisdiction offers subsidies and the other does not. These payoffs
represent benefits to the community, net of subsidy costs, from new
investment. The upper right and lower left comers of figure reveal that if only
one community uses subsidies, all development will occur in the subsiding
community, at the expense of its neighbor. When both communities choose
the same subsidy strategy, the incentive effects are neutralized and each can
expect to attract investment with probability .5 (upper-left and lower-right
Depending on the values of the payoffs parameters, communities may choose
to provide subsidies even though they would be better off if both agreed not
to. This claim holds only under very restrictive assumptions about the
(c/2, c/2) (a,0)
(0, a) (b/2, b/2)

sensitivity of investment to price changes. If investment occurs strictly at the
expense of competing jurisdictions, the subsidies have had no effect on the
size or composition of the investment, only its location. It is far more likely
that the quantity and composition of investment is endogenous to the choice of
subsidy used. Investment is price sensitive, so industrial subsidies will not
only change the location of investment, but will, in addition, yield a greater
quantity of investment. As a result, not all jurisdictional gains in this game are
at the expense of the non-subsidizing locality (Wolkoff, 1990).
After having discussed substantial critiques developed against economic
development incentives, we can simply ask if reality is that clear why state or
local governments still pursue offering incentives. It is argued that in a world
where other localities are pursuing development, it would appear remiss not to
follow suit, whether or not the policies have any practical effect (Keating,
1995). If development does occur, then the local political leader can take
credit only if there is a policy in place. Conversely, if the development goes
elsewhere, the politicians in the losing community will be blamed for not
having a policy. Keating illustrates this in the form of a matrix as the
Policy No Policy
Development A B
No Development C D
Figure 2.3: Why local governments pursue offering incentives (Keating,
In cases A and C, there is a policy in place. If there is development, the
politicians take the credit. If there is no development, they can say that they
are responding to the problem to the best of their ability. In cases B and D,

there is no policy in place. If development occurs, the politicians can take no
credit. If there is no development, they will be blamed for not having a policy.
Indeed, empirical study of the enactment of various economic development
policy instruments has demonstrated that some of most expensive policies,
such as tax abatements for business, are the most popular. Rubin (1988)
argues that many practitioners implement economic development policy
without much information about their effectiveness. This author names this
situation shoot anything that flies; claim anything that falls. At the same
time, some analysis has suggested that certain relatively inexpensive policy
instruments may actually be more effective. The continued popularity of these
expensive policies may be related in part to the competitive environment
facing state and local governments, desperate to do something regardless of
cost as well as the high visibility of many of these policies. Highly visible
incentives, especially those that can be given out in an ad hoc fashion may
provide local officials an opportunity for credit claiming( Feiock, 1989).
Finally, Luger and Bae (2005: 327) argue that most state tax incentive
programs have been enacted and implemented for political rather than cost-
benefit or cost-effectiveness reasons. Similarly, Brunori (2001 in Luger and
Bae, 2005) argues that most tax incentives are the product of interstate
competition to attract businesses from other states, or what he calls an arms-
race mentality2.
2 Arms race is known as security dilemma in the international theory. As there is no weapon
that can be used for solely defensive or aggressive purposes, the military defensive measures a
state takes may easily perceived as offensive by the other state. This reality leads to an arm
race in which the safety of one party as a rule creates insecurity in the other, which,
ultimately, increases rather than reduces insecurity in both camps (Isakovic, 2000:215).

Methods to Evaluate Effects of Economic Development Incentives
There are two basic quantitative methods used to assess the relative merits of
an economic development program or project. One, economic impacts,
focuses on the economic performance while the other, fiscal impacts, focuses
on the net revenue consequences for public agencies resulting from the
activity. Economic impact analysis, quite simply, can be used to examine the
changes that occur as a result of economic development investments. These
changes can be measured in different units, such as new employment or new
gross regional product. Fiscal Impact Analysis involves examining the costs
of public policies and assessing their relative value to the taxpayer. Two basic
strategies can be employed in conducting a fiscal impact analysis: a cost-
effectiveness analysis and a cost-benefit analysis. Cost-effectiveness analysis
examines the relative costs of providing a specific number of units of a
public service. Those units are often measured in terms of outcomes. For
instance, what is the number of jobs created per dollar of public funds
invested? Hybrids of these methods integrate efforts to assess both economic
and fiscal impacts. Each provides policy makers with an important gauge of a
programs merits, but no one method can provide precise results about what
will happen in the future. The data required for both methods involves a
combination of secondary and primary data (Poole et al., 1999). Below
various types of evaluation methods are discussed briefly.
Experiments represent a way to study an economic event or intervention by
controlling for all variables except the one being studied. An economic
development experiment might involve making an incentive available to
certain companies but not making the incentive available to similar companies
to assess whether there are any differences in firm behavior.

Structured Surveys involve asking and analyzing questions about a subject to
identify trends or relationships between different observations. Most
economic development surveys are cross-sectional, meaning they are
conducted once and the data are used to compare the subjects of the survey to
one another at one point in time.
Case Histories involve the collection and analysis of a few case examples or
histories. For instance, a company selecting a site for a new production line
might be studied as a case because the decision making process is quite
complex and there is only one relevant observation (or case).
Archival Studies often involve the use of secondary data (such as employment
or tax revenues) and the analysis of that data through a variety of qualitative
and quantitative techniques. Frequently, data gathered about a project are
analyzed as inputs to an economic or fiscal impact model. These models are
created to estimate variables (as in econometric models) or compare the
relative value of variables (as in cost-benefit models).
Each of these strategies is particularly useful for addressing certain types of
policy questions. For example, experiments or case studies may be appropriate
in addressing questions that challenge the notion that a program intervention
results in a certain benefit. Questions that merely ask about the impacts of a
program often use survey and archival evaluation strategies (Poole et al.,
Key actions that should be taken for undertaking program evaluation can be
summarized as the following:

Identify the programs or activities to be evaluated;
Articulate the reasons for conducting the evaluation;
Develop goals to be achieved by the evaluation;
Identify actors and their respective roles in the evaluation process;
Identify planned uses of the evaluation results;
Establish decision rules for judging program performance;
Determine the data needs and sources;
Determine the analytic tools to be used; and
Determine the performance time period to be assessed (Poole et al.,
Existing research suggests that future economic development incentive
research should (1) carefully select appropriate control groups (i.e., use an
experimental evaluation research design); (2) use several economic
performance measures and examine trends over the long term (i.e., use time
series methods); (3) use net change indicators to judge impact; (4) implement
basic cost-benefit analyses as the first step, but incorporate opportunity cost
analysis in defining the cost factors; (5) attempt to research externalities and
spillover effects resulting from changes or interventions in the economy; (6)
examine incentive impacts on business profitability levels; and (7) examine
program impacts on particular industry sectors (Poole et ah, 1999: 38).
However, it should be remembered that incentives represent only one of many
factors that may influence economic growth and development. As a
consequence, only limited information exists about whether these incentive
programs have a causal effect on business behaviors and economic growth
(Poole et ah, 1999). For example, effect of a small tax incentive program on
local economic growth is hard to isolate even though the program may be

effective, because program effects can be dominated by uncontrolled local
factor and characteristics (Bartik, 1991 in huger and Bae, 2005).
The European Commission has developed MEANS approach to evaluation for
the structural funds. This approach recommends a combination of top-down
and bottom-up evaluation techniques (for cross checking purposes). Top-
down methods analyze secondary data sources, while bottom-up methods
collect and analyze primary data through sample surveys of beneficiary
businesses and individuals. Top-down techniques seek to estimate global
impacts (for example GDP increases or employment), whereas the bottom-
up technique proceed by identifying microeconomic impacts on individuals
and firms. The micro impacts are then aggregated to give global effects
(Armstrong et al., 2002).
Future of Economic Development Incentives
It is argued that the problem is that many incentives currently being offered in
the United States have costs that exceed benefits. For example, in 2001
Chicago awarded large incentives to the Boeing Corporation for relocating its
headquarters, even though the jobs would go to relocated workers, which
eliminates many of the labor market benefits (Bartik, 2005).
If all states had rational incentive regimes, on the margin investors in each
state would be charged a tax rate net of incentives that would reflect the
marginal public service and environmental costs, net of any employment
benefits, that the investment caused. Under those conditions, a state incentive
that attracts a marginal plant that would have otherwise gone to another state
causes no net cost for that other state. Of course, states do not usually have
rational incentive regimes, and so it is likely that attracting this marginal plant

would cause net social costs (or benefits, depending upon the net effect in the
other state) for the state that otherwise would have received this investment.
But this externality is not the fundamental problem rather it reflects the
fundamental problem: each state lacks, from its own self-interested
perspective, a rational incentive regime that maximizes the interests of all
state residents (Bartik, 2005).
Alternative Incentive Reforms
Maintain traditional state and local policies towards business, but remain
competitive in the global economy: State and local governments would
maintain their traditional business tax systems and not offer any incentives or
other business cost reductions to improve the local business climate (Oakland
and Testa, 1996 in Bartik, 2005). According to Bartik, that is not politically
viable. He argues that the argument for doing something will win out over the
argument for doing nothing.
Localism: Rather than competing for mobile capital, local areas could just
say no, eliminate incentive for mobile corporations, and rely on locally-
owned capital. Self sufficiency and import substitution are emphasized in
localism. The main problem with this approach is that greater reliance on local
capital and local production would significantly reduce an areas real per
capita income because there are static and dynamic gains from trade and
capital mobility (Bartik, 2005).
Develop unique local assets that yield economic rents: One unique asset
would be a unique cluster of industries that increases productivity by resulting
in more new ideas, and greater availability of specialized inputs such as
workers with special skills (Rosenfeld, 2002 in Bartik, 2005: 148). Another

unique local asset would be distinctive local amenities that attract what
Florida (2002 in Bartik, 2005: 147) calls creative class, the professional and
technical workers who enhance productivity for many high-tech businesses.
The problem is that for most local areas, it is difficult to develop unique
industry clusters or local amenities.
Make discretionary incentives truly selective: Michigans MEGA program,
for example, requires that all projects present data showing that a non-
Michigan site would be more profitable without the incentive, and that all
projects be subject to fiscal impact analysis. The problem is that whether
government officials can determine if the incentive is needed to tip the
location decision is questionable (Bartik, 2005).
Metro-wide economic development programs, not within-metro competition:
State governments can require that incentives not be provided by individual
local governments, but only by the state as a whole or by metro-wide
organizations or coalitions. There is considerable disagreement within and
between states and localities as to how incentive programs should be used in
competing for new business investment, jobs, income, and taxes. The research
found no existing agreement at this time among the states or communities to
work toward a consensus on the use of incentive programs. Support for
federal intervention to coordinate state and local incentive activities in the
future is practically nonexistent. At the same time, voluntary agreements
among states on the use of incentive programs to pursue economic
development opportunities seem impractical and idealistic to many policy
makers and practitioners, given the political realities and the competition for
investment and job creation (Poole et al., 1999).

Better benefit-cost analysis: State legislatures require that all economic
development incentive offers to be subject to a prospective benefit-cost
analysis to estimate whether the incentive offer is cost-efficient. This analysis
would estimate: employment benefits, including what proportion of the new
jobs would likely to go to local residents, particularly unemployed local
residents; wage effects, including the wage rate paid on the jobs for workers
of given credentials versus local jobs held by local residents with similar
credentials; fiscal effects, including local as well as state effects, effects on
required public expenditure as well as taxes, and analysis of the capacity of
existing infrastructure to accommodate job growth; and environmental effects.
If the estimates are high quality, they increase the likelihood of the right
incentive choices (Bartik, 2005). However, it should be noted that these
estimations are very difficult and complex.
Clawbacks: The net benefits of an incentive regime are increased when some
of the incentive can be recovered if the business receiving the incentive does
not provide the planned social benefits. For example, if the business relocates
or the number of jobs created fall short of projections. This can be dealt with
by legally-binding clawback provisions, which recover some portion of the
up-front incentives if the business does not meet performance goals. The main
potential problem with clawbacks is that they being onerous, they may be a
disincentive to attracting businesses. However, if clawbacks are designed so
that incentives and clawbacks are clearly related to the social benefits
associated with the business, the businesses that expect to make a long-term
investment in the community should not perceive such as clawback as a huge
disincentive to their location decision.

To overcome barriers to the efficient working of local labor markets,
economic development incentives should be used to overcome barriers
preventing the use of brownfields, and to encourage used of underutilized
public infrastructure (Bartik, 2005). It is argued that social benefits will also
be greater if the current public infrastructure is utilized, which allows
increased employment and population growth to increase tax revenue without
a commensurate increase in public infrastructure costs.
Although there are problems with current incentive programs and solutions to
these problems are still immature, what it has been recently across the county
is that states began to be aware of importance of offering incentives. For
example, New York continues to offer incentives to firms that are making
location decisions they believe that they are likely to lead to decisions in favor
of New York and where state officials can be sure that the benefits to state
residents outweigh the costs. In support that effort, Empire State Development
Corporation (ESD) has implemented the use of a formal cost-benefit analysis
model in making decisions about incentives offered (Wiewel, Persky &
Felsenstein, 1994 in Bacheller, 2000).
Most recent studies indicate that states are more careful than before in offering
economic development incentives. For example, the criteria of New Yorks
Empire State Development determine that an economic incentive is
If a competing state has offered the firm an attractive economic
incentive package,
If the firm can show that it would be operating at a severe competitive
disadvantage without the incentive package,

If the firm has a viable business plan, but has not attracted private
New Yorks economic development agency also uses industry cluster analysis
as an organizing concept to guide it marketing, business assistance, and
infrastructure development efforts. Among the approaches used by ESD to
maximize the effects of its business assistance programs is the provision of
assistance to groups of firms collaborating to improve their competitiveness
(Bacheller, 2000). Michigans MEGA program, for example, requires that
all projects present data showing that a non-Michigan site would be more
profitable without the incentive and all projects are subject to fiscal impact
analysis (Bartik, 2005: 149). A Vermont law requiring that an economic
impact statement be conducted to evaluate large development projects
provides example where a more expansive understanding of the external costs
imposed by private investment leads to innovative policy. For example, based
on the economic impact statement required by this law, a citizen review
commission rejected Wal-Marts attempt to open a large store in state because
when the commission examined the full range of economic effects applying
the type of full-cost accounting used by these impact statements (an
accounting method not unlike a public balance sheet), it concluded that this
prospective investment would cost the community $3 for every $1 in benefits
it created, as a result of its adverse (external) impact on local businesses and
infrastructure (Morris, 1998 in Imbroscio, 2004: 89).
Long and sometimes inconclusive discussions in the previous sections reflect
the complexity of economic development and seem to continue in the future
too. However, in the literature it is well documented that local governments

offer many business incentives not necessarily based on cost-effectiveness,
but most likely for political reasons. This concluding section briefly examines
reasons that economic development incentives have become so common on
the ground of market failure and suggests some policy directions for local
Scarcity of Information
Since very little systematic information exists on how public officials think
about local economic development and how they make policy decisions in
that realm (Wolman, 1988 in Bumier, 1992), some see market failure a main
reason for ineffective economic development policies and they argue that
intervention to market will lead more effective and efficient economic
development. It is argued that in contrast to an arms race of incentives, an
arms race of information between localities could improve efficiency by
promoting socially optimal location choices through better comparative
information. In fact, the focus on marginalized areas may reveal the hidden
potential of previously ignored regions. In particular, the provision of
information on both private and social rates of return may serve to spark
beneficial entrepreneurship. First, adequate information on private costs and
benefits in otherwise thin markets may be enough to motivate private
investors, promoting private market efficiency. Second, high social rates of
return may provide a rationale for the public support of otherwise private
ventures, resulting in socially beneficial entrepreneurship (Weiler, 2000).
Classical economics theory of the market assumed perfect knowledge among
buyers and sellers. According to the Austrian School, markets are processes of
discovery that mobilize dispersed information (Jacobson, 1992 in Bovaird,
1993: 642), but Malizia and Feser (1999) argue that even in an age of

information technology, the assumption of perfect information is unrealistic.
Given its cost, many buyers and sellers must act with information that is
highly inadequate. The inadequacy results in markets that are imperfect and
fail to allocate resources optimally or in many cases even satisfactorily.
Information is one of the clearest of public goods and it is in the effective
performance of the citys economy that it should be developed and publicized.
It is argued that the citys economy can be conceptualized in terms of various
demand and supply curves for a variety of goods and services. Many of these
demand and supply curves are implicit rather than explicit and hence fail to
produce a desirable response (Long, 1987).
It is argued that The economic rationality of the profit-maximizing for
private corporation does not necessarily lead to welfare maximization for
individuals and communities, since the market does not adequately measure,
and in fact explicitly ignores, the public costs and benefits of private
investment decisions (Smith, 1979 in Imbroscio, 2004: 84).
The public balance sheet technique demonstrates how the actions of the
private sector impose significant costs on the public (and the larger
community). Conventional (i.e., neo-classical) economic theory treats these
social costs as externalities costs that stem from market processes but
that will not be experienced by market actors. Instead, these costs are
externalities that is, transferred to others, usually the community at large.
The presence of externalities creates an instance where the competitive market
will fail to maximize social welfare: Because of the divergence between
private and social (i.e., total community) returns, write Stokey and
Zeckhauser (1978 in Imbroscio, 2004) in their classic statement of neo-

classical theory, uncoordinated individual actions lead to less optimal
results. Together with public goods/bads (essentially a variation of the
externality problem), the need to correct for this market failure provides
the basic justification in conventional economic theory for government
intervention into the workings of the market economy. As such, this reasoning
underlies neo-classical economics normative theory of the state, as Bobrow
and Dryzek correctly note (1987 in Imbroscio, 2004).
Competitive Advantages
It is suggested that to have any real advantage, a local industry has to be the
most efficient producer in the world. Since there is no real difference between
local and world prices, a local industry competes globally with others in the
same industry to gain a competitive advantage. In the old comparative
advantage system, a local industry competed locally against, other local
industries to show that it was relatively more efficient. In the old system, it did
not matter if other of an industry in other local economies were absolutely
more efficient; now it does (Rosentraub and Przybylski, 1996). Also the idea
of successful economic development from going with your strengths is not
The ability of any area to sustain economic development based on
comparative advantages is becoming more and more difficult; thus, there is a
question of the value or merit in any set of publicly financed subsidies.
Technology shortens the term of many local advantages in factors of
production. Areas with advantages from inexpensive labor are replaced by
other regions even lower-priced labor. Areas with high-cost labor can offset
this comparative disadvantage with technology that increases productivity.
Local assets, such as natural resources or proximity to large markets, have

declined in importance as changing extraction and transportation technology
makes even comparative advantages based on location less and less critical.
Subsidies (incentives) from the public sector, designed to enhance
comparative advantages or offset comparative disadvantages, are only
effective to the point that technology is static. Further, a subsidy based on cost
factors that exist today will have to be increased tomorrow, when new
technologies redefine comparative advantages. As a result, locational
decisions based on subsidies designed to affect factor-based comparative
advantages are temporary and bound to require an expansion. Any firm that
receives a subsidy has a reduced incentive to identify new or enhanced
technologies since the existence of the subsidies reduces the immediate need
for the new approach or technique. Such an outcome, of course, reduces the
firms competitiveness since non-subsidized competitors will seek new
procedures or technologies in an attempt to reduce any comparative
advantages that result from subsidies.
The decreasing importance of factor-based comparative advantages has led to
surge of interest in formulating policies that enhance competitive advantage.
Porter, as the leading advocate of this perspective, is quite persuasive in his
observation that subsidies to change factor-based comparative advantages will
generate, at best, temporary gains. Porter, as well as others, maintains that an
economys advantages come not from factor-based comparative advantages
but from the consistent ability to innovate and lead in certain industries. This
competitive advantage still emphasizes an areas strengths, but these assets are
defined in terms of the competitive edge developed by local industries
compared to others in their industry throughout the nation and world.

A focus on competition instead of factor-based comparative advantages
accepts two basic premises. First, any economic advantage that rests on
factor costs is vulnerable to even lower factor costs somewhere else, or
government will to subsidize them. Todays low cost country is rapidly
displaced by tomorrows. The lowest-cost source for a natural resource can
shift overnight as new technology allows the exploitation of resources in
places heretofore deemed impossible or uneconomical. Second, an economic
advantage based on a firms competitiveness and the culture of competition
created by a cluster of firms is not easy to replicate or to subsidize into
existence (Porter, 1994 in Rosentraub and Przybylski, 1996).

Enterprise zones (EZs) are geographic areas which are targeted for
government-sponsored business and economic development programs. This
geographical or spatial targeting is the unique aspect of EZs (Wilder and
Rubin, 1996). Enterprise zone areas are characterized by unemployment, low
incomes, inappropriate infrastructure for new businesses, and low tax base.
The enterprise zone concept originated in the United Kingdom and first
proposed by Peter Hall (1977 in Belasich, 1993). The goal was to turn around
derelict areas (land severely damaged by previous generations of economic
activity) of cities by removing any and all government control. Industrial
location theory, business cycles, and the observation of rapid growth in Third
World Economies constitute theoretical basis of enterprise zones. In the
United States, the concept was introduced by congressmen Jack Kemp and
Robert Garcia in the early 1980s. The concept of public-private partnership
and limited government intervention into economic affairs are two focal
points or policy streams that have guided enterprise zone program
development in the U.S. (Brintnall and Green, 1988). 3
3 After returning from a tour of the tax-free zones in the Far East in the mid-1970s, Hall
envisioned the creation of Asian free-ports inside the most economically distressed inner
city areas of Great Britain, similar to those that led to the remarkable development in Hong
Kong and Taiwan in the 1950s and 1960s (Brintnall and Green, 1988).

Although the concept imported from the Britain, the American conditions and
goals of enterprise zones are largely different from the British experience.
American goals are directed to business expansion and employment of target
population rather than to the land. Derelict land is not an American problem,
but depressed inner cities and rural economies are. EZs are also seen as one
tool for fostering economic development, often in combination with other
traditional state and local programs in many states in the U.S. (Wolf, 1990).
Enterprise zone programs generally seek to influence the development of a
more positive business climate through two elements: tax incentives to
businesses located in a zone, and economic development incentives which can
create a more favorable zone business climate (Belasich, 1993). The first type
of subsidies is an upfront subsidy that reduces the firms initial investment
required to relocate or remain in the zone, such as subsidized land and
building costs and investment tax credits. The second is subsidies to labor and
capital that offset the higher operating costs of businesses in the zone. This
may include employee tax credits, reduced regulatory requirements, lower
interest rate financing, and reduced state, local property tax rates (Mauer and
Ott, 1999).
Review of the Enterprise Zone Evaluation Literature
Although the economic development literature often discusses the potential
effects of enterprise zones, empirical research on, analysis of zone programs is
limited. According to Wilder and Rubin (1996), the modest amount of
empirical research is due to two basic constrains: (1) the lack of reliable
quantitative data to evaluate zone performance, and (2) the difficulty of
isolating the effects of zone designation and incentives from these of other
economic development factors and initiatives.

The first direct study of an enterprise zones economic effects was carried out
by Earl Jones (1985) in a case study of Bridgeport, Connecticut. Jones
compared the number of property transfers and building permits issued within
the zone to those in a similar, but nonzone area. He found only marginal
differences in transfer and building activities between the zone and nonzone
One of the most comprehensive studies of enterprise zone impacts was carried
out by the U.S. Department of Housing and Urban Development (1986). HUD
assessed ten enterprise zone programs in nine states, using data derived from a
series of interviews with approximately 200 state and local officials, as well as
with business and neighborhood representatives. The survey results showed
that 263 zone firms were responsible for over $147 million in new investment,
and 7,348 new and retained jobs. The HUD study acknowledged the difficulty
of obtaining reliable data on zone effects and cautioned that the observed
outcomes for jobs and investment could not be attributed solely to the
enterprise zone programs (Wilder and Rubin, 1996)
The issue of data bias and/or reliability can be raised with many of the studies
that rely on interviews and perceptions rather than verifiable data. For
example, Wilder and Rubin (1996) argue that HUDs study findings must be
qualified because it used interviews with zone officials, who are potentially
biased by a natural tendency to attribute any job growth within the zone
exclusively to the enterprise zones influence. Such officials tend to ignore job
losses, focusing on gross rather than net job growth.
In the subsequent years, an initial assessment of the California Enterprise
Zone Program was conducted by the Office of the State Auditor General in

1988. By using a sample survey of the firms in the ten zones, rather than
interviews with zone officials, this study employed a more reliable
methodology than that in the HUD study. It was concluded that economic
activity had increased somewhat in the states original ten enterprise zones
and three incentive areas. A subsequent study of zone firms by Litster (1990
in Wilder and Rubin, 1996) directly challenged the accuracy of the Auditors
report. Analyzing responses to a survey of zone businesses Litster found that
2,518 (rather than 3,600) new jobs had been created in the California zones.
A more recent evaluation of the California zone program was done Dowall,
Beyeler, and Wong (1994) by using County Business Patterns on employment
and establishments to estimate job generation and business development.
These researchers found job growth across most of the zones and incentive
areas. It is worth noting that this study introduced County Business Patterns (a
database which is more reliable than interviews with zone officials) as a new
data source (Wilder and Rubin, 1996).
According to Wilder and Rubin (1996), the exact nature of the assumed
relationships has not been adequately established in these previous studies.
Two important unresolved issues were raised: (1) What factors explain the
variations in zone effects? (2) Can observed changes in economic activity be
directly attributed to enterprise zone zones? A study by Sheldon and Elling
(1989) sought to address the first of these questions. The second question has
been addressed in studies by Rubin and Wilder (1989), and Erickson et al.
In an effort to explain the variability in zone program outcomes, Sheldon and
Elling (1989) studied the economic attributes and effects of 47 enterprise

zones in Illinois, Indiana, Kentucky, and Ohio. It was found that although
each of these states had succeeded in attracting some new investment and
employment to at least some of its zones, the researches also found a high
degree of intra-state variability. Within each state there were examples of zone
that had exceeded program objectives as well as of zone that had failed to
meet them. Sheldon and Elling concluded that variations in investment and
job generation across zones were most likely to be due to differences in: (1)
local economic conditions (e.g. costs and condition of transportation, labor
conditions, tax rates); (2) state and regional market conditions; and (3) state
economic development (e.g. technical assistance, location /site analysis,
special staffing).
The Sheldon and Elling analysis is useful in sorting out the variables that may
affect business investment and location. What is less clear is how these
variables affect job generation (Wilder and Rubin, 1996). That question has
been examined in two other studies. Wilder and Rubin (1988) undertook a
detailed case study of the Evansville, Indiana enterprise zone. Using firm level
survey data, they found that from 1983 to 1986, 47 new businesses had been
created in the zone, and 14 existing businesses had expanded. Wilder and
Rubin concluded that the relative success of the Evansville zone was due to:
(1) the economic viability of the zone area, (2) a high concentration of
industrial land use, (3) autonomy of the zones administration from local
government, (4) the credibility and competency of the zone manager, and (5)
the community organization role by the zone association (Wilder and Rubin,
1996). These researchers conclusion should be interpreted cautiously because
they examined only the 3-year effect of the enterprise zone program on
economic development. If the same study is repeated now, it is likely to find
opposite results because, as we will see in the case of California and

Colorado, results of short and long term evaluations of the same program may
be contradictory.
In a more comprehensive study, Erickson, Friedman, and McCluskey (1989)
carried out a cross-sectional analysis and analyzed HUD data for 357
enterprise zones in 17 states, and 2,014 business establishments. The authors
examined the relationship between the structural characteristics of enterprise
zone programs (e.g., number/type of incentives, number/size of zones) and the
observed changes in employment and investment. By interviewing
coordinators of 21 high performance zones, the researchers tried to identify
four common explanations for zone success:
The zone encompassed an area that was still economically viable,
Zone designation served as a catalyst/or stabilizer, but was not the sole
determining factor in re-vitalization,
The number and variety of zone incentives added to the programs
effectiveness. However, tax incentives (e.g. property, inventory, and
sales) were the most frequently used,
Strong local support from the private and public sectors increased
Although the researchers concluded that their analysis did not establish a clear
casual relationship between enterprise zones and observed economic growth,
they felt that considerable evidence pointed to a significant marginal and
catalytic role (Erickson et ah, 1989 in Wilder and Rubin, 1996). While the
results obtained by Erickson et al. imply that some zones are successful, there
are two caveats about their methodology. The first is that their data were