An analysis of the relationships between agriculture and the central place economies on the eastern plains of Colorado

Material Information

An analysis of the relationships between agriculture and the central place economies on the eastern plains of Colorado
Marx, Edward C
Place of Publication:
Denver, Colo.
University of Colorado Denver
Publication Date:
Physical Description:
112, A1-A2 leaves : charts, maps ; 28 cm

Thesis/Dissertation Information

Master's ( Master of Urban and Regional Planning)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
College of Architecture and Planning, CU Denver
Degree Disciplines:
Urban and Regional Planning
Committee Chair:
Schler, Daniel J.


Subjects / Keywords:
Agriculture -- Economic aspects -- Colorado ( lcsh )
Retail trade -- Colorado ( lcsh )
Agriculture -- Economic aspects ( fast )
Retail trade ( fast )
Colorado ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 110-112).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Urban and Regional Planning, College of Architecture and Planning
Statement of Responsibility:
by Edward C. Marx.

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:
15581055 ( OCLC )
LD1190.A78 1986 .M38 ( lcc )

Full Text

r- P

Edward C. Marx
Date Due

A thesis submitted 1n partial fulfillment of
the requirements for the degree of
Master of Urban and Regional Planning
College of Design and Planning
University of Colorado

To my wife, Aural1e without her support and
understanding this work would not have been possible.

This thesis examines the relationship between the regional
agricultural economy In eastern Colorado and the retail trade
economies of the small town central places 1n the region. Retail
sales and several Indicators of the agricultural economy are analyzed
over the period from 1963 to 1982. Other Important factors
Influencing retail trading patterns* Including population* Income*
transportation network Improvements and regional shopping center
development* are also examined.
Data 1s analyzed for sixteen non-metropolitan counties 1n eastern
Colorado. Trends are tracked based on data for the five years between
1963 and 1982 for which economic censuses were published. The
hypothesis tested 1s that differences and changes 1n the agricultural
economy* specifically 1n the number and average size of farms* would
affect the functional hierarchy of central places 1n the region. This
effect would be exacerbated during the study period by changes 1n
retailing generally* especially the development of regional shopping
centers. Based on central place theory 1t was expected that decreases
1n the farm population and the" development of regional shopping
centers just to the west of the study area would change the locations
at which certain retail goods were offered within the study area.
Data 1s reported for each county 1n the planning region and 1s
aggregated for the entire study area. Data 1s also analyzed for
combinations of counties by population of the largest city 1n the
county and by state planning region. These analyses point out
functional differences 1n the retail economy by size of central place

Significant change occurred 1n the farm economy 1n the region.
An Increase 1n Irrigation from deep groundwater and plowing of
rangeland for dryland farming both contributed to an Increase 1n the
acres of harvested cropland In the region despite a slight decline 1n
total farmland. The number of farms decreased and the average farm
size Increased 1n most of the area's counties.
Correlation analysis Is applied to Identify potential
relationships between the factors studied. The statistical analysis
leads to a rejection of the hypothesis that changes 1n retail trade
are related to change 1n farm size. Other factors# such as location
relative to transportation networks# proximity to larger urban
commercial centers and existing population 1n the central places
appear to be more Important 1n explaining differences 1n the
performance of the retail economy. However# the data does suggest a
relationship between the productivity of the agricultural resource
base and the central place population which can be supported In an
area. Part of the explanation for this relationship may be the fact
that a more productive resource base supports a higher level of farm
service employment. Also# a more productive agricultural resource
base tends to result 1n smaller farms and a more dense rural farm
Finally# 1t 1s observed that changes which occurred 1n
agriculture during the study period are unlike changes now occurring
due to the combined effects of low farm prices and high farm debt.
Therefore# the results of this analysis may not be able to predict the
likely Impact of on local communities of current changes 1n

and geographic differences within the agricultural resource base of
the region.
The specific Indicators examined to explain differences 1n retail
economies are total retail sales# per capita retail sales# percent
capture of resident retail expenditures and number of retail
establishments. Indicators of the farm economy are number of farms#
average farm size# acres of harvested cropland and value of
agricultural production. Also examined 1s employment In businesses
which provide goods and services directly to farmers. The development
of regional shopping centers and Improvements 1n highway
transportatlpn are evaluated by estimating travel time from the study
area counties to the nearest regional shopping center.
The analysis of trends 1n the retail economy shows that the there
was Increased outflow of residents' consumer expenditures from the
study area during the study period. The function of providing
shoppers goods such as apparel and furniture was Increasingly
concentrated 1n the larger communities In the region and communities
with fewer than I#000 residents also saw significant decline 1n
establishments providing convenience goods like food and drugs.
Constant dollar retail sales Increased 1n counties located on the
major highways 1n the region and decreased 1n counties without direct
access to these highways.

Population and Income 29
Retail Trade 34
Trends 1n Retail Markets 55
The Farm Economy 61
The Farm Support Economy 72
Summary of Regional Trends 76
Section 7. CASE STUDIES 89
Crowley County 90
K1t Carson County 93
Logan County 98
Section 9. CONCLUSION 108

Rural America Is today facing conflicting economic pressures.
Many small towns* particularly those close to major metropolitan
areas* are growing rapidly as people seek a lifestyle combining the
advantages of a small town residential environment with the cultural
amenities and employment opportunities of larger cities. This
contrasts with the situation in agriculturally-based rural America
where tough times 1n the farm economy are rippling through the
economies of many small communities. It 1s the latter trend which has
prompted this research although the former development 1s also
reflected 1n the results. This thesis Investigates the relationship
between the agricultural economy and one of the key aspects of the
economies of small towns* the retail trade component of the central
place economy.
Small towns In predominantly rural regions have traditionally
functioned as commercial trade and service centers for the farm
population. Economic geographers have theorized that a hierarchy of
communities exists 1n agriculturally based regions and that each level
1n the hierarchy performs a different mix of economic functions.
Central place theory* the theoretical description of this hierarchy*
was developed 1n depth by Walter Christaller based on h1s study of
such cities 1n Southern Germany 1n the 1930s. (Christaller* 1966)
Other well known economic geographers. Including August Losch (Losch*
1954) and Brian Berry (Berry, 1967), have built upon Christaller's

Central Place Theory 1s based upon the relationship between small
commercial centers and the hinterland# or rural region# which these
centers serve. Different goods and services are provided at different
levels 1n the central place hierarchy. For example# groceries may be
available 1n the smallest class of central places# clothing only at a
larger class of central place and furniture at only still larger
places. (Yeates# 1971) There are several factors which contribute to
this relationship. Foremost among these 1s the microeconomics of the
retail firms which provide the goods and services. By producing or
selling large quantities of a good or service a firm can often offer
the good at a lower cost than can an Individual or firm producing a
smaller quantity of the product; that Is# the firm can achieve
economies of scale. However# the cost differential must more than
offset the consumer's cost of travel to the market at which the good
1s offered. As travel costs Increase with distance from a market
center# Intervening opportunltes may arise for other merchants to set
up shop and offer goods at a lower combination of price and travel
cost. In this way a hierarchy of central places evolved 1n
agriculturally based regions as towns grew up to serve the surrounding
farm population. (Christaller# 1966)
The distribution of the providers of a good or service In a
region 1s based upon three factors: the degree to which economies of
scale can be achieved 1n providing the good; the density of population
1n the region demanding the good; and the relative costs of
transportalon to the market centers where the good 1s offered.
(Hellbrun# 1981) Goods for which significant cost reductions can be
realized by achieving economies of scale will be offered at more

dispersed trading centers; the more densely populated a region# the
closer comparable sized trade centers will be to one another; and the
greater the ease of tranportatlon within the region# the more
dispersed the trade centers are likely to be.
Another Important perspective on central place theory Is provided
by considering consumer behavior. In a very general sense# the less
frequently a good or service 1s purchased by a consumer# the larger
the central place at which the good or service 1s offered. Frequently
purchased Items are usually sought at locations close to home and
variability 1n price of these goods between establishments 1s
relatively low. Therefore# not only 1s 1t Inconvenient for the
consumer to travel long distances to purchase these goods but the
benefits 1n terms of reduced price of doing so are unlikely to offset
the cost of travel to more distant markets. By contrast# for less
frequently purchased goods# consumers are more willing to travel
longer distances and the relatively higher price of many of these
goods and greater price and qualitative differences 1n the
marketplace# make comparison shopping more desirable.
These aspects of Central Place Theory may be summarized In terms
of two fundamental concepts of retail marketing# the range and
threshold of a good. The range of a good 1s the distance that a
consumer Is willing to travel to purchase a good. (Watkins# 1980)
This distance 1s a function of the cost of travel to a market and the
price of the good at that market. If a consumer can obtain a good for
a lower combination of product cost and travel cost at an alternative
location he will# 1n theory# travel there to purchase the good.

The threshold of a good 1s the population of a market area
required In order for a good to be offered at a particular location.
The threshold essentially measures the demand necessary to keep a shop
owner offering a specific good or service 1n business. In order for a
good to be offered 1n a central place* the range of that good must
encompass an area with a population that provides a demand equal to or
greater than the threshold for the good. Thus* the localities at
which goods are offered depend upon the geographic distribution of the
population of an area. This fundamental relationship 1s a key
theoretical element of this analysis.
Economic geographers have conducted numerous studies of central
places* their distribution* and function over the last fifty years.
(Berry* 1965) This thesis accepts the validity of the essential
elements of central place theory: that 1n a predominantly agricultural
region a hierarchy of places has developed 1n which the larger places
offer a broader range of goods and services. Although the spatial
pattern of distribution of central places 1s an element of this
theory* It 1s the functional rather than the spatial aspect of the
central place hierarchy which 1s the primary focus of this analysis.
The thesis question examined Is whether the traditional
hierarchy of different sized central places 1s altered by differences
or changes 1n the underlying agricultural economy of a region. The
continuing decline 1n the number of farms 1n a region* for example*
could fundamentally change the geographic distribution of population
1n a predominantly agricultural area. The premise of this thesis 1s
that* absent other changes 1n the retail economy* the basic effect of
a decline 1n the number of farms would be to reduce the number of

goods and services which could be offered at a given level In the
central place hierarchy.
From 1963 to 1982# the period covered by this research# 1t was
anticipated that any Impacts of agriculture on central place economies
would Interact with two other factors: 1) Improvements 1n
transportation links to larger cities which have reduced travel time
and# thus# travel cost to these higher order central places; and# 2)
changes 1n the retail Industry toward offering goods 1n larger stores
and shopping centers drawing from larger geographic trade areas. The
lower prices and greater variety of goods offered by these larger
stores and centers# made possible by achieving economies of scale#
allow them to expand their range at the expense of smaller retail
trade centers while the threshold required to support these centers
weighs against such new retail development 1n smaller communities.
The general question being addressed In this thesis 1s not
entirely new. Rural sociologists have argued that there 1s a
relationship between the scale of agricultural production and the
community health of the small towns 1n a region. One measure of
community health often employed 1n studies purporting to show this
relationship 1s the level of services and range of goods available 1n
a community. Thus# a direct connection 1s suggested between the
scale of agricultural production 1n an area and the health of a
communitys central place economy.

Probably the most well known study of the relationship between
farm size and community welfare Is Walter Goldschmidt' s study In the
1940s of the California towns of Arlva and Dlnuba. (Goldschmldt*
1978) This study focused on two agricultural areas which had
developed under markedly different conditions of 1n terms of farm size
and* Goldshmldt argues* virtually all aspects of community life. In
one case agriculture had developed 1n a pattern with large
landholdings and hired farm labor. In the other* a system of family
farming with smaller acreages was the norm. Goldschmidt examined the
social and economic structure of the communities and concluded that In
the community surrounded by the smaller family run farms there was a
greater social and economic diversity and community cohesiveness than
In the town surrounded by large landholdings.
However* recent critics of this study convincingly argue that the
reason these areas developed differently was because they possessed
very different natural resource characteristics and that 1t was not
really meaningful to compare them as evidence of how the scale of
agriculture Influences communities. (Hayes1984) This research
attempts to avoid this potential analytical flaw by examining the
same area over a period of twenty years to show changes 1n the scale
of production which have occurred and by specifically examining the
role of the resource base 1n determining the scale of agricultural

In their 1982 article regarding farm size and rural community
Nucktln, Rochln, and Gwynn argue that the Issue of the relationship
between farm size and community health has been largely Ignored since
the work by Goldschmidt and call for a joint research effort between
agricultural economists and rural sociologists to analyze the overall
benefits and costs of Increasing the scale of farming operations.
Although this researcher has also uncovered little recent work focused
on this Issue, there has been research conducted over the last twenty
years which provides background and some Insight Into recent trends
among central places 1n rural regions.
One of the key factors tied to the long term decline of small
towns over the past forty years has been the historical trend toward a
declining rural population. (Johansen, 1973) Research has shown that
outmigration from rural market areas leads to a decline 1n the number
of retail stores 1n the communities serving these areas. (Walzer,
1979) This change can be explained under central place theory: the
decrease In demand for goods and services results In a range for goods
1n many small market centers which no longer provides the threshold
demand necessary to support the retailer offering the good.
Although decline In the rural population has historically been
the driving force behind decline 1n small towns, more recently other
factors have played Important roles 1n growth and decline 1n rural
regions. Research addressing change 1n rural communities during the
last twenty years, roughly the period covered by this thesis.
Identifies the size of the small town or city as a factor critical to
explaining changes 1n retail trade. Studies have documented declines
1n commercial functions 1n Midwestern towns with populations under

1000 (Hart# 1969)# and In Wisconsin communities of under 2500
population (Johansen# 1973). Other research has shown that the
probability of community economic decline 1n rural areas Increases as
the population of the community decreases. (Northam# 1969) A 1980
study of central places 1n northwestern Wisconsin found evidence of a
concentration of commercial activity In larger communities.
(McGranahan# 1980) As discussed above# the decline of rural
population generally suggests that this would be the case with the
smallest class of central places no longer able to attain the
threshold necessary to support most commercial activity. However#
other research has suggested that factors 1n addition to decreases 1n
rural population have Influenced the decline observed In the central
place functions performed by smaller communities.
A significant determinant of change 1n small towns has been
proximity to a larger urban area. Comparably sized communities close
to large urban areas have less commerce than do communities more
distant from metropolitan areas. (McGranahan# 1980) On the other
hand# population growth has been shown to be greatest In small towns
within twenty miles of cities of over 50000 population. From the
twenty mile mark decline Increases with distance from the urban center
to a distance of about 80 miles. Beyond 80 miles the probability of
decline decreases with further distance from and urban center.
(Northam# 1969) Communities close to large urban areas can attract
residential population growth but cannot compete for retail trade with
the larger concentrations of merchants 1n the larger city. (Johansen#
1979) Some researchers have argued that towns proximate to much
larger cities experience symbiosis# with the larger cities providing

goods and services and the smaller towns attracting residential
development. Small towns located near cities which are only slightly
larger compete for both population and retail trade. (Butler# 1970)
An Important component of the relationship between small towns
and large urban areas 1s the transportation networks which link the
communities. Advancements 1n automobile technology and the
accompanying Improvements 1n roads and highways have Improved access
to large urban centers. It has been argued that this Improved access
has led to a restructuring of regional retail economies# with major
Impacts on communities of less than 2#500 residents. An explanation
based on central places theory 1s suggested with an Increase 1n the
range of urban retailers who are able to extend their market areas at
the expense of small town retailers. (Johansen# 1973; Walzer# 1979)
All of the factors discussed above have led researchers to the
consensus that small communities are undergoing functional change.
The smallest class of communities Is losing the function of commercial
center and taking on a number of more limited or different roles. For
example# the lower cost of housing 1n Midwestern towns of under 1#000
population caused them to grow as residential communities 1n the
1960s. (Hart# 1969) In most cases# although there had been overall
decline In commercial activity# at least one business was prospering#
suggesting that a "regional" or "dispersed" city might be developing
1n some rural areas. Small towns found a market niche# offering one
or more goods or services# but no longer were able to support the
broader scope of economic activity traditionally associated with the
central place function.

Other recent research shows that small rural communities are
evolving from residential and retail service centers to simply
residential communities dependent on larger trade centers for goods
and services. This 1s particularly true for those communities with
good access to larger cities. (Johansen* 1979) The decline In
commercial functions may be accompanied by an Increase 1n social
functions such as clubs and churches. (McGranahan* 1980) Other
communities have been able to make the transition from farm service
centers to small manufacturing centers. (McGranahan* 1980) Among the
types of communities observed evolving 1n rural areas are bedroom
communities* resort towns and retirement centers* 1n addition to the
more traditional agricultural service center. The central place
function may persist 1n varying degrees 1n any of these communities
provided that they are accessible to rural residents of the region and
are sufficiently distant from major urban areas to avoid the affects
of competition for retail trade. (Dahms* 1986)
Finally* the research suggests that* at least 1n some areas* two
Important economic changes aside from agriculture are Impacting the
ability of central place theory to explain small town change. In some
areas the decentralization of manufacturing has transformed the rural
regional economy from a primary dependence on agriculture to a
dependence on Industry. (McGranahan* 1980) Also* 1t has been
suggested that the Initial retail decline caused by the decrease 1n
the farm population results 1n an eroded tax base 1n a community.
This strained fiscal situation makes 1t Impossible for many small
towns to provide the services necessary to attract further residential

growth or other employment opportunities* exacerbating and
accelerating outmigration from the region. (Walzer* 1979)
The research discussed above does not directly address the
relationship between agriculture and small central place economies.
However* the findings are Instructive; taken together this body of
research suggests that a decline 1n the rural farm population sets
Into motion a series of self-re1nforc1ng events leading to decline of
the retail economy 1n a small town. It also suggests that the hardest
hit small towns 1n terms of overall decline are those located close
enough to larger cities that occasional shopping trips are attractive
but too far away to serve as bedroom communities to the larger
employment centers.

This thesis examines the relationship between the scale of
agricultural production 1n a region and the health of the central
place economies 1n the region. The analysis 1s specifically aimed at
learning how changes 1n the number and size fo farms 1n a region might
Impact local retail trade economies.
The first step 1n developing the methodological approach was to
Identify the factors to be studied. The primary focuses of the
research are retail trade and scale of agricultural production.
Retail sales and number and types of retail establishments were
determined to be factors by which to gauge changes 1n retail trade.
Average farm size was selected as the primary Indicator of scale of
agricultural production. Also> 1n order to gain a more complete
picture of the agricultural economy other factors were analyzed
Including value of agricultural products sold acres 1n cultivation
and percent of land 1n agriculture. These additional agricultural
factors were added to further the understanding of the role of the
agricultural resource base.
Several other factors are critical to understanding the nature of
change 1n retail trade 1n the study area. Data on population and
Income was required to assess the opportunities for retail sales 1n
the local economies. Other elements of analysis considered as
potentially Instructive Included farm service employment and distance
from larger regional trade centers.

The second step 1n conducting this analysis was to Identify an
appropriate study area. The Eastern Plains region of Colorado was
considered most appropriate for this type of study. This area 1s and
has traditionally been dominated by agriculture; 1t 1s largely part of
a portion of the Great Plains known as the High Plains. Also# the
plains of eastern Colorado are very similar to many other major
agricultural regions 1n the United States 1n that many farmers there
are facing or have recently faced severe economic difficulties. It 1s
widely feared that gradual trends toward fewer farmers and larger farm
operations could be greatly accelerated by a rash of bankruptcies and
farm foreclosures. (University of Colorado# 1985) As many as one
third of Colorado farmers could eventually go out of business
according to estimates by some state agricultural officials.
Having determined a geographic region for the study# 1t was then
necessary to set specific boundaries for the study area. Most of the
data analyzed 1n this thesis Is only available on a county-wide basis;
therefore# the study area consists of a number of counties. Although
many counties along the Front Range of Colorado have Important
agricultural economies, these counties were excluded from
consideration because many other sectors are major contributors to the
local economies and would tend to obscure the relationships to be
studied. Finally# from a policy analysis perspective a study area
consisting of number of mutll-county state planning areas was thought
to be most appropriate. Applying all of these criteria led to
definition of a study area consisting of sixteen counties and three
state planning areas 1n eastern Colorado shown on the map on the
following page.


0 __________
50 miles

Several difficulties are encountered when conducting economic
analysis 1n rural areas. Many types of data are available only at the
county levels 1f at all* and thus relationships between Incorporated
and unincorporated areas 1n the counties may be difficult to
document. However# for this analysis It was assumed that In a
sparsely populated agricultural region the amount of retail trade
taking place outside of small towns and cities 1s Insignificant.
Therefore# the county data should provide a valid Indication of the
overall situation 1n the central places within the county.
Data sources were chosen which would offer the most consistent
data over the twenty year study period. Data from the economic
censuses# published by the Bureau of the Census of the Department of
Commerce# although not entirely free from problems with consistency#
offer the best overall source of data on the agricultural and retail
economies of the sixteen counties 1n the study area. These data
sources are available for areas all over the country which assures
that the results can be compared with other areas and that the
methodology developed for this research can be employed by small town
planners and other researchers to analyze their local economies.
The analysis was conducted on three levels. First# each of the
Individual factors was examined separately and trends analyzed over
the 19 year study period. Then statistical analysis was applied to
Identify relationships between the factors with a focus on Identifying
possible relationships between agriculture and changes 1n the
performance of local retail economies. Finally# three counties and
their primary central places were selected for brief case studies to

describe 1n qualitative terms and more quantitative detail the trends
Identified from the regional analysis.
The Census of Retail Trade was the primary source of Information
on the retail economy. This Census was published for the years 1963
1967 1972, 1977 and 1982. Retail trade Is the central focus of this
study and these are, therefore, considered the focal years for the
analysis. The retail census contains Information on retail sales and
number of retail establishments by county, reported 1n total and by
two digit Standard Industrial Classification (SIC) categories. One
difficulty encountered In the analysis was that, because of the small
number of businesses In any one category In many rural counties,
county data on retail sales by SIC code was frequently withheld from
the published data to avoid disclosure of Information about a single
establishment. Therefore, the analysis of retail sales was limited to
total retail sales. Data from state sales tax receipts was used to
supplement the Information from the Census of Retail Trade. Census of
Retail Trade data was available regarding changes In the number of
retail establishments 1n the SIC categories and this also offered some
Insight Into the nature of the change 1n the retail economy.
The other Information required to conduct the retail analysis was
total personal Income by county for each year studied and an estimate
of how much consumers spent on retail expenditures out of this
Income. County per capita Income data prepared by the Bureau of
Economic Analysis was available for all years except 1963; prior to
1965 per capita Income was calculated for only selected years 1963
and 1964 data Is not available. Therefore, 1963 per capita Income was
estimated based on the 1962 and 1965 figures.

In addition to per capita Income* population data was also
required to estimate total personal Income. Population data was
obtained from two sources: Bureau of Economic Analysis figures
published 1n conjuntlon with their Income estimates* and estimates
prepared by the Colorado State Demographer. The BEA population
figures were used to estlmat total personal Income because BEA Is the
source of the per capita Income estimates. Total personal Income by
county for each year was calculated by multiplying population by per
capita Income.
With total personal Income data and retail sales data for each
county In the study area and for the entire State of Colorado* 1t was
then possible to derive an estimate of the percent of total resident
Income which was spent on retail consumer expenditures for each county
during each year studied. First# total retail sales 1n the entire
State of Colorado In a given year were expressed as a percent of total
personal Income 1n the state during that same year. The ratio of
retail sales to total personal Income represents the average portion
of personal Income expended on retail purchases by each state
resident* assuming that residents' dollars spent outside the state
roughly equal expenditures 1n Colorado by out of state residents.
The statewide expenditure pattern was used to estimate the
expenditure patterns 1n the study area; this ratio or percent was
applied to the total personal income 1n the counties being studied to
estimate the amount spent by residents on retail goods 1n a given year
as demonstrated 1n Figure 2. This amount 1s referred to herein as the
resident retail expenditure potential. The resident retail
expenditure potential was compared with actual retail sales In the


county to determine whether there was a net Inflow of consumer dollars
Into the county or whether there was retail outflow or leakage.
Retail sales, expressed as a percent of resident retail expenditure
potential, yields a measure called percent capture of retail
expenditure potential also shown 1n Figure 2. Changes 1n this percent
capture were tracked for the five Census of Retail Trade years during
the study period to Identify changes 1n the performance of the central
place economies 1n the region.
Data on the agricultural economies 1n the study area counties was
also obtained from the economic census published by the Census Bureau
of the Department of Commerce. Until 1982 the Census of Agriculture
was published on a different schedule than the Census of Retail Trade,
for the years 1964, 1969, 1974, and 1978, compared to 1963, 1967,
1972, and 1977 for the retail census. Therefore 1t was not possible
to compare data for an Identical series of years. However, since the
objective was to look at broad trends over twenty years this was not
considered to be an obstacle to obtaining valid results. The Census
of Agriculture contains Information on farm size, number of farms,
type of agricultural production, and value of farm products, among
other data, which was used to document trends 1n the agricultural
economy of the study area counties.
Data on the farm sevlce economy was obtained from County Business
Patterns published by the Census Bureau; the economic censuses did not
contain sufflclenly detailed Information for this analysis.
Specifically, the number of establishments and employment in
agricultural service and farm wholesale sectors of the economy were
followed on a county by county basis to determine whether there was

any dlscernable change 1n the structure of this aspect of the regional
agricultural economy.
In addition to the analysis of the data on the retail and
agricultural economies# several other factors were considered as
potentially Important determinants of change 1n the central place
economies within the study area. First# Improvements 1n automobile
transportation networks were examined. As was discussed previously#
other researchers have suggested that Improvements In transportation
links to larger cities have affected the ability of small town
retailers to compete with larger regional retail trade centers. It
was also expected that a location on a major transportation route
could offer advantages to some towns because of Improved accessibility
to rural residents of the region and the potential for making sales to
travellers through the region. Changes 1n retailing# specifically
the development of regional shopping centers# were examined 1n the
larger cities along the Colorado Front Range. As explained ealler
these regional shopping centers would be expected to expand the retail
market areas for these cities at the expense of the central places In
the study area.
After examining each of the above factors Individually an attempt
was made to synthesize all of the major elements of the analysis using
the statistical method of correlation analysis. The relative
significance of the relationships between the variables and the
success of central place economies could thus be determined. It 1s
this level of analysis which would prove or disprove the hypothesis
that there Is a relationship between the scale of agrlcultrual

production 1n an area and the performance ot the central place
economies of the region's communities.
Examining secondary data sources on a regional basis provided at
best a rough sketch of what has actually occurred 1n the small town
central places of the study area. Therefore, three counties and their
primary central places which were thought to represent the trends
Identified for the study area were selected for case studies.
Additional research was conducted on the historical background of the
communities, changes 1n retail and farm service establishments, trends
1n agriculture, and current economic development efforts, 1n order to
gain a more complete understanding of the dynamics of change 1n these

The study area consists of sixteen primarily rural counties 1n
Eastern Colorado. The counties cover a land area of over 27*000
square miles* of Which 87 percent 1s 1n agricultural uses. The major
agricultural products Include winter wheat* corn* sorghum* soybeans*
sugar beets* hay and cattle. Winter wheat 1s the primary crop 1n the
region with ten of the sixteen counties producing at least six million
bushels per year In 1983. The vast majority of this crop Is produced
by dryland farm1ng(1.e. without Irrigation). By contrast* corn 1s
grown almost exclusively on Irrigated cropland with production
concentrated 1n the north and east central portions of the study
area. Six counties In this area each harvest three million or more
bushels of corn grain per year. Sorghum Is grown primarily 1n the
southeastern and far eastern counties of the region with production
divided almost evenly between Irrigated and non-1rr1gated cropland.
Sugar beet production 1s concentrated In the northeast but recently
has been 1n decl1ne. Hay 1s grown primarily 1n the river valleys of
the South Platte and Arkansas rivers and cattle are raised 1n
significant numbers throughout the study area with a concentration 1n
the northeast.
The study area contains sixty Incorporated towns and cities. The
largest city 1n the region 1s Sterling 1n Logan County which had a
1980 census population of 11*385 and an estimated current population
of about 11*700. The smallest Incorporated town 1n the study area 1s
Paoll 1n Phillips County with just 81 residents according to the
decennial census. Thirty-four of the communities lost population

between 1960 and 1980* twenty-five gained population and one
experienced no change.
Early agriculture In eastern Colorado was concentrated along the
Front Range piedmont area which lies to the west of the study area.
Much of this agriculture developed to serve the mining communities 1n
the mountains to the west. In the late 1870s and 1880s farming and
ranching spread downstream In the valleys of the South Platte and
Arkansas rivers.
The plains of eastern Colorado were settled primarily by
homesteaders. In fact 1t could be argued that the provisions of the
Homestead Act of 1862 destined this area to see a steady Increase 1n
farm size In this century. The first homesteaders were limited to
claims of 160 acres* a limit probably established based more on the
deep black soil and plentiful rainfall of the midwest* which was being
settled at the time of the passage of the Homestead Act* than on
conditions 1n the high plains. Although the fertility locked In the
prairie soil could be tapped Initially by fanners (providing they
began farming during an unusually wet spell* as did many of the early
homesteaders)* given the sporadic and generally sparse rainfall on the
high plains the deck was probably stacked against success for most of
these early tillers of the soil. On the other hand the Homestead Act
did not allow ranchers to file claims for sufficient land to run a
successful cattle operation; the 160 acres allowed provided barely
enough forage to support six steers. Cattle operations were therefore
largely dependent on the poorer unclaimed public lands for grazing.

Corporate farming was also Important In the early days of
Colorado agriculture. During the 1880s the eastern plains were
promoted as a grazing area and between 1880 and 1885 more than 150
corporations were organized to finance stock raising operations. Much
of this Investment came from the East and from Europeans. The Prairie
Cattle Company# one of the largest operations# ran 60000 head of
cattle on two million acres of open range.
In the 1880s windmills came to the plains bringing with them the
first Irrigation from shallow# alluvial groundwater and allowing more
productive agriculture In some areas. Also In the late 1800s#
Irrigation companies were established to build canals for Irrigation
1n the valleys of the South Platte and Arkansas rivers. In the 1890s
cattle operations were transformed as more and more acreage was fenced
1n; cattle were Increasingly fattened on alfalfa grown 1n the river
valley bottom lands. Many of the Issues still prominent 1n
agriculture 1n eastern Colorado today had their beginnings early 1n
the area's history: sodbustlng vs. grazing# the appropriate scale of
agricultural operations# and finding ways of providing Irrigation
water to Improve the productivity of the land 1n this sem1-ar1d
region. These three Issues combined with nationwide trends 1n
agriculture are constant and recurring themes 1n the history of
eastern Colorado agriculture.
In 1909 the Preemption Act allowed farmers to triple the size of
their homestead claims providing that ten acres were planted 1n
trees. Over time the size of agricultural operations Increased with
much of the land coming Into the ownership of ranchers and large
absentee landholders.

Much as 1s the case today the major crops raised 1n the early
part of this century were wheat and corn. Other major crops Included
pinto beans* kaffir corn, sorghum and millet. With the coming of
World War I the demand for wheat skyrocketed justifying Investment 1n
the first "modern" farm machinery, the steam tractor. The price of
wheat climbed from $.80 to $2.00 a bushel and this coincided with a
period of higher than average rainfall 1n the region. As a result
wheat acreage Increased by over 200 percent 1n Colorado. When the
Great Depression hit the price of a bushel of wheat dropped from $3.50
a bushel to forty cents and corn fell from $3.00 to a quarter. In a
scene which has unmlstakeable parallels to the present, grain
elevators overflowed with grain for which there was no market.
Then came the drought and resultant dust bowl of the early 1930s
precipitating the first massive move off of the farms of eastern
Colorado. Population and the number of farms both declined 15
percent. During this period the first application of land banking as
a tool of agricultural policy occurred with the United States
Department of Agricultures (USDA) Resettlement Administration
purchasing thousands of acreas of uneconomic farmland and pulling 1t
out of production. Some of this land Is now the Comanche National
Grassland, part of which lies within the study area.
The pattern of boom and bust 1n agriculture repeated Itself 1n
the 1940s and 1950s with high prices and a period of above average
rainfall 1n the 1940s followed by drought 1n 1953 and 1954. The next
fundamental change 1n farming on the eastern plains came about 1n the
mid-1950s when center pivot Irrigation was Invented 1n eastern
Colorado. This method, which any air traveller flying east from

Denver has undoubtedly noticed as the dries of green In a sea of
brown* opened up a new chapter 1n Increased productivity. Irrigation
allowed Increased yields on existing cultivated land and also made 1t
feasible to grow more of crops such as corn which could not survive on
only the sparse rainfall In the region. The advent of an economical
method of deep well Irrigation and high demand for exports In the
1970s also encouraged plowing of virgin grassland for crop
production. At the beginning of the study period In the early 1960s*
much of the study area was actually 1n the midst of an agricultural
transition brought on by central pivot Irrigation.
In addition to Its agricultural resource base* much of the study
area also lies above rich energy deposits* primarily oil and natural
gas. The Denver-Julesburg Basin boomed 1n the late 1950s and again 1n
the 1970s with dollars pumped Into the region by exploration, drilling
and from the royalties from oil and gas production.

In this section the discussion 1s focused on describing the
trends which have occurred 1n retailing and agriculture 1n the study
area over the nineteen year period from 1963 to 1982. Each of the
elements of the analysis 1s dealt with separately and then the overall
trends and relationships are summarized.
This discussion presents study area trend data In three ways.
First* data Is aggregated for the entire study area and tracked over
the five year Intervals of the economic censuses. Where appropriate
the study area trends are compared to trends 1n the State of Colorado
as a whole. Second* data 1s also aggregated for groups of counties
with similarly sized central places. The determinant for this
analysis Is the population of the largest city 1n the county and data
1s presented for groups of counties with cities of less than 500* 500
to 1000, 1000 to 2500, 2500 to 5000* 5000 to 10,000, and over 10,000
population. These classes have been Identified 1n prior studies as
delineating functional classes of communities 1n the central place
hierarchy. (Walzer, 1977) Although the largest and smallest classes
each Include only one study area county, and the numbers varied 1n the
other classes from two to seven, these classes were used to maintain
the comparability with other research results and to allow functional
change to be Identified. The populations used for this classification
were from the 1970 census because this data most closely approximated
the populations at the midpoint of the study period. Under central
place theory the size -of the largest city 1n the county should

determine the range of goods available 1n the county and* thus* the
potential for capturing residents'retail expenditures locally.
The third way the data 1s compared 1s by the three state planning
regions which make up the study area: Planning Region 1 (the
Northeast Region)* Planning Region 5 (the East Central Region)# and
Planning Region 6 (the Southeast Region). Several characteristics
which distinguish these regions are highlighted by this analysis.

Population and Income
The overall study area popuatlon Increased by two percent during
the nineteen year study period* compared with population growth of 59
percent 1n the State of Colorado as a whole. Ten counties lost
population* ranging from eighteen percent declines 1n Bent* Crowley
and Sedgwick counties to a less than one percent loss of population In
Logan County. Six counties gained population; the largest percentage
and absolute gain was registered by Elbert County with a 99 percent
Increase. Elbert County borders on the rapidly growing Denver
metropolitan area and this population growth Is largely 1n rural-
suburban bedroom type communities.
When population change 1s examined from the perspective of size
of largest city In the county only the groups of counties with city
populations under 500 and greater than 5000 showed aggregate positive
population growth. Since the only county with largest city population
under 500 was Elbert County* rural population growth was largely
limited to counties which already had a substantial city population*
Indicating an employment base with more significant non-farm
When population data 1s aggregated by planning region* a somewhat
different picture of the study area emerges with two of the three
planning regions showing positive population growth. The population
data suggests that very different trends have been at work 1n the
three planning regions: the East Central Region showed definite,
though relatively low* 19 percent positive population growth; the
Northeast Region was stable with a four percent population Increase;

and the Southeast Region registered an overall population loss of
seven percent. The population figures by county# size of largest city
and planning region are shown 1n Table 1.
Per capita Income figures were adjusted to constant 1982 dollars
1n order to eliminate the substantial Impacts of Inflation during the
study period. The Implicit price deflator developed by the Department
of Commerce for adjusting Income and consumer expenditure figures to a
constant dollar basis was used for this purpose. In 1982 dollars#
average per capita personal Income Increased by 91 percent 1n the
study area compared with a 78 percent Increase for the State of
Colorado. Per capita Income 1n the study area Increased from an
average level of $5324 1n 1963 to S10186 1n 1982. Despite this
Increase, which exceeded the rate for Colorado as a whole# income In
the study area 1n 1963 lagged so far behind the state that the 1982
average Income was still more than $2000 below the Colorado average.
No county 1n the study area equalled or bettered the state average.
In 1963 per capita Income In the study area ranged from a low of
$4#317 1n Bent County to a high of $6#565 In K1t Carson County. The
growth 1n personal Income from 1963 to 1982 varied widely by county
from a low of 39 percent real per capita Income growth 1n Cheyenne
County to a 157 percent Increase 1n Crowley County. As a result
Crowley County moved from next to last to second highest 1n study area
county per capita Income. In 1982 Elbert County had the highest per
capita Income of $11#783 and Cheyenne County was lowest with an
average Income of $8#126. The counties most proximate to the
metropolitan population centers of the Colorado Front Range# Elbert#
Crowley# and Lincoln# had the highest per capita Incomes 1n the study

1963 1967 1972 1977 1982 1963-1982
BACA 6101 5900 5700 5600 53 00 -13%
BENT 7102 6700 6400 6200 5800 -18%
CHEYENNE 2601 2400 2300 2100 23 00 -12%
CROWLEY 3668 3200 3200 3200 3000 -18%
ELBERT 3767 3900 4100 5400 7500 99%
KIOWA 2301 2100 2000 1900 1900 -17%
KIT CARSON 7066 7300 7500 7700 7700 9%
LINCOLN 5034 4600 5000 4800 4500 -11%
LOGAN 20039 19400 19300 19600 20000 0%
MORGAN 20170 19100 21600 21700 22900 14%
OTERO 23 3 6 8 22800 23700 23500 22500 -4%
PHILLIPS 43 3 4 4200 4000 4400 4600 6%
PROWERS 12967 12400 13500 13600 13300 3%
SEDGWICK 4035 3600 3400 3300 3300 -18%
WASHINGTON 6168 5600 5500 5000 5400 -12%
YUMA 8701 8600 8400 9300 9900 14%
TOTAL STUDY AREA 137422 131800 135600 137300 139900 2%
0 to 500 3767 3900 4100 5400 7500 99%
501 to 1000 4902 4500 4300 4000 4200 -14%
1001 to 2500 38041 35700 35200 35600 36000 -5%
2501 to 5000 14168 14000 13900 13900 13500 -5%
5001 to 10000 56505 54300 58800 5 8800 58700 4%
10001 + 20039 19400 19300 19600 20000 0%
NORTHEAST 63447 60500 62200 63300 66100 4%
EAST CENTRAL 18468 18200 18900 20000 22000 19%
SOUTHEAST 55507 53100 54500 54000 51800 -7%
Note: 1963 figures are author's estimate based on
data reported for 1962 and 1965.
Source: Bureau of Economic Analysis

area* perhaps Indicating a large percentage of residents who commute
to higher paying urban jobs.
No clear pattern emerges when average county per capita Income
1s considered by size of largest city 1n the county. Percentage
Increases varied from a high of 115 percent for Elbert County to a low
of 67 percent for counties with largest city 1n the 501 to 1*000
population range. Logan County* the only county with a 1970 city
population of over 10*000 population* had a higher than average Income
Increase of 104 percent* and was slightly above the study area average
1n 1982.
Considering the average county per capita Income by planning
region we see a convergence of Income levels caused by dlfferrlng
rates of Increase. Northeast region per capita Income Increased by
almost 99 percent from an average of $5*363 1n 1963 to $10*653 1n
1982* the highest for any planning region 1n that year. The East
Central region had the highest 1963 average Income of $6*028 but
Increased only 76 percent falling to second 1n 1982 with average per
capita Income of $10*618. The Southeast Planning Region lagged behind
the other areas throughout the study period with an average 1963 per
capita Income of $5*042* Increasing 87 percent to a 1982 level of
$9*407. Per capita Income figures are shown 1n the table on the
following page.

1963 1967 1972 1977 1982 1963-1982
BACA $5,865 $8,063 $8,193 $7,361 $9,073 54.7055
BENT $4,317 $7,614 $6,627 $6,772 $8,824 104.39%
CHEYENNE $5,840 $5,047 $8,168 $11,332 $8,126 39.14%
CROWLEY $4,571 $5,534 $8,934 $7,590 $11,732 156.69%
ELBERT $5,480 $5,688 $7,564 $7,567 $11,783 115.00%
KIOWA $5,430 $8,063 $10,203 $11,747 $11,060 103.69%
KIT CARSON $6,565 $7,614 $10,368 $8,327 $9,632 46.72%
LINCOLN $5,782 $6,345 $6,300 $8,052 $11,635 101.22%
LOGAN $5,096 $6,137 $9,744 $9,173 $10,411 104.30%
WRGAN $5,403 $6,507 $8,825 $9,012 $10,991 103.43%
OTERO $5,077 $5,861 $7,391 $8,085 $9,052 78.31%
PHILLIPS $5,759 $6,550 $10,881 $11,839 $9,989 73.44%
PROWERS $5,053 $6,122 $6,934 $7,320 $9,636 90.70%
SEDGWICK $6,529 $6,862 $9,622 $10,312 $10,407 59.40%
WASHINGTON $5,778 $6,479 $8,221 $8,205 $10,398 79.96%
YUMA $4,880 $6,028 $8,704 $9,548 $10,889 123.15%
STUDY AREA $5,324 $6,415 $8,475 $8,580 $10,186 91.31%
0 to 500 $5,480 $5,688 $7,564 $7,567 $11,783 115.00%
501 to 1000 $5,648 $6,454 $9,115 $11,529 $9,453 67.39%
1001 to 2500 $5,548 $6,577 $8,845 $8,992 $10,552 90.20%
2501 to 5000 $5,438 $7,614 $8,646 $7,634 $9,285 70.74%
5001 to 10000 $5,188 $6,148 $7,813 $8,250 $9,941 91.62%
L0001 + $5,096 $6,137 $9,744 $9,173 $10,411 104.30%
NORTHEAST $5,367 $6,342 $9,216 $9,341 $10,653 98.50%
EAST CENTRAL $6,028 $6,542 $8,945 $8,371 $10,618 76.13%
SOUTHEAST $5,042 $6,455 $7,466 $7,766 $9,407 86.59%
Source: Bureau of Economic Analysis

Retail Trade
The retail economies of the counties 1n the study area are the
primary focus of this analysis. It Is through this focus that changes
1n the central place function of the small town agricultural centers
are examined.
In current dollar terms (1.e. dollars not adjusted for the
effects of Inflation) total retail sales 1n the study area Increased
by almost 200 percent between 1963 and 1982. However* during this
same period retail sales 1n the State of Colorado were up 526
percent. In order to gain a more realistic and meaningful view of
what has happened In the retail economy all the data was transformed
to a 1982 constant year basis; that 1s# the figures for past years
were adjusted 1n order to account for the effects of Inflation. The
Implicit price deflator developed by the United States Department of
Commerce was used to calculate the adjustment factor. When retail
sales figures are adjusted for Inflation we find that sales 1n the
study area Increased by six percent from 1963 to 1982# a period during
which statewide sales Increased by over 127 percent. The largest
Increase 1n sales was 1n Lincoln County with a 71 percent gain and the
worst declines were 1n Kiowa and Crowley counties where retail sales
dropped by 61 and 47 percent* respectively. Among planning regions
the East Central Planning Region posted the largest real gain 1n
retail sales which were up 22 percent. Sales 1n the Northeast
Planning Region were up 16 percent while the Southeast Planning
Region suffered the only decline with sales down 16 percent.

Aggregating the data by size of largest city provides another
perspective. Counties with largest cities of between 500 and 1*000
population showed the greatest decline with sales down 45 percent.
Generally* total retail sales declined 1n counties with largest cities
of less than 5*000 population and Increased 1n counties with largest
cities above 5*000 population. The Increase In the 1*001 to 2*500
class 1s entirely due to Lincoln County; without Lincoln County there
1s a $4 million dollar decline. Sales 1n Logan County* the only
county In the region with a city of more than 10*000 population* were
up 22 percent* the largest Increase for any class of county. Data on
retail sales 1s presented 1n Table 3.
Even looking at adjusted retail sales figures tells us very
little about possible explanations for the decline 1n retail sales 1n
the study region. Declines 1n some areas are at least partly
attributable to losses of population. This Influence can be
eliminated from the figures by looking at per capita retail sales*
which show a somewhat different pattern of change. The five percent
study area Increase 1n sales per capita 1s slightly less than the
Increase In total sales* but compares somewhat more favorably with the
Increase for the state of 43 percent (as opposed to 127 percent for
total retail sales). Kiowa County still shows the greatest percent
decline but Elbert County Is second with a 45 percent drop 1n sales
per capita compared to a 16 percent Increase In total retail sales.
The largest Increase 1s still In Lincoln County.
The data by size of largest city shows a much more definite
pattern of change when sales are examined on a per capita basis. All
of the aggregate decline In sales per capita has occurred 1n counties

1963 1967 1972 1977 1982 1963-1982
IACA $25,230 $24,555 $22,332 $20,771 $22,054 -12.59%
IENT $18,490 $17,686 $19,074 $18,714 $14,859 -19.64%
HEYENNE $8,753 $6,327 $12,208 $8,648 $6,656 -23.96%
ROWLEY $6,332 $5,334 $7,496 $5,264 $3,355 -47.01%
LBERT $5,524 $4,472 $4,740 $7,983 $5,610 1.56%
IOWA $11,643 $5,369 $4,303 $5,149 $4,495 -61.39%
IT CARSON $35,798 $39,092 $36,130 $39,471 $37,699 5.31%
INCOLN $22,936 $23,205 $23,774 $28,695 $39,248 71.12%
OGAN $64,534 $77,611 $98,581 $104,651 $102,967 21.80%
ORGAN $92,394 $97,084 $103,818 $111,927 $115,674 25.20%
TERO $89,559 $90,544 $84,256 $93,217 $76,806 -14.24%
HILLIPS $21,841 $17,433 $14,309 $18,952 $18,367 -15.91%
ROWERS $65,231 $63,001 $73,326 $76,697 $70,926 8.73%
EDGWICK $19,257 $18,469 $23,191 $22,633 $17,148 -10.95%
ASHINGTON $16,739 $12,322 $12,690 $15,525 $14,666 -12.39%
UMA $34,129 $38,838 $32,618 $38,391 $43,720 28.10%
TUDY AREA $558,390 $541,342 $572,847 $616,689 $594,250 ' 6.42%
to 500 $5,524 $4,472 $4,740 $7,983 $5,610 1.56%
01 to 1000 $20,396 $11,696 $16,511 $13,797 $11,151 -45.33%
001 to 2500 $146,464 $140,156 $136,411 $150,232 $158,558 8.26%
501 to 5000 $54,288 $56,778 $55,204 $58,185 $52,558 -3.19%
001 to 10000 $247,183 $250,628 $261,400 $281,840 $263,406 6.56%
0001 + $84,534 $77,611 $98,581 $104,651 $102,967 21.80%
ORTHEAST $268,895 $261,757 $285,207 $312,079 $312,542 16.23%
AST CENTRAL $73,010 $73,096 $76,852 $84,797 $89,213 22.19%
DUTHEAST $216,485 $206,489 $210,787 $219,813 $192,495 -11.08%
Sources Census of Retail Trade

with a largest city population below 1#000. The planning regions also
exhibit a different pattern of change when we account for changes 1n
the population level. The Northeastern region registers the largest
Increase 1n per capita sales followed by the East Central region while
the Southeast still shows a decline but of a lesser degree. The order
of rank does not change among the three regions from 1963 to 1982 but
the disparity 1n per capita retail sales between the Northeast and
Southeast regions grows from $338 1n 1963 to $1000 1n 1982. Per
capita retail sales data 1s shown 1n Table 4.
The overall Increase 1n retail sales for the entire study area Is
less than might be expected given the small Increase 1n population and
91 percent Increase 1n per capita Income. This suggests that
residents of the study region were spending more dollars outside the
region 1n 1982 than 1n 1963. A tool of retail market analysis was
employed 1n order to begin to construct a better picture of what has
caused this change 1n the retail trade sector.
The first step of this analytical method Is to estimate the
amount of money spent by local residents 1n retail establishments!
regardless of whether those dollars were spent In the local community
In which the Individuals reside. Total retail sales 1n the State of
Colorado for each of the focal years of the analysis were compared
with total personal Income of state residents 1n that year. In this
manner the percent of personal Income spent on retail goods was
estimated. This percent was calculated for specific categories of
retail goods such as food and apparel as well as for retail trade as a
whole. Using sales data from the Census of Retail Trade and Income
data from the Bureau of Economic Analysis a retail expenditure profile

with a largest city population below 1,000. The planning regions also
exhibit a different pattern of change when we account for changes 1n
the population level. The Northeastern region registers the largest
Increase 1n per capita sales followed by the East Central region while
the Southeast still shows a decline but of a lesser degree. The order
of rank does not change among the three regions from 1963 to 1982 but
the disparity 1n per capita retail sales between the Northeast and
Southeast regions grows from $338 1n 1963 to $1000 1n 1982. Per
capita retail sales data 1s shown 1n Table 4.
The overall Increase 1n retail sales for the entire study area Is
less than might be expected given the small Increase In population and
91 percent Increase 1n per capita Income. This suggests that
residents of the study region were spending more dollars outside the
region 1n 1982 than In 1963. A tool of retail market analysis was
employed 1n order to begin to construct a better picture of what has
caused this change 1n the retail trade sector.
The first step of this analytical method Is to estimate the
amount of money spent by local residents In retail establIshments,
regardless of whether those dollars were spent 1n the local community
1n which the Individuals reside. Total retail sales 1n the State of
Colorado for each of the focal years of the analysis were compared
with total personal Income of state residents 1n that year. In this
manner the percent of personal Income spent on retail goods was
estimated. This percent was calculated for specific categories of
retail goods such as food and apparel as well as for retail trade as a
whole. Using sales data from the Census of Retail Trade and Income
data from the Bureau of Economic Analysis a retail expenditure profile

(1982 DOLLARS)
1963 1967 1972 1977 1982 1963-1982

BACA $4,135 $4,162 $3,918 $3,709 $4,161 0.62%
BENT $2,604 $2,640 $2,980 $3,018 $2,562 -1.60%
CHEYENNE $3,365 $2,636 $5,308 $4,118 $2,894 -14.01%
CROWLEY $1,726 $1,667 $2,343 $1,645 $1,118 -35.21%
ELBERT $1,466 $1,147 $1,156 $1,478 $748 -48.99%
KIOWA $5,060 $2,557 $2,152 $2,710 $2,366 -53.25%
KIT CARSON $5,066 $5,355 $4,817 $5,126 $4,896 -3.36%
LINCOLN $4,556 $5,045 $4,755 $5,978 $8,722 91.43%
LOGAN $4,218 $4,001 $5,108 $5,339 $5,148 22.04%
MORGAN $4,581 $5,083 $4,806 $5,158 $5,051 10.27%
OTERO $3,833 $3,971 $3,555 $3,967 $3,414 -10.93%
PHILLIPS $5,039 $4,151 $3,577 $4,307 $3,993 -20.77%
PROWERS $5,031 $5,081 $5,432 $5,639 $5,333 6.01%
SEDGWICK $4,772 $5,130 $6,821 $6,859 $5,196 8.88%
WASHINGTON $2,714 $2,200 $2,307 $3,105 $2,716 0.08%
YUMA $3,922 $4,516 $3,883 $4,128 $4,416 12.59%
STUDY AREA $4,063 BY SIZE $4,107 $4,225 OF LARGEST CITY IN $4,492 THE COUNTY $4,248 4.54%
0 to 500 $1,466 $1,147 $1,156 $1,478 $748 -48.99%
501 to 1000 $4,161 $2,599 $3,840 $3,449 $2,655 -36.19%
1001 to 2500 $3,850 $3,926 $3,875 $4,220 $4,404 14.39%
2501 to 5000 $3,832 $4,056 $3,972 $4,186 $3,893 1.60%
5001 to 10000 $4,375 $4,616 $4,446 $4,793 $4,487 2.58%
10001 + $4,218 BY $4,001 $5,108 $5,339 STATE PLANNING REGION $5,148 22.04%
NORTHEAST $4,238 $4,327 $4,585 $4,930 $4,728 11.57%
EAST CENTRAL $3,953 $4,016 $4,066 $4,240 $4,055 2.58%
SOUTHEAST $3,900 $3,889 $3,868 $4,071 $3,716 -4.72%

for Colorado residents was derived for each year of the retail census
from 1963 to 1982# as shown In Table 5.
These percentages of personal Income spent on various retail
categories and on total retail trade were applied to total personal
Income 1n each of the study area counties to derive an estimate of
expenditure potential for that county 1n the given year. Expenditure
potentials were then compared with retail sales In the respective
counties 1n each of the census years 1n the study period. By dividing
retail sales by expenditure potential(1.e.# an estimate of the amount
actually spent by county residents 1n that year) a number was derived
which# when expressed as a percent# represents the net percent of
retail expenditures by local residents which were purchases made 1n
the county of residence. This figure will be referred to throughout
this analysis as the percent capture of resident retail expenditure
potential or# simply# percent capture. It 1s Important to stress that
this Is a net percentage because 1n reality every county registers
some retail sales made to non-residents and some residents 1n every
county make retail purchases outside of the county.
When these percent capture figures are compared across the years
from 1963 to 1982 a clearer picture of what has been happening to the
retail economies of the study area counties emerges, as shown 1n Table
6. In 1963 there was evidence of a weak central place hierarchy. The
counties with larger populations 1n their largest city generally had
higher percent captures than did counties with smaller cities#
although a significant functional change appears to have occurred at
the 500 population mark only. In virtually every county In the study
area the percent capture of resident expenditures declined

1963 1967 1972 1977 1982
TOTAL PERSONAL INCOME $4,804,784,876 $6,269,862,000 $10,998,640,400 $19,133,512,000 $37,569,061,000
TOTAL RETAIL SALES $2,648,618 55.12* $3,280,672 52.32* $5,869,039 53.36* $9,824,594 51.35* $16,581,084 44.1
BUILDING MATERIALS $194,064 4.04* $209,631 3.34* $450,036 4.09* $675,161 3.53* $949,233 2.X
GENERAL MERCHANDISE $320,025 6.66* $456,758 7.28* $815,490 7.41* $1,204,007 6.29* $1,658,107 4.1
FOOD STORES $569,663 11.86* $693,804 11.07* $1,150,698 10.46* $2,008,048 10.49* $3,681,410 9.1
AUTO DEALERS $513,910 10.70* $633,823 10.11* $1,222,436 11.11* $2,080,810 10.88* $3,106,272 8.1
GAS STATIONS $219,749 4.57* $256,939 4.10* $416,813 3.79* $708,010 3.70* $1,429,634 3.X
APPAREL & ACCESSORIES $121,656 2.53* $143,299 2.29* $263,858 2.40* $437,814 2.29* $771,721 2.X
FURNITURE & HOME FURN. $122,849 2.56* $150,438 2.40* $297,987 2.71* $479,686 2.51* $770,605 2.X
EATING & DRINKING $203,147 4.23* $260,468 4.15* $492,243 4.48* $966,283 5.05* $1,822,770 4.1
DRUG & PROPRIETARY $108,567 2.26* $130,416 2.08* $179,627 1.63* $248,035 1.30* $353,508 O.X
MISCELLANEOUS RETAIL $208,981 4.35* $280,519 4.47* $580,550 5.28* $1,016,740 5.31* $1,665,995 4.1
NONSTORE RETAIL $65,987 1.37* $64,547 1.03*

1963 1967 1972 1977 1982 1963 -1982
BACA 128% 99% 90% 98% 104% -24%
BENT 109% 66% 84% 87% 66% -44%
CHEYENNE 105% 100% 122% 71% 81% -24%
CROWLEY 69% 58% 49% 42% 22% -47%
ELBERT 49% 39% 29% 38% 14% -34%
KIOWA 169% 61% 40% 45% 48% -121%
KIT CARSON 140% 134% 87% 120% 115% -25%
LINCOLN 143% 152% 107% 145% 170% 27%
LOGAN 150% 125% 98% 113% 112% -38%
MORGAN 154% 149% 102% 111% 104% -50%
OTERO 137% 129% 90% 96% 85% -51%
PHILLIPS 159% 121% 62% 71% 91% -68%
PROWERS 181% 159% 147% 150% 125% -55%
SEDGWICK 133% 143% 133% 130% 113% -19%
WASHINGTON 85% 65% 53% 74% 59% -26%
YUMA 146% 143% 84% 84% 92% -54%
STUDY AREA 138% 122% 93% 102% 95% -44%
0 to 500 49% 39% 29% 38% 14% -34%
501 to 1000 134% 77% 79% 58% 64% -70%
1001 to 2500 126% 114% 82% 91% 95% -31%
2501 to 5000 128% 102% 86% 107% 95% -33%
5001 to 10000 153% 143% 107% 113% 102% -51%
10001 + 150% 125% 98% 113% 112% -38%
NORTHEAST 143% 130% 93% 103% 101% -43%
EAST CENTRAL 119% 117% 85% 99% 87% -32%
SOUTHEAST 140% 115% 97% 102% 90% -51%

substantially from 1963 to 1982 and the hierarchy appears to have
developed Into three distinct tiers rather than the two evident 1n
1963 as shown 1n Figure 3. Only the counties with largest city of more
than 5000 maintained aggregate percent captures of over 100 percent
and counties with largest cities of 500 to 1000 population performed
much more poorly than 1n 1963 capturing only about 64 percent of the
retail potential based on the population and Income of county
residents. Thus, events 1n the last twenty years have changed the
hierarchy of central places 1n the study area and, 1n general, 1t
would appear that many of the smaller central places have lost any
role as regional trading centers. This trend 1s reflected 1n data for
the study area as a whole where the overall percent capture declined
from 138 percent of potential 1n 1963, Indicating net Inflows of
retail dollars Into the region, to 95 percent 1n 1982, Indicating a
net outflow of retail expenditures from the study area.
The percent capture figures used here may 1n fact be a little
higher than the percent of actual resident expenditures captured 1n
the counties 1n the study area. Because of the lower average Income
1n the study area counties residents there may spend a higher percent
of personal Income on retail expenditures. This would have the effect
of Increasing the expenditure potential and lowering the percent
capture. However, because figures are compared for counties which
have similar Income characteristics and for the same set of counties
over time, this does not significantly affect the analysis. The
higher than 100 percent overall capture 1n 1963 probably represents
some Inflows from the adjoining states of Nebraska, Wyoming, Kansas
and Oklahoma. While this Inflow may still exist, 1t appears that 1t

r r_ r; x, n_ i \j i x,m. n ukl ur KLl a IL h" u I L |\| I IA L.
1 <5 0 % -
150% -
1 40% -
1 30% -
120% -
110% -
1 00% -
90% -
80% -
70% -
60% -
50% -
40% -
30% -
20% -
10% -
0% -



\ -
A A.

A -
sy Yi
V/ Af

\ X/-



A v
' A Ax
' Ay
A A\
'A An
' A, \/
/ x
A A.
' A\x
A A,
A A.

A A.,
A A.'
A A.
Y '
\ '-yxl
'A An
Ar N
A Ny
\ v
c. f
9 67
^2 1972
ES3 19
1XXI 1978

1s now more than offset by outflows from the study area# primarily to
the Colorado Front Range counties.
In 1963 thirteen of the sixteen study area counties registered
net Inflows of retail sales while 1n 1982 only seven counties had
Inflows. However* the decline 1n percent capture 1s not a straight
line trend. The study area percent capture declined from Its 1963
high to just 93 percent 1n 1972* rose to 102 percent 1n 1977 and then
dropped back to 95 percent 1n 1982. The mid-1970s were a period of
major agricultural export activity and Increased exploration for
energy resources. Both of these factors may have contributed to the
general upswing 1n many of the numbers for 1977.
The location of the major retail trade centers 1n the study area
also changed from 1963 to 1982. In 1963 Prowers* Kiowa* Phillips*
Morgan and Logan counties stood out as centers of retail trade* all
with percent captures of over 150 percent. By 1982* the leading
counties were Lincoln with 170 percent and Prowers with 125 percent.
All of the other counties were at 115 percent or less and two of the
former leaders* Kiowa and Phillips* had fallen below 100 percent. As
will be shown later 1n this discussion* location on a major
transportation network seems to be the factor which best explains this
When percent captures are calculated by planning region* the
Northeast and Southeast Planning Regions showed almost Identical
average captures of Just over 140 percent In 1963. The East Central
Planning Region trailed with a capture of 119 percent. By 1982 only
the Northeast Planning Region had an overall average capture of over
100 percent but even here the capture rate was just 100.62 percent.

The East Central and Southeast Planning Regions had fallen to 87 and
90 percent* respectively. The Northeast region has maintained a net
balance 1n retail trade while the East Central and Southeast regions
are now showing net outflows.
It Is worthy of note that only Lincoln County registered an
increase In the net capture of retail expenditure potential. It 1s
perhaps significant that Lincoln County 1s one of only three counties
1n the study area which a recent USDA study Identified as having a
diversified economy not driven primarily by agriculture. (Bender*
The aggregate retail sales capture figures discussed above show
more clearly how the changes 1n retail sales In the region have
affected the central place hierarchy of the regions communities.
However* aggregate numbers still tell only part of the story of the
retail changes which have occurred. One way of gaining a better
understanding of retail economies 1s to focus on sales 1n the
different categories of retail goods. Data 1s reported In the Census
of Retail Trade by two digit Standard Industrial Class1f1cat1on(SIC)
codes under the following categories:
Building Materials, Hardware* Garden Supply and Mobile
Home Dealers
General Merchandise Stores
Food Stores
Automotive Dealers (except Gasoline Service Stations)
Gasoline Service Stations
Apparel and Accessory Stores
Furniture, Home Furnishings, and Equipment Stores

Eating and Drinking Places
Drug and Proprietary Stores
Miscellaneous Retail Stores
The types of retail shops and stores typically found 1n downtown
shopping districts or shopping centers may be categorized 1n two
subgroups* convenience goods and shopping goods* as shown below:
Convenience Goods Shoppers Goods
Food General Merchandise
Drug and Proprietary Apparel
Liquor Furniture and Home
Miscellaneous Shopping
Goods Stores
Convenience goods are defined as those Items which are purchased
frequently and which are generally sought at locations close to the
consumer's place of residence. Shoppers goods are less frequently
purchased Items for which consumers often seek comparison shopping
opportunities. It was Initially hoped that 1t would be possible to
use the retail census data to compare percent capture of resident
retail expenditure potential 1n these two categories. Changes 1n
retail trading patterns caused by Improved transportation networks and
the development of regional shopping centers along the Front Range of
Colorado would be expected to show up primarily 1n the shoppers goods
category. If counties showed a decline 1n the convenience goods
categories 1t would Indicate even more serious deterioration of the
central place function. However* due to the relatively few
establishments 1n any retail trade category In many of the study area

counties, most of the detailed data by two digit SIC category has not
been reported 1n the retail census. According to the census, data 1s
withheld to avoid disclosure of Information about any single
business. These serious gaps 1n the data base made 1t Impossible to
conduct this aspect of the analysis using Census of Retail Trade data.
Another source of data on retail sales 1n the study area counties
was therefore sought. Comprehensive data 1s reported by the Business
Research Division of the Graduate School of Business Administration at
the University of Colorado but was not available for all of the years
of the study period. This data Is based on sales tas reports filed
with the State of Colorado and the figures are not directly comparable
to the Census of Retail Trade. However, 1n an attempt to gain some
further Insight Into the current makeup of the retail economies 1n the
study area, this data was used to develop an Independent set of
percent capture figures for the year 1982. Percent captures were
calculated for all of the major categories of retail trade.
Although not totally consistent with the census data, this
analysis, which 1s summarized 1n Table 7, does point out some of the
functional differences among classes of central places 1n the study
area. Shoppers goods sales are clearly concentrated 1n communities of
over 5,000 population, the only classes of counties with 100 percent
or better captures. On the other hand, resident demand for food, a
category representative of convenience goods generally, are
substantially provided 1n counties with largest city populations above
1,000. Eating and drinking sales are concentrated 1n counties with
communities with more than 2,500 residents. Building materials and
auto related sales at first glance seem to contradict any hierarchy of

BACA 243% 192% 59% 85% 33%
BENT 46% 129% 22% 97% 20%
CHEYENNE 105% 113% 28% 53% 89%
CROWLEY 80% 46% 4% 116% 13%
ELBERT 16% 37% 4% 35% 19%
KIOWA 54% 169% 10% 49% 32%
KIT CARSON 357% 186% 83% 90% 133%
LINCOLN 262% 363% 31% 86% 92%
LOGAN 127% 137% 108% 107% 78%
MORGAN 171% 213% 70% 113% 67%
OTERO 117% 83% 110% 95% 62%
PHILLIPS 414% 247% 28% 35% 39%
PROWERS 244% 140% 141% 58% 85%
SEDGWICK 482% 262% 43% 125% 91%
WASHINGTON 103% 97% 29% 74% 60%
YUMA 389% 169% 67% 87% 52%
STUDY AREA 189% 154% 74% 89% 65%
0 to 500 16% 37% 4% 35% 19%
501 to 1000 78% 143% 19% 51% 59%
1001 to 2500 293% 194% 43% 84% 55%
2501 to 5000 230% 163% 58% 93% 87%
5001 to 10000 168% 151% 100% 94% 69%
10001 + 127% 137% 108% 107% 78%
NORTHEAST 217% 179% 74% 99% 67%
EAST CENTRAL 187% 163% 37% 65% 77%
SOUTHEAST 150% 115% 92% 84% 56%

retail function with the strongest percent captures 1n counties with
communities 1n the 1#000 to 5,000 population range. However# building
materials and auto related establishments 1n smaller communities often
also sell farm supplies. This may explain the high percent captures
1n these categories for these size classes.
Although the data for a single year cannot tell us much about
change which has occurred 1n the region# 1t does point out the
functional differentiation which currently exists 1n the retail trade
sector of the region's communities. Cities with a population above
5#000 capture large percentages of residents' purchases across all
retail categories. Counties with cities of 1000 to 5000 population
capture significant percentages of consumers' retail dollars only 1n
the convenience categories. Counties with cities below 1#000
population do not provide for a high percentage of residents' retail
needs 1n any category.
The final Indicator examined regarding changes In retail trade
was the number of retail establishments by SIC category. The Census
of Retail Trade reports the number of establishments with payroll for
all study area counties at the two digit SIC level. Establishments
were aggregated by the shopping goods and convenience goods categories
and tracked over the study period. In order to provide a more
meaningful relative measure across all study area counties and
planning regions the numbers were converted to the number of
establishments per 1000 residents.
In the shoppers goods categories total retail establishments 1n
the study area declined from an average of 2.2 per thousand residents
1n 1963 to 1.5 per thousand population 1n 1982. This compares with

statewide Colorado figures of 1.3 establishments per thousand
population 1n 1963 and 1.2 1n 1982. Although the relative decline 1n
the study area was much greater than for the stater the study area
still had .3 more establishments per thousand population 1n 1982. Two
forces appear to be at work which would explain the change 1n this
Indicator. The first trend 1s the dramatic decline 1n trade 1n
shoppers goods 1n many of the smaller communities 1n the study area.
At the beginning of the study period all classes of county by size of
largest city had approximately two establishments per thousand
residents 1n the shopping goods categories. The greatest declines
between 1963 and 1982 were 1n the classes of counties with largest
cities below 5r000 population. Although counties with cities above
5,000 population also registered declines 1n numbers of shoppers goods
establishments per thousand residents# the declines were not as great
and by 1982 the counties with larger cities retained the largest
numbers of establishments relative to population.
The second factor which explains some of the decline 1n the
counties with larger cities 1s that general trends In retailing have
been toward offering shoppers goods 1n larger stores such as discount
department stores. Only the larger cities In the region have the
necessary population base to support these higher order retailers.
When this data 1s considered by planning region# the Northeastern
and Southeastern regions show similar patterns of change with the
number of shopping goods establishments declining by .6 per thousand
residents 1n both cases as shown 1n Table 8. The East Central
Planning Region experienced the greatest decline from 1.8
establishments per thousand residents In 1963 to just .6

1963 1967 1972 1977 1982 1963-1982
BACA 2.3 2.4 2.6 2.9 1.5 -0.8
BENT 1.3 0.9 0.6 1.1 0.7 -0.6
CHEYENNE 0.8 1.3 2.2 1.9 0.4 -0.3
CROWLEY 0.5 1.3 1.3 0.9 0.3 -0.2
ELBERT 0.5 0.5 0.7 1.3 0.0 -0.5
KIOWA 3.5 1.0 1.0 1.6 0.0 -3.5
KIT CARSON 2.5 2.1 2.1 2.3 0.9 -1.6
LINCOLN 2.4 3.0 1.4 1.7 1.1 -1.3
LOGAN 2.2 2.3 2.1 2.2 1.9 -0.4
MORGAN 2.0 2.2 1.8 1.8 1.7 -0.3
OTERO 2.1 1.9 1.7 2.0 1.5 -0.6
PHILLIPS 2.3 2.6 2.5 2.3 1.3 -1.0
PROWERS 2.3 2.4 2.2 2.0 2.1 -0.2
SEDGWICK 4.0 3.1 3.5 3.3 1.2 -2.8
WASHINGTON 2.1 1.8 1.6 2.6 1.3 -0.8
YUMA 3.0 3.1 3.1 3.0 2.2 -0.8
TOTAL STUDY AREA 2.2 2.1 1.9 2.1 1.5 -0.7
0 to 500 0.5 0.5 0.7 1.3 0.0 -0.5
501 to 1000 2.0 1.1 1.6 1.8 0.2 -1.8
1001 to 2500 2.4 2.5 2.4 2.5 1.5 -1.0
2501 to 5000 1.9 1.5 1.4 1.8 0.8 -1.1
5001 to 10000 2.1 2.1 1.9 1.9 1.7 -0.4
10001 + 2.2 2.3 2.1 2.2 1.9. -0.4
NORTHEAST 2.4 2.4 2.2 2.3 1.8 -0.6
EAST CENTRAL 1.8 1.9 1.6 1.9 0.6 -1.3
SOUTHEAST 2.0 1.9 1.8 1.9 1.4 -0.6

establishments 1n 1982. The East Central region has two
characteristics which may help explain these figures. First* there
are no cities with populations of more than 5000 residents In this
region. Second* this 1s the region Just east of the extensive
comparative shopping opportunltes available In the Denver metropol1tan
area. The construction of most of Interstate 70 during the study
period has Increased the accessibility to these opportunities for
residents of the East Central region.
The change 1n numbers of convenience goods establishments
exhibits a somewhat different pattern* shown 1n Table 9. Total
establishments per thousand residents 1n the study area declined form
2.2 In 1963 to 1.3 In 1982. During the same time period convenience
goods establishments 1n Colorado declined form 1.4 tp .8 per thousand
residents. The study area pattern has generally followed the trend
for the state* and although the number of convenience establishments
remains above the state average* the difference 1s less 1n 1982 than
1t was 1n 1963.
When change 1n the number of convenience establishments 1s
considered by size of largest city 1n the county 1t can be seen that
although the decline 1n number of establishments has been greatest 1n
the counties with cities below 2*500 population* the counties 1n the
500 to 5*000 range still maintain a higher proportion of convenience
establishments relative to population than do counties with largest
city populations over 5*000. This suggests that virtually all
counties 1n the study area still have establishments offering goods
such as groceries* but that 1t 1s primarily 1n the larger areas that
chain supermarkets and drug stores have opened.

.... 1.963 1967 1972 1977 1982 1963-1982
BACA 4.3 3.4 2.1 2.1 1.5 -2.8
BENT 1.8 1.3 1.1 1.0 1.6 -0.3
CHEYENNE 5.4 3.3 4.3 3.3 1.7 -3.6
CROWLEY 1.6 2.2 2.5 1.3 0.3 -1.3
ELBERT 2.1 2.3 1.0 1.1 0.7 -1.5
KIOWA 2.6 2.9 2.0 2.6 3.2 0.6
KIT CARSON 1.8 2.3 1.9 1.9 1.8 .0
LINCOLN 2.2 2.8 2.4 1.9 1.6 -0.6
LOGAN 1.4 1.3 1.2 1.2 0.8 -0.6
MORGAN 1.5 1.9 1.5 1.0 1.2 -0.4
OTERO 2.6 2.4 1.8 1.6 1.4 -1.2
PHILLIPS 2.5 2.4 2.3 3.0 1.3 -1.2
PROWERS 2.9 1.9 2.0 1.5 1.4 -1.6
SEDGWICK 2.2 3.1 2.6 2.7 2.1 -0.1
WASHINGTON 1.8 1.3 1.5 1.2 1.5 -0.3
YUMA 2.1 1.5 1.9 1.7 1.4 -0.7
TOTAL STUDY AREA 2.2 2.0 1.8 1.5 1.3 -0.9
0 to 500 2.1 2.3 1.0 1.1 0.7 -1.5
501 to 1000 4.1 3.1 3.3 3.0 2.4 -1.7
1001 to 2500 2.4 2.3 2.1 1.9 1.4 -1.0
2501 to 5000 1.8 1.9 1.5 1.5 1.7 -0.1
5001 to 10000 2.3 2.1 1.7 1.4 1.3 -1.0
10001 + 1.4 1.3 1.2 1.2 0.8 -0.6
Planning Region 1 1.7 1.7 1.6 1.4 1.2 -0.5
Planning Region 5 2.5 2.6 2.1 1.9 1.4 -1.1
Planning Region 6 2.7 2.3 1.8 1.6 1.4 -1.3

The data aggregated by planning region shows that the greatest
change occurred 1n the regions with the most establishments at the
beginning of the period* perhaps Indicating that chain convenience
goods establishments were later 1n coming to the East Central and
Southeast than to the Norhteast region. This 1s consistent with the
fact that the largest cities 1n the region are 1n the northeast; these
were probably the first communities to attract larger convenience
goods chain retailers.

Trends In Retail Markets
One factor which has clearly Influenced retail markets 1n
Colorado over the last twenty years has been the development of
regional shopping centers. Although this type of shopping center
traces Its beginnings to the 1930s 1t first became prominent
nationally 1n the 1950s and has come to Colorado largely during the
period covered by this study. Regional shopping centers generally
require a trade area with a population of about 150,000 within a
thirty mile radius of the center site. (The Urban Land Institute,
1985) The shopping mall, the most common current form of regional
center, typically 1s anchored by multiple department stores and offers
a wide range of consumer goods, primarily shoppers goods, 1n numerous
specialty shops complementing the full-line department stores. The
first true regional shopping centers 1n Colorado opened 1n the 1950s
In Denver. Since then 21 other regional shopping centers have opened
along the Colorado Front Range, most of them 1n the Denver
metropolitan area. (The Urban Land Institute, 1985) Most significant
In terms of Influence on the study area, however, are the Greeley
Mall, opened In 1973, the Pueblo Mall, opened In 1976, and the Citadel
and Chapel Hills Malls 1n Colorado Springs which opened 1n 1972 and
1981, respectively.
Although regional shopping centers typically have a primary trade
area extending to a radius of about 30 miles or less, they frequently
draw shoppers from far beyond this range. Sales from outside the area
could make up a fairly small percentage of the total sales at a
regional shopping center and at the same time represent a substantial

drain on the retail economies of small central places 1n outlying
rural areas. Thus# the development of the numerous regional shopping
centers along the Colorado Front Range over the past thirty years may
have contributed to the decline of small trade centers 1n the rural
region of eastern Colorado. The large concentrations of retail
establishments offering a wide range of shoppers goods# and oriented
toward highway access# appear to have extended the range of the retail
trade area of the larger Front Range communities at the expense of
retailers 1n the smaller communities 1n rural areas.
Another factor contributing to the extension of the retail market
areas of these larger communities has been the Improvements 1n major
highways completed during the same period as the development of
regional shopping centers. Almost all of Interstate Highway 70 east
from Denver to the Kansas border was built since 1963# as was
Interstate 76 1n northeastern Colorado and Improvements on stretches
of U.S. Highway 50 1n the southeastern part of the state. The
Improvements to these major transportation networks linking the study
area to the larger cities of the Front Range and the Improved highway
acces provided by the highway locations of the new regional shopping
centers have decreased the travel time from many rural areas and small
towns to the regional shopping centers. Thus# 1t has become easier
for the study area residents to take advantage of the more extensive
shopping opportunities provided by these larger, trade centers.
Conversely# 1t has become more difficult for retail establishments 1n
small towns to compete with the regional malls.

These two factors have been Integrated Into this analysis by
estimating the effect on travel time to the nearest major regional
shopping center from the largest city 1n each study area county. In
1963 only Denver offered the type of highway oriented shopping
opportunities typically found 1n regional shopping centers. During
the 1970s and early 1980s these shopping facilities were brought
closer to many eastern plains residents when new regional shopping
centers opened up 1n Pueblo# Colorado Springs and Greeley. The result
In terms of decreased travel time from the largest city 1n the county
to the nearest regional shopping center are shown In Table 10.
In addition to facilitating travel to Front Range cities#
Improvements 1n major highways also Increase the accessibility to
communities located on those highways. As shown 1n Figure 4 the total
retail sales per capita have Increased In aggregate among counties
with cities located on the major transportation networks and decreased
among those counties without major highway access. This has occurred
despite the fact that the distribution of population among counties
with and without major regional highways has remained unchanged as
shown In Figure 5.

COUNTY DISTANCE CENTER 1963 1967 1972 1977 1982 1963-1982
BACA 172 PUEBLO - 3.39 3.39 3.38 3.38 3.36 -1.94
277 DENVER 5.29 5.29 5.29 5.29 5.27
BENT 86 PUEBLO 1.67 1.67 1.67 1.67 1.67 -1.91
191 DENVER 3.57 3.57 3.57 3.57 3.57
CHEYENNE 175 DENVER 3.50 3.39 3.39 3.34 3.34 -0.51
139 COLO. SPRINGS 2.99 2.99 2.99 2.99 2.99
CROWLEY 48 PUEBLO 1.01 1.01 1.01 1.01 1.01 -1.91
153 DENVER 2.92 2.92 2.92 2.92 2.92
ELBERT 50 DENVER 1.00 1.00 1.00 1.00 1.00 0.00
46 DENVER 0.92 0.92 0.92 0.92 0.92
137 COLO. SPRINGS 2.96 2.96 2.96 2.96 2.96
173 DENVER 3.46 3.35 3.35 3.30 3.30
.INCOLN 72 COLO. SPRINGS 1.44 1.44 1.44 1.44 1.44 -0.34
89 DENVER 1.78 1.67 1.67 1.73 1.73
.OGAN 125 DENVER 2.43 2.35 2.34 2.33 2.30 -0.62
96 GREELEY 0.00 0.00 1.84 1.84 1.81
40RGAN 81 DENVER 1.55 1.55 1.55 1.53 1.50 -0.58
50 GREELEY 0.00 0.00 1.00 1.00 0.97
)TER0 67 PUEBLO 1.29 1.29 1.29 1.29 1.29 -1.91
172 DENVER 3.20 3.20 3.20 3.20 3.20
>HILLIPS 173 DENVER 3.39 3.31 3.30 3.29 3.26 -0.62
144 GREELEY 0.00 0.00 2.80 2.80 2.77
ROWERS 123 PUEBLO 2.41 2.41 2.40 2.40 2.38 -1.93
228 DENVER 4.31 4.31 4.31 4.31 4.29
iEDGWICK 180 DENVER 3.53 3.45 3.34 3.32 3.30 -0.69
153 GREELEY 0.00 0.00 2.87 2.87 2.85
fASHINGTON 113 DENVER 2.19 2.19 2.19 2.17 2.14 -0.58
82 GREELEY 0.00 0.00 1.64 1.64 1.61
fUMA 141 DENVER 2.75 2.75 2.75 2.73 2.70 -0.58
110 GREELEY 0.00 0.00 2.20 2.20 2.17
* Indicates change 1n travel time to nearest regional shopping center.

n i____i \ '-.-ry r i i i rt i\l miu
I >-_J £_

1/ I
\ \ \ \ \ \ \ \ X \ \ X \ X \ \
\ X X X \ X \ X X X X X X X X X
L X \. X. x\ X X X X. X X X X X X N
X X \ \ \ \ \ \ \ \ X X X X X X
->>>>> > >>>>>>>> >
/ / X / X X XX XX XX X x X X X X
X X X X X X X X X X ///////,
V X X X X X X X X X //////// A
X X X X X X X x X X X X X X X X X X
I x z_z_/ ; /_z_ \ \ \ \ \ \ \ \ \ \ X \ \ \ \ '
k X X X X \ X X X \ X X X X X X
X X X X X \ X X \ \ X X X \ X
N X X X \ \ X X \ \ X \ \ \ \ \ I
\ X
}}>}>}?>>>)/ // z
X V///Z///A
/ // // // // // A
X X \ \ \ \ X X X X X X X X X X 1
X X X X \ X X \ \ X X X X X X X I
k X X X x \ \ \W X X X X X X N|
\ \
/ / ? / / / /
Y / / / / /
/ / / / /
V /
X \ X x X X
. / / / / / / /
/ / / / / / / / / .
/ / / A
/////////////////// I
/ .A
\ \ X \ V\ \ \ X X X \ \ \ >
\ X X X X X X X \ X X \ X X X
\ X X x\xX X
^ "> ^ N
/ // / / / / / /
//////// /
< "U
£ >
5 m
i A'

1 0
1 0

The Farm Economy
In order to gain an understanding of changes which have occurred
1n the agricultural sector of the study area economy# several
Indicators were selected from the data published 1n the Census of
Agriculture. First and foremost from the perspective of this analysis
were changes In the number and size of farms. The number of farms
declined 1n thirteen of sixteen counties with the greatest decline of
40 percent occurring In Crowley County. The number of farms Increased
1n Elbert and Kiowa counties and remained essentially unchanged In K1t
Carson. County. The growth In number of farms In Elbert County
reflects the suburbanization and exurbanization of parts of that
county with an Increase In the number of part time or "hobby" farms.
Overall the number of farms declined by 12 percent 1n the study area
as a whole compared with a decrease of nine percent 1n the State of
When the data 1s aggregated by the size of largest city 1n the
county# only counties with largest towns below 1000 population showed
an Increase 1n the number of farms. In general# the counties with the
largest cities showed the greatest decline In number of farms. All
planning regions 1n the study area lost farms with the smallest
decline of just two percent occurring 1n the East Central region#
which also had the largest Increase 1n population# and the largest
decline of 18 percent 1n the Southeast region# which experienced the
greatest population loss. Data on the number of farms 1s presented 1n
Table 11.

1964 1969 1974 1978 1982 1963-1982
3ACA 707 771 754 718 643 -9%
3ENT 414 366 309 336 301 -27%
3HEYENNE 320 370 364 336 307 -43
'ROWLEY 362 309 243 249 218 -40%
ELBERT 579 566 513 518 611 6%
CIOWA 304 412 386 366 327 6%
:IT CARSON 760 844 842 802 763 0%
.INCOLN 534 480 444 475 466 -13%
.OGAN 1092 1082 1031 1013 908 -17%
ORGAN 1056 1038 934 939 846 -20%
)TERO 647 578 550 576 536 -17%
HILLIPS 460 466 444 445 427 -7%
'ROWERS 683 669 618 578 518 -24%
IEDGWICK 356 327 302 316 253 -29%
IASHINGTON 885 935 906 894 854 -4%
TJMA 1009 1023 1065 1060 996 -1%
OTAL STUDY AREA 10168 10236 9705 9621 8974 -12%
to 500 579 566 513 518 611 6%
01 to 1000 624 782 750 702 634 2%
001 to 2500 4313 4311 4158 4157 3857 -11%
501 to 5000 1174 1210 1151 1138 1064 -9%
001 to 10000 2386 2285 2102 2093 1900 -20%
0001 + 1092 1082 1031 1013 908 -17%
ORTHEAST 4858 4871 4682 4667 4284 -12%
AST CENTRAL 2193 2260 2163 2131 2147 -2%
OUTHEAST 3117 3105 2860 2823 2543 -18%

Average farm size showed a similar though as would be expected
opposite trend with farm size Increasing 1n 11 of 16 counties. The
largest Increase was 1n Crowley County where the acreage of the
average farm grew by 83 percent. By contrast the largest decrease 1n
farm size was nine percent 1n Elbert County. For the study area as a
whole average farm size Increased by eight percent, compared with a
four percent decrease statewide. Trends by class of city and for the
planning regions were essentially the opposite of what was shown for
the number of farms: areas that had declines In the number of farms
had corresponding Increases 1n farm size and vice versa. It 1s
Interesting to note that average farm size 1n the study area rose
from about thirty percent larger than the statewide average 1n 1964 to
almost fifty percent larger 1n 1982 when the average study area farm
was 1.835 acres compared to a 1.237 acre average for the State of
Colorado. This reflects the growing predominance of the eastern
plains as the location for large scale agriculture In the state.
Total farm acres 1n the study area decreased five percent from
1964 to 1982. a period during which acreage statewide decreased by 12
percent. The Northeastern Planning Region showed the greatest decline
1n total farm acres, down six percent, followed by the Southeast
region with a five percent decline and East Central region which lost
Just three percent of Its acres of farmland.
A very different perspective on agriculture 1n the area 1s gained
by looking at harvested cropland. This 1s a significant Indicator of
the farm economy because cropland generally brings a much higher
return to the farmer per acre than does pasture or grazing land.
Acres of harvested cropland Increased 1n every county In the study

1964 1969 1974 1978 1982 1963-1982
BACA 1833 1677 1856 1856 1996 9%
BENT 1989 2269 2560 2173 2528 21%
CHEYENNE 3025 2440 2485 2715 3080 2%
CROWLEY 1127 1403 1796 1795 2066 83%
ELBERT 1816 1794 2106 2107 1654 -9%
KIOWA 2909 243 8 2596 2741 2785 -4%
KIT CARSON 1675 1617 1631 1593 1701 2%
LINCOLN 3037 3261 3636 3382 3202 5%
LOGAN 1082 1058 1115 1118 1180 9%
MORGAN 775 779 801 828 823 6%
OTERO 1213 924 1159 1325 1164 -4%
PHILLIPS 1042 1009 1052 1078 1085 4%
PROWERS 1509 1379 1652 1687 1797 \9%
SEDGWICK 953 1057 1132 1071 1284 35%
WASHINGTON 1617 1518 1525 1534 1599 -1%
YUMA 1443 1382 1346 1390 1422 -1%
TOTAL STUDY AREA 1559 1511 1626 1633 1676 8%
0 to 500 1816 1794 2106 2107 1654 -9%
501 to 1000 296 8 2439 2542 2729 2928 -1%
1001 to 2500 1630 1610 1701 1696 1762 8%
2501 to 5000 1786 1814 1880 1764 1935 8%
5001 to 10000 1104 991 1145 1202 1185 1%
10001 + 1082 1058 1115 1118 1180 9%
NORTHEAST 1174 1150 1179 1194 1246 6%
EAST CENTRAL 2241 2145 2299 2294 2211 -1%
SOUTHEAST 1677 1616 1849 1860 1951 16%

area and many of these Increases represented major changes 1n the
nature of the agriculture 1n the county. Cropland more than doubled
1n Kiowa County with a 126 percent Increase and nearly tripled In
Cheyenne County with an Increase of 198 percent as shown 1n Table 13.
The East Central Planning Region posted the greatest Increase of
72 percent* followed by the Southeast region* up 61 percent* and the
Northeast, where harvested cropland acres grew by 38 percent. In the
study area as a whole harvested cropland Increased by 52 percent
compared with a statewide Increase of just 28 percent. In 1964 the
study area accounted for Just over fifty percent of the harvested
cropland 1n the state. By 1982 this had Increased to 60 percent, and,
more significantly* the 1.2 million acre Increase 1n harvested
cropland within the study area accounted for 95 percent of the
Increase 1n harvested cropland In the entire state.
The Increases 1n cropland resulted from two factors: first*
acreage Irrigated from groundwater grew significantly spurred by the
Introduction of central pivot Irrigation to the region In the late
1950s; second* the strong export market for grain In the mid-1970s
encouraged the plowing of grazing land for dryland wheat farming.
These figures point out the Increasing predominance of the study area
as the agricultural center of Colorado.
Harvested cropland as a percent of total farmland Increased from
15 percent for the study area 1n 1964 to 24 percent 1n 1982. The
Northeast region remained predominant with 31 percent of Its farmland
1n cropland 1n 1982 compared to 21 percent 1n 1964. The other two
planning regions showed virtually Identical trends Increasing from 11

1964 1969 1974 1978 1982 1963-1982
BACA 236,797 278,326 414,731 385,773 371,756 57%
BENT 50,245 61,044 74,101 66,997 71,659 43%
CHEYENNE 80,340 151,273 195,815 184,382 239,463 198%
CROWLEY 29,172 37,382 50,436 39,289 34,192 17%
ELBERT 81,589 81,610 94,521 85,967 90,633 11%
KIOWA 96,730 180,3 80 240,627 195,092 218,981 126%
KIT CARSON 265,468 296,215 402,652 350,394 418,519 58%
LINCOLN 128,051 163,036 185,486 167,471 208,389 63%
LOGAN 233,221 281,064 284,267 282,840 268,684 15%
MORGAN 175,567 177,694 197,084 215,735 201,666 15%
OTERO 55,982 55,763 55,720 61,080 59,325 6%
PHILLIPS 136,770 180,480 217,896 213,263 239,242 75%
PROWERS 162,438 211,336 283,164 260,345 258,443 59%
SEDGWICK 103,998 113,108 121,076 126,193 129,506 25%
WASHINGTON 286,172 311,881 380,293 366,157 401,630 40%
YUMA 264,891 285,826 365,958 413,221 419,954 59%
TOTAL STUDY AREA 2,387,431 2,866,418 3,563,827 3,414,199 3,632,042 52%
0 to 500 81,589 81,610 94,521 85,967 90,633 11%
501 to 1000 177,070 331,653 436,442 379,474 458,444 159%
1001 to 2500 1,185,851 1,370,039 1,735,876 1*711,367 1,804,669 52%
2501 to 5000 315,713 357,259 476,753 417,391 490,178 55%
5001 to 10000 393,987 444,793 535,968 537,160 519,434 32%
10001 + 233,221 281,064 284,267 282,840 268,684 15%
NORTHEAST 1,200,619 1,350,053 1,566,574 1,617,409 1,660,682 38%
EAST CENTRAL 555,448 692,134 878,474 788,214 957,004 72%
SOUTHEAST 631,364 824,231 1,118,779 1,008,576 1,014,356 61%

and 12 percent of total farmland 1n cropland In 1964 to 20 percent 1n
The final factor examined 1n this sketch of changes In the farm
economy 1s the market value of agricultural products sold# shown 1n
Table 14. In constant dollar terms there were Increases 1n the value
of farm production 1n every county 1n the study area* although the
degree of Increase varied widely from county to county. Increases of
twenty percent or less were posted 1n Logan and Morgan counties while
other counties registered much more Impressive gains: Crowley County*
up 884 percent; Phillips County* up 365 percent; and Prowers County*
up 303 percent* to name Just a few. In the study area as a whole the
market value of agricultural products sold was up 121 percent 1n
constant 1982 dollars from $604 million 1n 1964 to over $1.3 billion
1n 1982. These figures accounted for over 36 percent of the value of
agricultural products 1n the state 1n 1964* rising to more than 45
percent of the statewide total 1n 1982. The peak year for value of
agricultural production 1n the study area was 1974; from 1974 to 1982
the constant dollar market value of agricultural products declined by
almost $300 million.
Because the counties 1n the study area vary 1n land area and 1n
acres of farmland, a value of agricultural products sold per acre of
farmland was calculated as an Indicator of the relative productivity
of the land. For the study area as a whole the market value of
prodcuts sold per acre of farmland increased from $38 to $88 during
the study period as shown 1n Table 15. The Northeast region was most
productive 1n 1964 with $63 of products per acre. The East Central
and Southeast regions were far behind with per acre production of $19

(1982 S1000)
1964 1969 1974 1978 1982 1964-1982
BACA $24,638 $59,519 $78,847 $75,171 $60,903 1475?
BENT $17,582 $27,001 $38,552 $32,427 $38,344 1185?
CHEYENNE $9,761 $17,541 $34,636 $28,141 $36,682 2765?
CROWLEY $9,135 $13,526 $61,958 $97,319 $89,904 8845?
ELBERT $18,004 $23 ,984 $26,957 $32,083 $25,078 395?
KIOWA $9,484 $22,790 $40,697 $24,902 $23,629 1495?
KIT CARSON $41,352 $86,185 $172,564 $151,542 $143,511 2475?
LINCOLN $22,415 $30,134 $47,888 $64,614 $48,667 1175?
LOGAN $111,485 $192,909 $256,806 $189,416 $131,389 185?
MORGAN $139,517 $223,465 $263,365 $276,347 $166,959 20%
OTERO $54,575 $96,888 $97,541 $97,739 $90,313 65%
PHILLIPS $16,409 $25,423 $79,738 $61,065 $76,347 365%
PROWERS $35,398 $59,933 $133,379 $105,969 $142,704 303%
SEDGWICK $24,962 $29,438 $49,789 $40,987 $39,714 59%
WASHINGTON $30,661 $42,590 $87,168 $69,243 $72,306 136%
YUMA $38,998 $97,881 $160,878 $249,876 $149,573 284%
TOTAL STUDY AREA $604,377 $1,049,208 $1,630,762 $1,596,841 $1,336,023 121%
0 to 500 $18,004 $23,984 $26,957 $32,083 $25,078 39%
501 to 1000 $19,245 $40,332 $75,333 $53,043 $60,311 213%
1001 to 2500 $167,219 $298,511 $566,266 $658,276 $537,414 221%
2501 to 5000 $58,935 $113,186 $211,115 $183,969 $181,855 209%
5001 to 10000 $229,490 $380,285 $494,285 $480,055 $399,976 74%
10001 + $111,485 $192,909 $256,806 $189,416 $131,389 18%
NORTHEAST $362,032 $611,706 $897,744 $886,934 $636,288 76%
EAST CENTRAL $91,533 $157,845 $282,045 $276,380 $253,93 8 177%
SOUTHEAST $150,813 $279,657 $450,973 $433,527 $445,797 196%

1964 1969 1974 1978 1982 1964-1982
BACA $19.01 $46.03 $56.34 $56.41 $47.45 150*
BENT $21.35 $32.51 $48.74 $44.41 $50.39 136*
CHEYENNE $10.08 $19.43 $38.29 $30.85 $38.79 285*
CROWLEY $22.39 $31.20 $141.97 $217.74 $199.61 791*
ELBERT $17.12 $23.62 $24.95 $29.40 $24.82 45*
KIOWA $10.72 $22.69 $40.61 $24.82 $25.95 142*
KIT CARSON $32.48 $63.15 $125.66 $118.62 $110.57 240*
LINCOLN $13.82 $19.25 $29.66 $40.22 $32.62 136*
LOGAN $94.36 $168.52 $223.39 $167.25 $122.63 30*
MORGAN $170.47 $276.36 $352.03 $355.43 $239.79 41*
OTERO $69.54 $181.41 $153.02 $128.06 $144.75 108*
PHILLIPS $34.23 $54.07 $170.71 $127.30 $164.79 381*
PROWERS $34.35 $64.96 $130.64 $108.68 $153.31 346*
SEDGWICK $73.58 $85.17 $145.64 $121.11 $122.25 66*
WASHINGTON $21.43 $30.01 $63.09 $50.49 $52.95 147*
YUMA $26.78 $69.23 $112.23 $169.59 $105.61 294*
TOTAL STUDY AREA $38.14 $67.83 $103.33 $101.63 $88.81 133*
0 to 500 $17.12 $23.62 $24.95 $29.40 $24.82 45*
501 to 1000 $10.39 $21.15 $39.51 $27.69 $32.49 213*
1001 to 2500 $23.78 $43.01 $80.05 $93.39 $79.08 233*
2501 to 5000 $28.11 $51.56 $97.54 $91.63 $88.33 214*
5001 to 10000 $87.13 $167.88 $205.39 $190.82 $177.69 104*
10001 + $94.36 $168.52 $223.39 $167.25 $122.63 30*
NORTHEAST $63.45 $109.19 $162.58 $159.15 $119.21 88*
EAST CENTRAL $18.63 $32.56 $56.72 $56.55 $53.50 187*
SOUTHEAST $28.85 $55.73 $85.29 $82.56 $89.88 212*

and $29 of farm products* respectively. In 1982 the Northeast
remained predominant with $119 of products per acre followed by the
Southeast with $90 and the East Central region with $54.
What 1s most Interesting about the data on value of agricultural
production Is not the regional differences but the differences
observed when the data 1s aggregated by size of largest city 1n the
county. As shown 1n Figure 6 there 1s a direct correlation evident
between the productivity of the land and the size of the largest city
1n the county. This 1s evidence of how the underlying resource base
has determined the location and growth of communities 1n the region.
In an area dominated by agriculture 1t Is the richness of the
agricultural resources which determines the level of the population
base that can be supported.

$2 40
j $1 80
C $1 60
i $1 40
3 $1 20
j $1 oo
l $80
C $60
$ AO




S -

TLX /\'/A
X \>

/AY \x
7 A

. V'\N,

s /

\ -V s


's\. A
' \Y
[771 1 964
Y\ 19 69 7/7 1974 ^ 1 978 XX
1 9 82

The Farm Support Economy
Another factor considered relevant to the relationship between
the farm economy and the economies of small town central places 1s the
farm support sector of the economy. For purposes of this analysis the
"farm support" sector 1s defined to Include agricultural services# and
wholesale trade 1n farm machinery# farm supplies# and farm products.
These sectors cover the support services which a farmer would
routinely rely upon 1n operating a farm: the providers of direct
farming services such as custom combining and aerial spraying; the
suppliers of machinery and supplies; and the purchasers of farm
products. By Including these farm support activities 1n the analysis
1t was thought that a relationship might be seen between where farmers
go to conduct farm-related business and where they and their families
make their retail purchases.
The economic censuses# such as the Census of Wholesale Trade# did
not contain sufficiently detailed data to allow analysis of the level
of sales In these agricultural support activities by county over
time. Therefore# data on employment and number of firms from the
census publication County Business Patterns was examined. Even here
the level of detail of data appears to have changed somewhat over
time. Still# the data 1s believed to accurately reflect the relative
concentration of farm support sector activity 1n the study area during
the selected years. The numbers reported here are aggregates of data
reported separately 1n the four sectors Identified above.

The first point which becomes apparent when examining this data
1s that the farm support sectors of the economy of the study area grew
during the study period. Estimated total employment 1n these sectors
1n the study area Increased from 1069 1n 1964 to 2224 1n 1982.
Although some of this Increase may be due to changes In the way firms
were reported there 1s no question that the farm support sectors have
expanded over the last twenty years. A comprehensive analysis of the
reasons for this Increase are beyond the scope of this study, but 1t
most likely reflects the Increasing capital Intensity of agriculture
with more and more dollars expended on the additional machinery and
agricultural chemicals which have accompanied Increases In
production. (Gregor, 1982)
The number of firms providing these farm support services also
Increased slightly during the study period but much less than did
employment; the major Increase was 1n the number of employees per
firm. In 1964 202 firms were reported with an average of
approximately five employees per firm while by 1982 there were 209
firms with average employment per firm of more than ten employees.
Contrary to what might Initially have been expected, but not
suprising given the rapid Increase 1n activity 1n these sectors, the
location of farm services was actually much more geographically
concentrated 1n 1964 than 1n 1982. In 1964 Logan, Morgan, and Otero
counties accounted for 68 percent of the firms and 70 percent of the
employment reported 1n these standard Industrial classification
categories. By 1982 this had decreased to just 45 percent of the
firms and 53 percent of the employment. Although these counties
remained the top three 1n employment, K1t Carson and Yuma counties

also developed Into major centers of activity. Farm support
employment remained concentrated 1n counties with a largest city of
over 5000 population throughout the study period# although a few
counties with largest cities of 1#000 to 5000 population registered
significant employment gains In these sectors.
When employment 1n the farm service sectors 1s aggregated by
planning region# the Northeast Planning Region appears to have
strengthened Its role as the agricultural service center of the
region. The Southeast Planning Region remained second In employment
throughout the study period but significant employment growth
occurred In the East Central region# the only region to experience a
net Increase 1n farm support businesses. Farm support employment data
1s presented 1n Table 16.

1964 1967 1972 1977 1982 CHANGE 1964-1982

BACA 0 0 2 5 8 8
BENT 6 16 8 3 0 -6
CHEYENNE 0 0 0 3 3 3
CROWLEY 0 0 0 0 0 0
ELBERT 0 0 0 0 15 15
KIOWA 0 0 0 3 5 5
KIT CARSON 68 144 166 216 258 190
LINCOLN 6 14 14 9 12 6
LOGAN 164 109 217 288 364 200
MORGAN 288 286 252 474 496 208
OTERO 300 340 198 259 324 24
PHILLIPS 14 118 16 3 86 72
PROWERS 140 44 143 151 129 -11
SEDGWICK 2 2 2 12 57 55
WASHINGTON 45 0 0 77 172 127
YUMA 36 52 65 177 295 259
STUDY AREA 1069 1125 1083 1680 2224 1155
0 to 500 BY SIZE OF LARGEST CITY 0 0 0 IN THE 0 COUNTY 15 15
501 to 1000 0 0 0 6 8 8
1001 to 2500 103 186 99 283 630 527
2501 to 5000 74 160 174 219 258 184
5001 to 10000 728 670 593 884 949 221
10001 + 164 109 217 288 364 200
NORTHEAST 549 BY STATE 567 PLANNING REGION 552 1031 1470 921
EAST CENTRAL 74 158 180 228 288 214
SOUTHEAST 446 400 351 421 466 20
Source: County Business Patterns

Summary of Regional Trends
Several regional trends become apparent from the proceeding
description of retail and agricultural factors. Some of the trends
are best explained In terms of what has happened In the three state
defined planning regions which make up the study area. The northern
third of the study area was stable throughout the study period. The
Northeastern Planning Region which encompasses this area experienced a
small Increase In population and was able to maintain the best overall
balance In retail trade. The farm economy underwent the least change
here* which primarily reflects the already well developed and
productive agricultural economy at the beginning of the study period.
The Northeast region had* and still has* both the most productive
farmland and the most extensive farm service economy. Also of
significance 1s the fact that Sterling and Fort Morgan* the largest
cities 1n the study area* are In this planning region. While 1t 1s no
coincidence that the most productive agricultural area supports the
largest cities* Sterling and Fort Morgan also obviously have broader
economic bases to support their larger populations and this regional
economic base has contributed to the stability of the Northeast
The East Central Planning Region has undergone a different set of
changes with significant Increases In population* Irrigated farm
acreage and farm support businesses. This area has been Influenced by
the construction of Interstate 70* the major east-west Interstate
through Colorado* allowing cities such as L1mon and Burlington to
capture highway related trade. At the same time the highway has made

1t easier for residents of the region to travel to larger retail trade
centers# especially Denver. Although the major cities In the area
have remained relatively prosperous# the region as a whole has seen
declines In retail trade relative to the potential which exists In the
region as a result of Increased population and Income. In terms of
1970 population# Burlington with 2828 residents was the largest city
1n the planning region but ranked only seventh 1n population among
study area cities. In terms of the effects on retail trade# the
combination of no major city and the Improved access through the area
seem to have offset gains 1n population# Income# farm service
employment# and agricultural production which have occurred 1n the
The Southeast Planning Region Is an area which has experienced
overall decline throughout the study period. Not only has this area
seen the only planning region-wide loss of population but 1t has also
registered the greatest relative decline 1n the performance of the
retail economy. The decrease 1n number of farms was also greatest 1n
this area and average farm size grew more than In any other planning
region. There was virtually no growth 1n farm service employment. On
the other hand# the Southeast region recorded large Increases 1n acres
of cropland and 1n the market value of agricultural products sold.
The Southeast region differs from the other two regions In two
major respects: first# 1t does not have a four lane highway running
all the way through the region; and# second# the settlement pattern Is
marked by a number of medium sized towns with four cities of between
4#000 and 8#000 population. It appears that a contributing factor to
this area's retail decline could be that these cities are all

competing for a retail sales base which 1s not growing very rapidly.
None of these cities has reached the population threshold necessary to
support the size and dlverstly of establishments required to serve a
more regional market. As a result* lower Income growth* population
decline and Increased outflow have led to a reduction In constant
dollar retail sales.
Other Important perspectives on the study area are gained by
focusing on the size of the largest city In the county. The number of
farms Increased only 1n counties with largest cities of less than
1*000 population* the group of counties with the largest declines In
every measure of the retail economy. The counties with the largest
cities 1n the study area experienced both the largest decline 1n
numbers of farms and the largest Increases 1n retail sales.
Clearly* factors other than farm size and number of farms are
more Important as determinants of changes 1n retail trade In the study
area Counties. Counties with cities of more than 5*000 population and
counties which are located on the major regional highways consistently
show better overall performance ot the retail economy* both 1n terms
of absolute measures of retail trade In a given year and 1n terms of
changes which occurred during the study period. However* the major
urban populations and highest absolute levels of retail sales are
still 1n areas where the agricultural resource base 1s most productive
and the average farm size 1s relatively small.

The descriptive analysis 1n the prior section suggests trends
which may explain differences and changes 1n retail trade 1n the study
area. In this section statistical techniques are applied to determine
the degree of relationship between variables and help more clearly
Indentlfy those factors which may be responsible for the differences
1n performance of retail economies. For purposes of this statistical
analysis eight factors were selected from the data analyzed 1n the
previous subsections covering the critical aspects of retail trade*
central place theory* the farm economy* the farm support economy*
competition from larger regional trade centers* and Improvements In
The selection of the variables for this analysis was based upon a
review of prior research analyzing central place functions and retail
trade* the objectives of this particular study* and the results of the
analysis conducted 1n the prior section on trends 1n the study area.
Retail sales per capita was selected as the variable by which to
compare differences 1n retail trade between the study area counties.
This was considered the most appropriate measure for purposes of this
analysis because this factor Indicates how retail trade varies between
areas* while controlling for differences In population level. Per
capita retail sales 1s an appropriate factor for correlation
analysis with other factors which have been hypothesized as
determinants of levels of retail trade 1n an area and has been used
by other researchers conducting similar research. (Walzer* 1979)

Two population factors were Included 1n the analysis. First#
population of the largest city 1n the county was selected as a
variable Indicating the level of central place hierarchy present In
the county. This factor 1s generally Included 1n one form or another
In all research regarding elements of central place theory. As
discussed 1n considerable detail earlier 1n this analysis# the size of
the largest city 1n the county would be expected to be the primary
determinant of the variety of goods available In the county and of the
opportunity to capture residents' retail expenditures locally. In
addition to the size of largest city# total central place population
1n the county was also Included as a separate variable In the
analysis. This variable was Intended to account for the possibility
of the development of a "dispersed central place#" as suggested by
some researchers# where different communities specialize In one or
more categories of goods. (Hart# 1969) If this were the case# the
size of the largest city might not be the best Indicator of consumer
opportunities 1n the county. Instead# the total central place
population might be a better measure of the range of goods offered.
Per capita personal Income was Included as a measure of the
purchasing power of the residents of a county. Higher Incomes would
be expected to result In a higher level of retail sales per capita.
This factor has been shown to be an Important determinant of levels of
retail trade In other studies of the central place function In rural
communities. (Walzer# 1979)
Average farm size was chosen as the best Indicator of the scale
of agricultural production In a county. This variable was employed In
earlier research regarding the Influence of scale of production on

rural communities. (Goldschmidt* 1978) Most of the change 1n farm
size 1n the study area has resulted from a consolidation of farms;
change In farm size has consistently exhibited an Inverse relationship
to change In the number of farms. As will be shown later* this
variable 1s also useful 1n terms of understanding the relationship
between the agricultural resource base and the development of central
place economies.
Average market value of farm products sold per acre was selected
as an Indicator of the relative productivity of the resource base.
This 1s an Important component of the effort 1n this analysis to
understand how the resource base Influences central place functions
and avoid the pitfalls which might be encountered by trying to draw
conclusions from a comparison of data regarding markedly different
agricultural areas.
In addition to the factors regarding the agricultural base and
scale of production a variable was Included to show any links between
the locations at which farm support goods and services are available
and the location of centers of retail trade. The variable chosen was
"farm support" employment as that term was defined earlier In this
Finally* travel time to the nearest regional shopping center was
Included 1n this analysis to show the Influence of change 1n larger
cities' retail economies upon retail trade In small towns and rural
areas. This 1s of particular Interest because of the development of
regional shopping centers 1n many larger cities proximate to the study
area during the period covered by the analysis and the Improvements In
transportation networks linking the study area to these cities.

A review of texts on the statistical techniques commonly applied
by geographers to analyze spatial data Indicated that correlation
analysis was the most useful and valid statistical tool to be applied
1n this research. (Taylor* 1977; Krueckeberg# 1978) The primary goal
of the research was to determine whether the agricultural resource
base and farm economy of a region might Influence the relative level
of retail trade 1n the area's local economies. Based on prior
research and central place theory, 1t was also recognized that several
other factors are likely to Influence the level of retail trade In an
area. By evaluating variables representing the key factors Identified
1n other research with Indicators of the farm economy* correlation
analysis can help to show which factors are most closely related to
the performance of the retail economy 1n the counties 1n the study
Regression analysis was also explored as a potential tool of
measuring the degree and causation of relationships between the
variables studied 1n this research. This technique has also been used
by researchers addressing questions similar to the Issues examined
here. However* due to the small number of observations for any single
year (the 16 counties 1n the study region) and the relatively large
number of variables to be analyzed* this statistical technique was
found to be less applicable. Several regressions were run but none of
the variables were found to be significant at the .05 confidence
level. Therefore* the discussion here 1s limited to the results of
the correlation analysis.

Correlations were run using the Number Cruncher Statistical
System software package for microcomputers. A separate correlation
was conducted for each of the five focal years of the study period.
The results of this analysis are presented 1n the table on the
following pages. Several points are apparent from the results of the
correlation analysis. The supr1s1ng finding 1s the degree to which
many of the correlations vary between the years of the analysis. Per
capita Income shows the strongest correlation with per capita retail
sales 1n 1963 but becomes much less significant 1n the later years of
the analysis and shows essentially no correlation at all to per capita
retail sales 1n 1982. Population of largest city, total central place
population# value of agricultural production, and farm support
employment all show similar patterns across the years of the study
period. These factors appear to Increase 1n Importance during the
late '60s and '70s. They consistently show the highest correlation
with retail trade of any of the Identified variables. By contrast,
farm size shows a very weak correlation Indicating that 1t has not
been an Important determinant of per capita retail sales 1n the
communities 1n the study area.
Travel time to the nearest regional shopping center shows Its
strongest correlation to retail sales 1n 1963 when Denver was the only
regional trade center as defined for this analysis. The only other
time a similar correlation 1s observed 1s 1n 1977 which 1s the first
year of the analysis that the newer regional shopping centers 1n other
Colorado Front Range cities were open. Still, travel time does not
show a strong correlation to retail sales during any study period

1 2 3 4 5 6 7
1 Retail Sales Per Capita 2 Population of Largest City 0.27
3 Central Place Population 0.29 0.95
4 Per Capita Income 0.45 -0.33 -0.28
5 Average Farm Size -0.01 -0.46 -0.48 0.11
6 Value of Ag Products/Acre 0.28 0.69 0.70 0.00 -0.64
7 Farm Support Employment 0.25 0.85 0.95 0.19 -0.49 0.78
8 Travel Time to Reg. S. C. 0.37 0.04 -0.03 0.06 0.06 -0.29 -0.18
1 2 3 4 5 6 7
1 Retail Sales Per Capita 2 Population of Largest City 0.44
3 Central Place Population 0.45 0.94
4 Per Capita Income 0.24 -0.17 -0.18
5 Average Farm Size -0.23 -0.51 -0.54 0.14
6 Value of Ag Products/Acre 0.30 0.41 0.43 -0.02 -0.39
7 Farm Support Employment 0.42 0.66 0.83 -0.10 -0.57 0.60
8 Travel Time to Reg. S. C. 0.14 -0.10 -0.07 0.43 -0.04 -0.40 -0.23
1 2 3 4 5 6 7
1 Retail Sales Per Capita 2 Population of Largest City 0.37
3 Central Place Population 0.31 0.94
4 Per Capita Income 0.16 -0.16 -0.19
5 Average Farm Size -0.21 -0.49 -0.51 -0.29
6 Value of Ag Products/Acre 0.38 0.67 0.65 0.31 -0.76
7 Farm Support Employment 0.37 0.89 0.90 0.04 -0.54 0.77
8 Travel Time to Reg. S. C. 0.26 -0.04 -0.01 -0.05 -0.04 -0.18--0.15

1 2 3 4 5 6 7
1 Retail Sales Per Capita 2 Population of Largest City 0.41
3 Central Place Population 0.36 0.94
4 Per Capita Income 0.18 -0.18 -0.19
5 Average Farm Size -0.23 -0.51 -0.53 -0.03
6 Value of Ag Products/Acre 0.23 0.56 0.59 -0.01 -0.68
7 Farm Support Employment 0.36 0.83 0.86 -0.10 -0.59 0.75
8 Travel Time to Reg. S. C. 0.33 -0.28 -0.32 0.47 0.04 -0.35 - -0.30
1 2 3 4 5 6 7
1 Retail Sales Per Capita 2 Population of Largest City 0.33
3 Central Place Population 0.30 0.94
4 Per Capita Income 0.02 -0.13 -0.15
5 Average Farm Size 0.02 -0.53 -0.57 -0.10
6 Value of Ag Products/Acre 0.11 0.55 0.59 0.13 -0.67
7 Farm Support Employment 0.28 0.79 0.82 0.03 -0.71 0.59
8 Travel Time to Reg. S. C. 0.17 -0.28 -0.32 -0.54 0.17 -0.25 -0.27
1 2 3 4 5 6 7
1 Retail Sales Per Capita 2 Population of Largest City 0.01
3 Central Place Population 0.01 0.84
4 Per Capita Income 0.11 0.04 0.1
5 Average Farm Size 0.12 -0.4 -0.46 0.23
6 Value of Ag Products/Acre -0.16 0.19 0.17 0.2 0.54
7 Farm Support Employment 0.14 0.56 0.61 0.1 -0.33 0.04
8 Travel Time to Reg. S. C. 0.15 0.37 0.2 0.07 -0.6 0.4 0.45

None of the Individual variables alone correlates strongly with
per capita retail sales. It 1s likely that several of the factors
play a role 1n explaining the differences 1n retail sales 1n an
agriculturally based region and that retail trade 1n each community 1s
affected by Its unique combination of characteristics. Location on or
off a major regional highway* a factor not specifically Identified as
a major component of the hypothesis tested In this thesis and not
amenable to the correlation analysis* may be more Important than any
of the factors analyzed here. Further research 1s recommended to
determine which factors best explain the differences 1n retail sales
1n rural communities. Such research should examine retail trade In a
larger number of rural counties so that regression analysis techniques
can be successfully applied.
Although the data does not support the hypothesis that farm size
1s an Important determinant of differences 1n retail trade 1n the
study area during the years from 1963 to 1982* the correlations
between the agricultural variables and size of largest city 1n a
county are relatively strong throughout all the years of the
analysis. As would be expected based on the earlier discussion of
trends 1n farm support employment* this variable 1s shown to correlate
strongly with size of largest city 1n the county. There 1s also a
relatively strong correlation between size of largest city and value
of agricultural products per acre as well as a weaker Inverse
correlation between average farm size and size of largest city. These
relationships suggest that the agricultural resource base 1s Important
1n determining the pattern of central place development 1n the

There 1s also an Inverse correlation between farm size and value
of agricultural products per acre. Although correlation analysis does
not show causation, this relationship among the varalbles suggests an
explanation of central place development based on agricultural
productivity whereby productivity of the resources base determines
farm size and the level of demand for farm support services. Lower
farm size also means higher rural population density and more demand
for consumer goods and services. The higher demand for both farm and
consumer products leads to the formation of larger agricultural
support communities In an area.
Before concluding this analysis 1t 1s Important to note that farm
size 1n the study area actually remained relatively stable during the
study period compared to what could happen 1f predictions that as many
as one-th1rd of current farm operations will go out of business are
accurate. Farm size appears to have been related to the formation and
growth of central places 1n the study area. Any drastic change 1n
farm size could severely Impact the current central place hierarchy.
As one state agricultural official noted, very large corporate farmers
can afford to go to Denver, Greeley or Colorado Springs for a better
buy on a tractor and often have their own trucks to haul products as
far as necessary ,to find the highest price. In an area dominated by
such large operations, the farm support sector, an Important component
of the economies of many eastern Colorado cities, could be severely
This analysis also suggests that changes 1n farm size and number
of farms 1n the study area during the study period were related to
changes 1n agricultural productivity and 1n the agricultural resource

base. There 1s evidence that current changes 1n the farm economy are
more closely related to the level of debt Incurred 1n the 1970s than
to changes 1n productivity or 1n the resource base. Therefore*
changes occurring today 1n the farm economy may have very different
effects than did changes which occurred during the previous twenty

The analysis to this point has been focused on a regional
perspective. However* the data clearly shows that each county 1n the
region has a unique set of characteristics and economic
circumstances. Although regional trends are Important for the future
economic welfare of all the people 1n the study area* 1t 1s only by
focusing on specific commultles that the nature of the economic change
that has taken place can be fully appreciated. In this section three
counties and their primary central places which exhibit very different
characteristics have been chosen as the focus. In this way 1t 1s
possible to better understand the economic forces at work 1n the small
town economies of the study area.
In addition to the sources utltUzed 1n the regional analysis*
resources of the Denver Public Library's Western History Department
were also used to document changes which have occurred 1n these
communities. Of particular assistance were the newspaper clipping
files which chronicled major events 1n the development of these
communities and telephone and city directories by which changes 1n
numbers and types of firms were analyzed.
The three counties chosen for case studies are Crowley County*
K1t Carson County* and Logan County. The primary central places 1n
these counties are the Incorporatd communities of Ordway* Burlington
and Sterling. They are located 1n different state planning regions
and are at the east central* southwestern and northern boundaries of
the study area. In almost all of the characteristics examined thus
far 1n this analysis they have exhibited markedly different patterns

of change and between them they demonstrate the significance of these
factors to Individual communities.
Crowlev Countv
Like much of rural America# Crowley County has experienced steady
population decline dating back to 1930 when the dlcennlal census
recorded 5,934 persons living 1n the County. By 1963 the County's
population had dropped to 3668 and this decline continued throughout
the study period to a 1982 population of about 3000. The largest
town 1n the County 1s Ordway which experienced a similar population
decline from 1,254 residents 1n 1960 to a population of 1,135 1n
1980. The 19 percent decline 1n population for the county as a whole
1s the largest percentage decline for any county 1n the study area.
The Town of Ordway, county seat of Crowley County, was
Incorporated 1n 1900, founded on land claimed by George W. Ordway when
he came West at the end of the C1v1l War. Major settlement of the
area surrounding what would become the town of Ordway first occurred
during the 1880s. The early settlements were spurred by development
of Irrigation canals enabling the establishment of a sugar beet
processing Industry 1n the area. One of the other small towns 1n
Crowley County, Sugar City, was the site of an early sugar factory.
Other early agricultural products Included sheep, alfalfa, honey and
turkeys. In addition to the sugar factories two alfalfa mills were
built 1n the area.

Agriculture 1n Crowley County has changed 1n more recent times
with many of the farmers selling their water rights to the growing
Front Range cities of Colorado Springs and Aurora. As a result the
nature of agriculture 1n the area changed from Irrigated farms growing
cantalopes, onions and sugar beets to dryland farming and cattle
ranching. This latter develoment supports a feedlot Industry, the
county's largest employer.
In 1982 farming employed 504 of 1114 persons with Jobs 1n Crowley
County. That was followed by government with 193 Jobs, services with
119, and retail trade with 114 employees. Agriculture and central
place functions have been the dominant forces 1n the local economy.
That will change to some extent soon when a new state prison opening
1n Ordway will become the major employer with over 200 jobs. Still,
agriculture accounted for 68 percent of the earnings 1n the county 1n
In 1963 there were sixteen retail establishments 1n Ordway: two
service stations, two car dealers, a hardware store, three grocery
stores, a fruit market, three furniture and appliance stores, a liquor
store and three cafes. By 1982 the number of stores dropped to twelve
with the mix remaining much the same. The fruit market, two grocery
stores, a car dealer and two of the furniture and appliance stores
were gone. Additions Included a clothing store, a bakery and a fabric
shop. A Dairy King had replaced one of the cafes. Based on retail
establishments alone the town seems to have a slowly declining retail
economy with reductions in both shopping and convenience goods. Given
the major declines in retail sales for the county, retail trade 1n

smaller towns 1n the county has probably been even more adversely
af fected.
Farm service establishments were not a major component of the
Ordway economy 1n 1963 and the situation has remained unchanged.
There was one farm machinery dealer and one full time farm supply
dealer In 1963 and there were two farm supply dealers In 1982. It
appears that In recent times Ordway has not been a focal point for
regional farm service activity.
Despite the decline 1n population, Crowley County has experienced
the most dramatic Increase 1n per capita personal Income 1n the study
area, from Just $5,365 (second lowest 1n the study area) 1n 1963 to
$11,732 per capita (fourth highest 1n the study area) 1n 1982. The
rise 1n Income, however, has not translated Into retail sales.
Constant dollar retail sales registered the second largest decrease
for any county In the study area, down 47 percent, and the net percent
capture of potential retail purchases by county residents remained
second worst throughout the study period dropping from a poor 69
percent 1n 1963 to a dismal 22 percent 1n 1982.
Crowley County 1s also remarkable for having recorded the largest
percentage decline 1n number of farms of any county 1n the study
area. Of the 362 farms 1n the county 1n 1964 only 218 remained 1n
1980, a decline of 40 percent. At the same time the average farm size
Increased from 1,127 acres to 2,066 acres, an Increase of 83 percent,
by far the largest Increase 1n the study area. The amount of farmland
1n the county actually Increased by ten percent from almost 408,000
acres 1n 1964 to over 450,000 acres 1n 1982. At the same time the
amount of harvested cropland 1n the county Increased a relatively