The Social Science Journal 49 (2012) 108–113

Contents lists available at ScienceDirect

The Social Science Journal

j our na l ho me p age: www.elsev ier .com/ locate / sosc i j

Stemming retail leakage with a sense of community: Leveraging thelinks between communal ties and shopping decisions

Katherine Chalmersa,∗, Mary Gessnerb, Linda Venturonib, Stephan Weilerc

a Department of Economics, California State University at Sacramento, 6000 J Street, Sacramento, CA 95819-6082, USAb Venturoni Surveys and Research, Inc., Dillon, CO 80435, USAc College of Liberal Arts, Department of Economics, Colorado State University, Fort Collins, CO 80523, USA

a r t i c l e i n f o

Article history:Received 4 August 2010Received in revised form 14 June 2011Accepted 28 June 2011Available online 9 November 2011

a b s t r a c t

Retail business development is a broad goal for both private business interests as well aslocal policymakers, yet the goal of retail opportunities for local residents themselves isoften seen as secondary. This paper considers the argument that retail opportunities andsense of community are in fact linked in important ways, links that reinforce the socialfabric of a community and/or neighborhood. The paper first briefly reviews the inherentlinkages between retail shopping and local development patterns, and then considers the

sense of community in the context of Garfield County in western Colorado. Based on the keyquestions derived from this background, we formally test the inter-relationship betweenlocal retail spending and sense of community from detailed survey data, then more broadlyconsider the factors that critically shape a locality’s “sense of community.” These findingsshape several important policy implications.

ocial S

© 2011 Western S

1. Perspective

Despite often widely varying geographic, policy, andresource backgrounds, both urban and rural communi-ties share a range of fundamental priorities. The broadgoal of economic prosperity is common to the smallestof towns to the largest of cities, as well as individualneighborhoods within each of these areas. The successfulsustainable development of private business is the com-mon means to these ends, across a range of sectors. Whileoften derided as a secondary, low-wage industry on thesurface, retail development in particular is often key toattaining a community’s fiscal goals as a primary conduitfor sales taxes as well as a major indicator of local quality-

of-life through the convenient and proximate availability offood, clothing and other vital necessities. A variety of retailshopping options in a given community is further seen by

∗ Corresponding author. Tel.: +1 916 278 7080; fax: +1 916 278 5768.E-mail address: chalmers@csus.edu (K. Chalmers).

0362-3319/$ – see front matter © 2011 Western Social Science Association. Publidoi:10.1016/j.soscij.2011.06.010

cience Association. Published by Elsevier Inc. All rights reserved.

many residents and visitors as simply an attractive featurein itself, with world-class cities boasting well-known corri-dors of niche boutiques, delicatessens, and complementaryservice outlets.

Local government is thus often faced with the ques-tion of whether it can help foster such desirable formsof retail development. Individual businesses often seekparticular favorable property tax terms, other forms ofdirect subsidy, or cost-reducing consumer-friendly ameni-ties such as free proximate parking. Others underscorethe complementary provision of public goods, such asparks, museums, regulations to harmonize architectureand building standards, and community policing. Similarly,business owners and community leaders often encouragean outward-orientation towards visitors and tourists, withretail effectively becoming a type of export industry evenin smaller towns/communities.

Yet providing local shopping opportunities for local res-idents may in itself be a worthwhile rationale, from bothfiscal and competitive perspectives. Having residents shoplocally avoids the leaking of valuable income-generating

shed by Elsevier Inc. All rights reserved.

l Science

serdail

maRbwIfbiiCrcac

lbnwfammgtts

rsTfinfctb

2

2

siMpmnwae

K. Chalmers et al. / The Socia

pending, thus raising the multiplier effect of any givenxternal stimulus to the economy. Furthermore, localesidents make up the most devoted, consistent, and pre-ictable market for retail establishments. These featuresre particularly noteworthy, and are in fact reinforced, dur-ng periods of increased transport costs, such as the recenteap in gasoline prices.

These characteristics hint at the virtuous cycle that canake retail development for residents a self-sustaining

nd valuable component of economic development itself.esidents’ ties to a place and its people are strengthenedy local shopping habits, commitments and relationshipshich in turn reinforce residents’ ties to a community.

n that sense, a sense of community itself can developrom local retail options, laying the foundation for furthereneficial relationships which themselves heighten a bind-

ng sense of community. While such a virtuous cycle mayndeed reflect a form of a social capital investment (e.g.,oleman, 1988; Putnam, 2000), our focus here is on the nar-ower, arguably more quantifiable, community psychologyoncept of a locality’s sense of community which provides

set of eventual foundations and incentives for such socialapital investments (Pooley, Cohen, & Pike, 2005).

This paper evaluates the relevance of this link betweenocal shopping patterns and sense of community. Given thisackground, we explore the concept of “sense of commu-ity” in greater detail in the context of Garfield County inestern Colorado, with an eye towards those quantifiable

actors that indicate such communal ties. The analysis takesdvantage of a particularly detailed survey of rural com-unities’ assessment of their local quality-of-life in theountain region of Colorado. Rural areas provide a natural

eography for testing shopping patterns and communityies, given their relatively well-defined and atomistic spa-ial structure. These results can thus also help illuminatehopping decisions in more complex urban environments.

The following section provides a brief background onural shopping patterns, along with additional detail for ourense of community metric for Garfield County, Colorado.he third section then explicitly tests two hypotheses byrst assessing the relationship between sense of commu-ity and local shopping patterns before considering what

actors determine a town/neighborhood’s own “sense ofommunity.” A fourth section outlines the implications ofhe findings for both private and public decision-makersased on these Garfield County findings.

. Shopping and sense of community

.1. Rural retail shopping and local development

Retail development strategies traditionally focus on theector’s potential role as a source of regional export earn-ngs through visitors and tourists (Thilmany, McKenney,

ushinski, & Weiler, 2005). However, local customersrovide a much more proximate, stable, and consistentarket. As noted above, outflows of local income to exter-

al retail businesses create a distinct form of retail leakage,hich not only drains potential economic vitality from

community but also effectively reduces the multiplierffect of other unrelated stimuli to the local economy.

Journal 49 (2012) 108–113 109

Studies show that such retail leakage is in fact an impor-tant characteristic in both urban neighborhoods and ruralregions, constraining potential growth opportunities (e.g.,Mushinski & Weiler, 2002).

Such leakages occur not only in lower-margin, high-turnover items, but also higher-margin goods and servicesthat can help reinforce each other. Multipurpose shoppingtrips in fact have become the norm in an era of higher trans-port costs, alongside a constantly rising opportunity costof time through higher consumer wages and a range ofalternative recreational activities. The evolutions of shop-ping patterns and good/service complementarities tend toheighten the stakes involved in retail capture, with certainareas/communities having first-mover advantages in cre-ating retail/service corridors that can entrench shoppinghabits to the detriment of those areas that cannot providethese opportunities.

How can communities re/capture the loyalties of localcustomers? The key to answering this question is under-standing the factors determining shopping behavior. Inmost settings, the closest available opportunities that meetconsumers’ needs tend to be preferred, ceteris paribus.Such proximity is again reinforced by recent increases intransportation costs. Such factors are similar in both ruraland urban settings, with rural areas experiencing evengreater sensitivity to transport costs given the longer traveldistances involved. Yet urban shopping can be similarlyfraught, due to greater amounts of congestion (particularlyat peak traffic/shopping periods) and the higher opportu-nity costs of time in high-wage/high-cost urban settings.Anecdotal evidence indeed suggests that the bonds createdby regular shopping in local establishments provides con-sumers with a more secure sense of place in the spirit ofBolton’s seminal work (1992).

In that sense, local customers are potentially the mostloyal. This geographic loyalty has an additional featurethat makes local shopping capture even more attractiveto businesses. First, local shopping opportunities provide asmall but significant amount of pricing power, as local cus-tomers face a transport-cost-wedge between local storesand more distant options. This pricing power could allowsmaller shops to compete with higher volume stores thatmight otherwise have prohibitive price advantages. Yet thislocal market area and its greater grip on local customersalso can benefit these shoppers in more subtle ways. Clearrecognition of a local clientele can allow local stores to bet-ter target the needs/desires of those consumers, orientingproduct and service lines more precisely to these shoppers.Such orientation benefits both customers and store own-ers, allowing simultaneous maximization of both consumersurplus, through specific targeting of product variety, andbusiness profits, through the leveraging of geographic mar-ket power. In contrast to the usual pernicious effects ofmarket power, this local framework actually reduces riskand raises both utility and profits by providing both busi-nesses and customers with a stable, focused market.

2.2. Sense of community within the survey

This keenly felt sense of customer loyalty can haveadditional nonpecuniary effects as well. First described by

110 K. Chalmers et al. / The Social Science Journal 49 (2012) 108–113

Table 12008 Garfield County community survey.

elivered a

070

dents’ sense of security and sense of community go handin hand. The Traffic variable also falls into this group withthe same direction of relationship.

Table 2Comparison of our survey and SoC components.

Issue addressed insurvey question

Coordinating SoC component

Q96—Public safety Integration and fulfillment of needsQ83—Scenic/visual quality Shared emotional connectionQ13—Sense of community Shared emotional connectionQ6—Why live in Garfield Membership

Samples Sample frame Mailed D

Voter registration 22,622 4,275 4,

a 95% confidence level.

McMillan and Chavis (1986), sense of community is a con-cept that attempts to capture both a territorial as well as arelational cohesiveness. Through such variables as mem-bership, influence, integration and fulfillment of needs,and shared emotional connection, community residentsencounter a feeling of belonging and commitment.

With this concept in mind, the residents of GarfieldCounty, Colorado were asked to participate in a communitysurvey to assess the strength of their sense of communityas defined by McMillan and Chavis (1986). In 2008, GarfieldCounty contracted with Venturoni Surveys & Research, Inc.(VSR) to perform the survey analysis. The Garfield Countyvoter registration list was used to sample local residents.

It should be noted that voter registration by itself couldbe an indication of the survey respondent’s attachmentto a particular location. The state of Colorado considersemployment, ownership of real property, continued res-idence even during a period of nonenrollment in school,the acceptance of future employment within the state, andvehicle registration, as evidence that an individual is alegal resident in Colorado, in addition to voter registration.1

Therefore, while voter registration is not the only require-ment for legal residency in Colorado, the fact that allsurvey respondents are also registered voters in GarfieldCounty cannot be ignored. Voter registration is not the onlyrequirement for legal residency in Colorado, but neverthe-less signals an incremental sense of locational attachmentby the citizen in question.

While using voter registration lists for sampling to testa population’s Sense of Community may therefore intro-duce some limited bias to the sample itself, voter rollsare nevertheless one of the most reliable representativeand comprehensive lists of involved residents. Their votingcommitment to the locality may in fact make such surveysespecially compelling to local policymakers, who recognizethat their own political fortunes will depend on respond-ing to the sentiments of this particular audience. The choiceof voter registration lists in this case was simply the bestof several imperfect options for such microlevel sampling,one where at least the noted advantages and disadvantagesare relatively clear and well-known.

Random sampling techniques were employed to selectthe samples. Letters were sent to potential respondentsdirecting them to go to the internet to fill out the survey.All respondents were assigned identification numbers toassure that no duplicate responses were counted. After twoweeks, if the potential respondent had not filled out theinternet survey, they were sent a reminder letter, paper

copy of the survey and postage paid return envelope.

Table 1 details the size of the original sample frame,the number of surveys that were mailed, the number of

1 Colorado Department of Higher Education (CDHE, 2011),http://highered.colorado.gov/Finance/Residency/statutes.html.

Completed surveys % Margin of error

1,048 25.7 2.96%

surveys delivered (surveys not delivered were returnedby the post office as “undeliverable”), the number of sur-veys completed and the number of surveys returned, thesurvey response rate, and resulting margin of error of2.96%.

Residents were asked to assess the importance of 21issues using a 5 point Likert scale and then asked toassess how successfully their own community was inaddressing these issues. The issues included in the sur-vey included: Economic Development, Affordable housing,Open space, Animal control, Sense of Community, Publicsafety, Air quality, River access, County road maintenance,Public transportation, Cultural integration, Water qual-ity/quantity, Scenic/visual quality, Shopping opportunities,Historic preservation, Public land access, Biking trails,Recreational opportunities, Traffic, Hiking trails, and Recy-cling services.

Table 2 illustrates how specific issues identified in thesurvey questions directly correlate to specific componentsof McMillan and Chavis’s sense of community concept.

Sense of community encompasses a variety of expe-riences and the survey attempts to capture respondents’attachment to Garfield County within a variety of per-spectives. For example the membership component canbe seen in questions regarding river access and residents’friendliness. River access is an example of a freely availablelocal recreational amenity. It provides experiences that areshared by several groups – fishermen, boaters, picnickers –and consequently creates a common experience, regardlessof a specific activity. Likewise, survey questions regardingrecycling services and growth control indicate that respon-dents are both forward-looking and altruistic. In this sense,they feel they have input in shaping the community. Inte-gration and fulfillment of needs is shown by respondents’feelings regarding the issue of Public Safety in that respon-

County? A15 friendlinessQ88—Traffic Integration and fulfillment of needsQ16—River access MembershipQ98—Historic preservation Shared emotional connectionQ87—Recycling services InfluenceQ50—Growth control A3

about the same rate ofgrowth as at the present

Influence

l Science

odqtwsb

2

obtldtCpc

dpfaGtatssau

3

nWioni

bwfahta

h

sarft

K. Chalmers et al. / The Socia

Finally, the ways in which residents bond with eachther indicate a shared emotional connection. Respon-ents indicated how important such issues as scenic/visualuality, valued sense of community, and historic preserva-ion questions were as well as how well their communityas addressing these issues. Clearly, attempts to preserve

hared history and maintain shared values increase theonds of connections.

.3. Garfield County—background and motivation

Garfield County, located in Western Colorado, stretchesver 3,000 square miles from Glenwood Canyon to the Utahorder and is known for year-round recreational oppor-unities and related services. Essential components of theocal economy include agriculture, mineral and energyevelopment and limited manufacturing and construc-ion activities. There are six municipalities within Garfieldounty; Rifle, the largest, has a population just under 9,000eople and the estimated 2008 population of the entireounty is only 58,120 people.

However, the Colorado State Demography office pre-icts Garfield County will eventually grow to an eventualopulation of 119,216 people by 2,030 posing real issuesor open space, water, economic stability and recreationalvailability.2 With an expected influx of new residents,arfield County needs to know what issues are important

o its residents in contributing to its sense of communitynd how it can build on its current strengths to addresshese coming challenges. An assessment of a region-wideense of community by evaluating intra- and inter-regionalhopping behavior with other sense of community char-cteristics may provide a uniquely valuable perspective tonderstand critical community ties (Chavis & Pretty, 1999).

. Analysis

The survey determines how strong the sense of commu-ity is while also quantifying the amount of retail leakage.e believe that a strong sense of community is signif-

cantly and negatively correlated with retail leakage. Inther words, the more strongly a local resident feels con-ected to his or her community, the less likely that person

s to shop outside his or her community.To delineate communities within Garfield County,

oundaries were drawn to better reflect each municipalityithin Garfield County to which residents in that area may

eel most connected. By considering topography, restrictedccess due to public land, the location of the municipality,

ousing clusters and the road networks in an area, we kepthe number of households within each municipality as highs possible to avoid nonviable results from our survey.3

2 Colorado State Demography Office (CSDO, 2011),ttp://dola.colorado.gov/demog/pop cnty forecasts.html.3 Prior to analysis, standard data preparation steps were taken. Each

urvey question was reviewed to understand the data type of the possiblenswers: binary, ordinal, continuous, write-in. Questions with write-inesponses were not incorporated into the analysis. Next, it was notedor each question how non-answers or missing responses were coded sohat a missing response could be standardized as such across all survey

Journal 49 (2012) 108–113 111

To statistically test the relationship between sense ofcommunity and retail leakage, the survey’s shopping ques-tions were transformed to create a binary indicator of arespondent’s inclination to shop locally. When answer-ing the shopping questions, the respondent indicated theregion in which he predominantly purchased goods by cat-egories such as groceries, gasoline, office supplies, etc. Thechoices were each of the 6 areas encompassing a munic-ipality within Garfield County, the closest urban area tothe west (Grand Junction), the closest urban area to theeast (Denver), by mail order/internet, or by some other,unnamed method. The respondent’s answer was recodedas a binary indicator of shopping locally. That is, the trans-formed answer was a ‘1’ if the respondent shopped for thecategory of merchandise in his own area; otherwise itsvalue was a ‘0’. Also, a transformation was done to indi-vidual categories of merchandise. They were grouped byfrequency of occurrence shown in Table 3.

The goal of creating the binary indicator for shoppinglocally and for grouping the categories of merchandisewas to obtain a single data field that could representRetail Leakage or those frequent retail purchases takingplace outside the local community within Garfield County.Those individuals who exclusively purchase Groceries,Gasoline/Diesel, Medical Prescriptions, Entertainment andRestaurants locally would receive a 4 while a resident whoexclusively purchases such goods outside Garfield Countywould receive a 0. In this way, we are able to measure theextent of retail leakage along a scale of 0–4.

We find that there is a significant, negative, linear rela-tionship between retail leakage and a sense of communityrating in our survey results. In other words, if a resident fre-quently purchases goods locally, he or she also has a higherSense of Community (SoC) rating. We used two statisticalmeasures to determine this relationship. We used the Pear-son correlation to measure a linear relationship betweenretail leakage and sense of community. Our Pearson corre-lation measure, r, is equal to 0.20 which is significant at the0.05 level. Similarly, the chi-squared statistic, a measureof a potentially nonlinear relationship between these vari-ables, is equal to 61.1 and is significant at the 0.05 level. Thisresult indicates the distribution of survey responses is sig-nificantly different from what is expected if a respondent’sSense of Community and Retail Leakage were independentand unrelated.4

4. Implications: emphasis on sense of community

Our statistical analysis gave us the empirical footholdwe initially sought, namely that retail leakage and senseof community were significantly negatively correlated.When retail activity leaks away from a region, the region’s

sense of community is systematically lowered, and like-wise, where this sense of community is hampered, moreretail leakage is also likely. These two characteristics clearly

questions. Lastly, the distributions of each question’s responses were tab-ulated, mainly as a reference for determining the questions/answers thatwere rare or did not have meaning across all respondents.

4 For a comparison of the survey responses and retail leakage, pleasesee Appendix A.

112 K. Chalmers et al. / The Social Science Journal 49 (2012) 108–113

Table 3Local retail shopping categories.

Shop for frequently Shop for periodically Shop for rarely

Groceries Clothing AppliancesGasoline/diesel Office supplies AutomobilesMedical prescriptions Hardware and building suppliesEntertainment and restaurants Gardening supplies

ecreatio

Sports and r

are linked, each endogenously shaping the other. Further-more, by identifying specific components that contributeto an individual’s perception of sense of community, wemay addresses methods by which local governments mayattempt to strengthen communal ties. This can be done bybuilding on the factors we identified to strengthen com-munity ties.

Typically, rural resort communities see orienting theirgoods and services towards outsiders as the way to growthe local economy through export earnings. Yet this narrowperspective ignores their most committed potential cus-tomer base, the local residents themselves. That locals oftenmust travel further to satisfy basic shopping needs leads tocritical leakages from the local economy. With relatively lit-tle effort, shops can recapture this lost demand, increasingthe local spending multiplier while also helping to foster astronger sense of community. This reorientation can lever-age local shop owners’ superior knowledge of local tastes,allowing simultaneous increases in both consumer happi-ness and potential business profits, while simultaneouslyreinforcing a locality’s sense of community. This mutualfulfillment of needs and bidirectional influence satisfiestwo components of sense of community—customers cankeep local businesses open while local businesses remainopen by stocking the goods and services their customersneed.

The town of Carbondale provides an excellent exam-ple of this phenomenon. Carbondale residents reportedthe highest sense of community in Garfield County. Localsrated survey questions illustrating the membership com-ponent of sense of community, namely “Friendliness” and“Recreational Amenities,” as important reasons for liv-ing in Garfield County. Interestingly, a broad array ofrecreational activities were also important, with AlpineSkiing, Hiking, Road Biking, Walking/Jogging, and Raft-ing/Kayaking/Boating, among the most popular. It should

be noted that several of these activities usually require spe-cialized equipment that makes it easy to spot others whoalso do these activities, thereby building a common symbolsystem whereby members can recognize other members.

#Merchandise categories purchased locally Likert scale

Observed sense of community

1 2 3 4

0 6 21 30 15

1 20 27 62 32

2 22 44 72 52

3 10 50 72 69

4 12 44 143 112

Total 70 186 379 280

nal equipment

This sense of belonging and identification helps estab-lish membership within a group and ties locals not onlyto each other but also to the natural amenities that sur-round them. In addition, survey respondents also feel morepositive about their sense of community, citing specifi-cally contentment with the rate of economic growth and ageneral feeling that the county is moving in the right direc-tion. This positive perception of Carbondale’s direction alsoillustrates the bidirectional nature of influence and sharedvalues. Residents feel that the values of the communitymatch their own and that their collective action is movingeveryone in the right direction.

Local governments may also play a crucial networkingrole by fostering membership within the community. Retailactivity’s social context engenders feelings of belonging bystrengthening residents’ social ties with one another. Thesesocial ties can further be strengthened if the local gov-ernment pursues programs that simultaneously promoteboundaries (making local residents perceive themselves asa distinct group from the outside world) while also invitingtourists to visit. In this way, the local government can focuson its natural comparative advantage, namely providingpublic goods (e.g., coordinating farmer’s markets, commu-nity events, cultural/historical links, etc.). These seeminglymutually exclusive goals can encourage a sense of com-munity. By allowing local residents to see their own rolein sustaining local, private businesses, local governmentcan simultaneously strengthen its residents’ economic andsocial ties to one another.

Appendix A.

Survey respondents’ reported sense of communityLikert measures are correlated with their observed retailleakage. We contrast these with the expected numbers thatwould be expected based upon the row and columns in theobserved section. Because these values are significantlydifferent, we argue that there is a pattern between anindividual’s decision to shop locally and their perceptionof sense of community.

Expected sense of community

5 Total 1 2 3 4 5 Total

2 74 5.1 13.6 27.8 20.5 7.0 748 149 10.3 27.4 55.9 41.3 14.0 149

12 202 14.0 37.2 75.8 56.0 19.0 20227 228 15.8 42.0 85.6 63.2 21.4 22846 357 24.7 65.7 134.0 99.0 33.6 357

95 1,010 70 186 379 280 95 1,010

l Science

R

B

C

C

C

C

community. New York: Simon & Schuster.Thilmany, D., McKenney, N., Mushinski, D. & Weiler, S. (2005). Beggar-thy-

neighbor economic development: A note on the effect of geographic

K. Chalmers et al. / The Socia

eferences

olton, R. (1992). ‘Place prosperity vs. people prosperity’ revisited: An oldissue with a new angle. Urban Studies, 29, 183–205.

havis, D. M. & Pretty, G. M. (1999). Sense of community: Advances inmeasurement and application. Journal of Community Psychology, 27(6),635–642.

oleman, J. (1988). Social capital in the creation of human capital. AmericanJournal of Sociology, 94, 95–120.

olorado Department of Higher Education. Colorado Residency Stat-

ues, http://highered.colorado.gov/Finance/Residency/statutes.html(accessed June 13, 2011).

olorado State Demography Office. Population Totals for ColoradoCounties, http://dola.colorado.gov/demog/pop cnty forecasts.html(accessed June 13, 2011).

Journal 49 (2012) 108–113 113

McMillan, D. W. & Chavis, D. M. (1986). Sense of community: A definitionand theory. Journal of Community Psychology, 14, 6–22.

Mushinski, D. & Weiler, S. (2002). A note on the geographic interdepen-dencies of retail market areas. Journal of Regional Science, 42(1), 75–86.

Pooley, J. A., Cohen, L. & Pike, L. (2005). Can sense of community informsocial capital? Social Science Journal, 42, 71–79.

Putnam, R. (2000). Bowling alone: The collapse and revival of American

interdependencies in rural retail markets. Annals of Regional Science,39(3), 593–605.