number 12, april 2006 tions food availability a ul &...
TRANSCRIPT
This special food assistance policy series has been funded in part by the USDA Economic Research Service and the Farm Foundation,
in partnership with the Southern Rural Development Center.Fo
od
Ass
ist
An
ce n
ee
ds
oF t
he s
ou
th
’s V
uln
er
Ab
le P
oP
ul
At
ion
s
Troy Blanchard, Mississippi State University, and Thomas Lyson, Cornell University
Food Availability & Food Deserts
in the Nonmetropolitan South
Over the past thirty years, the structure of food retailing in the United States has changed dramatically. Local grocery stores that once served a small community or neighborhood are increasingly being replaced by regional or national chain grocers. In addition, big box general merchan-disers have also entered the retail grocery sector with the advent of hybrid superstores that combine groceries with a wide array of product lines. A key consequence of this restructuring is the growing uneven distribution of food retailers across rural America. For example, Kaufman reports that rural counties in the Lower Mississippi Delta average one supermarket per 190.5 square miles. Additionally, over 70 percent of the low income populations in this region must travel 30 or more miles to access the lower food prices offered by a supermarket or large grocery store. The remaining options included small grocers or convenience stores where consumers are likely to pay substan-tially higher prices for a smaller variety of lower quality foods [7].
Todate,researchershaveyettounderstandtheimplicationsofruralretailrestructuringonfoodaccessamongruralpopulations.Priorresearchonthede-terminantsoffoodintakeanddietaryqualityhavefocusedalmostexclusivelyontherelationshipbetweenindividuallevelcharacteristicssuchasfamilystructure,race,ageandfoodsecurity[1,2,9,10,11,12].Althoughthesestudiesinformourunderstandingoftheindividuallevelprocessesthatinfluencefoodsecurity,priorstudieshaveoverlookedtheroleofcommercialfooddistributionintheUnitedStatesandthemannerinwhichthestructureofretailfooddistributionconditionstheaccessibilityandavailabilityoffoodwithinlocalcommunities. StudiesundertakenintheUnitedKingdomhavedescribedareaswithlimitedaccesstofoodas“fooddeserts”[6].IntheUnitedKingdom,someeconomicallydisadvantagedneighborhoodslostallgrocerystoresandmarkets,creatingafoodinsecurepopulation.Althoughresearchershavedocumentedruralpopulationswithrestrictedaccesstolowcost,highqualityfood,theconceptoffooddesertshaseludedpolicymakersandresearchersintheUnitedStates.Onepossibleexplanationfor
Number 12, April 2006
thisoversightisthatunliketheUnitedKingdom,theprolifera-tionofconveniencestoresandgasstationsensurethatsometypeoffoodisaccessibletoalmosteveryone.However,thequalityandpriceoffoodproductsvariesdramaticallybythetypesoffoodretailers.Con-sumerswhoareforcedtopurchasefoodatsmallgroceryorconve-niencestoresoftenpayapremiumforfoodproductsthatmayormaynotcontributetohealthydiet. Inthisreportwedocumentthepresenceoffooddesertsinthenon-metropolitanSouth.Ouranalysisaddressestwoissuesregardingfooddeserts.First,weexaminethedistri-butionoffooddesertcountiesacrossthenonmetropolitanSouth.Second,weexaminethefoodretailenviron-mentinfooddesertcountiestounderstandthenatureandrangeofproductsavailabletolocalresidents.
How Are Food Deserts Distributed in the Non-metropolitan South? Toourknowledge,thisisthefirstattempttoidentifyU.S.fooddeserts.Toaddressthisissue,wecreatedafooddesertclassificationbasedonthepercentageofacounty’spopulationwith“convenienceofaccess”tolargefoodretailers.Largefoodretailersincludesupermarketswith50ormoreemployees,super-centers(hybridretailersofferinggeneralmerchandiseandgroceries),andwholesaleclubs.Usingdatafromthe2000CensusofPopulationandHousingandthe1999ZipCodeBusinessPatterns,wedeterminedthelevelofaccesstoalargefoodretailer.Residentswithconvenientaccessaredefinedaspersonswhoresidenomorethan10milesfromalargefoodretailer.Personstravel-
ingmorethan10milesareclassifiedashavinglowaccesstoasupermar-ket,supercenterorwholesaleclub.Usingthiscriterionforidentifyingresidentswithlowaccesstosuper-markets,supercentersandwholesaleclubs,wetabulatedthepercentageofthepopulationwithlowaccessforeachcountyanddesignatedcountiesasfooddesertornon-fooddesertcounties(seeFigure1foramapofthesepercentages).Acountyisclas-sifiedasafooddesertif50percentormoreofthepopulationexperi-enceslowaccesstoasupermarket,supercenterorwholesaleclub[3,4]. Accordingtoourclassificationscheme,256ofthe873nonmetroSouthcountiesarefooddeserts.AmapofthesecountiesispresentedinFigure2.Amongthesouthernstates,Texas,Alabama,ArkansasandOklahomahavethehighestpercent-ageofnonmetrocountiesthatare
Page �
Figure 1. Percent of County Population with Low Food Access in the Nonmetropolitan South
Nonmetropolitan Counties, 2003 0.00% 0.01-12.9% 13.0-70.9% 71.0-100%
Page �
classifiedasfooddeserts.Clearly,thelargestfooddesertregioninthenon-metroSouthislocatedinthewesternportionsofTexasandOklahoma.Smallerclustersoffooddesertcoun-tiesarefoundalongtheMississippiDelta,theAppalachianregioninKentuckyandWestVirginia,andthebandofcountiesoftenreferredtoasthe“BlackBelt”thatstretchesfromsouthwesternLouisianathroughthecentralportionsofMissis-sippi,AlabamaandGeorgia,totheeasternshoreofNorthCarolina. Thepresenceoffooddesertcountiesinthesethreeregions—theDelta,theBlackBeltandthenorth-ernportionsofAppalachia—areespeciallyimportantbecauseoftheirhighratesofpoverty.Theaveragepovertyratein1999fornonmetrocountiesinKentuckyandWestVirginiawasapproximately21percent[13].SimilarhighlevelsofpovertyarefoundintheMissis-sippiDeltaandBlackBeltstates.Theaveragepovertyratein1999forLouisianaandMississippiwasnearly25percent[13].Forresidentsoftheseregions,especiallythepoor,thelackofaccesstosupermarketsandsupercenterspresentsahealthriskbecauseofthescarcityoflowcost,highqualityfoodretailers.What Types of Food Retail-ers Exist in Food Deserts? InFigures3–5wepresentmapsthatidentifythetypesoffoodretail-ersfoundinfooddesertcounties.FooddesertcountiesthatcontainasupercenterorwholesaleclubareidentifiedinFigure3.Onlyoneofthe256fooddesertcoun-tiescontainasupercenterstore,
suchasaWal-MartSupercenterorSuperK-Mart,orawholesaleclub,suchasSam’s.Thus,virtuallyallfooddesertpopulationslackac-cesstoasupercenterorwholesaleclubintheircountyofresidence. InFigure4,weexaminethepresenceoflargesupermarkets
(supermarketswith50ormoreemployees)infooddesertcoun-ties.Thismapindicatesonlyasmallfractionoffooddesertcounties(10of256)containasupermarket.Thedistributionofsupermarketsdiffersfromthatofsupercentersandwholesaleclubsinonerespect:
Figure 2. Food Desert Counties in the Nonmetropolitan South
Figure 3. Presence of Supercenter/Wholesale Club in Food Desert Counties
Nonmetropolitan Counties, 2003 FoodDesertCounty Non-FoodDesertCounty
Supercenter/Wholesale Club County Yes No
Page �
somefooddesertcountiescontainmultiplesupermarkets.Asecondfindingregardingsupermarkets,supercentersandwholesaleclubsisthatallcountiescontainingthesestoresareadjacenttoametroarea. Giventhelackoflargeretailoutletsinfooddesertcounties,wealsoexploredthepossibilitythatthesecountiesmaycontainalter-nativesourcesofnutritiousfoods,suchasfreshfruitandvegetablemarkets.Inotherwords,thelackoflargeretailersmaybeoffsetbyproducemarkets,whichprovidehealthyfoods.InFigure5,wemaptheprevalenceoffruitandvegetablemarketsforfooddesertcounties.Twoimportantfindingsemergefromthismap.First,only12ofthe256nonmetroSouthfooddesertcountiescontainafruitandvegeta-blemarket.Second,muchlikeothertypesoffoodretailing,fruitandvegetablemarketsaremostprevalentincountiesadjacenttoametroarea. Tocompleteouranalysisoffood
retailersinthenonmetroSouth,weexaminedtheavailabilityofsmallgrocers,conveniencestores,gassta-tionconveniencestores,fastfoodrestaurants,andfullservicerestau-rantsinfooddesertandnon-fooddesertcounties(seeTable1)[a].Shoppingatsmallgrocers,conve-niencestores,andgasstationstoresisnotanoptimalsolutionforconsum-ersbecauseofthehigherfoodprices
andsmallerselectionofproducts.Inaddition,theavailablefoodsmaybeofalowerqualityornotmeettherequirementsofahealthydiet.Allcountiescontainedatleastonesmallgrocerorconveniencestore.Furthermore,fooddesertcoun-tiescontainednearlytwiceasmanysmallgrocersasthenon-fooddesertcounties.Weinterpretthisfindingtomeanthatsmallgrocersarelessviableincountieswithlargefoodretailers.Forfooddesertresidents,smallgrocerystoresmaybetheonlyoptionforobtainingfood.Bothfooddesertandnon-fooddesertcountiescontainasimilarnum-berofconvenienceandgasstationstores.Incontrast,thepresenceoffastfoodrestaurantsvariesdra-maticallybetweenthetwotypesofcounties.Non-fooddesertcoun-tiescontainasubstantiallylargernumberoffastfoodrestaurantsthanfooddesertcounties.However,fooddesertandnon-fooddesertcountiescontainasimilarnumberoffullservicerestaurants.Thesefindingssuggestthatsmallfood
Figure 4. Presence of Supermarket in Food Desert Counties
Figure 5. Presence of Fruit and Vegetable Market in Food Desert Counties
Fruit and Vegetable Market County Yes No
Supermarket County Yes No
Page �
retailers,especiallysmallgrocers,fillthegapsleftbylargerretailers.
Conclusions ThisanalysiswasdesignedtodevelopandmeasuretheemergingconceptsofretailfoodaccessandfooddesertsinnonmetrocountiesintheSouth.Overthepast20years,theUnitedStateshaswitnessedatransitionfromsmallerscaleretaildistributionofgroceriestolargerscalesupermarketsandsupercenters.Thistrendhassignaledadeclinein"momandpop"storesandthein-creasingpenetrationoflargenationalandmultinationalcorporationsintoruralcommunities.Althoughthisprocessinvolvesmanyactors,rang-ingfromwholesalers,processors,retailersandconsumers,wefocusedonthespatialrelationshipsbetweentypesofretailfoodoutletsandcon-sumers.Ourfindingsshedlightontheimportanceoffooddesertsintheexplanatoryframeworkemployed
infoodassistanceresearch.Tobesure,allcounties,eventhoseclassi-fiedasfooddeserts,containedsomeretailsourceoffood.However,thequalityofthefoodretailersacrossthesecountiesisnotconsistent. Aprimaryfindingfromourstudyisthatpopulationsinasub-stantialnumberofcountiesinthenonmetroSouthexperiencelimitedaccesstosupermarkets,supercent-ersandwholesaleclubs.Thereareanumberofimplicationsthatspeaktotheissueoffoodsecurityandfooddesertification.First,individualsliv-inginareaswithlowaccesstolargefoodretailersarelikelytopayhigherpricesforgroceriesand/orincuragreatertravelcosttoaccessthelargefoodretailer.Thetravelcostmayoffsetthesavingsavailableatthesestores.Thisisespeciallytroublingforeconomicallyvulnerableseg-mentsofthepopulationintheimpoverishedregionsoftheSouth,suchastheBlackBelt,theMisssissip-
piDeltaandAppalachia.Forindi-vidualslivingintheseregionsitmaynotbefeasibleorpracticaltoshopatalargefoodretailerbecauseoftravelcostandtimeconsiderations.With-outaccesstothelargefoodretailer,theseindividualsarelefttoshopatconveniencestores,gasstationsandsmall"momandpop”grocerystores. Fooddesertsalsohaveconse-quencesforthenutritionalhealthofnonmetropopulationsintheSouth.Inanexploratoryanalysisnotpresentedhere,weappliedourcon-ceptoffooddesertstoexplainfruitandvegetableintakeamong3,322nonmetroMississippiresidents.Ourfindingsshowedthatpersonsresid-inginfooddesertcountieswere23.4percentlesslikelytoconsumetherecommendedfiveormoreservingsoffruitsandvegetablesperdaythanthoseinnon-fooddesertcoun-ties[b].Giventheimportantroleofdietinreducingtheprevalenceofdiabetes,heartdisease,stroke
Table 1. Other Food Stores and Restaurants by Type of County
Food Desert County Non-Food Desert County
Page �
andcertaintypesofcancer,fooddesertspresentaclearthreattothehealthofnonmetropopulations. GiventheprevalenceoffooddesertsinthenonmetroSouthandthehealthriskassociatedwithliv-inginafooddesert,policiesdirectedatalleviatingfoodinsecurityintheUnitedStatesshouldconsidertheroleofthelocalcommunitycontext.Forexample,personsreceivingfoodstampsstandlittlechancetoachievefoodsecuritywithoutconvenientaccesstofoodretailerstoredeembenefits.Inaddition,theFoodStampProgramlimitstotalassetsheldbyprogramparticipantstonomorethan$6550,limitingtheabilityofafamilytosimultaneouslyownreliabletransportationandreceivefoodstampbenefits.Beyondthebasicissueofac-cesstofoodretailers,foodstamppar-ticipantsinfooddesertareaswouldhaveasmallerselectionoffoods.Thepoorerqualityoffoodsatsmallergrocersplaceprogramparticipantsatadisadvantagerelativetopartici-pantsinnon-fooddesertcounties. Oneapproachtoremedyingthefooddesertprobleminvolvesspatiallymatchingfoodretailersandconsum-ers.Becausefooddesertcountiescon-tainalargenumberofunderservedfoodconsumers,ruraleconomicdevelopmenteffortsdirectedatthecreationofproducemarketsandotheralternativefoodretailerscouldyieldbeneficialresultsforthedietaryhealthofthesepopulations.Forconsumersinfooddeserts,acentralconcernistransportationtoandfromthenear-estlargeretailsupermarket,super-centerorwholesaleclub.Policiesdirectedatcreatingshuttlesoperatedbylocalcommunityorganizationsor
carpoolingcanassistdisadvantagedresidentsinaccessinglargeretailers.
Endnotes[a]ThefindingspresentedinTable
1arereportedinthenumberofstoresper10,000persons.Wechosethismeasureratherthantherawnumberofstoresbecausetheaveragepopulationsizeoffooddesertandnon-fooddesertcountiesdifferedsubstantially.Becausethescaleofretailactivityisstronglyrelatedtothepopula-tionsize,comparingthenumberofstoresbetweenfooddesertandnon-fooddesertcountiescouldyieldmisleadingresults.
[b]Weobtainedthesefindingsaftercontrollingfordifferencesinfruitandvegetableintakeattributabletoage,sex,race,andeducation.
References[1]Alaimo,K.,R.RBreifel,E.A.
Frongillo,andC.M.Olson.1998.“FoodInsufficiencyExistsintheUnitedStates:ResultsfromtheThirdNationalHealthandNutritionExaminationSurvey.”American Journal of Public Health88:419-426.
[2]Bickel,G.,S.Carlson,M.Nord.1998.MeasuringFoodInse-curityintheUnitedStates:HouseholdFoodSecurityintheUnitedStates1995-1998.USDA,ERS,Washington,D.C.
[3]Blanchard,TroyandThomasLyson.2002.“AccesstoLowCostGroceriesinNonmetro-politanCounties:LargeRetailersandtheCreationofFoodDes-erts.”MeasuringRuralDiversityConferenceProceedings,No-
vember21-22,2002,EconomicResearchService,WashingtonD.C.http://srdc.msstate.edu/measuring/blanchard.pdf.
[4]______.2003.Retail Concen-tration, Food Deserts, and Food Disadvantaged Communities in Rural America.FinalReportforSouthernRuralDevelop-mentCenter-EconomicRe-searchServiceFoodAssistanceGrantProgram.http://srdc.msstate.edu/focusareas/health/fa/blanchard02_final.pdf.
[5]FoodandNutritionService.2000. Strategic Plan 2000-2005.U.S.DepartmentofAg-riculture,WashingtonD.C.
[6]Furey,Sinead,ChristopherStrugnell,andHeatherMcIl-veen.2001.“AnInvestigationofthePotentialExistenceof‘FoodDeserts’inRuralandUrbanAreasofNorthernIreland.”Agriculture and Hu-man Values18(4):447-457.
[7]Kaufman,PhillipR.1998.“RuralPoorHaveLessAccesstoSupermarkets,LargeGro-ceryStores.”Rural Develop-ment Perspectives 13(3):19-26.
[8]Molnar,Joseph.2001.“PrivateFoodAssistanceinaSmallMet-ropolitanArea:UrbanResourcesandRuralNeeds.”Journal of Soci-ology and Social Welfare 187-209.
[9]Nord,Mark.2001.“FoodStampParticipationandFoodSecu-rity.”Food Review 24(1):13-19.
[10]Nord,Mark.2002.“RatesofFoodInsecurityandHungerUn-changedinRuralHouseholds.”Rural America 16(4):42-47.
[11]Nord,Mark,NaderKab-bani,LauraTiehen,Margaret
Page �
Andrews,GaryBickel,andStevenCarlson.2002.House-hold Food Security in the United States, 2000.USDA,ERS,Washington,D.C.
[12]Olson,Christine,BarbaraS.Rauschenbach,EdwardFrongillo,andAnneKendall.1997.“FactorsContributingtoHouseholdFoodInsecurityinaRuralUpstateNewYorkCounty.”Family Economics and Nutrition Review10(2):1-22.
[13]U.S.CensusBureau.2002.Census2000SummaryFile3,UnitedState.DepartmentofCommerce,WashingtonD.C.
About the AuthorsTroy Blanchard is an Assistant Profes-sor in the Department of Sociology, Anthropology, and Social Work and a Research Fellow in the Social Science Research Center at Mississippi State University. Thomas Lyson is Liberty Hyde Bailey Professor of Develop-ment Sociology and Director of the Community, Food, and Agriculture Program at Cornell University.
A publicAtion of the
Southern rural Development CenterBox 9656
Mississippi State, MS 39762Phone: (662) 325-3207Fax: (662) 325-8915
http://srdc.msstate.edu
For more inFormation, ContaCt:Lionel J. “Bo” Beaulieu, Director
Emily Elliott Shaw, Program [email protected]
NoNprofit org.U.S. poStage
paiDpermit No. 39
miSSiSSippi State, mS
The Southern Rural Development Center does not discriminate on the basis of race, color, religion, national origin, sex, age, disability or veteran status.
p.o. Box 9656miSSiSSippi State, mS 39762