economic dependence, over-urbanization, and economic growth

8/12/2019 Economic Dependence, Over-Urbanization, And Economic Growth

1/20

Economic Dependence, Overurbanization, and Economic Growth: A Study of Less DevelopedCountriesAuthor(s): Michael Timberlake and Jeffrey KentorSource: The Sociological Quarterly, Vol. 24, No. 4 (Autumn, 1983), pp. 489-507Published by: Blackwell Publishingon behalf of the Midwest Sociological SocietyStable URL: http://www.jstor.org/stable/4106021.

Accessed: 01/09/2011 21:33

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at.http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

Blackwell PublishingandMidwest Sociological Societyare collaborating with JSTOR to digitize, preserve and

extend access to The Sociological Quarterly.

http://www.jstor.org
http://www.jstor.org/action/showPublisher?publisherCode=blackhttp://www.jstor.org/action/showPublisher?publisherCode=msshttp://www.jstor.org/stable/4106021?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/4106021?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=msshttp://www.jstor.org/action/showPublisher?publisherCode=black


2/20

The Sociological Quarterly24 (Autumn 1983):489-507EconomicDependence,Overurbanization,andEconomicGrowth:A Studyof Less DevelopedCountries*

MichaelTimberlake,MemphisStateUniversityJeffreyKentor, Johns Hopkins UniversityExcessively large and rapidly growing urban populations characterizesocial struc-tural change in less-developedcountries. This is in contrast to the developmentalpat-terns which accompanied structural transformation in the presently rich, industri-alized countries of the world. Many less-developed countries are thus said to beoverurbanized. n this paper, quantitativecross-national data are brought to bearon the hypothesis that overurbanizationhas been fueled by the economic depen-dence to which these countries are, to varying degrees, subject. The hypothesis thatthe degree of overurbanizationand changes therein inhibit economic growth is alsoexamined. But the overurbanizationquestion is often framed as a matter of the rela-tive distributionof the labor force into service and manufacturingoccupations. Weherefore examine parallel hypotheses which define overurbanization n terms of theurban labor distribution. Constraints on the availability of labor force data, how-ever, relegatethese findingsto a subordinaterole.

Panel regression analysis provides support for the proposition that dependenceupon foreign capital leads to overurbanization (defined as either the urban/devel-opment relationship or in terms of labor structure imbalance). Furthermore,rela-tive increases in overurbanizationare consistently accompanied by relative declinesin per capita economic growth, though the effects of higher levels of overurbaniza-tion do not appearto impedeeconomic growth.Excessively argeand rapidlygrowingurbanpopulationsandexcessiveemploy-ment in serviceoccupations relativeto manufacturingccupations)are amongthe tendenciesmost often said to peculiarlycharacterizeocial structural hangein less-developed ountries.Thesepatternscontrastwith the developmental at-ternswhichaccompanied conomic ransition n the presently ich,industrializedcountriesof the world.Both of these conditionsare frequentlyregardedas di-mensionsof overurbanization. anyhave argued hatthis structural onditionimpedeseconomicgrowthin less-developedcountries(LDCs). More recently,othershaveargued hatthis imbalance s likelyto accompanyperipheral egions'economic dependenceon foreign capitol from the core regions of the worldeconomy.?1983 by The Sociological Quarterly. All rights reserved. 0038-0253/83/1500-0489$00.75

*This research was supported in part by a grant from the National Institute of Mental Health, ADAMHA,HHS, PHS 5 T32 MH1487-06. An earlier version of the paper was presented at the annual meeting of theAmerican Sociological Association, San Francisco, 1982. The authors are grateful for the helpful com-ments of Lynn Weber Cannon, Christopher Chase-Dunn, and reviewers for The Sociological Quarterlyon an earlier draft of the paper. They also thank Sondra Phillips and Elise Bennett for their help in pre-paring the manuscript. Mike Timberlake's address is Department of Sociology and Social Work, MemphisState University, Memphis, Tennessee 38152.


3/20

490 THE SOCIOLOGICALQUARTERLYIn this paperwe will attempt o answer wo specificquestions.First, is over-urbanization ueledby the economicdependence o whichless-developed oun-tries are subjectto varyingdegrees?Second,does overurbanization-conceived

of as either the relativesize of the urbanpopulationor the labordistribution-appear o present tselfas a barrier o economicgrowth? f economicdependenceis linked to these imbalancesand if overurbanizations linkedto economicstag-nation,it is quitepossiblethatoverurbanization ediates ome of the previouslydiscoverednegativeeffects of dependenceupon economicgrowth.The modelsuggesteds shown nFigure1.Figure1. TheoreticalRelationships mongEconomicDependence,Overurbanization,ndEconomicGrowth.

DEPENDENCEN + ECONOMICFOREIGNAPITAL GROWTH

OV R-URBANIZATIONThe underlying onceptualnotion of overurbanizations evaluative,andit hasbeen criticizedas such (Sovani, 1964). It is used to characterizehe fact thatmanyLDCshavelargerurbanpopulationshancan be effectively mployed--orsupported-by the structureof the economyor the level of development.Im-plied or explicit operationaldefinitionsof overurbanizationave included:theratioof the percentage f a country's otalpopulationwhich lives in urbanareasto some measureof level of development(Hibbs, 1973); the ratio of percenturbanto secondaryemployment(Kentor, 1981; Bairoch, 1975); the ratio ofurbanpopulation o secondaryand tertiary mployment Moir, 1975); the ratioof the rate of urbanization, suallythroughrural-urbanmigration, o the rateofeconomicdevelopment Wellisz, 1971); andthe ratio of serviceto manufactur-ing, or tertiary o secondary mployment.Statisticalcomparisonshave been made between the degreeof overurbaniza-tion in LDCs todayandmore-developedountries(MDCs) at earlier imes,andthey indicatea disparitybetweenthe two experiences.Generaldescriptive tate-ments about LDCs mask important ntercountrydifferences.Nevertheless,asmanyhavepointedout that,overall,LDCstodayhavea muchlargerproportionof theirpopulationsn cities than did the presentlydevelopedcountriesat simi-lar levels of economicdevelopment cf., Ward, 1969; Hawley, 1971:272; Bai-roch, 1975; U.N., 1976). Impliedby the importanceattached o such observa-tions is that there is a symbioticrelationshipbetweenurbanization nd level ofeconomic developmentwhich has a balancing point at which developmentalefforts are facilitated and developmentalproblemsare minimized.It has beenarguedthat urbanizations necessary or economicdevelopmentbecauseit re-


4/20

A Study of Less Developed Countries 491ducesthe frictionof space (Davis andGolden,1956), allowing he juxtaposi-tion of complementaryunctionsrequiredby the industrializationrocess (Hig-gins, 1967; JakobsenandPrakash,1971:21-22). On the otherhand,it hasbeenmaintained hat too much urbanizationmay presentobstacles to industrialdevelopmentf it comestoosoonandtooquickly Higgins,1967).The descriptionof less-developedcountries s similarwhenwe focus on im-balances n the urban abordistribution, atherthanon the relativesize of theurbanpopulation tself. Three broad distinctionsare commonlymade betweendifferent ypes of economicactivities.Primaryoccupationsare mainlyagricul-tural. The secondary abor sector includesoccupations n manufacturing,on-struction,mining,and utilities,but it is primarily houghtof as the manufactur-ing sector. The tertiarysector includesjobs in commerceand transportation/communications, s well as government, ommercial,andpersonalservices.Thetertiary s oftencharacterized s the servicesectorbecausemanyof the jobsin itservice ctivitiesof theother wosectors.The most striking eatureof the laborstructuren the ThirdWorld s that thetertiaryandsecondary ectorsareverysmall,withmostworkersbeingemployedin the agricultureector.Moresurprisingo mostobservers,however, s the largesize of the tertiarysector, relativeto the level of industrializationhat has oc-curred.This structuraleatureof LDCs hasbeen variously ermed hypertrophyof the tertiary (Amin, 1976), overurbanization, nd the pseudo-tertiary(Germani,1973). The last termis used to emphasize hecontention hat theter-tiarysector s largelycomposedof marginally mployedor underemployed ork-ers, who seemto abound n mostlargecitiesof the ThirdWorld-the operatorsof automatic levators, he vendorsof singlesticksof chewinggumand of singlecigarettes,car watchers,and so on. Again, these generalizations elie strikingdifferences etweenThirdWorldcountriesn thesize of thenonagriculturalaborforce andits demographic omposition,particularlywithrespectto the extenttowhichwomen areemployed n nonagricultural age labor (e.g., Boserup,1970;Youssef, 1976).The standardagainstwhichthesefeaturesof LDCs'occupational tructure recompared s the patternof labor compositionwhich historicallyaccompaniedeconomic expansion in the early industrialcountries. Jakobson and Prakash(1971:24), summarizingJean Fourastie'sthesis, argue that under normalconditions i.e., thosebasedon theexperiences f the early ndustrializers) ur-ing the take-off tageof economicgrowth,employmentn the secondary ectorwill rise to around35 percent, he tertiarywill risemoreslowlyto a lowerlevel,and employment n the primarysector will decrease.Duringthe expansionstage,secondary mploymentpeaksandbeginsto decrease,as labor-savingech-nology is developed;primaryemploymentdecreases o a minimum, tablelevelof around10 percent;andthe relativesize of the tertiary ectorcontinues o in-crease,exceeding he employmentevel of the secondary ector.Ultimately, ec-ondaryemploymentevels off, and tertiaryemployment tabilizesat a relativelyhigh level, providingas much employmentas the other two sectorscombined.However, these patterns rarely characterize what has occurred in LDCs.The descriptive data presented in Table 1 show that the ratio of tertiary tosecondary workers in MDC's was about 1.00 (equal employment in the two sec-tors) as recently as 1950. In 1850 the ratio was only .68, indicating that there


5/20

492 THESOCIOLOGICALUARTERLYwere aboutseventertiaryworkers or every 10 secondaryworkers n MDCs. Incontrast,the tertiarysectorin LDCs has been substantiallyarger han the sec-ondary aborforce sinceas long ago as 1900. Duringthis time, there wereover12 tertiaryworkers for every 10 secondaryworkers.The ratio rose to 1.7 in1970. More detaileddata describing he laborforce distributionn subdivisionsof the secondaryandtertiary ectorsallowus to compare heratio of workers nservices o workers n manufacturingor 1960 and 1970. This is moretrue to theunderlyingoverurbanizationoncept,but data coverageis not nearly as com-plete.These dataindicatethat, in 1960, LDCshad about 13 workers n serviceoccupations or every 10 in manufacturing.At the same time, in MDCs, therewereonly eightworkers n services or every 10 in manufacturing. y 1970, theratio had increasedto 1.5 in LDCs, comparedto .94 in MDCs. Observationssuch as these raise serious doubts about the likelihood that the patternsofchangein occupationalstructurewhich accompanieddevelopment n the earlyindustrial ountrieswillbe repeatednLDCs.The questionhas been raisedregardinghe significance f comparingevelsofoverurbanizationn LDCs today to the patternfollowed by more-developedcountries.Amongthe criticismsof the conceptare, first,thatthere s no a priorireason to expect that the developmentalpatternof the West is an appropriatemodelfor the less-developednationstoday (Sovani, 1964). It wouldbe equallyappropriate o arguethat Westernnations are underurbanized, ompared oLDCs, Sovanipoints out. Yet, the notionpersiststhat overurbanization in-dersdevelopment.EvenSovani,whoBose argues haseffectivelydemolishedhe'overurbanization'hesis (1971:99), admitsthat thereare severalpossiblere-searchquestions hat could be pursued egardingherelationship etweenurbani-zation and economicgrowth(Sovani, 1964:1213). On view of the coexistenceof rapidurbanization nd economicstagnation n manyThirdWorldcountries,it seemsreasonable o ask whether he two are causallyrelated.Hence, regard-less of any implicitcomparison o the presentlydevelopedcountries,one mightjustifiablyask whethercountries that are more overurbanized implying acontinuum,not a criterion)arenot alsoless successfulat economicdevelopment.Table 1. Tertiary/Secondary atios for MDCsand LDCs 1850-1970 (PartA),andService/Manufacturingatiosfor MDCs and LDCsin 1960 and1970(Part B).*A. Tertiary/SecondaryRatios

1850 1880 1900 1920 1930 1950 1960 1970MDCs .68 .87 .81 .92 1.09 1.05 1.10 1.20LDCs 1.25 1.26 1.33 1.68 1.61 1.70B. Services/ManufacturingRatios, 1960 and 1970

1960 1970MDCs .81 .94LDCs 1.26 1.51*Adapted from Lunday and Timberlake (forthcoming), who, in turn, rely upon data presented by Bairochand Limbor (1968), as well as ILO data.


6/20

A Study of Less Developed Countries 493The suspicion that too much urbanization nhibits economic developmentwhen it occurstoo early n the developmental istoryof a countryhas beenprop-agatedby studentsof developmentn all global regions.Overurbanization,ri-

macy, and optimalcity size are often discussed as one problem.For example,Ward (1969), after showingthat less-developedcountriesare overurbanized,suggestedpartof a solutionwould involve urbanplanningaimedat producingdecentralizedurban growth in various regionalgrowthcenterschosen on thebasis of theirability o satisfycertain unctions.The idea thata solution o over-urbanizationwill involveplanning n rural,as well as urban,areasis also foundin the literature.Gugler(1982) is among he mostrecentof manyobserverswhohave pointedout that increased nvestment n rural,ratherthan urban,areaswould probablybenefitdevelopmental ffortsand decreasethe degreeof over-urbanization(c.f., Wellisz, 1971; Sovani, 1964; Firebaugh,1979). Bairoch(1982) focuseson the need to constrainpopulationgrowth n the largestcities.Similarly,Wellisz,in discussingoverurbanizationn Asia, suggeststhat optimalcity size m;ghtbe a usefulapproach o the phenomenon.But he clearlyarguesthat unwieldyurban populations-however defined-will bog down develop-ment: Overurbanization,n short, standsfor a 'perverse' treamof migration,sappingthe economicstrengthof the hinterlandwithoutcorrespondinglyargebenefits o urbanproduction. nsteadof beinga signof development, verurbani-zation is a sign of economic illness (Wellisz, 1971:44). Lander and Funes(1975:333) assertthatoverurbanizations supportive f the dualism f LatinAmericancountries n general,particularlyVenezuela,and that, together,theywill limit the effectivenessof developmental fforts.Portes and Walton (1976:169-78) submitthat rapid,unbalancedmodernizations both a result of eco-nomicstagnationanda conditionsupporting tagnation. Thus,the solution iesin checkingthe uncontrolledgrowthof a few metropolitan entersand concen-tratingon efficientagricultural roductionand the developmentof alternativeindustrial olesof growth.The distinctionbetweentwo primarydimensionsof overurbanizations clearin the literaturewe havereviewed o far.Onedimensionemphasizeshesize andgrowthof the proportion f a country'spopulation ivingin cities,relativeto thelevel and growth n the level of economicdevelopment f the country.The sec-ond dimension uggests hatthe balanceof serviceto manufacturingccupationsis crucial.In the remainder f thepaperwe will occasionallydistinguishbetweenthese two dimensions.

OverurbanizationndMarginalityOn the whole, the researchmentionedabove is not tremendously heoretical.Generalizationsuggestedby the literaturemainlyconsist of variationson theassumptions hat urbanizationand rapid growthof the tertiarysector are, inhealthyeconomies,concomitants f the growthof modern ndustry. ncreasesnurbanizationor tertiaryemployment n the absence of economicgrowthor agrowingsecondarysector are seen as pathologicaland are used to explaintheapparent marginality of the large numbers of urban poor who are observedsubsisting in the slums and squatter settlements of Third World cities. The mar-ginality argument often focuses on the labor force imbalance dimension of over-


7/20

494 THE SOCIOLOGICALQUARTERLYurbanization. t is usually claimed that workers are unemployedor underem-ployed in numerous informal ectoractivities.Rapidurbanpopulationgrowth(throughimmigrationand naturalincrease) and economicstagnationare seenas the immediateconditionscreating his imbalance.Overurbanization,tself, inturn,servesas an impedimento further conomicgrowth, n partbecauseof thesocialwelfare/socialcontrolburdens hat it placesuponthe state.Countriesmostin need of domestic nvestmentsor development rethus drainedof a significantportionof themeager conomic urpluseswhicharegenerated.These views,oversimplified ere,have been criticizedon a numberof points.Thereare at least two generalobjections o the proposition hat the overurban-ized portionof the urbanpopulation n LDCs is marginal. irst is criticismof the assumption hat workin the tertiaryor informalsector-the occupationsprevalentamongresidentsof slums andsquatter ettlements-are dead-end obswhichdo nothingbutensure he continuedpovertyof the peoplewho holdthem.Such criticismsusually have been foundedon surveysof individuals n ThirdWorldcitiesor on in-depthcase studiesby participant-observersf communitieswithina city (e.g., Peattie, 1968; Perlman,1976). Together, hesestudieshaveshownthat the move into the city andsubsequent mploymentn these so-calledmarginal obsoftenrepresentan economic mprovementor manyindividuals.That is, serviceemployment,and even informalsectoremployment,may repre-sent upwardsocial mobilityfor peoplemovinginto these occupations.Further-more, it has been shownthat even for those alreadyemployed n these occupa-tions, there is substantialupwardmobilitywithin the sector, sometimesalmostequal o thatwithin hesecondary ector.These studiesare importantn indicating he need to distinguishbetweenthesignificance f a givenstructuralphenomenon or microsocialprocesses,as op-posedto macroprocesses.t is the latterwith whichwe aredealing n thepresenteffort.It is not at all contradictoryo note that,on one hand,tertiaryor informaloccupationsare a route to upwardmobility or individualswithincountriesand,on the other hand, maintain hat the largerelativesize of this sectorserves toinhibitoveralleconomicgrowth.It is the overallshapeof the occupational truc-ture whichconstrainshe overallamountof mobility(Boudon,1973). Compara-tive status-attainmenttudiesusuallycontrolfor shapeof the occupational truc-turewhenseeking o determine heamountof puremobility whichexists n onesystem,compared o another.That is, mobilitydueto factorsother thanshifts nthe relativenumbersof positionsat different evels of the statushierarchys ex-amined n these studies.However,thereis a growingbody of evidencethat theshapeof the structure ndchanges hereinactuallyaccount or most of the over-all mobilty (e.g., Hope, 1982). This is similar to the point, made by Gugler(1982), thatthere is a paradoxregardingmigrationrom ruralto urbanareas nThirdWorldcountries: t is rational or the individualmigrant,but it is clearlyirrational or the economyas a whole. His resolutionof the paradox s to suggestthatrural-urbanmigration presumably esultingn moreinformal ectoror ser-vice sectoremployment) hasa redistributiveffect (Gugler,1982:186), withrespect to the vast rural-urban ifferencesn employmentopportunities,wagelevels, etc. Some individuals stand to gain, but the economy, as a whole, does notbenefit.From our point of view, it is extremely important to understand the structural


8/20

A Study of Less Developed Countries 495limitationsplacedon individualmobility.This is not to deny thatThirdWorldworking-classpeople and peasantsare remarkably reative in their struggle oimprovetheircondition.For some, obtainingemploymentn the overgrown er-vice sectoror informalsectormay represent uccessin this regard.Instead,ouremphasisassumes he importanceattached o structural onstraints n mobility.Wheneconomicgrowth s slower in one country hanin another, t is likelythatthe numberof mobility successstories will be correspondinglyewer (but byno meansabsent).A secondcriticismof muchof the early writingon overurbanizationndmar-ginalityis based on a steadily growingbody of researchwhich shows that theproblemsof urbanization nd underemploymentn Third Worldcountriesaregender-specific,nvariably o the disadvantage f women (e.g., Boserup, 1970;ChanneyandSchmink,1976). In addition, here aretremendous egionaldiffer-ences in the extentto whichwomenareemployed n nonagriculturalccupations(Boserup, 1970; Youssef, 1974), and when they are employed n these areas,it is disproportionatelyn reproductive abor. Recenttheorizing eemsto sug-gest that the interactionof capitalist penetrationwith culturallyspecific pre-capitalistmodes of production includingsex-baseddivisionsof labor) accountsfor much of the regionalvariation n this regard(cf., Beneriaand Sen, 1981).However,whether he functionsof a givenlaborsector (e.g., the servicesector)will vary with its sex composition s open to question,and, in any case, it is amattermuchbeyond hescopeof thepresenteffort.A thirdcriticismof theviewthattheurbanpoor in theThirdWorldare mar-ginal involvesa structural rgumenthat will move us towardconsideration fthe structure f relations n the largerworldeconomy.Portes (1981) articulatedthe positionthat the informalsectorperformsa crucialrole for the worldecon-omy in that it functions o subsidizereproduction f the formalsector aborforcewithinLDCs. This helpsto explain why wages are lower in the periphery hanin the core,even whenproductivitys heldconstant.Thisprocess s viewedas animportantmechanismhroughwhichcore institutions, uchas multinationalor-porations,exploitThirdWorldpopulations.This argumenteads us to considerthe recentresearchwhichhas linkedurbanization rocessesto dependenceandposition nthe worldeconomy.

Dependenceand OverurbanizationDependency heoristsarguethat the economicpenetration f ThirdWorldcoun-tries (in theformof tradepatterns, oreignaid, andinvestments romenterpriseslocated n therich,corecountries)hasprofoundlynfluenced heeconomicstruc-ture of poor countries.Early dependency heoristspointedout that economicinteractionbetweenrich andpoorcountries endedto consist of the exploitationof naturalresourcesandotherprimaryproductsof the latterby the former(cf.,Baran,1957; Frank,1972). Ledogar(1975) contended hatforeign nvestmentin agriculture timulatesrural-to-urbanmigrationby reducingthe amountofland available for peasant agriculture. But tertiary or pseudo-tertiary jobs areoften the only jobs in town. More recently, industrialization in the periphery hasbeen acknowledged by critics of the early dependency theorists (e.g., Warren,1973). The response has been to argue that the nature of dependent industriali-


9/20

496 THE SOCIOLOGICALQUARTERLYzation s not conduciveo the symbiotic ependencemongabor orcesectorsthoughto characterizehe Westernndustrializedations.The manufacturinglaborsectorsof dependentountries o not supportertiaryectorsof thesizeand nature ound n MDCs.First,dependentndustrializationoes not createa sufficientomesticurpluso sustainhesupport etworks ssociated ith n-dustrializationDelacroixand Ragin,1980). Globalcorporationsncourageindustrial roductionor export;consequently,nternalmarkets re not ade-quately eveloped.econd,he industrializationhichdoesoccur s oftenbaseduponthecapital-intensiveechnologyf the West(cf., Merhav,1969). Third,thehigh-technologyertiary ccupations, hicharesymbioticallyinked o de-pedentndustrializatione.g.,research nddevelopment),emainocatedntherichcountries,longwith heheadquartersf themultinationalirms.Withrespecto thedistributionf thelabor orce, he neteffectof economicandtechnologicalependencen Westernountriesndmultinationalorpora-tions s to constrainmploymentpportunitiesn thesecondaryector, ince heproducer oodscomponent f this sector s largely uppliedhrough mports(see alsoNewfarmerndMueller, 975;Sunkel,1973:145).As a consequenceof this,a largereserveabor orce is leftunemployedr underemployed,ostoften n theovergrownertiary.Wehavealreadyuggested hyoverurbanization,n general,s likely o con-tributeo sloweconomicgrowthn LDCs,but it maybe morecompellingoseekto linktheseeffects pecificallyo imbalancesn the labor orcestructure.If it is accurateodescribehetertiarynmanyLDCsascomposedfcarwatch-ers, otteryicketvendors, ndtheinnumerabletherswhoproduceor or ser-vice heformal conomy,hen t is quitereasonableo assume hat hissector'slackof productivityervesasanimpedimento developmentalfforts.Certainly,theinhabitantsf squatterettlements ndothersuchsubsistenceommunitiesexist n sufficientumbershat heymustbe reckoned ithwhen ocial ervices,howevermeager, redistributed.urthermore,hismarginal,argely rbanpop-ulation s likely o be perceiveds a potential oliticalhreat, ndrevenuesmaybe usedto incorporateome of its membersnto the national ureaucracyrmilitary, r used to toughennternalecurityorces n order o moredirectlycounterhethreatt representscf., Nelson,1969;Sunkel,1973). To thisex-tent,potentialnvestmentavingswhichwouldaiddevelopmentrespent n-steadon thewelfareproblem Schweinitz, 964). Onthe otherhand,somehavesuggestedhattertiarymployment ayaid in developmentalfforts, venin the absence f a secondaryectorof themagnitudehat existedamong heearlydeveloperse.g.,Moir,1976:132).Recentresearchhas supportedhe hypotheseshatdependenceontributesto bothdimensionsf the overurbanizationhenomenon.vansandTimber-lake (1980), usingtwomeasures f dependence,howed hat thegreaterhedegreeof economic enetrationf LDCs, hegreaterherelative rowth f thetertiaryector rom1950to 1960.These indings,owever, requalifiedythesmallnumberfcases ntheiranalysis,around 5) and he imited imeperiod.In a much largersampleof LDCs, Kentor (1981) found that dependencehada positiveeffecton rates of urbanizationrom 1955 and 1960 to 1970. Further-more, his more detailed analysis showed that these effects on urbanizationseemedto havebeen mediated hrough he positiveeffects of investmentdepen-


10/20

A Study of Less Developed Countries 497denceon growthof the tertiary aborsector,1950-70. It was shownthatdepen-dence did not affectincreases n urbanization y contributingo secondarysec-tor employment.In fact, these effects were negativeover the same period oftime,indicating hatmoredependentcountrieswere less successfulat generatingjob opportunitiesn the secondarysector. Whenthe effects of dependenceonoverurbanization-measured y the ratioof percenturbanto secondaryemploy-ment-were estimated,Kentorfound that greater evels of dependencecontrib-uted to increases n overurbanization. ince there is strongevidencethatwhenforeigninvestmentsn LDCs aregreater,economicgrowth s limited(cf., Born-schieret al., 1978), it is possiblethat overurbanizationmediatesthese effects.

The ProposedAnalysisWe will restrict our analysisto less-developedcountries,since our concerniswith extendingpreviousresearchwhichhas linkeddependence o overurbaniza-tion andeconomicstagnationn the peripheryamongLDCs.Wefocuson 1965-75. For the present,we distinguishbetween mbalance n the labor force struc-ture and the narrowerconceptionof overurbanization-the size of the urbanpopulationrelativeto the level of economicdevelopment.We are distinguishingthem mainly as operationalizations f differentdimensionsof a single, under-lying concept. Hence, we are essentiallyconducting wo paralleltests of eachhypothesis,but the limitedcoverageof the most appropriateabor force datareducesto a very few the numberof cases in that partof the analysis.We willthereforeemphasize he findingspertaining xplicitly o the urbanization imen-sion of overurbanization. o the extentthe results of the paralleltests are mu-tuallysupportive,we willbe more confidentn ourfindings.First,we willexplorethe relationshipbetween penetrationby foreign capital and overurbanization,hypothesizing: 1) the greaterthe level of penetrationof foreign capitalinto agiven less-developedcountry,the greaterthe degree of overurbanization. ec-ond, we will attemptto evaluate the notion that overurbanizationnhibitseco-nomicgrowh.Thus,we hypothesize:(2) the greatera country'sdegreeof over-urbanization,he less theeconomicgrowth.

Data and MeasurementThe samplesize was limited,by the availabilityof data,to between20 and 111cases, dependingupon which measuresare being analyzed.In order to maxi-mize the numberof cases,we used a liberalcriterion or defining ess-developedcountries,excludingthose countrieswith 1965 per capita GNP in excess of$2,000. This resulted n the inclusionof a few countriesof westernEuropeandelsewherewhich would undoubtedlybe regardedas semi-peripheral y worldsystem theorists.Investigationof scatter plots of bivariaterelationships ndi-cated that these countriesdo not presentthemselvesas outliers n our analysis.The indicatorof dependencewe usedwas measured n 1967, and,ideally,wewould have examined change in the dependent variable from this time point tosome later time point. We were able to obtain urbanization data and GNP datafor a reasonably large number of countries for 1965 and 1975. We have 69LDCs in the analysis involving this dimension of overurbanization. The appro-priate labor force data are available yearly, but for only a few countries, and we


11/20

498 THE SOCIOLOGICALQUARTERLYwere unfortunately left with a sample size of only 20 cases in our analyses in-volving change in the labor force ratio. The time points for the service andmanufacturing labor force data are around 1968 and around 1977. Other timepoints would have been more appropriate from the standpoint of matching themeasurement times of the other variables, but this would have resulted in aneven smaller sample size. The dependence measure, though, is highly stable overa few years, since it represents the value of total stock of foreign investmentswhich had been accumulating in a country over a long period of time. Thus, the1967 measurement time is not likely to be problematic. Because of the smallsample size, the findings regarding the service/manufacturing labor structureand dependence must be interpreted with caution. We believe they are, none-theless, instructive insofar as they are consistent with the findings regarding theother dimension of overurbanization.The measure of development used here is per capita GNP (in constant dol-lars) obtained from Bornschier and Heintz (1979). We use logarithmic trans-formations (LGNPPC65 and LGNPPC75) of their 1965 and 1975 measures.The labor force data were obtained from the yearbooks of the InternationalLabor Organization (various years). The secondary sector of the labor forcecustomarily includes the categories of manufacturing, electrical and other utili-ties, construction, and, often, mining. Tertiary occupations include those in com-merce; transportation, storage, and communications; and government and otherservices. In the initial analysis, we used only the manufacturing subsector of thesecondary and the services subsector of the tertiary because they are closer tothe underlying conceptualization of the labor force dimension of overurbaniza-tion. We constructed ratios of services to manufacturing employment for around1968 and around 1977, which are used as our measures of labor structure im-balance in most of the analysis (SERMFG68 and SERMFG77). Labor forcedata which describe the distribution of the labor force into the three broad sec-tors are available for a much larger number of countries in 1960 and 1970.These are contained in the Labour Force Estimates and Projections of the Inter-national Labour Organization (1977). (Data at later points are given, but theseare estimates which are inappropriate for use in panel regression.) We also referto the 1960 and 1970 data (111 cases). These data pertain to the distributionof the labor force into the entire tertiary sector and entire secondary sector,rather than just the service and manufacturing subsectors. Thus, we sacrificedconceptual precision for the sake of gaining more cases-many more cases.One common conceptualization of overurbanization involves simply a com-parison of the size of a country's urban population to its level of economic de-velopment. The most obvious way to operationalize this concept would be to usethe ratio of percent urban to some appropriately scaled measure of per capitaGNP. We would be reluctant to use a development measure other than GNP(e.g., energy consumption) because, at least in principle, countries with highlevels of per capita productivity could support large urban populations, regard-less of the reasons for productivity (manufacturing automobiles or producingoil, for example).A problem which is likely to occur with such a ratio measure, however, isthat of spurious ratio correlation (c.f., Bollen and Ward, 1979). For example,the regression of per capita GNP on an overurbanization ratio such as the one


12/20

A Study of Less Developed Countries 499suggestedabove wouldbe likely to producea negativeestimateof the effectofoverurbanizationimplybecauseper capitaGNP is commonto both terms.In-steadof usingthe ratio,then,we decidedto producea measureof overurbaniza-tion whichcaptures he notionof the relationship f percenturbanto per capitaGNP, but which is statistically ndependentof the latter. We accomplishedhisby usingthe residualdifferencebetweenpercenturban n a givenyearanda pre-dictedvalue of percenturban obtained romthe regressionof percenturbanonper capita GNP in the same year. Thus, OVURB65 and OVURB75 measurethe level of urbanization elativeto the level of GNP per capita,but they areneverthelessorthogonal o LGNPPC65and LGNPPC75,respectively.The dataon urbanizationwere obtainedfrom the United Nations (1980). Becausetheanalysisis confined to less-developedcountries-the countriesof the worldtowhichdependency heoryandthe overurbanizationypothesesapply-there arenone of the problemswith ceilingeffectsin the measureof urbanizationwhichwouldexist if the full rangeof countrieshad been included n the analysis.Investmentdependence PEN1) measures he dollarvalueof stock of foreigninvestments n 1967 from the 16 countrieswithmembershipn the DevelopmentAssistance Council of the Organization or Economic Developmentand Co-operation (OEDC, 1972). These are essentiallythe developedwesternEuro-pean and North American countriesand Japan. We use Bornschier's 1980)transformation f this measure,which controlsfor the size of the penetratedcountry'seconomy.Panel regressionanalysiswas used to evaluatethe hypotheseswhichwe for-mulated.This involved ncludinga laggedmeasureof the dependentvariableasan independentvariable n a regressionequation.In this way the estimatedef-fects of the other independentvariableson change in the dependentvariablewere obtainedwithoutthe problemsof usingsimpledifferencesor gain scoresto measurechange (c.f., Bohrnstedt,1969). If this modelcreated he possibilityof serialcorrelationof errors, t wouldresult n a moreconservative est of ourhypothesesbecausetheeffectof thelaggeddependentvariable(overurbanizationor LGNPPC65) wouldbe inflatedat the expenseof the estimatedeffectsof thevariablesof theoretical nterest(Kesslerand Greenberg,1981:87-89).

The ResultsFirst,we examine he hypothesis hatdependenceeads to higher evelsof urban-ization, relative to levels of development.The resultsin column 1 of Table 2show that the hypothesizedpositiveeffectsof dependenceon overurbanizationare corroboratedamongthe 69 LDCs for which we have data on these mea-sures. LGNPPC65is not includedin this equationsince this overurbanizationmeasure s constructed o that it is almostcompletelyorthogonal o it. This isdescribedabove. The resultsindicatethat the greaterthe level of dependence,the greaterproportionof the populationresides in cities, compared o the levelof development.Next, we turn our attention to the hypothesis that investment dependenceleads to greater levels of employment in services, relative to manufacturing.Table 2, column 2, shows the estimates of the panel regression of the services/manufacturing ratio on measures of dependence, levels of development, and the


13/20

500 THE SOCIOLOGICALQUARTERLYTable2. PanelRegression f Overurbanization975 (OVURB75)orUrbanLaborForce Imbalance1977 (SERMFG77)onInvestmentDependence1976 (PEN1), Per CapitaGNP (LGNPPC65),

andLaggedDependentVariable OVURB65or SERMFG68).Dependent Variable OVURB75 SERMFG77IndependentVariablesLaggedDependentVariable b .7642*** .3541*Beta .8026 .3386t 10.14 1.40

b .3778** .0986**PEN1 Beta .1380 .5812t 1.74 2.19LGNPPC65 b -.4345B -.2178t -.94Intercept - 2.2248 1.5265AdjustedR2 .598 .559n 69 20

*p < .10**p ? .05 ) b = regression coefficient***p < .01 ) one-tailed Beta = standardized regression coefficient****p < .001 ) t = t statisticearlierimbalancemeasureamongthe 20 LDCs for whichwe have laborforcedata. (When earlier levels of employment n the two labor sectorswhich arecomponentsof the imbalancemeasurewere includedas separatecontrolvari-ables,theyhad no significant ffectson the ratio.) The hypothesis s supported:the greater he dependenceof LDCs, the moreimbalanceoccurs n the ratio forthe 1968-77 time period. Together,these findingsare importantbecausetheycorroborate he hypothesis hat investmentdependenceexacerbatesoverurbani-zation. At the sametime,the resultsregardinghe laborstructuredimensionsofoverurbanization orroborateearlier cross-nationalresearch which employedsomewhatdifferentmeasuresof laborstructurembalanceover differentperiodsof time (e.g., Evans and Timberlake,1980; Kentor, 1981).Do these structuralmbalances nhibit economicgrowth?First,we estimatedthe effectsof OVURB65 on economicgrowth.Table 3, column1, presents heresults of the regressionof LGNPPC75 on the 1965 level, OVURB65, andPEN1. Contrary o the overurbanization ypothesis,the simplelevel of over-urbanizationhas no significanteffect on economic growth. In this equationPEN1 has an estimatednegativeeffect on economicgrowth,whichothershavereported.Before dismissing he hypothesisaltogether,we examined he respon-siveness of changein GNP levels to change n levelsof overurbanization.In Table 3, column2, we estimatethe effects of changein overurbanizationfrom 1965 to 1975 on economic growth by includingboth OVURB65 andOVURB75 in the equation.Because differencescores are inappropriatemea-sures of change n independent ariables,we included nsteadboth the earlyandlatermeasuresof laborstructurembalance n the equation.The estimated ffects


14/20


15/20

502 THE SOCIOLOGICALQUARTERLYtionshipsmay reflecttemporary ssociations.For example, heymaybe peculiarto the particularcycle of the world economy duringthe time period underconsideration.

Next, we examinethe correspondingorm of the hypothesiswith the laborforce dimension of overurbanization.Again, the coverage for the labor forcesubsectordata is very limited.We used them mainlyas a check on the resultswe obtainedusing the alternative pecification f the overurbanizationoncept.The estimatedeffect of labor force imbalanceon change in levels of develop-ment is estimated in an equation which includes LGNPPC65, PEN1, andSERMFG68. Table 3, column 3, shows the results of the regression ofLGNPPC75 on these variables.PEN1 has the statistically ignificantnegativeeffects on developmentwhich othershave found,but whenit is in the equation,the effects of labor structure mbalanceare not statisticallysignificant.Poorcountrieswith excessive ervice-to-manufacturingmploymentratios do notseem to suffersignificantconsequencesbecauseof it-at least not in termsofeconomicgrowth.The possibilityremained,however,that the processof increasing mbalancein service to manufacturingmploymentdoes inhibiteconomicgrowth.Table3,column4, presents he resultsof an equationwhichestimates he effectof changein the laborstructureratio on economicgrowth, ndependently f PEN1. Withboth SERMFG68 and SERMFG77in the equation,our hypothesispredictedthat the estimatedeffect of the latterwouldbe negative,andthis is, in fact, whatwe find. Furthermore,he effect of PEN1 is not statistically ignificant n thisequation.Because the samplesize is exceedinglysmall and the cases shiftfromcolumn3 to column4, these findingsare presented n their capacityas poten-tially supportiveof the findings n columns 1 and 2. Thus, we tenuouslycon-clude that increasing abor structure mbalancemay mediatesome of the pre-viously discoverednegativeeffects of dependence.We should emphasizethatthere s no support or thehypothesis hat the level of excessserviceemploymentto manufacturingmploymenthas the negativeeffecton economicgrowthwhichhas beenattributedo overurbanization. ut change n the ratiodoes seemto in-hibiteconomicgrowth,at least in theshortrun.Substitutinghangescores for thetwo level of SERMFGscoresproducedsimilarfindings, ust as it did with theanalysis involvingthe other dimension of overurbanization. luctuationsoverlongerperiodsof timewillbe importanto study n thefuture.We also estimatea versionof our model with the 1960 and 1970 ILO esti-matesof tertiaryand secondaryemployment ILO, 1977). These are less satis-factory than the ILO yearbookdata (variousyears) in that the occupationalcategories are broader,reportingmeasures of labor force participationn alltertiaryand all secondaryoccupations,and not just the serviceandmanufactur-ing subsectorsof each. The earliermeasurementperiod also qualifies indingswhich used the derivedoverurbanizationmeasuresin an equationwith PEN1,which was measuredonly threeyearspriorto the end of the 10-year ag. Fur-thermore,PEN1, itself,serves as a limitingvariable,withrespectto data cover-age. Nevertheless,we mentionthat an OLS estimateof the effectof PEN1 onincreasing tertiary to secondary employment imbalance produced findings simi-lar to and supportive of those shown in Table 2. We decided to focus instead


16/20

A Study of Less Developed Countries 503on an incompleteversionof our modelwhich excludesPEN1. This allowedus tolook at a much broader sample of less-developed ountries(n = 111). Table4, column1, reports he regressionof LGNPPC70on its 1960 measureandtheratio of tertiaryto secondaryemployment n 1960 (TERSEC60). The resultsindicatethat LDCs with higherlevels of tertiary o secondaryemployment x-periencerelatively lowereconomicgrowth.The equationestimated n column2of Table 4 also includes he tertiary/secondaryatioin 1970, thus providinganestimate of the effectof changein the ratio on changein per capitaGNP. Theresultsof this indicate,once again, that relative ncreases n laborstructurem-balance are negativelyassociated with economicgrowth.Again, we estimatedalternative pecifications f this model,namelyone with the componentsof theratio added to the equationimpliedin column 1 of Table 4 and one in whichpercentchangein the ratio replacedTERSEC60andTERSEC70in column2.The resultsof these regressionestimatesdid not differsignificantlyrom thosereported n Table4.The resultsof the largersample,1960-70 labor force estimatesanalysis,aregenerallysupportiveof the smallersample, 1968-77 estimates, nsofar as thenegativeeffects of changing abor structure mbalanceon economicgrowtharereproduced. t is important o point out, however,that we used differentopera-tions of laborstructurembalance,and we did not includethe measureof inves-ment dependence n the largersampleprocedures.We thereforeemphasize hefindingspertainingto the urban/developmentdimensionof overurbanizationandpointout thatthe findingsregardinghe laborforce dimensionaregenerallyconsistent.Despite deficienciesof various labor force measures,we thoughtitwas betterto use the availabledata thanto ignorethem.Ourprocedures ouldhave produced indings hatwereinconsistent,undermininghe positedhypothe-

Table4. PanelRegressions f PerCapitaGNP (LGNPPC70) onLaggedDependentVariable(LGNPPC60), andUrbanLaborForce Imbalance(TERSEC60or TERSEC60and TERSEC70).DependentVariables LGNPPC70 LGNPPC70IndependentVariables b 1.1448**** 1.1468****LGNPPC60 Beta .9592 .9609t 41.81 42.93

b - .0258** .0484TERSEC60 Beta - .0426 .0800t - 1.86 1.46b - .0819***TERSEC70 Beta - .1326t -2.45

Intercept - .0685 - .0599AdjustedR2 .949 .951n 111 111*seenotesto Table 2.


17/20

504 THE SOCIOLOGICALQUARTERLYses. Instead,the analyses pertaining o the labor force supported he analyseswhich focusedon the relationship f urbanizationo development.Both proce-duresprovidedevidencethat increases n overurbanizationre accompaniedbyless economicgrowth.

DiscussionThe findingspresentedhere corroboratehe hypothesis hatpenetrationby for-eign capitalleads to overurbanization,efinedeither as a laborforce imbalanceor as an imbalance n the urbanization/developmentelationship.The findingsextend and supportearlierresearch inkingdependence o the structureof thelabor force. Using a dependencemeasurewhichindicates1967 levels in stockofforeign nvestments n Third Worldcountries,we haveshownthat whenthe rel-ative amountof this stock wasgreater,relatively ncreasingevelsof overurbani-zation occurredbetweenthe mid-1960s and the mid-1970s. We say relativelyincreasing ecause,with our regressionanalysis,whathas reallybeen shownisthatpenetrations positivelyassociatedwithchanges n these measuresof over-urbanization.The actual changes which are experienced n more dependent,compared o less dependent,countriesoverallmay range,in principle, romin-creasingmoreto decreasingess.But do thesestructural eaturesreally presenta problem or LDCs?After all,thereare manywho wouldargue hatthe increasing izeof the service aborsec-tor is a healthysign. In the absenceof the detaileddata thatwouldbe requiredto show that this sector really is overrepresentativef workersin marginaloccupations,we had to entertain he possibility hat thisviewis correct.Perhapsfindingpositiveeffectsof foreigncapitalpenetrationon the relativesize of thissector wouldbe an indication hat the infusionof capitalinto the ThirdWorldis accompaniedby changetoward the postindustrialaborstructureenjoyedinthe highly-developedountries.Our findingsleave us very skepticalthat this is the case. Higher levels ofservices-to-manufacturingmploymentwere not associatedwitheconomicgrowth(no statisticalrelationship),and increases n these levels over the time periodstudied were accompaniedby slower growthor stagnation(negativerelation-ship). Furthermore,when we examined a more broadlydefinedlabor imbal-ance, the ratio of tertiary o secondaryemployment,we found that both levelsandchanges n levels have estimatednegativeeffectson economicgrowth n the1960s. Again,we mustpointout that the significance f the growing ervicesec-tor maybe quitedifferentat otherlevels of analysis.Withincountries,entry ntothe informalsectoror the service sectormay represent outesto upwardmobil-ity for largenumbersof individuals.But as a structuraleatureof the economiesof Third World countries,it appearsto representquite the opposite. To theextentthateconomicgrowth s limited, t is likelythatoverallmobility-in termsof increases n averagenationalincome and shifts in the shape of the occupa-tionalstructure-will alsobe thereby imited.Similarly, the level of urbanization/development imbalance has no effect oneconomic growth, but changes in levels of this dimension of overurbanizationimpede economic development in a manner consistent with our hypothesis. In-creases in overurbanization (or the failure to reduce the degree of overurbani-


18/20

A Study of Less Developed Countries 505zation) are associatedwith slower economicgrowthduring 1965-75. We areleft, then,witha model whichlinkspenetrationby foreigncapitalto increases noverurbanizationnd service/manufacturingmbalance,andthese increasesare,in turn,seen as ways in whichdependencemay impedeeconomicgrowth.Finally, we must end with the usual cautionsand call for furtherresearch.Among the most useful endeavorswould be the compilationand analysisofmuchmore detailed aborforcedata.We need to be able to distinguishbetweenthe vendorsof single sticks of chewinggum and professionalbureaucrats ormanymore countries han we nowcan. Buteven with the kind of aggregatedatanow available,there is much work to be done. If these same laborforce mea-sureswere availablefor more countries,we would be able to proceedwith thenecessary analysisof the relationshipbetweenthe two indicatorsof structuralimbalancewhichwe are examining.Hypertrophicationf the tertiaryandover-urbanization re not alwaysclearlydistinguishedromone another n the litera-ture. We have discussed hem and examined heir effectsas if they were some-what distinct,but it is very likely that they are in many ways redundant.Adetaileddescription f the structure f the nonagriculuralaborforce is largelyadescription f theemployedurbanpopulation.Unfortunately,he available aborforcedata do not covera largeenoughnumberof countries o allowus to exam-ine this relationship.Futureresearchshouldalso examinethe role of gender-specific ifferencesnurbanization nd nonagriculturalmployment.Dependence n the world econ-omy has been hypothesized o undermine he employment tatus of womeninperipheral conomies(e.g., Ward,1982). Sex differencesn labor forcepartici-pation may mediate some of the dependencyeffects on economicgrowth.Ofcourse, it remains to be shown whethersuch differencesare explainedby theuniqueway in whichwomen are incorporatednto the urban abor force or bypatternsof sex exploitationwhichtranscend he urbanization rocess. Certainly,womenare super-exploitedduringthe developmentprocessin ruralsectorsofThirdWorldeconomies,as well (c.f., Loutfi, 1980). Furthermore, ome femi-nist scholars suggest that patternsof sex exploitationattributedby others tocapitalistdevelopmentare,in fact, transcendent f any specificmodeof produc-tion. (BeneriaandSen, 1981).Finally,workby Lipton(1977), GilbertandGugler(1982), Gugler(1982),and others has suggested hat overurbanizationan be regardedas an outcomeof policiesfavoringthe organized nterestsof economic and politicaleliteswhocontinueto favor urbanareasover ruralareas. Migration o cities and subse-quent servicesector and informalsector employmentare rationalresponsestothisdualism.If politicalandeconomicdecisionmakingwere to beginto redressthe urban/rural nequity,then presumably herewould be fewer incentives ormigration o cities, and thus less expansionof the relativelyunproductive er-vice andinformal aborsectors.

REFERENCESAmin, Samir. 1976. Unequal Development: An Essay on the Social Formations of Peripheral Capitalism.New York: Monthly Review Press.


19/20

506 THE SOCIOLOGICALQUARTERLYBairoch,Paul. 1975.The EconomicDevelopmentof the Third World Since 1900.Berkeley,Calif.: Univer-sityof CaliforniaPress.- . 1982. Employmentand large cities: problems and outlook. InternationalLabour Review 121(5):519-33.Bairoch, Paul, and J. M. Limbor.1968. Changes n the Industrialdistribution f the world labor force byregion,1880-1960. nternational abourReview98 (4):311-36.Baran,Paul A. 1957.The Political Economyof Growth. New York: MonthlyReview Press.Beneria, Lourdes,and Gita Sen. 1981. Accumulation,reproduction,and women's role in economic de-velopment:Boseruprevisited. Signs: Journalof Women in Culture and Society 7: (2):279-98.Bohrnstedt,George W. 1969. Observationson the measurementof change. Pp. 113-33 in Edgar F.Borgotta, d., SociologyMethodology, 969.San Francisco:JosseyBass.Bollen, Kenneth,and Sally Ward, 1979. Ratio variables n aggregatedata analysis: their uses, problemsandalternatives. ociologicalMethodsandResearch7:431-500.Bornschier, Volker. 1980. MultinationaleKonzerne, Wirtschaftspolitikand Nationale Entwicklung imWeltsystem.Frankfurt:CampusVerlag.Bornschier,Volker, and Peter Heintz, eds. 1979.Compendium f Data for World-SystemAnalysis. Bulle-tin. Zurich:Soziologisches nstitutderUniversitatZurich.Bornschier,Volker, ChristopherChase-Dunn,and Richard Rubinson. 1978. Cross-national vidence ofthe effects of foreign investmentand aid on economic growth and inequality: a survey of findingsandreanalysis. AmericanJournalof Sociology84:651-83.Bose, A. 1971. The urbanizationprocessin south and southeast Asia. Pp. 81-110 in Leo JakobsenandVed Prakash,eds., Urbanizationand National Development.BeverlyHills, Calif.: Sage Publications.Boserup,Ester. 1970.Woman'sRole in EconomicDevelopment.London: Allen and Unwin.Boudon, R. 1973. Education,Opportunity, nd Social Equality.New York: John Wiley.Browning, Harley. 1967. Urbanizationand modernizationin Latin America: the demographicper-spective. Pp. 71-92 in Glenn Beyer, ed., The Urban Explosion in Latin America. Ithaca, N.Y.: St.Martin'sPress.Chaney, Elsa, and MarianneSchmink.1976. Women and modernization:access to tools. Pp. 160-82 inJune Nash and Helen Safa, eds., Sex and Class in Latin America.New York: Praeger.Davis, Kingsley, and Hilda Golden. 1954. Urbanizationand the developmentof preindustrialareas.EconomicDevelopment ndCulturalChange3 (October) 6-29.Delacroix,J., and C. Ragin. 1980. Structuralblockage: a cross-national tudy of economic dependency,state efficacyand underdevelopment. mericanJournal of Sociology.86 (May):1311-47.Evans, P., and M. Timberlake.1980. Dependence, nequalityand the growth of the tertiary: a com-parativeanalysisof less developedcountries. AmericanSociologicalReview45:531-52.Firebaugh,Glenn. 1979. Structuraldeterminants f urbanization n Asia and Latin America, 1950-1970.AmericanSociologicalReview44 (April):199-215.Fourastie,Jean. 1963.Le GrandEspoirdu XX Siecle. Ed. definitive.Paris: Gallimard.Frank,Andre Gunder.1972. The developmentof underdevelopment. p. 3-17 in James Cockroftet al.,eds.,DependenceandUnderdevelopment. ardenCity,N.Y.: Doubleday.Germani, Gino. 1973. Urbanization,social change, and the great transformation. Pp. 3-58 in G.Germani,ed., Modernization,Urbanization,and the Urban Crisis. Boston: Little, Brown.Gilbert, Alan, and Josef Gugler, 1982. Cities, Poverty, and Development: Urbanizationin the ThirdWorld.London: OxfordUniversityPress.Gugler, Josef. Overurbanizationeconsidered. Economic Development and Cultural Change 31 (1):173-89.Hawley, Amos. 1971. Urban Society: An Ecological Approach.New York: Ronald Press.Hibbs, Douglas A. 1973. Mass Political Violence: A Cross-NationalCausal Analysis. New York: Wileyand Sons.Higgins, Benjamin. 1967. Urbanization, ndustrializationand economic development. Pp. 117-55 inGlenn Beyer, ed., The Urban Explosionin Latin America. Ithaca, N.Y.: Cornell UniversityPress.Hope, Keith. 1982. Vertical and nonvertical class mobility in three countries. American SociologicalReview47 (February) 99-113.InternationalLabour Organization.(Various Years.) Year of Labor Statistics. Geneva: InternationalLabourOrganization.

. 1977.LabourForce Estimatesand Projections.Geneva: InternationalLabourOrganization.Jakobsen, Leo, and Prakash, Ved. 1971. Urbanization and urban development: proposals for an inte-


20/20

A Studyof LessDevelopedCountries 507grated policy base. Pp. 15-38 in Leo Jakobson and Ved Prakash, eds., Urbanization and National De-velopment. Beverly Hills, Calif.: Sage Publications.Kelly, Maria. 1981. Development and the sexual division of labour: an introduction. Signs 7 (2):268-78.Kentor, Jeffrey. 1981. Structural determinants of peripheral urbanization: the effects of internationaldependence. American Sociological Review 46:201-11.Kessler, Ronald C., and David F. Greenberg. 1981. Linear Panel Analysis: Models of Quantitative Change.New York: Academic.Kuzmin, S. A. 1982. Structural change and employment in developing countries. International LabourReview 121 (3):315-26.Lander, Luis, and Julio Funes. 1975. Urbanization and development. Pp. 287-337, in Jorge Hardoy, ed.,Urbanization in Latin America: Approaches and Issues. Garden City, N.Y.: Anchor Press.Ledogar, Robert J. 1975. Hungry for Profits: U.S. Food and Drug Multinationals in Latin America. NewYork: IDOC/North America, Inc.Lipton, Michael. 1977. Why the Poor Stay Poor: A Study of Urban Bias in World Development. Cam-bridge, Mass.: Temple Smith/Harvard University Press.Loutfi, Martha. 1980. Rural Women: Unequal Partners in Development. Geneva: International LabourOffice.Lunday, James, and Michael Timberlake. 1984. Labor force structure in the zones of the world-economy1950-1970. Forthcoming in M. Timberlake, ed., Urbanization in the World-Economy. New York: Aca-demic Press.Merhav, Meir. 1969. Technological Dependence, Monopoly, and Growth. Oxford: Pergamon Press.Moir, Hazel. 1976. Relationships between urbanization levels and the industrial structure of the laborforce. Economic Development and Cultural Change 25: (October):123-35.Nelson, Joan. 1969. Migrants, Urban Poverty, and Instability in Developing Nations. Occasional Papersin International Affairs, No. 22. Cambridge, Mass.: Harvard University Center for International Affairs.Newfarmer, Richard S., and Willard F. Mueller. 1975. Multinational Corporations in Brazil and Mexico:Structural Sources of Economic and Noneconomic Power. Report to the Subcommittee on MultinationalCorporations of the Committee on Foreign Relations, United States Senate. Washington, D.C.: U.S.Government Printing Office.Organization for Economic Cooperation and Development, Development Assistance Directorate. 1972.Stock of Private Direct Investment by D.A.C. Countries in Developing Countries, End 1967. Paris: OECand D.Peattie, Lisa R., 1968. The View from the Barrio. University of Michigan Press.Perlman, Janice E. 1976. The Myth of Marginality, Urban Poverty and Politics in Rio de Janeiro.Berkeley: University of California Press.Portes, Alejandro. 1978. The informal sector and the world-economy: notes on the structure of sub-sidized labour. University of Sussex Institute for Development Studies Bulletin 9 (June):35-40.Portes, Alejandro, and John Walton. 1976. Urban Latin America. Austin, Tex.: University of Texas Press.Portes, Alejandro, and John Walton. 1981. Labor, Class, and the International System. New York:Academic Press.De Schweinitz, Karl. 1964. Industrialization and Democracy. New York: Free Press.Sovani, N. W. 1964. The analysis of 'overurbanization.' Economic Development and Cultural Change12 (January) :113-22.Sunkel, Osvaldo. 1973. Transnational capitalism and national disintegration in Latin America. Socialand Economic Studies 22 (March):132-76.United Nations. 1980. Yearbook of National Account Statistics, 1980. New York: United Nations.Ward, Barbara. 1969. The cities that came too soon. Economist 233 (December) :56-70.Ward, Kathryn. 1982. The influence of the world economic system on the economic status of women andfertility behavior. Presented at the American Sociological Association annual meeting, San Francisco.Warren, Bill. 1973. Imperialism and Capitalist Industrialization. New Left Review 81:3-46.Wellisz, Stanislaw. 1971. Economic development and urbanization. Pp. 39-55 in L. Jakobson and V.Prakash, eds., Urbanization and National Development. Beverly Hills, Calif.: Sage Publications.Youssef, Nadia. 1974. Women and Work in Developing Societies. Westport, Conn.: Greenwood Press.

economic dependence, over-urbanization, and economic growth

Documents