Globalization and rural poverty

World Development Vol. 34, No. 8, pp. 1393–1404, 2006� 2006 Elsevier Ltd. All rights reserved

0305-750X/$ - see front matter

doi:10.1016/j.worlddev.2005.10.010www.elsevier.com/locate/worlddev

Globalization and Rural Poverty

PRANAB BARDHAN *

University of California at Berkeley, USA

Summary. — In this paper, we provide an analytical account of the mechanisms through whichglobalization, in the sense of increased foreign trade and long-term capital flows, affects the livesof the rural poor in developing countries (in their capacity as workers, consumers, recipients ofpublic services, or users of common property resources). Globalization can not only cause manyhardships for the rural poor, but it can also open up some opportunities which some countriescan utilize and others do not, largely depending on their domestic political and economic institu-tions, and the net outcome is often quite complex and almost always context dependent, belyingthe glib pronouncements for or against globalization made in the opposing camps.
� 2006 Elsevier Ltd. All rights reserved.
Key words — absolute poverty, self-employed, wage employed, trade liberalization, globalization

* Final revision accepted: October 20, 2005.

1. THE CORRELATIONS

As is common in most contentious publicdebates, different people mean different thingsby globalization. Some interpret it to meanthe global reach of new technology and capitalmovements, some refer to outsourcing bydomestic companies in rich countries, othersprotest against the tentacles of corporate capi-talism or the US hegemony (economic, mili-tary, or cultural). In this paper, I shall limitmyself to interpreting globalization simply asopenness to foreign trade and long-term capitalflows. I shall ignore here the important issuesarising from the devastation caused to fragileeconomies by billions of dollars of volatileshort-term capital stampeding around the globein herd-like movements, or the substantialpoverty-reducing potential of international(unskilled) labor flows from poor to richcountries (even if allowed in temporary andregulated doses).

By poverty, I shall refer to absolute povertyin low-income countries. A large part of the dis-cussion around globalization is around its effecton relative inequality, which we will largelyignore in this paper. In many of these countries,the majority of the poor are in the rural sector,which will be our main focus. While what hap-pens to the urban manufacturing and servicessectors as a result of globalization has attracteda lot of attention, and can have a large impact

139

on the work opportunities of migrants from therural sector and thus their poverty, I shall lar-gely confine myself to the rural sector (bothagricultural and non-agricultural). For exam-ple, the role that globalization may have playedin weakening trade unions and thus the bar-gaining power of organized industrial workersin achieving improvements in their living stan-dards is an important topic, but since suchtrade unions are rare in the rural sector of poorcountries, we shall not discuss this topic here.

In this paper, I mainly provide a brief analyt-ical account of the various processes throughwhich globalization in our sense of the termaffects the lives of the rural poor. In general, Ibelieve that globalization can cause many hard-ships for the poor in these countries, but it alsoopens up opportunities which some countriesutilize and others do not, largely dependingon their domestic political and economic insti-tutions and policies, and the net outcome isoften quite complex and almost always contextdependent, belying the glib pronouncements foror against globalization made in the opposingcamps.

There have been attempts to positively relatetrade liberalization with economic growth, andrelate growth with poverty reduction on thebasis of cross-country regressions. The former

3

1394 WORLD DEVELOPMENT

relation has been found controversial, 1 whilethe latter is sturdier. In any case, there are deepmethodological–econometric flaws in suchcross-country regressions, apart from acuteproblems of reliability and comparability ofthe data for many countries. The results of amore micro-analysis of the impact of trade lib-eralization on total factor productivity growthat the enterprise level are mixed (and scantyfor the rural sector). Even for the relationshipbetween openness and levels of firm productiv-ity, the evidence is quite ambiguous, as can beseen in the survey by Tybout (2000). Whilethe long-run effect of growth on poverty reduc-tion is generally accepted, the usefulness of theaverage estimated value of the elasticity of thiseffect—taken to be 2 in an estimate reported inthe World Development Report 2001, that is, a1% increase in real per capita income has beenassociated with a reduction in the headcountincidence of poverty by 2%—is somewhat lim-ited, as the underlying causal model is under-specified. Also, the value of elasticity variesfrom country to country depending on initialconditions (particularly initial levels of incomeand the extent of social and economic inequal-ity), and, of course, varies a great deal, evenwithin (large) countries.

Most of the general statements one sees inpopular presentations on the impact of global-ization on poverty are essentially those ofcorrelation. Pro-globalizers point to the largedecline in poverty in China, India, and Indone-sia (countries long characterized by massiverural poverty) in the recent decades of interna-tional economic integration. Chen and Raval-lion have estimated that during 1981–2001,the percentage of rural people living below aninternational poverty line of $1.08 per day (at1993 purchasing power parity) declined fromabout 79% 2 to about 27% in China, fromabout 63% to about 42% in India, and 55% to11% in Indonesia. But, contrary to repeatedassertions in the international financial press,no one has yet convincingly demonstrated thatthis decline is mainly due to globalization. InChina it could instead be, to a large extent,due to internal factors like expansion of infra-structure or the massive 1978 land reforms orpolicy changes relating to grain procurementprices or the relaxation of restrictions onrural-to-urban migration. That the spurt inagricultural growth following the 1978 decol-lectivization and land reform may be largelyresponsible for poverty reduction in China issuggested by the fact that the substantial part

of the decline in poverty in the last two decadesalready occurred by mid-1980s, before the bigstrides in foreign trade or investment. 3 Simi-larly, rural poverty reduction in India may beattributable to the spread of Green Revolutionin agriculture, large anti-poverty programs orsocial movements in India, and not the tradeliberalization of the 1990s (in fact as we will dis-cuss later, there is some evidence of trade liber-alization slowing down poverty reduction inIndia). In Indonesia 4 sensible macroeconomicpolicies, an active rice price stabilization policy,massive investment in rural infrastructure, andthe Green Revolution played a substantial rolein the large reduction of rural poverty during1981–2001 (note that by early 1980s the oilboom was largely over and by 2001 the econ-omy had not fully recovered from the financialcrisis).

Those who are more dubious of globalprocesses point out that in the same decades,poverty has remained stubbornly high in sub-Saharan Africa; as Chen and Ravallion (2004)have estimated, during 1981–2001 the percent-age of people 5 living below the poverty lineof $1.08 per day (at 1993 purchasing powerparity) increased in sub-Saharan Africa fromabout 42% to about 46%. But this may havelittle to do with globalization, and more to dowith unstable or failed political regimes, wars,and civil conflicts which afflicted several coun-tries in Africa; if anything, such instability onlyreduced their extent of globalization, as itscared off many foreign investors and traders.

2. THE POOR AS SELF-EMPLOYEDWORKERS

If one goes beyond correlations, the causalprocesses through which international economicintegration can affect poverty primarily involvethe poor in their capacity as workers, as con-sumers, and as recipients of public services orusers of common property resources. Let us firsttake the case of poor workers in the rural sector.They are mainly either self-employed or wageearners. In the rest of this section, I shall discussthe self-employed poor, and the next section willbe on the poor as wage earners and as consum-ers. Section 4 will be on the poor as recipients ofpublic services or users of common property re-sources. Section 5 concludes.

The self-employed work on their own tinyfarms or as artisans and petty entrepreneursin small shops and household enterprises. The

GLOBALIZATION AND RURAL POVERTY 1395

major constraints they usually face are incredit, marketing and insurance, and infra-structure (like roads, power, extension service,and irrigation), and government regulations(involving venal inspectors, insecure landrights, etc.). These often require substantivedomestic policy and governance changes; for-eign traders and investors are not directly toblame. If these changes are not made and theself-employed poor remain constrained, then,of course, it is difficult for them to withstandcompetition from large agri-business or firms(foreign or domestic). Let us just cite two exam-ples. Using panel data for farm households inZambia, Deininger and Olinto (2000) show thatmany households could not reap productivitybenefits from external liberalization becausethey lacked key assets like draft animals andfarm implements. Similarly Lopez, Nash, andStanton (1995) show from panel data of farmhouseholds in Mexico that the supply responseto price incentives is much lower for house-holds with more limited access to capital.Opening the product markets internationallywithout doing anything about the weak or dis-torted factor markets like credit or infrastruc-tural services may thus be a sub-optimalpolicy for many poor farmers and artisans,both from the point of view of their exploitingnew opportunities and of social protection forthose who may need extra help to cope.

Measurement of the direct impact of trade re-form on poverty is actually quite tricky. Apartfrom the scarcity of detailed household data be-fore and after trade reform, it is often difficultto disentangle the effects of trade reform fromthose of other reforms and other events andshocks that affect the household povertydynamics. One of the few attempts to directlyrelate trade liberalization with household pov-erty in the rural sector is by Topalova (forth-coming): she finds that across rural districts inIndia trade liberalization (primarily agricul-tural tariff reduction) has significantly slowedpoverty reduction. Most existing attempts tomeasure are really with simulation models.The study of Litchfield, McCulloch, and Win-ters (2003) is among the first empirical at-tempts, using household survey data for morethan one period in time. For Vietnam, in the1990s, for example, they find in a multinomiallogit model that the trade variables have a po-sitive significant effect on a household’s chanceof escaping poverty.

It is not hard to see that openness to foreigntrade and investment may sometimes help in

relieving some of the bottlenecks in infrastruc-ture and services and in essential parts, compo-nents and other intermediate products likefertilizers and pesticides. Gisselquist and Gre-ther (2000), for example, show how farmers inBangladesh benefited as liberalization increasedthe availability of farm inputs. In a more gen-eral sense, international diffusion of technologyin agriculture, of which the Green Revolutionhas been a dramatic example, has led tolarge reductions in poverty, particularly inAsia, even though the larger dependence offarm households on purchased inputs thatbecame necessary increased the importance ofthe const-raints of credit and irrigation.

Small farms or firms that are not severelyhandicapped by the credit and other constraintsare sometimes more productive than their lar-ger counterparts, and are also sometimes moresuccessful in export markets. Small producersare often heavily involved in exports (for exam-ple, coffee producers of Uganda, rice growers inVietnam, shrimp farmers in coastal Bangladeshor India, garment producers in Bangladesh orCambodia). But in exports, the major hurdlethey face is often due to not more globalizationbut less. Developed country protectionism andsubsidization of farm and food products andsimple manufactures (like textiles and clothing)severely restrict their export prospects for poorcountries. 6 By estimates of the World Bank,based on the widely used GATP (Global TradeAnalysis Project) model, the total income lossesincurred by developing countries on accountof rich-country trade barriers on textiles andapparel amount to about $24 billion. Takingtariffs and tariff equivalent of subsidies inagriculture, Cline (2004) estimates that theoverall protection in agriculture is about 20%for the United States, 46% 7 for the EuropeanUnion, 52% for Canada, and 82% for Japan.The annual loss to developing countries fromagricultural tariffs and subsidies in rich coun-tries is estimated from a static CGE modeland the GATP trade and protection databaseby Cline (2004) to be about $45 billion (andmuch higher if dynamic effects are taken intoaccount).

I wish the anti-global protesters of rich coun-tries turned their energies toward the vestedinterests in their own countries which prolongthis protectionism and cripple the efforts ofthe poor of the world to climb out of theirpoverty. Pro-poor opponents of NAFTA, forexample, point out how competition fromnorthern agri-business is destroying the


livelihoods of small farmers in Mexico, withoutbeing equally vocal about the farm subsidiesand tariffs in the United States and Canada(now going to be even substantially largerunder the new US farm policy) which are, to alarge extent, responsible for this. The US wheatexport prices are estimated to be 46% belowcost of production, the US corn export pricesare at 20% below cost, and so on. 8 It is not sur-prising that the US subsidies in cotton provideda major flashpoint in the breakdown of theWTO’s ministerial negotiations in Cancun inSeptember 2003, as this crop is grown by farm-ers in some of the poorest countries of theworld. Of course, this is not to minimize theresponsibility of domestic governments. InMexico, for example, following the peso crisisof 1994 the government abandoned its plansto phase in the trade liberalization gradually;although the Procampo program providedsome compensation to the very poor farmersagainst the price decline, there was a lack ofpublic support infrastructure to enable thesmall farmers to adjust to new patterns ofproduction necessary to be competitive in thepost-NAFTA world.

Another increasingly important barrier totrade that many small farmers of developingcountries face in world markets is that richcountries now shut out many of these importsunder a whole host of safety and sanitary regu-lations (sometimes imposed under pressurefrom lobbyists of import-competing farms inthose countries). This may actually increasethe importance of the need for involving rich-country transnational companies in marketingpoor-country products. These companies candeal with the regulatory and lobbying machin-ery in rich countries far better than the smallproducers of poor countries can and at thesame time can provide to consumers credibleguarantees of quality and safety. Of course,these companies will charge hefty fees for thismarketing service (usually much larger thanthe total production cost), and sometimes im-pose costs which small farmers find difficultto bear. European supermarkets, for example,now insist on criteria for farmers to satisfy thatinclude health and safety rules, product testing,farm audits, and staff training. It has beenpointed out that farm audits alone cost around$500 per farmer, more than what many farmersearn in the supplying countries in Africa. Insome cases, tighter control by the retail chainsover suppliers to ensure standards and practiceshas led to a drastic decline in the proportion of

exports coming from smallholders—for anexample from the case of Kenyan horticultureexports, see Dolan and Sutherland (2002).

Similarly, it may be very difficult, costly, andtime consuming for small producers of manu-factures or services in developing countries toestablish brand name and reputation in qualityand timely delivery, which are absolutely cru-cial in marketing, particularly in internationalmarkets (much more than comparative costsof production which traditional trade theoryemphasizes). This is where multinational mar-keting chains with global brand names, mediat-ing between domestic suppliers and foreignbuyers, will play a dominant role for a longtime, and small producers can do worse thanpaying the high marketing margin they charge.At the same time coordinated attempts onthe part of developing countries, with technicaland financial assistance from internationalorganizations, to build international qualitycertification institutions and domestic coopera-tive marketing organizations for their productsshould be a high priority.

There is very little hard empirical evidence onthe precise figures of marketing margins. Thereare occasional newspaper reports, for example,that for a 44 lb box of bananas which sell forabout $25 in the US supermarkets, the produc-ers in Ecuador get only $2 or $3. 9 Similarlythere are reports that for a shirt that sells forat least $20 in Gap stores in the United States,the producer in Hong Kong gets less than $1.Of course, much of the difference is made upof transportation, distribution, and inventorycosts, but the marketing margins are likely tobe substantial. Morisset (1998) points out thatthe spread between world and domestic pricesalmost doubled over 1975–94 in all major com-modity markets leading to several billions ofdollars of lost revenue for commodity-export-ing countries. He suggests that the marketpower of international trading companies couldbe the major reason, after showing why changesin trade and tax policies, or factors such astransport, processing, and market costs cannotprovide a systematic explanation. Let us alsogive the examples of two major beverage mar-kets. The coffee market is dominated by fourtransnational retail companies. In the early1990s, the coffee earnings of exporting coun-tries were 10–12 billion dollars, while retailsales were around $30 billion; by 2002, retailsales more than doubled, but coffee-producingcountries received about half their earnings ofa decade earlier. Three companies control more


than 80% of the world tea market. Many in thetea industry in India believe that the cartels ofthe big buying companies push down priceson the tea auction floors; a 2003 report in Delhistates that while the tea price in the retail mar-ket was around Rs. 160 per kg, in the auctionsit was less than Rs. 50 per kg (and while auctionprices have fallen, retail prices of tea continueto rise). In recent years through mergers, acqui-sitions, and business alliances the agri-food cor-porations have concentrated enormous marketpower: companies like Monsanto, Cargill, Nes-tle, and Wal-Mart have come to dominatesupply chains for food and agricultural goods,from seed to supermarket shelfs. Five compa-nies control 90% of the world grain trade; sixcorporations control three-quarters of the glo-bal pesticides market; Wal-Mart controls 40%of Mexico’s retail sector; Nestle has establisheda virtual monopoly of the UHT milk market inPakistan and controls around 80% of Peru’smilk production; DuPont and Monsanto dom-inate the world seed markets for corn (65%)and soya (44%); and so on. 10

Those who are thus justifiably outraged bythe extremely high marketing margins that themonopoly multinational companies currentlycharge the poor producers, their price-fixingcartels, or by their efforts to push out small pro-ducers from the supply chains should agitatemore for anti-trust action and not anti-tradeaction. There should also be more energeticinternational attempts to certify codes againstinternational restrictive business practices andto establish an international anti-trust investi-gation agency, possibly under WTO auspices.Even if such an agency may not have muchenforcement powers, internationally publicizedreports of anti-trust investigations by a recog-nized international body will have some impacton rapacious monopolies, and strengthen thehands of domestic Competition Commissionsin developing countries.

Trade liberalization, even when increasingthe mean incomes of the poor, may heightentheir vulnerability, particularly by increasingthe variance of prices or income sources. Theo-retically, there can be conflicting factors work-ing here, and whether in a particular casevariability increases or not can only be resolvedempirically for different cases. For a briefsummary of the empirical literature on thisquestion, see Winters, McCulloch, and McKay(2004). For example, they cite a study of howtrade liberalization may have helped to mitigatethe post-flood food crisis in Bangladesh in 1998

with private imports stabilizing prices; onthe other hand, they cite evidence from Coted’Ivoire that the ending of domestic marketingarrangements with liberalization may haveincreased the variance of prices. There is, ofcourse, general agreement on the low capacityof the poor to cope with negative price andincome shocks.

There is also the issue of commodity concen-tration of exports. More than 50 developingcountries depend on three or fewer primarycommodities for more than half of their export.Exports of such products are often a curse aswell as a blessing for these countries, as theirprices fluctuate wildly and the economy is toodependent on them. As a result of recent casesof elimination of the erstwhile inefficiently runmarketing boards and the dismantling of waste-ful stabilization schemes, farmers in many Afri-can countries now receive a higher fraction 11

of a more volatile (and in some cases, lower)world market price. 12 International commod-ity agreements among these countries to con-trol their supply in the world market have notworked very well in the past. For reducing theireconomic vulnerability, there is probably notmuch alternative to attempts at diversificationin production and skill formation, and gradualmovement up the supply chain toward activitieswith more value addition for the same com-modity and arranging at an international levelinstitutions of insurance for farmers in poorcountries.

With the opening of the economy, just as ex-port crops face new opportunities, potentiallylifting their producers from poverty, cropswhere the country may lack comparativeadvantage will lose out and push their smallproducers into poverty, if, in a situation of per-vasive failure of credit and insurance markets,there is no vigorous program of public adjust-ment assistance and extension services to helpproducers to reallocate their resources. Thepoor growers of traditional crops are often ill-equipped to shift by themselves to the new com-mercial products like fruits, vegetables, flowers,dairy products, processed foods, etc. Theseproducts require new storage and transportinfrastructure, large setup costs, marketingconnections, and new legal rules and institu-tional structures that can facilitate contractfarming and agro-processing in a way that doesnot expose small producers to exploitation bylarge marketing chains. This is clearly not anargument against globalization but for pro-active public programs to help poor farmers


adjust and coordinate. International agencieswhich preach the benefits of free trade havean obligation to contribute to such programswith financial, organizational, and technicalassistance.

What has been said in the preceding para-graphs about self-employed farmers is also lar-gely valid for those who are self-employed innon-agricultural activities in the rural sector.Some firms adjust well to new trade opportuni-ties, while others find it difficult to cope withthe competition, depending on their initialasset, credit, and other infrastructural condi-tions. Parker, Riopelle, and Steel (1995) in theirstudy of small enterprises in five African coun-tries show that firms that adapted quickly bene-fited from import liberalization, while those ill-prepared to face competition lost out. Whatis called for is therefore liberalization to beaccompanied by a comprehensive policy pack-age for enhancing the capability of latter firmsand a safety net for people who lose in the pro-cess.

In rural industrialization, the most successfulrecent case with a major role of exports andforeign direct investment is, of course, that ofthe township and village enterprises in China,whose phenomenal growth in the 1980s andthe 1990s may have played an important partin the reduction of poverty in China. Exportsof apparel and light manufactures also led toa significant reduction of poverty in Viet-nam—for a measurement of the poverty impacton the basis of a micro-simulation model, seeHertel, Ivanic, Preckel, Cranfield, and Martin(2003). Across states in India, Ravallion andDatt (2002) find that the elasticity of povertyreduction with respect to non-farm outputgrowth varies depending on initial conditions,like literacy or land distribution.

3. THE POOR AS WAGE WORKERS

Turning to poor wage earners, the literatureon how international trade affects the absolutelevel of the real wage or employment of un-skilled workers is extremely small relative tothe one on wage inequality (which, thoughan important issue, is not directly relevantfor my concern with absolute poverty here).Empirically it is hard to disentangle the effectson wages of trade reform from those flowingfrom macroeconomic policy changes or otheron-going deregulatory reforms and technologi-cal changes.

The traditional international trade theorysuggests that the workers in a poor country(presumably having abundant supplies of un-skilled labor) having a comparative advantagein products intensive in unskilled labor shouldbenefit from trade liberalization. The improve-ment in wages and employment of garmentworkers in Bangladesh or Mauritius or Viet-nam with expanding exports is an obviousexample. The matter is, of course, complicatedfor some developing countries (say, Brazil orMexico or Turkey) which may import labor-intensive products from even poorer countries(say, China or Indonesia or Bangladesh), sothat trade, consistent with the traditionaltheory, may lead to lower wages in the formerset of developing countries, for which thereseems to be some evidence. 13 Similarly, if apoor country has large supplies of non-laborfactors of production (like land or mineral re-sources), trade liberalization may not benefitthe labor-intensive sectors.

On the basis of household survey data, Hertelet al. (2003) estimate that global trade liberal-ization leads in the long run (i.e., when laborand capital are mobile across sectors) to a de-cline in poverty for all strata of the population;this is largely because of increased demand forunskilled labor which lifts income even of someof the formerly self-employed who now moveinto the wage labor market. Edmonds andPavcnik (2003) also note how Vietnam’s liberal-ization of rice trade in the 1990s led to a gainfulreallocation of labor of the poor from house-hold occupations to the wage labor market.

In some cases, however, intersectoral mobil-ity is limited for prolonged periods. If some fac-tors of production are intersectorally immobile,and some goods are non-traded, real wage ofan unskilled worker in a poor country maynot go up with trade liberalization even in anotherwise standard model of trade theory. Takea three-good model in a hypothetical Africancountry: one is a non-tradable good (say, a sub-sistence food crop) which is largely grown bywomen who for various social and economicreasons cannot move to other sectors, anothergood (say, an exportable tree crop) producedlargely by men in a capital-intensive way(maybe simply because tree crops lock up cap-ital for a long period), and the third good isan importable (say, processed food) which issomewhat substitutable in consumption forthe subsistence food. In this three-sector model,it is not difficult to show that the real wage ofwomen may go down when the importable pro-


cessed food is made cheaper by trade liberaliza-tion (under the condition that the elasticity ofsubstitution in consumption of the two foodsis sufficiently high). What we have said aboutpoor African women here is equally true forother people anywhere who are mobility con-strained (old workers and people who do nothave the collateral to raise capital to start newventures or move to new sectors, etc.).

It is often suggested that globalization associ-ated with more ‘‘informalization’’ may worsenthe conditions of workers. If large firms facingmore foreign competition and pressure to re-duce costs outsource activities to smaller firmsor household enterprises in the informal sec-tor, 14 the average wage (of those formerly em-ployed in the formal sector) may go down, butthis need not impoverish workers in general ifthe poorer informal workers get more employ-ment this way.

Let us now discuss the case of the poor asconsumers. Whether they gain as consumersfrom trade depends on whether they are netbuyers of tradable goods—for example, thelandless laborers in east or south India whoare net buyers of rice may gain from importsof cheaper rice from Thailand, but may losefrom higher prices of medicine as the Indiandrug market becomes internationalized (withthe laws changing in 2005 from recognizingonly process patents to the internationalproduct patent system under TRIPS), or howmonopolistic is the retail market structurewhich often blocks the pass-through from bor-der prices to domestic prices—for example, inMexico after NAFTA the cartelized tortillasector largely maintained prices even with theavailability of cheaper North American corn.In one of the most disaggregated exercises inthe empirical literature, with the use of Moroc-co’s household survey of living standards and ageneral-equilibrium simulation of trade policychange, Ravallion and Lokshin (2004) showthat liberalization of cereal imports in thatcountry (which does not have a comparativeadvantage in water-intensive cereals produc-tion) leads to a rise in rural poverty, with thelosses to the net producers of cereals outweigh-ing the gains to the net consumers among thepoor.

Whether developing countries are net import-ers or exporters of agricultural products variesa great deal from country to country. FromFAO data sources, Valdes and McCalla(2004) compute that of the 115 low-incomeand low-middle-income countries, 62 are net

agricultural good importing countries, and 53are net agricultural good exporting countries.In general with the expected price rise fromagricultural trade liberalization in the form ofreduction of agricultural tariffs and subsidiesin developed countries, the former set of coun-tries is likely to lose and the latter to gain. Socontrary to the impression one gets from advo-cates of agricultural trade liberalization, manypoor countries will not gain from this liberal-ization. 15 In particular, of the 46 least-devel-oped countries (by the UN classification) 30are net agricultural good importing coun-tries, 16 and it is unlikely that with liberal-ization some of the latter will transformthemselves into large agriculture-exportingcountries. Even in the case of the fewer agricul-ture-exporting least-developed countries manyof them are likely to lose the special preferentialstatus they enjoy under the current regime insome developed markets; for example, manyleast-developed countries in Africa have duty-and quota-free access to the EU market so thatthey currently sell in this market at the high EUinternal prices. This, of course, does not applyto the recently publicized case of poor countriesexporting cotton, as the highest domestic subsi-dies (depressing world price) are in the UnitedStates.

4. THE POOR AS USERS OF PUBLICSERVICES AND COMMON RESOURCES

Let us now briefly turn to the case of the pooras recipients of public services. In the low-income developing countries the poor, parti-cularly those who are in the preponderantinformal sector, do not receive much of effec-tive social protection from the state, but thepublic sector is usually involved in basic ser-vices like education and health and publicworks programs. Cuts in public budgets onthese basic services are often attributed to glo-balization, as the budget cuts to reduce fiscaldeficits often come as part of a package of mac-roeconomic stabilization prescribed by interna-tional agencies like the IMF. Trade reforms canbring about a decline in customs revenue(which is usually a substantial source of totalgovernment revenue in low-income countries)due to tariff cuts, to the extent these are notcompensated by the replacement of the pre-existing quotas by tariffs. But Pritchett andSethi (1994) analyze the experience ofJamaica, Kenya, and Pakistan on their tariff


reductions and found that revenues often fellsubstantially less than tariff rates did. Much de-pends on the nature of customs administration,the degree of complexity of the tariff structure,and the scope for expansion of the revenue basefollowing trade reform.

While there is a lot of scope for improvementin the internationally prescribed (occasionallyideologically blinkered) stabilization programsto minimize their adverse impact on the poor,one should keep in mind that the fiscal deficitsin these poor countries are often brought aboutin the first place more by domestic profligacy inmatters of subsidies to the rich, salaries for thebloated public sector, or military extravaganza.Faced with mounting fiscal deficits, the govern-ments often find it politically easier to cut thepublic expenditures for the voiceless poor(along with public investment programs), andthat is primarily due to the domestic politicalclout of the rich who are disinclined to sharein the necessary fiscal austerity, and it is alwaysconvenient to blame an external agency for aproblem that is essentially domestic in origin.

The low quality and quantity of public ser-vices like education and health in poor coun-tries is not just due to their relatively lowshare in the public budget. To a large extenteven the limited money allocated in the budgetdoes not reach the poor because of all kinds oftop-heavy administrative obstacles and bureau-cratic and political corruption. The develop-ment literature is full of accounts of targetingfailures in social expenditures. 17 Again this isa domestic institutional failure, not largely anexternal problem. The major effort requiredhere is to strengthen the domestic institutionsof accountability.

Apart from basic public services, the poor arealso users of common property resources, thedecline in which is not usually taken into ac-count in the standard estimates of poverty,based as they are on either household surveysof private consumer expenditure or nationalincome accounts. Environmentalists arguethat trade liberalization damages the poor byencouraging overexploitation of the fragileenvironmental resources (forestry, fishery, sur-face and groundwater irrigation, grazing lands,etc.) on which the daily livelihoods of, particu-larly, the rural poor crucially depend. Here alsothe answers are actually complex and meretrade restriction is not the solution. The envi-ronmental effects of trade liberalization on therural economy depend on the crop patternand the methods of production. Take, for

example, an African rural economy where theexportable product is a capital-intensive treecrop (like coffee or cocoa), the import substi-tute is a land-intensive crop (like maize), andthere is a labor-intensive subsistence (non-traded) crop (like roots and tubers). The econ-omy may have a comparative advantage in treecrops. In this case, under a trade protection re-gime an increase in import substitution leads toan expansion of cultivated land under the land-intensive crop as well as a shortening of thefallow period, leading to depletion of naturalvegetation and biomass. Trade liberalizationin this context, through encouraging the pro-duction of the less land-intensive tree crop,can significantly improve the natural biomass,as has been shown by Lopez (2000) for Coted’Ivoire in the latter part of the 1980s, usingthe data from the Living Standards Surveyand some remote sensing data from satelliteimages.

One reason why land-intensive crops maylead to overuse of land and depletion of naturalvegetation (or that expansion of the agricul-tural frontier in general leads to deforestation)is the lack of well-defined property rights orlack of their enforcement in public or commu-nal land. In such cases private cost of expand-ing production is less than the social cost andthere is overuse and degradation of environ-mental resources. If the country exports suchresource-intensive products, foreign trade maymake this misallocation worse. Internationaltrade theorists point out that trade restrictionis not the first-best policy in this situation, cor-recting the property rights regime is (includingcommunity based regulations and coordina-tion). But the latter involves large changes inthe legal regulatory or community institutionalframework which take a long time to imple-ment, and given the threshold effects and irrev-ersibilities in environmental degradation (e.g., aforest regeneration requires a minimum stock),one may not afford to wait. In that case someprogram of (time-bound) trade restriction cou-pled with serious attempts at the overhaulof the domestic institutional framework maybe necessary. In other cases domestic policychanges can be implemented much morequickly, and restricting trade is unnecessaryand undesirable. For example, when coastalshrimp ponds in a shrimp-exporting countrylike India or Bangladesh pollute the watersupply and destroy surrounding mangroves,domestic taxes on the basis of ‘‘polluter pays’’principle are imperative. In some cases domes-


tic government policies are primarily responsi-ble for environmental degradation. For exam-ple, administered under-pricing of preciousenvironmental resources (irrigation water inIndia, energy in Russia, timber concessions inIndonesia and the Philippines, etc.), prolongedby the pressure from powerful political lobbies,is a major cause of resource depletion. Domes-tic vested interests, not globalization, areresponsible for the continuation of such so-cially damaging policies.

In the case of some resource-intensive ex-ports, it is difficult for a country by itself toadopt environmental regulations if its interna-tional competitors do not adopt them at thesame time and have the ability to undercutthe former in international markets. Here thereis an obvious need for coordination, in the envi-ronmental regulation policies of the countriesconcerned. Given the low elasticity of demandfor many resource-intensive primary exportcommodities from developing countries in theworld market, 18 such coordinated policies,while raising prices and the terms of trade neednot lead to a decline in export revenue.

A common charge against multinationalcompanies is that they flock to developingcountry ‘‘pollution havens’’ to take advantageof lax environmental standards. In one of thevery few careful empirical studies on the ques-tion, Eskeland and Harrison (2003) examinethe pattern of foreign investment in Mexico,Venezuela, Morocco, and Cote d’Ivoire. Theyfind no evidence that foreign investment inthese countries is related to pollution abate-ment costs in rich countries. They also find thatwithin a given industry, foreign plants are sig-nificantly more energy-efficient and use cleanertypes of energy compared to their local peers.They find no evidence that foreign investmentin these countries is related to pollution abate-ment costs in rich countries. They also findthat within a given industry foreign plants aresignificantly more energy efficient and use clea-ner types of energy compared to their localpeers.

5. CONCLUSION

In general, the debates on globalization ofteninvolve a clash of counterfactuals. On one side,those who are against the pace of business-as-usual global trade and investment are makinga plea for doing something about the jobs andentrepreneurial opportunities for the poor and

for small enterprises that are being wiped out,and against the monopolistic practices of giantmultinational companies and the environmen-tal damages caused by the economic expansion.So their counterfactual is the world of moresocial justice and less dominant trading andinvestment companies, which gives some morebreathing space to the poor producers andworkers. On the other side, the counterfactualfor pro-globalizers is the case when there isno (or limited) trade or foreign investment, aworld which may be worse for the poor (as itis in the extreme cases of the closed economiesof North Korea and Burma). The way out ofthis clash of counterfactuals is to insist thatthere are policies that may attempt to helpthe poor without necessarily undermining theforces of globalization. In this paper, we haveemphasized that in the medium to long run,globalization need not make the poor muchworse off, if appropriate domestic policies andinstitutions are in place and appropriate co-ordination among the involved parties can beorganized. If the institutional prerequisites canbe managed, globalization opens the door forsome new opportunities even for the poor. Ofcourse, domestic institutional reform is noteasy and it requires political leadership, popu-lar participation, and administrative capacitywhich are often lacking in poor countries.One can only say that if we keep the focus onagitating against multinational companiesand international organizations like the WTO,attention in those countries often gets deflectedfrom the domestic institutional vested interests,and the day of politically challenging them getspostponed. In fact, in some cases, opening theeconomy may unleash forces for such a chal-lenge. So instead of pushing for anti-globaliza-tion policies if the requisite institutions andpolicies are not in place, pushing for a packagethat contains both open-economy policies andthose for support infrastructure and social pro-tection may be more successful (both politicallyand economically).

As in the debates several decades back around‘‘dependency’’ theories in development sociol-ogy, there is often a tendency to attribute muchof the problems of underdevelopment to theinexorable forces of the international economicand political order, ignoring the sway of thedomestic vested interests. In many countriesrural poverty alleviation in the form of expan-sion of credit, marketing, and extension facili-ties, or land reform, or public works programsfor the unemployed, or provision of education,


vocational training, and health need not beblocked by the forces of globalization. This, ofcourse, requires a restructuring of existing bud-get priorities and a better and more accountablepolitical and administrative framework, but theobstacles to these are often largely domestic(particularly in countries where there are somecoherent governance structures in place). Inother words, for these countries, globalizationis often not the main cause of their problems,contrary to the claim of critics of globalization;just as globalization is often not the main solu-tion of these problems, contrary to the claim ofsome over-enthusiastic free traders.

All this, of course, does not absolve theresponsibility of international organizationsand entities in helping the poor of the world,by working toward a reduction of rich-countryprotection on goods produced by the poor, by

energetic anti-trust action to challenge themonopoly power of international (producingand trading) companies based in rich countries,by facilitating international partnerships inresearch and development of products (e.g.,drugs, vaccines, crops) suitable for the poor,and by organizing more substantial (and moreeffectively governed) financial and technologytransfers and international adjustment assis-tance for displaced workers, and help in (legaland technical) capacity building for poorcountries in international negotiations andquality certification organizations. Globaliza-tion should not be allowed to be used, eitherby its critics or by its proponents, as an excusefor inaction on the domestic as well as the inter-national front when the matter involved is thatof relieving the crushing poverty in the lives ofbillions of people in the world.

NOTES

1. See, for example, Rodrik and Rodrıguez (2000).Warner (2003) has in turn refuted some of the criticismsof the earlier literature made by the latter. Wacziarg andWelch (2003) shift the focus from cross-section to time-series and panel analysis and seem to support the viewthat trade liberalization has a positive impact on growth.

2. This figure actually relates to China in 1980.

3. Ravallion and Chen (2004) note that mean tariffrates in China fell only slightly in the 1980s andnon-tariff barriers actually increased, and show econo-metrically that growth in the primary sector (mainlyagriculture) rather than in the secondary or tertiarysectors is largely responsible for the decline in poverty.One of their conclusions: ‘‘our data do not suggest thatexpanding trade can explain China’s progress againstpoverty.’’

4. See, for example, Timmer (2004).

5. This relates to the total population; they do not yethave a separate estimate for rural poverty.

6. This is, of course, not to minimize the trade barriersimposed by developing countries on imports of otherdeveloping countries, which are often higher than thoseimposed by rich countries. There are some conflictingestimates of the welfare gains of the reduction in tradebarriers imposed by developing countries themselves inrelation to that for reduction in trade barriers imposed

by industrial countries. A convincing estimate by Cline(2004) suggests that industrial-country liberalizationprovides from about half to two-thirds of the totalpotential welfare gains to developing countries fromtrade liberalization all around.

7. Adjusting for preferential entry of farm productsfrom some countries, the agricultural protection for theEuropean Union goes down to 34.5%.

8. See, for example, the recent Oxfam Report, Rigged

Rules and Double Standards: Trade, Globalization, and

the Fight against Poverty, 2002.

9. Similarly, there are reports that in the UnitedKingdom for every £1 that shoppers spend on looseEcuadorian bananas, around 40 pence goes to super-markets, while plantation workers receive just 1.5 pence.See www.bananalink.org.uk/tuforum/split.htm. Fivecompanies control over 80% of the global market.

10. Much of the information in this paragraph is froma summary report by Action Aid International (2005).The original sources are cited there.

11. Unless the public monopsony is replaced by privatemarketing cartels.

12. See, for example, Gilbert and Varangis (2003) forthe case of cocoa. For a whole range of crops in Africa,see the analysis in Townsend (1999).

http://www.bananalink.org.uk/tuforum/split.htm


13. This was emphasized by Wood (1997). For detailedevidence from Colombia, see Goldberg and Pavcnik(2005).

14. Attanasio, Goldberg, and Pavcnik (2004) find someevidence that the increase in the size of the informalsector in Colombia toward the end of the 1990s is relatedto increased foreign competition.

15. See Panagariya (2004).

16. In terms of population, roughly one-fifth of thetotal population of these least-developed countries is in

one country, Bangladesh, which is a net importer ofagricultural goods.

17. See, for example, Lanjouw and Ravallion (1999).

18. Repetto (1995) puts together the estimates of worldelasticity of demand for some of the natural resourceintensive export commodities of developing countries.For the eight commercial agricultural commoditiesconsidered by him the absolute value of the elasticitydoes not exceed 0.5. For tropical timber, it is 0.16 fornon-conifer logs, 0.74 for non-conifer sawn wood, and1.14 for non-conifer plywood.

REFERENCES

Action Aid International (2005). Power hungry: Sixreasons to regulate global food corporations. London:Action Aid International.

Attanasio, O., Goldberg, P. K., & Pavcnik, N.(2004). Trade reforms and wage inequality inColombia. Journal of Development Economics,74(2), 331–366.

Chen, S., & Ravallion, M. (2004). How have the world’spoorest fared since the early 1980s? World BankResearch Observer, 19(2), 141–170.

Cline, W. R. (2004). Trade policy and global poverty.Washington, DC: Center for Global Development,and Institute for International Economics.

Deininger, K., & Olinto, P. (2000). Asset distribution,inequality, and growth. World Bank policy researchworking paper 2375. World Bank: Washington, DC.

Dolan, C., & Sutherland, K. (2002). Gender andemployment in the Kenya horticulture value chain.Globalization and poverty discussion paper 8. Uni-versity of East Anglia: Norwich.

Edmonds, E., & Pavcnik, N. (2003). Trade liberalizationand the allocation of labor between households andmarkets in a poor country. Mimeo. New Hampshire:Dartmouth College.

Eskeland, G., & Harrison, A. (2003). Moving to Greenerpastures? Multinationals and the pollution havenhypothesis. Journal of Development Economics, 70(1),1–24.

Gilbert, C. L., & Varangis, P. (2003). Globalization andinternational commodity trade with specific referenceto the West African cocoa producers. NBER work-ing paper 9668. NBER: Cambridge, MA.

Gisselquist, D., & Grether, J.-M. (2000). An argumentfor deregulating the transfer of agricultural technol-ogies to developing countries. World Bank EconomicReview, 14(1), 111–127.

Goldberg, P., & Pavcnik, N. (2005). Trade, wages, andthe political economy of trade protection: Evidencefrom the Colombian trade reform. Journal of Inter-national Economics, 66(1), 75–106.

Hertel, T. W., Ivanic, M., Preckel, P. V., Cranfield, J. A.L., & Martin, W. (2003). Short- versus long-run

implications of trade liberalization for poverty inthree developing countries. American Journal ofAgricultural Economics, 85(5), 1299–1306.

Lanjouw, P., & Ravallion, M. (1999). Benefit incidence,public spending reforms, and the timing of programcapture. World Bank Economic Review, 13(2),257–273.

Litchfield, J., McCulloch, N., & Winters, L. A. (2003).Agricultural trade liberalization and poverty dynam-ics in three developing countries. American Journal ofAgricultural Economics, 85(5), 1285–1291.

Lopez, R. (2000). Trade reform and environmentalexternalities in general equilibrium: Analysis for anarchetype poor tropical country. Environment andDevelopment Economics, 4(4), 337–404.

Lopez, R., Nash, J., & Stanton, J. (1995). Adjustmentand poverty in Mexican agriculture: How farmerswealth affects supply response. World Bank policyresearch working paper 1494. World Bank: Wash-ington, DC.

Morisset, J. (1998). Unfair trade? The increasing gapbetween world and domestic prices in commoditymarkets during the past 25 years. World BankEconomic Review, 12(3), 503–528.

Panagariya, A. (2004). The tide of free trade will notfloat all boats. Financial Times, 3 August.

Parker, R. L., Riopelle, R., & Steel, W. R. (1995). Smallenterprises adjusting to liberalization in five Africancountries. Washington, DC: World Bank.

Pritchett, L., & Sethi, G. (1994). Tariff rates tariffrevenue and tariff reform: Some new facts. WorldBank Economic Review, 8(1), 1–16.

Ravallion, M., & Chen, S. (2004). China’s (uneven)progress against poverty. Unpublished draft.

Ravallion, M., & Datt, G. (2002). Why has economicgrowth been more pro-poor in some states of Indiathan others. Journal of Development Economics,68(2), 401–420.

Ravallion, M., & Lokshin, M. (2004). Gainers and losersfrom trade reform in Morocco. World Bank policyresearch working paper 3368. World Bank: Wash-ington, DC.


Repetto, R. (1995). Trade and sustainable development.In M. G. Quibria (Ed.), Critical issues in Asiandevelopment. Berlin: Oxford University Press.

Rodrik, D., & Rodrıguez, F. (2000). Trade policy andeconomic growth: A skeptic’s guide to the cross-national evidence. In B. Bernanke, & K. S. Rogoff(Eds.), Macroeconomics annual 2000. Cambridge,MA: MIT Press, for NBER.

Timmer, C. P. (2004). The road to pro-poor growth: TheIndonesian experience in regional perspective. Centerfor global development working paper 38. Center forGlobal Development: Washington, DC.

Topalova, P. (forthcoming). Trade liberalization, pov-erty and inequality: Evidence from Indian districts.In A. Harrison (Ed.), Globalization and Poverty.University of Chicago Press: Chicago.

Townsend, R. F. (1999). Agricultural incentives in sub-saharan Africa: Policy challenges. World Banktechnical paper 444. World Bank: Washington, DC.

Tybout, J. R. (2000). Manufacturing firms in developingcountries: How well do they do, and why? Journal ofEconomic Literature, 38(1), 11–44.

Valdes, A., & McCalla, A. F. (2004). Where the interestsof developing countries converge and diverge. In M.D. Ingco, & L. A. Winters (Eds.), Agriculture and thenew trade agenda (pp. 136–150). Cambridge: Cam-bridge University Press.

Wacziarg, R., & Welch, K. H. (2003). Trade liberaliza-tion and growth: New evidence. Stanford University.Unpublished draft.

Warner, A. M. (2003). Once more into the breach:Economic growth and integration. Center for globaldevelopment working paper 34. Center for GlobalDevelopment: Washington, DC.

Winters, L. A., McCulloch, N., & McKay, A. (2004).Trade liberalization and poverty: The evidenceso far. Journal of Economic Literature, 42(1),72–115.

Wood, A. (1997). Openness and wage inequality indeveloping countries: The Latin American challengeto East Asian conventional wisdom. World BankEconomic Review, 11(1), 33–57.

Documents

Globalization and rural poverty