new pressures in labor markets: integrating large emerging … · 2019-08-20 · increase exports....

101

New Pressures in Labor Markets:Integrating Large EmergingEconomies and the GlobalSourcing of Services

4

Rapid technological progress, trade in goods,and international sourcing of services come to-gether to put new pressures in labor markets,pressures that will only become more acute inthe next 25 years. Through these channels,globalization is creating a progressively moreintegrated global market for labor. The impactis tempered by differences in the skills, tech-nology, and know-how available to workers.

Globalization offers opportunities for ex-port growth and access to a wider range ofcheaper imported products that can fuel pro-ductivity growth and rising average livingstandards. But globalization also imposesadjustment costs on certain groups withincountries, primarily through labor markets byinfluencing wages and job security and bydemanding retraining, and the upheaval ofmoving between jobs. The unskilled have seentheir wages worsen relative to skilled workersand their jobs become less secure. This is trueeven in developing countries—contrary to ex-pectations that the unskilled benefit relativeto the skilled as labor-intensive manufacturingmoves to low-wage countries. The projectionsin this report offer little reason to believe thatthis will change in the coming decade.

Two challenges are particularly demanding:one is the rise of China, India, and other emerg-ing economies as manufacturing powerhouses,and the other is the emergence of global sourc-ing of services. While the qualitative implica-tions of increasing exports of manufacturedproducts from India and China are the same as

for the emergence of the Asian tigers, India andChina’s sheer size raises the specter of surgingnew export competition. Many developingcountries fear that exports from these largenew players may swamp their domestic mar-kets, squeeze them out of the global market,foreclose avenues of diversification in manu-factures as a road to higher growth, and gobbleup all the investment flows. And high-incomecountries worry that if the large emergingeconomies can readily acquire and master thenewest technologies, their exports may soontake over high-tech markets.

Global sourcing of services exerts pres-sures in the same direction. The transfer ofrelatively skilled service activities to firms indeveloping countries is putting new pressureson white-collar employment in both the high-income countries and advanced developingcountries. This puts higher-paying and higher-skill jobs at risk in both high- and middle-income countries. Unlike displacement inlow-skilled manufactures trade, services off-shoring has the potential to destroy the previ-ous investments of white-collar workers infirm-specific knowledge.

The analysis here suggests that three fac-tors are likely to mitigate these effects in themedium and long term.

• First, the growth of the Chinese, Indian,and other emerging markets offers enor-mous offsetting opportunities for otherdeveloping and developed countries to

increase exports. As China and India in-crease their exports, they will have toincrease imports of intermediate inputs,energy, technology, and investmentgoods. Driven by China, Asia was theprincipal destination for accelerated ex-ports from Africa and Latin America inthe late 1990s and the early 2000s.

• Second, accompanying the rising value ofexports and domestic living standards inemerging economies will be rising wages.This—together with the inevitable ex-change rate adjustment to the rise inglobal demand for these countries’ prod-ucts and services—will create space forlow-income countries to move into thelowest-skill activities vacated by produc-ers in the large emerging countries.

• Third, developing the social institutionsthat support a dynamic market economyin China and India will take time,providing an opportunity for smaller,more flexible countries to progress fasterin institutional development—and forrich countries to continue to lead inproductivity-enhancing innovation. Theflow of services activities from rich topoor countries, which entails some trans-fer of know-how, will be slowed to theextent that institutional frameworks dis-courage foreign direct investment (FDI)and in particular fail to protect the own-ership of such assets.

The policies that countries adopt willdetermine whether they will be able to takeadvantage of these new opportunities. Effec-tive policy responses will need to positioncountries to harness the opportunities fromglobalization while also addressing the adjust-ment tensions that inevitably arise from theunprecedented magnitude and speed ofchange in labor markets.

Policies to embrace, rather than resist, globalintegration will lay the foundations for futuregrowth and job creation. Openness to tradeand FDI will become ever more critical if thepoorest countries are to absorb technologies

and know-how from abroad and seize the op-portunities created by rising demand from—and production shifts in—India and China.But openness will not foster integration in theabsence of an attractive investment climate,one with sound institutions and policies thatallow labor, capital, and knowledge to flowfrom low-return to high-return sectors. Devel-oping knowledge-intensive activities as futuredrivers of growth will require investing in theinstitutions and policy frameworks that fosterinnovation, and in education and lifelonglearning for all workers. Developing countrieswith wages currently higher than those inChina and India will have to place greater at-tention on their institutions and on educationpolicies to create a climate for greater innova-tion and skill enhancement.

Social policies should focus on protectingworkers rather than protecting jobs. Even in themost propitious policy and institutional envi-ronments, rapid growth, globalization, andlabor-market flexibility are likely to quicken thepace of job creation and job destruction. Thisdemands policies to cushion the adjustmentcosts associated with increased volatility and in-voluntary dislocation. The returns to skilledlabor will continue to increase faster than thoseto unskilled labor, perpetuating a natural wage-widening tendency in many (if not most) coun-tries and underscoring the need for measures tosupport workers at the low end of the scale.Rising wage inequality, together with volatilelabor markets, are heightening insecurityamong workers throughout the world.

The impact of globalization: the story so far

Globalization, coupled with technologicalchange, has driven growth in the world

economy, bringing new employment opportu-nities and enabling millions of people to escapeabsolute poverty. That said, impacts havevaried across and within countries—and not allworkers have benefited equally. While manycountries have seized the opportunities offered

G L O B A L E C O N O M I C P R O S P E C T S 2 0 0 7

102

by greater integration of markets in goods andservices, others, especially in Sub-SaharanAfrica, have remained marginalized. Mean-while demand for skilled labor has increased inboth developed and developing countries andgreater global competition has become associ-ated with a growing sense of insecurity formany workers worldwide.

Product markets are rapidly integrating . . .with a geographical redistribution ofmanufacturingDeveloping countries’ trade has acceleratedover the last few decades. In the markets ofdeveloped countries, the share of developingcountries in imports of manufactured prod-ucts grew from barely 14 percent in 1973 tonearly 40 percent in 2003 (figure 4.1).1

Imports of developing countries have grownjust as quickly as their exports to the rest of theworld (Ghose 2003). Developing-country im-ports grew at about 2 percent per year duringthe 1980s, accelerating to 9.5 percent per yearduring the 1990s (Bhorat and Lundall 2004).

This increased two-way trade reflects thegrowth of outsourcing and global productionchains (see Global Economic Prospects 2003).Enabled by falling barriers to trade and FDI,lower transport and communication costs, andnew technologies, global chains break downgoods into their constituent parts, each pro-duced where it can be done most efficiently andat least cost, whether by an affiliate or by an in-dependent supplier. Ghose (2003) sees a corre-lation between countries’ share of world mer-chandise exports and their share of FDI inflows:between 1982 and 1999, foreign affiliates oftransnational corporations increased their shareof world exports from 31 percent to 45 percent.

At the same time, manufacturing employ-ment has been redistributed between developedand developing countries. While the precisenumbers are debated, the gain in the latter hasbeen much larger than the loss in the former(Sapir 2005; Ghose 2003).2 Overall, employ-ment in manufacturing in developed countriesdeclined from 28.7 percent in 1995 to 24.8 per-cent in 2005, while most developing regionssaw gains (table 4.1).

Not all developing countries have experi-enced gains in manufacturing employment,however. Consider, for example, the striking dif-ferences between East Asia and Latin America.Over the 1990s, employment in manufacturingincreased in China (by just under 15 percentcumulatively), India (by about 38 percent),Malaysia (40 percent), and Thailand (about49 percent). In Latin America, however, aggre-gate manufacturing employment declined overthe 1990s; increases in Chile (about 10 percent)were more than offset by declines in Brazil(about 50 percent) and Argentina (14 percent)(Bhorat and Lundall 2004).3

Services employment has increased inboth developed countries and all developingregions, except the Middle East andNorth Africa, where it has remained the same(table 4.1). Over the 1990s, large increasesin services employment were seen in Brazil(57 percent) and Mexico (62 percent), withsmaller increases in China (about 13 percent)and India (25 percent) (Bhorat and Lundall

N E W P R E S S U R E S I N L A B O R M A R K E T S

103

Figure 4.1 Developed countries’ importsof manufactures increasingly come fromdeveloping countries

0

5

10

15

20

25

30

35

40

45

Percent

Share of high-income countries’ imports ofmanufactures originating in developing countries

Sources: Data from World Bank World Integrated TradeSolution (WITS) and staff calculations.

Africa

Latin America

Middle East

Europe and Central Asia

China

Other Asia

2003199319831973

Note: Excludes trade among EU15 countries.

2004). Also worthy of note are the relativelyhigh levels of female employment in the sector.

To some extent, this increase may reflectchanges in business organization, where func-tions once performed inside manufacturingcompanies are now outsourced to other firmson a contract basis, resulting in their reclassifi-cation as services. This change in businessorganization has also crossed borders, withmultinational companies sourcing activitiesfrom subsidiaries or external firms around theglobe. (The global sourcing of services will berevisited below.)

Globalization has generally beenassociated with rising average wages—butnot all workers are benefiting equally . . .While an economy’s openness to trade and in-vestment is in general associated with fastergrowth of average wages over the longerterm, short-term impacts can vary. Althoughthe initial impact of trade liberalization onwages may be negative in some countries, itbecomes significantly positive over time. ForFDI, the picture is reversed: an initial positiveeffect on wages is reduced to nothing afterfive years. This highlights the importance of

the investment climate—if opening the econ-omy does not attract FDI, potential short-term wage losses from opening the economymay not be offset (World Bank 2002).

While average wages rise more rapidly inopen economies than in closed ones, increasingrelative demand for skilled labor is wideningthe wage gap between skilled and unskilledworkers in both developed and developingcountries.4 The latter is contrary to expecta-tions based on traditional trade theory thatglobalization will increase the relative returnto abundant unskilled labor in poor countries.While available evidence attributes wagewidening primarily to technology, trade is alsoimportant. The relative impacts are hard todisentangle since technology can lead to trade,and technological innovation in turn can be aresponse to increased competition from trade(box 4.1).

A widening wage gap between skilled andunskilled workers is particularly evident in theUnited States, where lighter labor-marketregulation permits faster adjustments inwages. In Europe, where labor markets aremore tightly regulated, the outcome of risingrelative demand for skilled labor has been


104

Table 4.1 Employment in developing countries has shifted out of agriculture intomanufactures and servicesTrends in sectoral shares in employment, 1995–2005 (percent)

Agriculture Industry Services

World region 1995 2005a Change (%) 1995 2005a Change (%) 1995 2005a Change (%)

World 44.4 40.1 �9.7 21.1 21 �0.5 34.5 38.9 12.8East Asia 54.4 49.5 �9.0 25.9 26.1 0.8 19.7 24.4 23.9South East Asia and 55.3 43.3 �21.7 15.4 20.7 34.4 29.3 36 22.9

the PacificSouth Asia 64.1 61.2 �4.5 13.4 14.1 5.2 22.5 24.6 9.3Latin America and the 23.4 17.1 �26.9 20.2 20.3 0.5 56.4 62.5 10.8

CaribbeanMiddle East and 30.8 26.3 �14.6 20.3 25 23.2 48.9 48.7 �0.4

North AfricaSub-Saharan Africa 70.1 63.6 �9.3 8.2 8.9 8.5 21.7 27.5 26.7Developed economies and 5.1 3.7 �27.5 28.7 24.8 �13.6 66.1 71.4 8.0

the European UnionCentral and Eastern 27.9 22.7 �18.6 27.5 27.4 �0.4 44.6 49.9 11.9

Europe and CIS

Source: ILO 2006; Bank staff calculations.CIS: Commonwealth of Independent States.a. Indicative.

reflected in higher unemployment among theunskilled.5 In the United States, increased de-mand for skilled workers since the mid-1980shas led to a relative increase in their employ-ment and wages (Katz and Autor 1999).

Hence, while average U.S. real wages did notchange significantly between the late 1970sand the mid-1990s, real wages of high-wageworkers increased and those of low-wageworkers declined (Helpman 2004)—despite


105

Traditional theory expects trade with low-wagecountries to result in a shift in the composition

of employment toward skilled labor between sectors(as industries expand or contract in response to for-eign competition). In both developed and developingcountries, however, labor composition within sectorshas moved toward skilled labor (also reflected in adramatic increase in their relative wages), suggestingthat technological change has been the major force atwork. Moreover, the sectors shifting toward skilledlabor in developing countries in the 1980s had doneso in the United States in the 1960s, suggesting amigration of technological change from developedto middle-income countries.

Technological change is generally viewed as themost important force in terms of the rising demandfor skilled labor (Krugman 1995), as evidenced by thepositive correlations between technology and growthof employment of skilled workers within industries,and by the fact that increases in the relative wages(cost) of skilled workers have been accompanied byan increase in their relative demand (Helpman 2004).

Trade, by contrast, is a less important force—although estimates vary. Feenstra and Hanson (2003)conclude that the offshoring of manufacturing ac-counts for 15–24 percent of the shift toward moreskilled labor. Anderton and Brenton (1999), on theother hand, find that, when only offshoring to low-wage countries is included, trade may actually accountfor about 40 percent of the rise in the wage bill shareof skilled workers and approximately one-third of theincrease in their employment in the U.K. textiles sec-tor. The OECD (2005d) finds that the average declinein employment in 15 Organisation for Economic Co-operation and Development (OECD) countries was27 percent in industries characterized by high interna-tional competition compared with 16 percent in totalmanufacturing. Evidence from developing-countrystudies also suggests that technology and FDI, rather

Box 4.1 What causes the gap between skilled andunskilled labor—technology or trade?

than trade, is the most important factor in wageinequality—for example in Chile (Reinecke andTorres 2001) and South Africa (Edwards 1999).

Trade plays an important role in disseminatingnew technologies, however. Robbins (1997), for ex-ample, finds that the amount of capital equipmentimported into a subset of developing countries is asignificant factor in raising the demand for tertiary-educated workers relative to demand for those com-pleting only primary school. Moreover, technologicalupgrading can itself be a response to trade competi-tion; there is substantial evidence that firms improveproductivity following competition from imports(Hoekman and Winters 2005). Companies in high-wage countries facing import competition fromlower-cost developing-country suppliers may engagein “defensive innovation,” moving up the valuechain and into more capital-intensive production.They may also respond by offshoring more produc-tion to reduce costs—generally the low-skilled activi-ties, with the high-skilled activities remaining in thehome market (Anderton, Brenton, and Whalley2006).a This can occur in both high-skill-intensiveand low-skill-intensive sectors so that trade withdeveloping countries can thus potentially have animpact on a wide range of sectors—and even withinlow-skill-intensive sectors, the higher-skilled activitiescould still expand. In this sense, offshoring can havethe same effect as technology in reducing the relativedemand for unskilled labor within an industry(Feenstra and Hanson 2003), and reallocating awayfrom high-wage economies to low-wage economies.

aCurrency and exchange rate movements can also prompt ashake-out that leaves only higher technology firms in a sector.Disproportionate increases in offshoring can be seen during largeexchange rate appreciations and the costs and difficulty of rever-sal may see offshoring remain after the currency has stabilized.

the fact that the relative supply of skilledworkers grew over the same period.6

The same decline in the relative wages oflow-skilled workers is found in other devel-oped countries, although to a lesser extent.While the gap increased by 29 percent in theUnited States and 27 percent in the UnitedKingdom (figure 4.2), it was only 15 percentin New Zealand, 14 percent in Italy, and9 percent in Canada (Katz and Autor 1999).

The relative wages of skilled workers in-creased in developing countries in the late1990s, along with a rise in their relativeemployment levels (Majid 2004; Bhorat andLundall 2004). There are several possiblereasons for this.

First, in some developing countries, notably inAfrica, increased demand for low-skilled labordid not lead to wage increases because there wasa large surplus of labor to be absorbed (Ghose2003; Wood 1997; Fox and others 2004). Stud-ies in Latin America and the Caribbean also sug-gest that the increased participation of womenin the labor market may have contributed to awidening wage distribution. In that region,

women’s skills tended to be lower than averageand hence their entry in the 1990s depressedearnings to lower-skilled workers (althoughoverall it raised the income of poor households)(de Ferranti and others 2002).

Second, demand for skilled labor increased(World Bank 2002). The share of skilled work-ers in total employment and the relative de-mand for these workers increased between theearly and late 1990s in a range of countries—including Brazil, Chile, Malaysia, the Philip-pines, Singapore, and Thailand, with Mexicoexperiencing a slight decline. In all cases, in-cluding Mexico, the growth rate of relativewages of skilled workers exceeded that ofunskilled workers.

The demand for skilled labor depends on theskill intensity of the export sector. In many coun-tries in Latin America, increased exports basedon abundant natural resources raised demandfor complementary skilled labor and capital(Perry and Olarreaga 2006). Furthermore, ac-tivities considered relatively low skill in devel-oped countries may nonetheless be relativelyskilled in developing countries, especially inmanufacturing. Transfer of activities consideredrelatively low skill in developed countries to de-veloping countries raises the relative demand forand relative earnings of high-skilled workers inthe latter (Feenstra and Hanson 2003).

Additionally, enhanced competition fromtrade can affect the relative demand forskilled labor by reallocating resources withinsectors away from less productive, unskilled-labor-intensive firms toward more productivefirms using more skilled labor. Trade also facil-itates the transmission of skill-biased techno-logical change. Protection in many developingcountries favored unskilled-labor-intensive sec-tors, hence liberalization led to expansion ofskilled-labor-intensive industries (Perry andOlarreaga 2006). Foreign direct investment canalso increase the demand for skilled labor.In Eastern Europe, for example, privatizationand FDI helped bias employment compositionand relative wages significantly toward skilledlabor as production facilities were upgradedwith new technologies.7


106

Figure 4.2 In many developed countriesthe gap between high- and low-incomeearners has widened

�10

�5

0

5

10

15

20

25

30

United

Sta

tes

United

King

dom

Nethe

rland

s

Franc

e

Norway

Germ

any

Percent change in gap between 90th- and10th-percentile earners, late 1970s to mid-1990s

Percent

Source: Katz and Autor 1999.

Canad

aIta

ly

New Z

ealan

d

Even outside the traded sector, globalizationmay have an impact on low-skilled workers inother ways. In all countries, wages in the non-traded sector may be affected by wages and em-ployment in the traded sector if there is mobilitybetween the two. In addition, international

mobility in the form of migration may also havean impact on low-skilled workers, althoughthere is considerable debate on this issue. Moststudies focus on developed countries and tend tofind small overall impacts—or indeed positiveimpacts—frommigrationonwages (seebox4.2).


107

Even where workers do not experience increasedcompetition from international trade, they may

be affected by an increase in the supply of workersresulting from migration, although this impact maybe positive. As immigration increases labor supply,and wages are reduced, more capital may be attractedand more jobs created. Moreover, consumption bymigrants also increases the overall demand for nativelabor and capital. That said, many studies identifyredistributional effects, with the impact of immigra-tion concentrated on the lowest-skilled workers. Theimpact depends critically on the extent to whichmigrants and natives are substitutes for one another;that is, whether they are competing for the same jobsor operating in segmented markets (for a fullerdiscussion, see Global Economic Prospects 2006).

Borjas (2003) finds negative impacts on workers upto some college level, with immigration harming theemployment prospects and lowering the wage of com-peting native workers (a 10 percent increase in supplyreduces wages by 3–4 percent). The lowest-skilledwere hardest hit: wages fell by 8.9 percent for highschool dropouts, 4.9 percent for college graduates,and 2.6 percent for high school graduates while barelychanging for workers with “some college.” However,this analysis ignores, among other things, the long-runcapital adjustments induced by immigration. If thecapital stock does adjust, overall wages are unaffectedand the loss of wages to high school dropouts is cut tobelow 5 percent (The Economist, “Economics Focus:Myths and Migration,” April 8, 2006).

Borjas, Freeman, and Katz (1997) argue that the ef-fect of immigrant-induced increases in relative laborsupply are strongly concentrated on U.S. workers withless than 12 years of schooling, many of whom work inthe nontraded sectors. Migration increased the supplyof workers with less than high school education by 15 to20 percent over 1980–95, leading to a 27–55 percent

Box 4.2 Workers in the nontraded sector—the role of migration

decline in the relative wages of high school dropoutsover 1980–95 (depending on wage elasticity). Theeffect of immigration is diffused throughout the econ-omy, as natives move in response to immigration.Large immigrant flows to one region may discourageflows to that region of native workers, but may encour-age flows of capital. Ottaviano and Peri (2005) findthat migration increases total employment by 10 per-cent, and increases U.S.-born workers’ wages by3–4 percentage points, largely because U.S.- andforeign-born workers are not perfectly substitutable,even when they have similar observable skills.College graduates, high school graduates, and collegedropouts all gain (about 2.4 percent real wage in-crease), but the low-skilled lose (by the same amount).

Cortes (2005) also sees sizeable wage effects on thelow-skilled, but these are concentrated on other immi-grants, as low-skilled immigrants and low-skillednatives are far from perfect substitutes. A 10 percentincrease in the number of low-skilled immigrants ina city reduced the wages of low-skilled natives by0.6 percent and of low-skilled immigrants by 8 per-cent. But migration also reduces the prices of non-traded goods and services. A 10 percent increase inthe average city’s share of low-skilled immigrants inthe labor force decreases the price of immigrant-intensive services such as housekeeping and gardeningby 1.3 percent, with about 50–80 percent of this neteffect caused by reduction in wages.

Card (2005) argues that the wages of natives withless than a high school education relative to native highschool graduates have remained nearly constant since1980. This is despite immigrant inflows that have in-creased the supply of workers with less than highschool education and despite the growing wage gap be-tween other education groups. Most of the absorptionof unskilled workers occurs in the form of city-specific,within-industry increases in low-skilled intensity.

. . . and workers are feeling less secureIndividuals are concerned not only about thelevel but also the security of their earnings.Greater global competition, along with morerapid technological change and diffusion, canincrease wage and employment volatility,8

although separating the effect of trade fromtechnological change in volatility is difficult.

Labor turnover is high in many countries,fueling individuals’ perceptions of economicinsecurity. Available data show gross sec-toral rates of job creation and destruction ofbetween 5 and 20 percent, adding up to an an-nual job turnover of up to 40 percent in somecountries. A significant part of that turnover(often 30–50 percent) can be traced to theentry and exit of firms, important for outputand productivity growth. About 20 percent offirms are created and destroyed each year inmany countries, involving 10–20 percent ofthe workforce (World Bank 2005). Whileevidence is not uniform, some studies ofOECD countries find that increased trade ex-posure is associated with more labor churning(Hoekman and Winters 2005). Overall it isestimated that 3 to 5 percent of the OECDworkforce experiences involuntary layoff inany given year (Kuhn 2002).

In the United States, more than 7 millionjobs have been destroyed on average everyquarter over the last decade as a result of thenormal functioning of the economy (OECD2005b), matched for the most part by equal orgreater job creation. Among the unemployed,about 45 percent were laid off, 10–15 percentare persons voluntarily between jobs, and theremainder are persons entering or reenteringthe labor market as new job-seekers (Kletzer2001). Voluntary attrition may account for upto two-thirds of employment reduction in theUnited States (OECD 2005d).

Falling transport and communication costsare creating new opportunities for developingcountries to participate in global productionchains by providing specific activities and tasks(see box 4.7). However, discrete activities arelikely to be more footloose than whole sectors,so that while globalization can bring better

job prospects to developing countries, it canalso bring greater volatility and insecurity.Bergin, Feenstra, and Hanson (2006) find thatwhile offshoring of production from theUnited States to Mexico has been an impor-tant source of growth in Mexico, there is ahigh degree of volatility in these activities.Domestic demand shocks in the United Statesare amplified when they are transmitted to theoffshored activities in Mexico. In this way, off-shoring has led the United States to export toMexico some of the employment fluctuationsthat it experiences over the business cycle.

In Latin America, overall labor turnover ishigher than in OECD countries. However,turnover depends on education, per capitaincome, and other demographic and growthvariables. For example, young workers changejobs more frequently than older workers andlower levels of education can imply lower lev-els of firm-specific capital and hence a higherincidence of voluntary separation. Adjustingfor these factors, the region does not showconditionally higher turnover. There is onlymixed evidence that either greater trade liber-alization or exposure to technological changeleads to greater overall turnover in the region;however, to the degree that trade liberalizationhas expanded the share of tradables in totaloutput, it may have led to more churning inthe job market (de Ferranti and others 2000).

To the extent that it reflects labor-marketflexibility and the reallocation of resources tomore productive sectors, increased turnovercan be a sign of healthy adjustment, linkedto further growth and job creation. However,churning can also be negative—for example,where job creation lags well behind jobdestruction, where high turnover lowersworkers’ and employers’ incentives to investin education and training (thus ultimatelyreducing productivity in a sector), or wherechurning results in labor moving into lessproductive sectors. In the absence of socialsafety nets, workers in developing countriesmay be unable to finance job searchesand may be forced into the informal sector(where productivity is generally lower) or


108

into low-productivity, relatively low-growthsectors such as agriculture.

The extent and nature of churning dependon the policy environment. Onerous labor-market regulations and restrictions on entryand exit of firms can discourage firms fromhiring new regular workers, and workers fromsearching for jobs in the formal sector. Theycan also limit the movement of resources outof low-productivity sectors. Overly restrictiveemployment protection (such as restrictionson hiring and firing) tends to have the effect ofprotecting only some workers (insiders, usu-ally prime-age males) at the expense of others(outsiders, usually youth, women, and low-skilled workers). Strict employment protectionis associated with higher income disparitiesand a greater incidence of informal work(World Bank 2005). It also raises the costs ofworkforce reorganizations, thereby reducingincentives for innovation and implementationof new technologies (Arias and others 2005).9

The precise impact of strict employmentprotection on job creation depends on whobears the cost: where wages absorb less of thecost than firms, the disincentives to create em-ployment are greater. In Latin America, firmscan bear up to 50 percent of the cost of non-wage benefits, resulting in reduced wages,greater informality, or both (World Bank2005). In the OECD countries, partial reformshave tended to reinforce labor-market in-equality, with temporary contracts for newentrants (youth or women) but only limitedaccess to more permanent jobs. Strict employ-ment protection is also associated with agreater feeling of insecurity, perhaps becauseworkers realize that their chance of long-termunemployment is higher (OECD 2005d,2004).10 A benefit of globalization is the pres-sure it exerts on institutions that cramp pro-ductivity growth and on governments to de-velop efficient safety nets that cushionworkers from the worst aspects of economicinsecurity, while preserving job creation andflexibility.

Today’s global labor market is characterizedby volatility, shifts in employment between

developed and developing countries, andincreasing wage gaps between low- and high-skilled workers worldwide. What will be theimpact of the key challenges now facing globallabor markets—namely absorption of largeemerging economies and the global sourcingof services?

New challenge I—absorbing large emerging economies into the global market By 2030 China and India together willaccount for about 40 percent of theworld’s workforce, which will remainpredominantly unskilledBy 2030 the world’s labor force will numbersome 4.1 billion workers, 90 percent of whomwill live in the developing world. The globallabor force is predicted to grow by about1 percent per year over 2001–30, with highergrowth in developing countries offset by somecontraction in developed countries (table 4.2).East Asia, the Pacific, and South Asia togetherwill account for just over half the world’sworkforce, with China and India alone repre-senting 40 percent—although China’s laborforce will grow far more slowly than that ofIndia. Sub-Saharan Africa will experience thehighest rate of growth (about 2.4 percent peryear) and will be the third-largest developingregion.

Worldwide, the supply of skilled workers islikely to grow faster than that of unskilledworkers, but the vast majority of the world’sworkforce will remain unskilled in 2030.11 Inthe developing world, rates of growth in thenumber of skilled workers will be highest inSub-Saharan Africa, South Asia, and theMiddle East and North Africa. Given the largepool of unskilled labor, however, these in-creases will raise the share of skilled workersin developing countries’ workforces onlyslightly (from 9.6 percent to 11.3 percent).

There will be significant regional variationsin the developing world. The Middle Eastand North Africa, Latin America and the


109

Caribbean, and Europe and Central Asia con-tinue to have relatively high rates of skilledworkers (30, 21, and 18 percent, respectively),compared to East Asia and the Pacific (9 per-cent), South Asia (8 percent), and Sub-Saharan Africa (7 percent). But in absolutenumbers, India and China each have moreskilled workers than Europe and Central Asiaor Sub-Saharan Africa, and almost as many asthe Middle East and North Africa. Overall,developing countries have more than twice asmany skilled workers as developed countries,even though the proportion of skilled workersin the workforce is four times higher in thedeveloped world.

Agricultural workers will constitute ashrinking share of the world’s labor force,declining from about 43 percent in 2001 toabout 30 percent in 2030. While the share ofagricultural workers will fall by about half indeveloped countries, the stark decline is froman already low base (from 4 to 2.6 percent).The more significant change will occur indeveloping countries, where agriculturalworkers will shift from about 50 percent ofthe workforce in 2001 to 34 percent in 2030.The most notable shifts will occur in Sub-Saharan Africa (61 to 47 percent), East Asiaand the Pacific (62 to 39 percent), and SouthAsia (55 to 35 percent)—with the latter

two driven by large changes in China (67 to42 percent) and India (54 to 34 percent).

Moreover, while average incomes will con-tinue to increase with new opportunities forgrowth, the skill premium—the ratio of skilledwages to unskilled wages—will also increase.Projections from the model developed in chap-ter 2 suggest that the skill premium in develop-ing countries will rise from 3.5 on average in2001 to 4.2 in 2030. In India the premium risesfrom 4.3 to 4.9 in 2030 while in China the in-crease is even larger, from 5.4 to 7.7. Develop-ments in Sub-Saharan Africa are similar, with arise from 5.1 to 6.8. The Middle East and NorthAfrica sees only a modest increase in the skillpremium from 1.3 to 1.5, while the premiumremains constant at 2.2 in Latin America.

Pressures on unskilled workers willintensify in both developed anddeveloping countries . . .Between 1995 and 2005 the global labor force(employed and unemployed) grew by some438 million workers, or 16.8 percent (ILO2006). However, the effective increase in theglobal labor market is considerably larger,because many workers in the emergingeconomies were previously only weakly con-nected to the global economy. Freeman (2005)calculates that the integration of China, India,


110

Table 4.2 In 2030 most workers will be in developing countries and unskilled Growth in the global labor force 2001–30

All workers (millions) Unskilled workers (millions) Skilled workers (millions)

Growth Growth Growth(% per (% per (% per

World region 2001 2030 year) 2001 2030 year) 2001 2030 year)

World total 3,077 4,144 1.03 2,674 3,545 0.98 403 598 1.37

High-income countries 481 459 �0.16 327 276 �0.58 154 183 0.60Developing countries 2,596 3,684 1.21 2,347 3,269 1.15 249 415 1.78East Asia & the Pacific 1,060 1,279 0.65 988 1,163 0.56 71 117 1.70China 773 870 0.41 740 816 0.34 33 54 1.72South Asia 632 1,005 1.62 589 925 1.56 42 81 2.27India 473 712 1.42 441 653 1.36 32 59 2.10Europe & Central Asia 236 233 �0.04 195 192 �0.06 41 41 0.02Middle East & North Africa 119 205 1.88 87 144 1.74 32 61 2.25Sub-Saharan Africa 313 617 2.36 293 573 2.33 20 44 2.74Latin America & the Caribbean 236 345 1.32 194 273 1.19 42 72 1.85

Source: World Bank staff calculations.

and the former Soviet Union has led to a“great doubling” of the global labor force.

The increasingly competitive global marketfor labor may be the most important issuefacing workers worldwide. Freeman (2005)argues that because the workers in these coun-tries brought little capital with them into theglobal labor force there has been a massivedrop in the overall global ratio of capital tolabor. In response to the huge amounts of newlow-wage labor, therefore, capital shouldhemorrhage from rich countries and flow toChina, India, and the ex-Soviet bloc. At themargin, new investment should take place inChina and India, where returns should behighest.

The prognosis from this view is that devel-oping countries with wages higher than those

in China and India risk losing ground follow-ing the entry of these countries into globalcommerce. The sheer size of China and Indiamay also preclude the diversification of thepoorest countries into manufactures and soclose off a route to growth and development(Cline 2006). In rich countries, low-skilledlabor is expected to lose as well, and futuregrowth opportunities will depend on whetherthe rich countries’ comparative advantage inhigh-technology sectors can be maintained.According to this view, it is the quantity ofnew entrants from China and India that risksswamping the global market, undermining theprospects of unskilled workers in all othercountries, both rich and poor. This may not bethe case, however; competition is not alwayswhat it appears (box 4.3).


111

In The World Is Flat, Thomas Friedman (2005) ex-amines the rise of China and India in global supply

chains for both goods and services, describing theincreasing pace and intensity of competition acrossskilled activities as the “flattening” of the globe. But, as Leamer (2006) asks, is flatness the rightmetaphor? What if the world is not flat, but justsmaller?

In the past, geography—physical, cultural, andinformational—had limited competition by creatingcost-advantaged relationships between proximatesellers and buyers. Three revolutionary forces arenow driving a smaller world: (a) the presence ofmore unskilled workers in the global labor marketresulting from liberalizations in China, India, theRussian Federation, and Latin America; (b) newequipment for knowledge workers (the Internet,computers) that has raised productivity, emphasizedtalent, and reduced the need for helpers; and(c) communications innovations that extend thegeographic reach of suppliers and the competitionfor routine work and standardized products. In asmaller world, exchanges are more contested andrelationships between buyers and sellers weaker.In a small world, wages in Los Angeles are set in

Box 4.3 Is the world flat . . . or just smaller?Shanghai. Does everyone now live in a world inwhich distance—physical, linguistic and cultural—nolonger isolates jobs from competition? Is the worldflat or are jobs protected from competition by rela-tionships and geography?

Competition is not always what it appears . . .Smallness may confer a larger market withoutgenerating many new competitors in sectors wherethere are highly localized economies of scale,agglomeration (or cluster) effects, and first-moveradvantages (consider the success of Hollywood inthe global market for cultural products). Whereyou are still matters. Economic activity is dispersingaround the globe, but with very strong clustering tobenefit from agglomeration effects. Commerce stilldeclines dramatically with distance (although cul-tural or linguistic forms of closeness can compen-sate for physical distance), and trade remains aneighborhood phenomenon, close to home bothgeographically and organizationally. Consumerpreferences and trust contribute to this pattern—U.S. Web surfers still favor foreign sites close tothe United States, particularly when financialtransactions are involved.

(continued)


112

Moreover, competition in knowledge products isnot necessarily a win-lose proposition: knowledgeproducts have their value enhanced by the existenceof other products (software is an example). Andnot all work is commoditized and sold in globalmarkets. Most exchanges still rely on long-termrelationships between buyer and seller—relationshipsthat create the language needed to communicate,that establish the trust needed to carry out theexchange, that allow ongoing servicing of implicitor explicit guarantees, and that monitor andenforce the truthfulness of both parties. This isthe difference between negotiated rather than con-testable exchanges. Reliability—and liability—formlimits to the contestability of high-skilled jobs. Todate, global sourcing of intellectual work has beena small drop in a very large bucket, and the devel-oped countries remain extremely well-positioned tocompete in the Internet-based segment of theeconomy.

. . . but competition—or the threat of it—matters for routine tasksGlobal competition is tight for standard tasks forwhich global markets exist, both in manufacturingand services. Movement of jobs is not the onlyindicator of global competition—contestabilitymay be reflected in a deterioration of wages andworking conditions, rather than the movement ofjobs. That is, the possibility of factor mobilitycreates competitive pressure even in the absence ofactual movement. Once this is factored in, the realeffect of contestability—of global competition—ishard to assess.

Innovation is key, but innovation movesaround the world, and its pace is quickeningIdeas stowaway with goods. As manufacturing workmoves to China, so naturally do process innovations—as those closest to production are best placed towork out how to do it better. But will product inno-vation also move? The Internet has increased thespeed and reduced the cost of distributing ideas(subject to the constraints of infrastructure andliteracy). Add the integration of former outsidersthat has increased the size of the global brain by

Box 4.3 (continued)two-thirds, and the pace of innovation in the 21stcentury will be unlike anything previously seen.

Education, infrastructure, and safety nets areessential, but technology guarantees thatinequality will persistGlobal sourcing of services presents issues similarto those posed by manufacturing. The lessons areclear—make the education and infrastructure invest-ments needed to keep high-paying, noncontestable,creative jobs at home, and argue for strong protec-tion of intellectual property rights (IPRs) to preservethe value of knowledge goods sold abroad. But it isimportant to recognize that technology can accentu-ate inequality by magnifying the importance of tal-ent and enabling it to reach a much larger customerbase. Education may help to remedy the income-inequality problems caused by technology, but thereare limits. First, if training is more effective for thetalented, they are likely to receive more of it—andthe amount of training needed to equalize incomesmay be enormous and a great social waste. (Howmuch training does it take to turn a World Bankeconomist into a Pavarotti? And is this a good useof resources?) Second, many jobs involve job-specific tacit knowledge gained only through on-the-job experience. But will workers invest in acquiringthese skills if the job is likely to disappear? Willthe incentives for skill acquisition also disappear?Policies are needed to facilitate the formation oflong-term relationships between workers andemployers and so instill the confidence to makerelationship-specific investments from which greatreturns can flow.

Metaphors matterThe landscape of global competition is not flat, atleast not much of it. The flat plains of open compe-tition for mundane tasks certainly exist, but much ofthe landscape is hills and mountains—where endow-ments, human capital, and policy matter. That land-scape is also constantly changing—today’s hill mightbe tomorrow’s plain, creating new opportunities andobstacles and demanding continual adaptation.

Source: Leamer 2006.

Productivity differences matter. Firms inrich countries combineunskilledworkers inpro-duction with more and better capital and techni-cal know-how than do firms in poor countries.What matters is whether the wage gap is greaterthan the difference in productivity—andwhether productivity differentials can be main-tained. Similarly, the leastdevelopedcountries inAfrica that have lower wages than China andIndia will be able to compete in the globalmarket—but only if their levels of productivityare close to those in India and China. Thesources of productivity differences across coun-tries will be discussed in more detail below.

There is another problem with the viewthat the global market will be swamped byproducts from China and India. The law ofcomparative advantage implies that there willalways be opportunities for other countries toexport, even though China and India willcome to dominate certain sectors. In general,as the global demand for Chinese manufac-tured products increases, dollar-denominatedwages in China will tend to increase, in re-sponse to higher wage demands from Chineseworkers (especially if the rural and urbanlabor markets remain partially segmented)and from the inevitable additional upwardpressure on the yuan.

There is evidence that this process isalready underway (figure 4.3). In 2004, realwages in China were 2.11 times the level of1989, and the rate of wage increase acceler-ated in 2004–05, especially in the coastal re-gions (Yusuf, Nabeshima, and Perkins 2006).In 2005 alone, according to the People’s Bankof China, average wages for Chinese workersrose by 14.8 percent (China Daily, “WorkerShortage Drives Salary Rise,” May 27, 2006).Thus China’s development should not keepthe poorest countries from being able toexport low-skill-intensive products, as long asthese countries can manage to create and sus-tain a business climate that supports invest-ment and trade. In Africa, competitivenessbased on low-cost labor is undermined byhigh indirect costs, with the main barriersbeing corruption, crime, and inadequate

infrastructure (Eifert, Gelb, and Ramachandran2005). The poor business environment leadsto lower returns to labor in production,depressing labor demand and real wages.

Even within sectors where China isexpected to dominate world trade, there areexamples of growing exports of other devel-oping countries. The removal of quotas in theUnited States and the European Union onimports of textiles and clothing products fromChina and India was expected by some todecimate exports of these products from otherdeveloping countries. For example, it was sug-gested that one million jobs would be lost inBangladesh and that half the factories in theindustry in Sri Lanka would close down(Oxfam 2004). However, exports of clothingfrom both of these countries to the UnitedStates have increased since the quotas weredismantled. Sri Lankan exports in 2005 were6 percent higher than in 2004 (with a furthergrowth of 3 percent over the first six monthsof 2006 relative to the same period in 2005).


113

Figure 4.3 Average wages in China haveincreased more than in other countries

30

1995

80

130

180

230

Internationally comparable average wage rates,indexed, 1998 � 100

Sources: China Statistical Yearbook 2005, People’s Bank ofChina, International Labour Organization (Philippines, SouthAfrica), IBGE (Brazil), Banco de Mexico, Ministry ofStatistics and Programme Implementation (India); exchangerates from IMF International Financial Statistics. Wages areaverage wages for China, the Philippines, and South Africa,average private sector wages in Brazil, and manufacturingwages for India and Mexico.

Note: 1998–2000 wages for the Philippines have beenestimated using observed wages from 2001 and projectingthem backward using GDP per capita growth rates.

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

China

Philippines

Mexico

South AfricaIndia

Exports of clothing from Bangladesh to theUnited States increased by 21 percent in 2005and by a further 28 percent over the first sixmonths of 2006.

Nevertheless, the growth of Chineseexports of textiles and clothing has had nega-tive impacts on other countries. Many jobshave been lost in Mexico’s maquiladorasbecause activities in sectors such as clothinghave been unable to compete with China inthe U.S. market. Clothing exports fromAfrican countries have declined substantiallysince 2004, amid reports of substantial lossof jobs in the sector. It is clear, therefore, thatthe emergence of China and India as majorexporters will entail significant adjustment insome sectors in some countries. The adjust-ment costs are likely to be higher in countriesthat offer a less favorable climate for businessand investment and that suffer from morerigidities in product and labor markets.

It should not be forgotten that trade andFDI have contributed to unparalleled reduc-tions in poverty in China and can continue todo so. The poverty rate (people living on lessthan $1 a day) in China fell from almost60 percent in 1980 to 17 percent in 2003.While lifting more than 400 million peopleout of poverty is a remarkable achievement,close to 200 million people still live on lessthan $1 a day, many of whom stand to bene-fit from China’s continued trading strength.

For other countries the impact of the inte-gration of the large emerging economiesshould not be qualitatively different from thepressures that globalization has exerted onlabor markets over the past 30 years, as sum-marized above. Unskilled workers in both richand developing countries are likely to facegreater volatility of employment and continu-ing downward pressure on relative wages.The following section will discuss how policymakers can help to ameliorate these costs.

. . . but opportunities for export andgrowth will remain for all countriesThe entry of large economic entities into theglobal market offers opportunities as enormous

as the challenges it poses for developed and de-veloping countries. The large markets of Chinaand India have changed the dynamic of South-South trade and offer developing countries aroute to decreased dependence on rich countries,whose demand for products produced in thepoorest countrieshasbeenrelatively stagnant foryears. Demand in Asia, and primarily in Chinaand India, has been the main source of the accel-eration in African exports since 1990. Asia alsohas been a key source of recent export growth forLatin America. Overall, China’s share of theworld’s non-oil imports grew from 1.8 percent in1990 to 6.5 percent in 2004, implying substan-tial opportunities for its trading partners to ex-pand exports and create jobs (figure 4.4).

As a result of Asia’s increasing demand forresources there is an increasing correlationbetween growth in China and India andgrowth in developing countries that have acomparative advantage in natural-resource-intensive products (Lederman, Olarreaga,and Soloaga 2006). Even resource-abundantcountries that have not increased exportsto Asia—such as Bolivia, Colombia, and


114

Figure 4.4 China’s imports fromdeveloping countries have surged overthe last two decades

Non-oil imports, $ billions

Sources: WITS and World Bank staff calculations.

01985 1990 1995 2000 2005

20

40

100

140

80

60

120

Africa

Middle Eastand North Africa

South Asia

Latin America

Europe and Central Asia

East Asia and the Pacific

Ecuador—have seen benefits from higherworld prices for their exports. Lederman,Olarreaga, and Soloaga (2006) also findfairly strong complementarity rather thansubstitutability between the exports of Chinaand Latin America to third markets. Theyattribute this complementarity to the growingimportance of production networks and theability of Latin American firms to jointhem, the impact of cheaper imports of inter-mediate inputs on export competitiveness, andlearning by exporting larger amounts toChina. Nevertheless, they suggest that if LatinAmerican and Caribbean countries were torefrain from protectionist policies that preventthem from using cheap inputs from Chinaand India and were to invest more in skills,research and development (R&D), and insti-tutions, they would be able to further exploitopportunities in the new global economy.

The surging demand in Asia for mineralshas been the primary driver of growing South-South trade. But China and India offer hugepotential for a range of other products as well,including agricultural products. However,trade restrictions keep many developing coun-tries from gaining market access for manysuch products.12 For most developing coun-tries multilateral trade negotiations are poten-tially a major route to better access to growingmarkets in Asia and to better prices for tradi-tional exports. The key feature of access tomarkets in Asia for developing countries inAfrica and Latin America is that accessshould occur on a most-favored-nationbasis—that is, each country should be entitledto the best trade terms an importer offers toany nation. Therefore market access is bestaddressed through multilateral trade negotia-tions, so that tariff concessions made byChina and India are immediately available toall developing countries, regardless of theirsize and global importance.

Lower duties in Asia can buoy the exportprospects of other developing countries inthree ways. The first is through lower tariffson products currently exported by these devel-oping countries or that are the focus of

export-diversification efforts. Markets in Asiaare very large, but key products for developingcountries face high tariff barriers. For exam-ple, cocoa beans face applied tariffs of 30 per-cent in India and 8 percent in China (incontrast to zero protection in developed coun-tries). Second, for traditional commodities,even if the reduction in tariffs in Asia does notlead to new exports for a specific developingcountry, there will be a positive effect throughthe impact on world prices. Third, it is im-portant to consider the tariffs on the finalproducts that use resource-intensive inputsexported from developing countries. Reducingsuch protection will expand the demand forthose inputs. It would also reduce tariff esca-lation and ease one of the constraints thatlimit higher value–added activities from beingundertaken in developing countries.

Moreover, both China and India havebecome significant sources of FDI for bothdeveloping and developed countries. India’soutward FDI stock grew from $0.6 billionin 1996 to $5.1 billion in 2003. China andIndia now occupy positions 54 and 72 (out of132 economies) in terms of outward FDI per-formance (UNCTAD 2005).13 About two-thirds of cumulative Indian FDI has gone toother developing countries, but developedcountries (in particular the United States) areimportant markets at the moment. The lead-ing developing country is Mauritius, whichattracts about 10 percent of Indian investmentflows. In the information technology sector,Indian firms’ success in global sourcing ex-posed them to new knowledge and businessmethods from developed-country companiesand induced outward FDI through demonstra-tion and spillover effects. Liberalization ofthe Indian government’s policies on outwardFDI since 2000 also proved critical. Restric-tions on maximum overseas investments as apercentage of net worth have been removed,as have the requirement to obtain prior ap-proval for investments from the Reserve Bankof India and prohibitions against overseasinvestments in the same activity as the com-pany’s core activity in India (UNCTAD 2004).


115

Fears that China and India will quicklydominate high-technology sectors aremisplacedSome worry about the impact of the risingnumbers of skilled workers in China andIndia.14 Freeman (2006) suggests that theincrease in these numbers together with in-creased capacity for technological advance-ment will undermine the advantage that richcountries have in high-tech, high-productivityactivities. Trefler (2005) has put this issueconcerning long-run comparative advantagein the following way: will China and Indiadominate high-tech goods and services to thewest, leaving, for example, “Americans tomend the socks of Chinese business execu-tives”? The prognosis that China and Indiawill swamp the global market not only withlow-skilled-intensive products but also skill-intensive high-tech products is based on theassumption that success in high-tech sectorsdepends on the absolute number of scientistsand engineers rather than the relative numberof such workers in the overall workforce. Thisview also fails to take account of a well-estab-lished literature that identifies the critical im-portance of domestic institutions in drivingand sustaining innovation-based growth.15

Much evidence supports the view thatgrowth and income levels do not dependsolely on the physical amounts of capital andlabor that are available in a country; instead,they depend on how those factors are com-bined in production. Cross-country varia-tions in per capita income cannot be ac-counted for by differences in endowments ofcapital and labor, but by variations in pro-ductivity. For example, in 1988 output perChinese worker was about 6 percent of thatof the typical U.S. worker. Most of that dif-ference was due to lower productivity inChina rather than lower capital per workeror lower levels of human capital. If produc-tivity levels had been the same, output perworker in China would have been more than50 percent of that in the United States (Halland Jones 1999).

Innovations matter. To understand differ-ences in levels of income across countries anddifferences in rates of growth of per capitaincome, it is necessary to explain the sourcesof variations in productivity. Innovation isat the heart of such explanations. In recentmodels of endogenous growth, innovationslead to new products and processes that are tosome extent protected by patents and otherinstitutional mechanisms that return profit tothe innovator and bolster the incentive toinvest. Where protection of the innovationis less than full, a certain amount of“disembodied” knowledge becomes accessibleto other innovators and so adds to the stock ofknowledge available to all, reducing the costsof future research and development (seeHelpman 2004 for a survey).

Some of a country’s R&D effort may thusbe accessed by other countries, even as itaugments the national stock of knowledge. Themain conduits for such technology transferare FDI and trade. In this way the innovativeefforts of rich countries push out the globaltechnology frontier and support the growthof their total factor productivity. Developingcountries, which invest little in R&D, canachieve long-run productivity growth througha process of continually catching up to thetechnology frontier. Policies that attract FDIfrom rich countries, openness to technology-intensive imports, and learning by exportinginto the most demanding markets are crucialfor this catching up. This learning can also beenhanced by temporary movement of people.16

Multinational firms exhibit the highestlevels of total factor productivity and createmore knowledge inputs than other types offirms. Criscuolo, Haskel, and Slaughter (2004)find that globally engaged firms generate moreideas than their purely domestic counterparts,not only because they employ more re-searchers, but also because they have access toa wider pool of knowledge. That pool is deep-ened by contacts with suppliers and customersand, for multinationals, by the intrafirmstock of ideas. Others (Bernard, Knetter, and


116

Slaughter 2004, cited in Criscuolo, Haskel,and Slaughter 2004) find that the parents ofU.S.-based multinationals perform about two-thirds of all private R&D in the United Statesbut are a small fraction of 1 percent of thetotal number of U.S. firms. Thus, it appearsthat openness to trade is important not onlyfor poor countries to absorb new technologiescreated by firms in developed countries, butalso for wealthier countries to stimulate in-vestment and productivity growth.

So do institutions. Even after taking intoaccount innovation efforts, a substantialamount of the variation in per capita incomelevels and growth rates across countriesremains unexplained. What accounts for therest of the variance? Institutions and insti-tutional quality are now accepted as the reasonwhy some countries have higher productivitythan others and why growth rates havediffered across countries, even when factorendowments and rates of innovation aresimilar (Helpman 2004). For example, Halland Jones (1999) conclude that “a country’slong-run economic performance is determinedprimarily by the institutions and governmentpolicies that make up the economicenvironment within which individuals andfirms make investment, create and transferideas, and produce goods and services.”

To compete with the United States, theEuropean Union, and Japan in innovation andhigh-tech products, China and India willrequire institutions similar to those of theOECD countries. The two countries are a longway from having such institutions at present.Moreover, building them takes a long timeand is unlikely to occur within 25 years(Trefler 2005). Thus the United States leads ininnovation-based growth not because it hasmore scientists and engineers, but because ithas an institutional framework that allowscompanies such as Microsoft, Apple, andYahoo to exploit new ideas.

Recent research has highlighted how institu-tional quality can determine comparative

advantage and so influence the commoditystructure of trade. Nunn (2005) shows thatcountries with a good institutional environ-ment for contract enforcement will tend to havea comparative advantage in producing and ex-porting goods that require relationship-specificinvestments.17 Countries with poor contractenforcement will suffer from underinvestmentand thus higher costs of production forgoods that require relationship-specific invest-ment. Such investments are more likely to benecessary in industries in which firms havesome form of firm-specific asset, which in turnare more likely to be high-technology andinnovation-intensive industries. The same isequally true for services.

The structure of all countries’ exports isthus influenced by the nature of domestic in-stitutions. The growth of exports from Chinaand India in some products and services thatrequire relationship-specific investment hasbeen facilitated by having good institutionsin particular enclaves of the economy such asspecial economic zones. The ability of thesecountries to substantially increase exports ofthese goods and services further will dependon the ability to engender economywideinstitutional change, something that will bemuch harder to achieve and will occur moreslowly.

Continual technological innovation andchanges in demand make comparativeadvantage a dynamic concept. It is very diffi-cult to predict in which sectors and taskscountries will be efficient producers. Thirtyyears ago, who could have predicted theemergence of the iPod or known how thevalue added in production and the returnto knowledge would be distributed acrosscountries (box 4.4)?

The opportunities of global productionchains will encourage the upgrading ofdomestic institutions, as countries competeon quality and efficiency as well as price—justas they have for another set of domestic insti-tutions related to labor standards. Ratherthan a race to the bottom, with declining


117

wages and standards as countries compete ontrade and investment, globalization is encour-aging gradually higher labor standards, bothdirectly in terms of attracting FDI and indi-rectly, through higher growth (box 4.5).18

Implications for middle-income countries.While China and India are unlikely to threatenWestern dominance of “big idea” innovation,Puga and Trefler (2005) suggest that the greatcapacity of these countries for lower-levelincremental innovations may have importantimplications for middle-income developingcountries. The presence of many well-trainedscientists and engineers in China and Indiameans that Western firms looking to invest willtend to be attracted to these countries—for theirgreater capacity to assist the firm in incrementalinnovation—rather than to other countries,such as Thailand and Mexico. Puga andTrefler refer to the rapidly increasing number

of U.S.-owned patents with at least one inventorwho is a resident of China or India. Thailandand Mexico have not witnessed such growth.

But Cravino, Lederman, and Olarreaga(2006) find no evidence that FDI by Westernfirms in China and India is displacing FDIin Latin America. In a detailed econometricexercise, Bravo-Ortega and Lederman (2006)find no statistical evidence that current patent-ing activity by China and India has had animpact on the number of patents of LatinAmerican countries. They do find some evi-dence, however, that the stock of patents towhich China and India have contributedis feeding the innovation process in LatinAmerica. In other words, innovators in LatinAmerica can learn from innovations under-taken in China and India.

Thus both economic theory and the avail-able evidence suggest that while the sheer sizeof new entrants into the global economy poses


118

Take just one component of the iPod nano, thecentral microchip provided by the U.S. company

PortalPlayer. The core technology of the chip islicensed from British firm ARM and is modified byPortalPlayer’s programmers in California, WashingtonState, and Hyderabad. PortalPlayer then works withmicrochip design companies in California that sendthe finished design to a “foundry” in Taiwan (China)that produces “wafers” (thin metal disks) imprintedwith hundreds of thousands of chips. The capitalcosts of these foundries can be more than $2.5 mil-lion. These wafers are then cut up into individualdisks and sent elsewhere in Taiwan (China) whereeach one is tested. The chips are then encased inplastic and readied for assembly by Silicon-Warein Taiwan (China) and Amkor in the Republic ofKorea. The finished microchip is then warehousedin Hong Kong (China) before being transported tomainland China where the iPod is assembled.

Working conditions and wages in China are lowrelative to Western standards and levels. Manyworkers live in dormitories and work long hours.

Box 4.4 Global production and the iPodIt is suggested that overtime is compulsory. Never-theless, wages are higher than the average of theregion in which the assembly plants are located andallow for substantial transfers to rural areas andhence contribute to declining rural poverty.PortalPlayer was only established in 1999 buthad revenues in excess of $225 million in 2005.PortalPlayer’s chief executive officer has argued thatthe outsourcing to countries such as India andTaiwan (China) of “non-critical aspects of your busi-ness” has been crucial to the development of the firmand its innovation: “it allows you to become nimblerand spend R&D dollars on core strengths.”

Since 2003, soon after the iPod was launched, theshare price of Apple, the company that produces andsells the iPod, has risen from just over $6 to over$60. Those who own shares in Apple have benefitedfrom the globalization of the iPod.

Sources: C. Joseph, “The iPod’s Incredible Journey,” Mail onSunday, July 15, 2006; “Meet the iPods’s ‘Intel,’” BusinessTrends 32(4)(April), 2006.


119

Looking at the relationship between core laborstandards (freedom of association and collective

bargaining; and elimination of forced labor, childlabor, and discrimination in employment) and trade,the OECD (1996) finds no empirical support forthe view that low-standards countries will enjoygains in export-market shares to the detriment ofhigh-standards countries. There is also no evidencethat low core labor standards are associated withlow unit labor costs: real wages actually grew fasterthan productivity growth in a number of low-standards countries from the mid-1980s to the mid-1990s. While core labor standards will notnecessarily affect comparative advantage negativelyand indeed may have a positive affect, noncore oreconomic standards such as working time and mini-mum wages may affect trade performance negatively(OECD 2000a).a However, the picture is not clear;Dehejia and Samy (2002) find no clear link betweenlabor standards and a country’s competitiveness.Rodrik (1996) finds that labor standards are a signif-icant determinant of labor costs when one controlsfor productivity, but not of comparative advantage,which is mostly determined by factor endowments.

Evidence on FDI also suggests that firms areattracted to countries with higher, not lower, laborstandards (OECD 2000b; Aggarwal 1995; Rodrik1996; Brown, Deardorff, and Stern 2002).b Multina-tionals invest principally in the largest, richest, andmost dynamic markets; with the significant exceptionof China, countries where core labor standards are notrespected receive a very small share of global flows.Even in China, the average foreign affiliate pays wages30 percent higher than the average in state-owned en-terprises and has higher occupational safety andhealth standards than Chinese-owned firms (Lardy2004). Overall, multinational firms provide incentivesto improve, rather than worsen working conditions;pay higher wages than alternative employment; andtend to promote, rather than repress worker rights(Brown, Deardorff, and Stern 2002).

In some countries, labor regulations do not applyand a range of labor standards issues still arise inexport processing zones (EPZs), which now employ

Box 4.5 Does globalization lead to a race to thebottom on labor standards?

around 50 million persons worldwide. That said, themajority of EPZs are covered by national labor laws,and physical conditions and wages tend to be betterthan in the rest of the economy (ILO 1998). EPZswith poor working conditions do not attract long terminvestment—“smart” EPZs have introduced measuresto continuously upgrade labor (OECD 2000b).

Why not a race to the top?Globalization may be forcing a race to the top, as itplaces a new emphasis on speed, efficiency, and qual-ity as well as cost, shifting the focus from cheaplabor to productive labor (ILO 1998; Aggarwal1995). Countries can gain an advantage by improv-ing labor standards. Strengthened core labor stan-dards can increase economic growth and efficiencyby raising skill levels, thereby creating an environ-ment that encourages innovation and higher produc-tivity (Stiglitz 2000; OECD 2000a). At the sametime, in the sectors with a reputation for poor laborstandards (clothing, footwear, and sporting goods)consumers are increasingly demanding products pro-duced under acceptable working conditions, withmonitoring and certification, often by trusted non-governmental organizations.

Indeed, efforts to promote labor standards at aglobal level have been increasing: examples includethe 1998 International Labour Organization (ILO)Declaration on Fundamental Principles and Rights atWork (under which monitoring and reporting oncore labor standards is extended to all members) anddevelopment cooperation programs to reduce childlabor. More controversially, trade agreements havebeen used to promote compliance with labor stan-dards. The United States suspends access under theGeneralized System of Preferences (GSP) in the eventof noncompliance, while the European Union grantsadditional access for compliance. Labor provisionsor side agreements figure in U.S. free trade agree-ments. Links to the World Trade Organization(WTO) have faced strong resistance from developingcountries.c Other market-based mechanisms, such aslabeling schemes or codes of conduct for firms (atthe OECD, ILO, and firm level) have expanded, with

(continued)

a number of challenges to other countries, bothdeveloped and developing, there are enormousopportunities. To grasp these opportunities re-quires that countries have in place a policy en-vironment that allows competitive advantagesto be exploited and the key sources of growthto flourish, while ensuring that those workersadversely affected are assisted in adjusting bymoving to new sectors and/or by augmentingtheir particular skill set. In other words, whileaggregate gains are available to all countries,some industries, firms, and workers will incursome pain. Appropriate policy responses arediscussed further below.

New challenge II—global sourcingof servicesWorkers in previously sheltered servicesface international competition The global competition in goods that has beenunder way for decades is now visible in ser-vices, as falling telecommunications costs andgreater openness to FDI enable different parts


120

of the services value chain to be performed indifferent locations around the globe—a phe-nomenon that has come to be known as “out-sourcing” or “offshoring,” but could perhapsbe most accurately termed “global sourcingof services.”19 Global sourcing has increasedcompetition in services markets for a wide va-riety of activities, from low-skilled functionssuch as data entry, word processing, and callcenters to higher-skilled activities such as soft-ware development, consultancy, medical ser-vices, and R&D. A range of services previouslythought to be nontradable are now being pro-vided electronically over large distances.

Global sourcing allows firms to benefitfrom around-the-clock production (for just-in-time delivery of both goods and services) andlower wage costs. Estimates of the total costsavings from global sourcing vary across awide range—for example, from 15–30 percent(Atkinson 2004) to 30–60 percent (industryestimates cited in Kirkegaard 2005).

While absolute numbers to date are notlarge, growth rates have been high, and globalsourcing of services is expected to grow by

most U.S. Fortune 500 companies now embracingsuch codes (see OECD 2000a; Stern 2003).d

Labor standards rise with income (Stern 2003;OECD 2000a), and the path to higher growth fordeveloping countries lies in seizing the opportunitiesof global production networks in goods and services.But this in turn requires efforts to raise productivityand create a stable and attractive environment forFDI. And the evidence suggests that improving corelabor standards and creating frameworks for soundand stable labor relations can contribute to both ofthese goals, with the potential to create a virtuouscircle of rising wages and standards for workers indeveloping countries.

aBates (2000) distinguishes between core labor standards,which are viewed as fundamental human rights and can createthe framework conditions for the economy to operateefficiently, and developmental or economic labor standards

Box 4.5 (continued)(for example, minimum wages), which will vary dependingon the level of income in a given society.

bData on freedom of association rights in 75 countries thatrepresent virtually all of world trade and all inward and out-ward FDI show no significant deterioration in these rights inany of the 75 countries between 1980 and 1999 (the periodduring which competition for FDI heated up). Data showsignificant improvement in those rights in 17 countries(OECD 2000b).

cMoran (2004) provides a persuasive analysis of thepractical problems of using dispute settlement under tradeagreements to enforce labor standards, given the lack of inter-national agreement on exactly what core labor standards meanand what is required for adequate implementation and thereliance on incomplete, nonrepresentative, noncomparable,and potentially biased sources of information.

dMore recently, attention has shifted beyond core laborstandards to the concept of decent work (work that is freelychosen, provides an income sufficient to satisfy basic economicand family needs, respect for rights and representation, basicsecurity through some form of social protection, and adequateconditions) (ILO 2004).

30 percent per year over 2003–08.20 While de-veloped countries still dominate the trade, somedeveloping countries experienced fast growth inexports of business services between 1994 and2003: nearly 700 percent for India; more than200 percent for China, Brazil, and Argentina;and more than 100 percent for Mauritius,Barbados, and Dominica (figure 4.5).

Sourcing locations expand and change overtime, and technology advances are likely toallow more services to be provided offshore. Ascosts in Ireland rose, activities moved to Indiaand the Philippines. Now, as costs in India rise,other locations, including some in Eastern Eu-rope, are becoming popular (Atkinson 2004).Language patterns influence location deci-sions, but these are not immutable.21 Againstthis backdrop, countries across all regions andlevels of development—from Senegal to SriLanka, Argentina to Zambia—are now seekingto become sites for services sourcing.

The number of service jobs that migratefrom rich to poor countries could be largeEstimates to date of the scale of potential jobmovements from global sourcing of servicesvary widely according to the definitions andmethodology used. A great deal of attentionwas initially given to a report by ForresterResearch (2002) that estimated that 3.3 mil-lion service jobs would move offshore from theUnited States by 2015. However, when putinto perspective—the U.S. economy createsabout 30 million jobs per year—that number isquite small. Even for the job categories deemedto be vulnerable to outsourcing, includingmanagement and computer operations, thepredicted impact amounts to just over 0.5 per-cent of existing employment (see Bhagwati,Panagariya, and Srinivasan 2004).

Blinder (2006) asserts that a much broaderrange of services will be liable to global sourc-ing (almost all of those activities that do notrequire direct personal delivery) as communi-cations costs decline further and technologycontinues to advance. His rough estimate isthat between 28 and 42 million jobs in theUnited States could move overseas as a resultof global sourcing. OECD (2005c) concludesthat close to 20 percent of total employmentcould potentially be affected by informationand communications technology–enabled off-shoring of services.

The potential size of the future market forglobal sourcing of services remains a matter ofdebate, reflecting uncertainties over how thedividing line between tradable and nontrad-able services will shift over time (see box 4.6).

Even if these higher predictions come tofruition they do not imply a correspondinglower level of employment. The impact ofglobal sourcing of services on the overall levelof unemployment in rich countries may besmall, especially in countries that are effectivein generating new jobs.

Global sourcing will benefit bothdeveloped and developing countriesThe shift in jobs resulting from global sourc-ing of services is unlikely to be a zero-sum


121

Figure 4.5 Developing-country exports ofbusiness services are growing rapidly

Growth in export of business services, 1994–2003

0 100 200 400 600500300 700

Source: Data from IMF Balance of Payment Statistics.

Percent

India

Estonia

Romania

China

Israel

Brazil

Argentina

Mauritius

EU15

United States

Canada

Barbados

Dominicaa

Australia

Ghana

Japan

Note: “Business services” are defined as total services minustransportation, travel, and government services.a. 1994–2002.


122

To identify the tradability of industries and occu-pations, Jensen and Kletzer (2005) use indicators

of regional concentration of production in the UnitedStates to group industries into three categories oftradability, leaving a similar number of industries ineach category (the more the geographical concentra-tion, the higher the degree of tradability). They thenuse this degree of tradability of the sector to estimatethe number of jobs that are prone to global sourcing.

Results are very sensitive to assumptions aboutwhat is considered tradable. The first figure belowpresents possible effects for the U.S. economy(excluding public administration), distinguishingbetween the agriculture and manufacturing sector,nonpersonal services, and personal services. Foreach category, the first bar shows the number ofworkers currently employed, while the second barshows the number of jobs that may potentially beglobally sourced under three possible scenarios. Thelowest part considers only the highly concentratedindustries as tradable; here fewer than 4.5 millionjobs in the nonpersonal services sector could belost overseas. In contrast, the middle part of each

Box 4.6 The number of services jobs liable to bemoved abroad: large or small?

second bar adds those jobs that could potentially belost if all jobs in all tradable sectors (including thosewhere production is only relatively geographicallyconcentrated) could be sourced globally. Thischanges the picture dramatically; more than half thejobs in nonpersonal services could be affected byglobalization (about 30 million jobs).

In Jensen and Kletzer’s analysis there is a distinctdifference in the vulnerability to global sourcing be-tween wholesale and retail activities, which are lessconcentrated and therefore deemed less prone toglobal sourcing (about 36 percent of jobs in thesesectors are at risk), and professional services (where71 percent of jobs are at risk). However, technologi-cal change may make global sourcing more relevantto the wholesale and retail activities. The top part ofthe second columns shows that an additional 9.7 mil-lion jobs could be lost if the sensitivity to globalsourcing for professional sectors were equal to thatof the other nonpersonal services (the light gray partin the second column for nonpersonal services). Thesecond figure below shows the same analysis for theEuropean Union (EU15). The figure suggests that the

Jobs and potential for outsourcing, United States

Millions of jobs

0

70

60

50

40

30

20

10

Agricultureand

manufacturing

Nonpersonalservices

Personalservices

Agricultureand

manufacturing

Nonpersonalservices

Personalservices

23

5755

0.84.36.8

3.625.213.5

9.7

Jobs and potential for outsourcing, EU15

Millions of jobs

0

70

60

50

40

30

20

10

36

5762

0.310.1 4.4

21.8 24.42.4

10.3

Sources: Jensen and Kletzer 2005; World Bank staff calculations.

Note: The first, black bar for each sector represents the number of workers currently employed in that sector. The second bar indicates the numberof jobs potentially globally sourced. The black section represents only workers in highly geographically concentrated industries subject to globalsourcing; the other two sections represent workers in less geographically concentrated industries (see text).


123

potential future adjustment for the EU15 is some-what skewed toward the agricultural and manufac-turing sectors, where currently a higher share of totalemployment can be found than in the United States.

A further important caveat to these estimates isthat only certain elements of production in tradablesectors can be sourced overseas. Equally, there couldbe activities in the production process of largely non-tradable services sectors that could be globally

Box 4.6 (continued)sourced. For these reasons it is preferable to identifytasks, rather than sectors, as tradable or nontradable(see box 4.7). Jensen and Kletzer make a crude at-tempt at this by repeating their exercise for broadoccupational groups. They find that 11 percent of totalemployment is represented by tradable occupations inindustries that are classified as nontradable. Similarly,about 22 percent of the total workforce is found innontradable occupations in tradable industries.

game, owing to the presence of significantoffsetting factors. While workers whose jobsare liable to offshoring will face lower labordemand and downward pressure on their rel-ative wages, workers who are complemen-tary to the offshored activities will see a risein their productivity and an increase in rela-tive wages. In addition, global sourcing willaugment the productivity of firms that utilizethe opportunities presented by lower laborcosts overseas. These firms are more likely toexpand than other firms, increasing theirdemand for labor, some of which will befor local tasks that can be fulfilled by thetype of worker affected by offshoring, thusoffsetting, to some extent, the impact ofoffshoring on wages (Grossman and Rossi-Hansberg 2006).

Moreover, demand is not inelastic—lowerwages for software workers in developingcountries raise global demand for software,benefiting all countries. Some OECD coun-tries are experiencing a net inflow of servicejobs from outsourcing (Amiti and Wei2004); investment by foreign companies inthe United States, for example, exceededinvestment by U.S. companies in foreign coun-tries every year over 1996–2001 (Atkinson2004),22 and several OECD countries haveexperienced double-digit growth in exports ofbusiness services. For example, exports grewat 11 percent in both the United States andAustralia (OECD 2005a).23

There is also little evidence to date thattradable service activities have lower employ-ment growth than other service activities,or that net outward investment or imports ofbusiness services are associated with signifi-cant declines in the share of employment po-tentially affected by outsourcing (Jensen andKletzer 2005; OECD 2005a; Amiti and Wei2004). However, growth is lower at the lowerend of the skill distribution—although thismay also indicate that these jobs are mostreadily substituted by technology. Worker dis-placement rates are higher in tradable ser-vices, but affected workers have higher skillsand higher predisplacement earnings thandisplaced manufacturing workers (Jensen andKletzer 2005).

The key labor-market issues raised in richcountries by the global sourcing of services arethe nature of the new jobs that will replacethose transferred overseas, and the difficultiesthat firms and workers may face in adjustingto this new facet of globalization. Workerspreviously sheltered from global competitionare facing greater job insecurity, downwardpressure on their wages, and potential costs ofadjustment in moving from one job to anotheror in upgrading their skills to obtain newemployment following displacement. Theseissues are explored in more detail below, fol-lowing a brief discussion of an appropriateframework in which to assess the nature andimpacts of global sourcing.

Global sourcing of services offersopportunities as well as challengesGlobal sourcing of services is creating consid-erable opportunities for development in poor,low-wage countries, through export possibili-ties and through access to cheaper service in-puts that raise productivity when used in othersectors. Global sourcing is providing impor-tant new employment—in India, employmentin the information technology (IT) sector isnow three million, although this is concen-trated in five or six urban centers (Yusuf,Nabeshima, and Perkins 2006). Employmentcreation is at a wide range of skill levels, re-flecting the range of activities open to globalsourcing. In the relatively low-value segmentssuch as call centers, wage costs are importantdeterminants of location (along with languageskills), and competition is fierce among devel-oping countries. At the high end, global sourc-ing of services may be reducing incentivesfor skilled migration by creating new opportu-nities at home. A large number of those em-ployed as a result of global sourcing arewomen, offering a different route to develop-ment than those based on the growth of agri-culture and manufacturing.

While India and China are likely to cometo dominate the market for global sourcingof services, comparative advantage will en-sure that there are opportunities for manydeveloping countries. Small island economiesin the Caribbean, for example, have beenable to attract certain back office activitiesfrom the United States, such as data entry.The services revolution and global sourcingare offering opportunities for new exportsand for attracting services-related foreigninvestment for a range of poor countries. ITand global sourcing offer new and alterna-tive drivers of development that circumventsome of the key constraints to growth drivenby the expansion of exports of agriculturaland manufactured goods.

This is most apparent for landlocked coun-tries and small (often island) economies thatface very high transport costs. For example,development in Rwanda has to confront an

extremely adverse location, one of the highestpopulation densities in the world, and a highpopulation growth rate. While increasing thequality and quantity of exports of traditionalagricultural exports (coffee) and minerals iscrucial to increases in incomes for the poor inthe short to medium term, the government ofRwanda has identified the provision of IT-intensive services, both locally and abroad, asa base for growth in the long run, to providefor employment and turn the country’s large,but very young, population into a driver ofdevelopment rather than a constraint.

The important new opportunities for devel-oping countries are accompanied by consider-able challenges related to the provision ofnecessary infrastructure, the design and imple-mentation of appropriate regulation, bettereducation to increase the supply of humancapital, and the creation of strong marketingprofiles and reputations for reliability.24 Ac-cess to relatively cheap and reliable electricity,a critical problem for many poor countries,will be necessary. High-quality telecommuni-cations infrastructure must be accompaniedby a competitive framework for the provisionof telecommunications services. Liberalizationof the trade and investment regime, comple-mented by an appropriate and effective regu-latory environment, can help ensure theefficient and competitive provision of thetelecommunications backbone services.

Many developing countries could assist theirnascent IT sectors by joining the InformationTechnology Agreement (ITA) of the WTO.The agreement covers the main categoriesof IT products, computers, telecommunica-tions equipment, semiconductors, semicon-ductor manufacturing equipment, software,and scientific instruments, and commits mem-bers to bind tariffs at zero on these items.Joining the ITA can provide a strong signal toinvestors, both domestic and foreign, of acountry’s commitment to an open IT environ-ment by ensuring access to necessary equip-ment at world prices. The ITA has 43 mem-bers (with the European Union treated asone), among them industrial countries and


124

some large and small developing countries,such as China, the Arab Republic of Egypt, ElSalvador, India, Mauritius, Moldova, andMorocco. As yet, however, none of the leastdeveloped countries is a member.

But does the global sourcing of serviceshave different implications for labor indeveloped and developing markets?Trade in services that use skilled labor inten-sively is not new. In the standard analysisof multinational firms, parent companies indeveloped countries are seen as exporting arange of services such as design, managementand engineering consultancy, marketing, andfinance to their overseas subsidiaries in poorercountries. What is new is trade in the oppositedirection, as services both within multination-als and through arm’s-length trade flow fromlow-wage countries to richer markets.

An immediate implication of this newdevelopment is that the standard factor-endowments model of trade (countries exportgoods and services that make intensive use offactors abundant in their country) cannot ex-plain why skilled-labor-intensive services arebeing exported from countries with veryscarce skilled labor. A common explanationinvolves the absence in developing countriesof the knowledge-based assets that are com-plementary to skilled labor. The lack of theseassets limits the use of skilled labor at homeand keeps such workers cheap even thoughthey are relatively more scarce than in richcountries. Globalization in the form of thetransfer of know-how to complement cheapskilled labor in poor countries leads to tradein skilled-labor-intensive services.

There is much discussion of whether theglobal sourcing of service activities to low-wage countries presents features and issues dif-ferent from those associated with global tradein goods. Bhagwati, Panagariya, and Srini-vasan (2004), for example, argue that globalsourcing of services has effects that are notqualitatively different from those emanatingfrom the sourcing of goods. In both cases, thereare gains from trade and national incomes

rise, but displaced workers face some costs ofadjustment.

For Trefler (2005), by contrast, the fact thatskilled workers lose their jobs when services aresourced from low-wage countries has impor-tant economic implications that do not arisewhen low-skilled jobs disappear. The loss ofrelatively high-wage jobs and the pressure onthe wages of high-skilled workers may reduceeconomic incentives to invest in and to acquireskills. In addition, in knowledge-intensive ser-vice activities skilled workers are more likely tohave obtained some industry- and firm-specificknowledge that is lost when the job is lost. Thismay have a direct negative impact on produc-tivity, especially if the knowledge is comple-mentary to other skills or factors. In developingcountries the opposite will tend to occur. Thetransfer of know-how, the increasing demandfor skilled workers, and the upward pressureon skilled wages will tend to increase the in-centives to acquire skills. This will increasedemands on the education system in develop-ing countries, which in many cases is likely tobecome a constraint on this process.

The rapid pace of change and flexibility de-manded by competitive global markets, alongwith new trends such as global sourcing of ser-vices, will lead to potentially rising adjustmentcosts falling on a wider range of—more highlyskilled—workers. These trends all argue forcountries to review their domestic policy andinstitutional frameworks to ensure that theiradvantages can be exploited and that affectedworkers are supported when they incur ad-justment costs.

Policies to confront the labormarket challenges of globalization

Focusing on factors that determine thegrowth of productivity will be key to con-

fronting the challenges of globalization with-out neutralizing its opportunities. This will re-quire a change of mindset by policy makers,who must grasp and internalize the fundamen-tal changes in the nature of international pro-duction and trade (see box 4.7).


125

In the new environment, productivitygrowth requires openness to new ideas and theability to exploit new technologies and oppor-tunities. Economies need to be sufficientlyflexible to enable resources to move from low-productivity to high-productivity tasks andactivities, which policy makers cannot identifybeforehand. This places a premium on institu-tions and policies that encourage innovation,

investments in human capital, and reductionsof barriers to the flow of knowledge, capital,and labor. This process is not without adjust-ment costs, and complementary policies areneeded to ensure that particular groups insociety do not bear a disproportionate shareof the pain.

The appropriate policy mix that provides aframework for productivity growth will vary


126

In the classic conception, international trade is theexchange of complete goods and services across

national boundaries. Countries gain from specializa-tion in particular sectors of the economy, such astextiles and steel. Within firms, gains are had fromhigher productivity, that is, by allowing workers tospecialize in particular tasks. In the past, effective co-ordination of these efforts, and the combination oftasks to produce a product, required proximity.Communication required physical presence and thetransportation of intermediate inputs was slow andcostly. Specialization led to geographic concentrationof production. International trade occurred if con-sumers lived in another country.

However, the nature of production has changed.Revolutions in transport and communications tech-nologies have led to enormous reductions in cost,allowing tasks to be separated in time and space, andweakening the link between specialization and geo-graphic concentration. Instructions and informationcan be effectively conveyed over long distances andintermediate inputs can be transported quickly andmuch more cheaply than before. Thus, increasingly itis tasks in addition to final goods and services that areexchanged across national boundaries, resulting inglobal production networks of activity in a widerange of sectors. In this new global economy there areadditional gains from specialization, as firms takeadvantage of differences in the cost of labor and skillsacross countries to allocate tasks internationally.

Some tasks can be offshored more easily than oth-ers. What matters is the extent to which a particulartask is contested globally (Leamer 2006). This ismore likely for standard, mundane tasks that can becoordinated through codifiable information and less

Box 4.7 Trading goods and services or trading tasks?likely for complex tasks that require tacit informa-tion. The latter often require relationships and areoften best performed in clusters of individuals. Yourneighbors matter. Even for some mundane tasks,such as mowing the grass, physical presence isrequired.

In this new global environment, interventions thattarget particular sectors will be ineffective relative toinitiatives to provide an environment that supportsactivities and tasks. This entails greater emphasis ona business environment that facilitates the entry andexit of firms across all sectors and policies and infra-structure and regulations that support the free flowand low cost of imported inputs (whether physical orinformation) to which domestic workers can con-tribute their tasks. What matters is the quality ofroads, ports, telecommunications, and electricity to-gether with relatively low tariffs on imported inputsand effective regulation of key backbone services.

Finally, the increasing importance of trade intasks creates a challenge for the measurement ofinternational trade flows. Currently imports of goodsare recorded according to their invoice value as theycross the border and the whole value of the import isattributed to the country in which the last substantialtransformation occurred. There is no system bywhich the countries that contributed value added tothe product are identified. Thus, for example, thevalue of the iPod discussed in box 4.4, when im-ported into the United States, is attributed to China,where it is assembled. Yet most of the value of theiPod is added by tasks undertaken in other countries.

Sources: Grossman and Rossi-Hansberg 2006; Leamer 2006.

over time and according to country character-istics, such as level of development and size.Nevertheless, key ingredients will be opennessto trade and FDI and an attractive climate forinvestment and for innovation, investing in ed-ucation, and repositioning labor-market poli-cies to focus on protecting workers, not jobs.Rich countries have a particular responsibilityto maintain and, indeed, increase the opennessof their markets to goods and services pro-duced in poor countries. A related issue is theimpact that globalization and openness mayhave on a country’s capacity to raise tax rev-enues to fund infrastructure for trade or train-ing for affected workers.

Supporting open access to markets,innovation, and a strong business climate Openness to trade in goods, services, andideas provides a critical stimulus to innovationand productivity growth, both for countriesat the global technology frontier and thosecatching up. But because trade and technologycan lead to lower relative wages and greateremployment volatility for some workers, pol-icy makers are often tempted to meet the chal-lenges of globalization by increasing tradeprotection. Doing so compromises a keysource of growth. As a former finance minis-ter of a developing country that undertooksuccessful reforms stated, “Trade shocks arebetter dealt with through more, rather thanless, trade.”25

Trade policies interact critically with otherelements of the business and investmentclimate. Reaping the benefits of globalizationrequires not only openness to trade and in-vestment but also physical infrastructureand a policy framework that enables actualand potential exporters to effectively exploittheir advantages. High costs of clearingcustoms, poor port infrastructure, weaktelecommunications services, and poor regula-tion, for example, raise costs and hampercompetitiveness. These major challenges fordeveloping countries, particularly the leastdeveloped, must be addressed if trade liberal-ization is to be effective in stimulating trade,

investment, and growth. Domestic policyreforms, underpinned where necessary by in-creased “aid for trade” from the internationalcommunity, will be essential in helping thepoorest countries benefit from the opportuni-ties of new global markets.

In addition, policies that affect innovationand access to technology are crucial. For theleast developed countries, moving up the tech-nology ladder by acquiring technologicalknow-how from overseas through trade andFDI will be a key driver of growth over thenext 20 to 30 years. Innovation and learningwill continue to play essential roles in raisingproductivity and sustaining growth in richcountries and increasingly in the middle-income countries, placing emphasis on theinstitutions that frame incentives to invest inR&D and in the acquisition and applicationof knowledge.26 In the middle-income coun-tries there will be opportunities to be had fromincremental innovations to processes thatimprove the tasks undertaken for foreignfirms and to products that can be tailored forgrowing domestic markets. In the richer coun-tries it is innovation and learning creating newgoods, services, and new processes for pro-ducing them, that will be of greater impor-tance. The key elements of a policy frameworkto support innovation and learning will differbetween countries according to level of devel-opment as well as size but to varying degreeswill include the following:

• Investing in human capital to overcomeshortages of skilled labor, including dueto migration. Learning-by-doing in firmsincreases with its workers’ human capital.

• Supporting public research throughuniversities and research centers, andfacilitating interaction with private busi-nesses to ensure dissemination of “basic”knowledge that stimulates research forcommercially exploitable innovations.

• Defining—and enforcing—adequate intel-lectual property rights to encourage do-mestic investment in innovation and ac-quisition of technology through FDI.27


127

• Promoting access to finance, especiallyfor new entrants and small and mediumenterprises (SMEs), which are morelikely to be innovative (Geroski 1990).

• Careful review of fiscal incentives tostimulate R&D and innovation, takinginto account that evidence of effective-ness is scarce (de Ferranti and others2002) and that there could be crowdingout or in of private investment (Jaumotteand Pain 2005).

Providing more people with lifelonglearningIn all countries investment in education willbecome an ever more critical determinant oflabor-market performance in the context ofgreater global competition and increasingrewards to skills. Higher-skilled workers arebetter at dealing with changes, includingadoption of new technologies, new workforceorganizations, and ongoing pressures for ad-justment and shocks, and also support the cre-ation of well-functioning institutions (WorldBank 2006; Hoekman and Javorcik 2006).28

Countries need to focus not only on enroll-ment in education but also on quality and rel-evance, a fact underlined by the prevalence ofyouth unemployment in both developed anddeveloping countries.

Education systems everywhere face newchallenges, however. In the face of rapidchanges in technology and business organiza-tion, these systems struggle to keep pace withdemand for new skills—a trend likely to be ex-acerbated by global sourcing of services. Forindividual workers, this means rapid changesin the value of their skills, demanding constantretraining and skill upgrading. But workersfacing more rapid obsolescence or devaluationof skills may have lower incentives for skillacquisition. Moreover, firms already facedwith competitive cost pressures and concernsthat they will not benefit from their invest-ment in training as their workers leave forother firms, will increasingly have access to aglobal pool of workers with the desired skillsto substitute for the existing workforce. This

could place greater pressure on the educationsystem to provide the industry-specific skillspreviously provided by firm-level training;while continuing volatility could also place apremium on providing workers with the gen-eral skills that enable continuous adaptation.

In sum, education systems will be expectedto provide more people with more opportuni-ties to learn across a broader menu of educa-tional and skill-development options at morestages of their lives than ever before. This willrequire a new model of education and train-ing, as well as ongoing reform of traditionalmethods, providers, and financing of educa-tion (box 4.8).

Protecting workers, not jobsWhile globalization offers new opportuni-ties for workers, it can also entail greatermovement—for example, between jobs, sec-tors, or regions—and this brings with it addi-tional risk. This calls for policies that shift theemphasis from measures designed to protectthose in employment—which, as discussedearlier, can discourage job creation—to mech-anisms aimed at ameliorating the potentiallynegative effects of greater labor movementthrough targeted labor-market policies andsocial safety nets. While the precise combina-tion of labor-market programs and incomesupport measures will need to be determinedat the national level—taking account of localcircumstances and involving all relevant stake-holders—and there can be important differ-ences in the types of programs that are mosteffective in developing and developed coun-tries, some general lessons can be drawn.

In all countries, income support programswill remain the core of worker assistance. InOECD countries, the redistributive impact ofthe tax-transfer system increased in the late1980s and 1990s (Brenton 2006). In develop-ing countries, the design of such programsraises specific challenges.

• Unemployment benefits can ease adjust-ment and maintain public support forstructural change, but if set too high


128


129

While some of the key challenges relate to prob-lems that education systems have traditionally

faced, such as increasing access to and quality of ed-ucation, others relate to revisiting the nature, type,and purpose of educational offerings to equip a glob-ally competitive workforce. In key respects, tradi-tional educational methods are ill-suited to providingthe lifelong learning that is necessary in the newglobal economy (World Bank 2003).

Increase access. In Sub-Saharan Africa and SouthAsia, more than 40 percent of those aged 25 and overin 2000 had not completed any formal education. Indeveloping countries, public funding—directedthrough public educational institutions or to individu-als (loans or vouchers)—can help expand access(World Bank 2005). Recent policies in Brazil that ad-dressed supply-side constraints in the education systemby establishing a minimum spending level per studenthave proven successful in increasing enrollment ratessubstantially (de Mello and Hoppe 2005). While thecentral government transfers funds to the local govern-ments in case these are unable to finance the prescribedspending levels, demand for education is increased byusing school attendance as a requirement for certaintypes of income transfers to low-income households.

Provide access at all ages. Preschool and earlychildhood programs establish a solid basis for subse-quent learninga while primary and secondary educa-tion give workers the basic skills that enable them tolearn new skills required by technology-inducedchanges (OECD 1996).b Lifelong learning helps work-ers to adjust, but government support (for instance,via a training levy) may be needed (World Bank 2005).In many emerging economies, improving the access tosecondary and postsecondary education will be criticalin view of the rising skill premium.

Improve quality.c Efficient increased spendingshould be combined with strengthened incentives toteach and learn, and with improved accountability(World Bank 2006). Quality assurance mechanisms(including regionally) and national qualificationsframeworks raise standards and facilitate interna-tional recognition.

Focus on learning to learn, equipping workersto learn throughout their working lives, and

Box 4.8 Key challenges for education systems in thenew global economy

continuously upgrade how they produce in whateversector they might be employed (de Ferranti and oth-ers 2002). This means moving away from a model inwhich the teacher is the source of knowledge to asystem where educators function as guides to multi-ple sources of knowledge.

Include a range of providers. Including privatesector as well as public sector providers can pro-mote greater access to education and greater varietyin educational offerings. Additionally, foreign insti-tutions can help upgrade standards, although soundregulation is needed to ensure quality and access,and to provide clear and nondiscriminatory condi-tions for investors. Here again, government mea-sures to ensure broad access may be necessary.

Strengthen the links between education andwork. The mismatch between graduates’ skillsand labor-market needs in many developingcountries argues for greater links between theprivate and public sectors. In the Middle East andNorth Africa, skills geared toward public sector jobsare ill-suited to the needs of industry, while the tra-ditional focus on law, philosophy, and theology ineducation in much of Latin America and theCaribbean is argued to have slowed the developmentof natural resource sectors (de Ferranti and others2002). Postsecondary education should be balancedbetween academic and technical-vocational training(OECD 1996), with the latter assessed, certified,and formally recognized.

aIn developing countries, early childhood and preschool pro-grams show returns of $2–5 for every $1 invested (World Bank2006).

bFor a full discussion of ensuring access and quality in edu-cation, see World Bank (2006). For a discussion of issues in thedelivery of basic education services, see World DevelopmentReport 2004: Making Services Work for Poor People.

cChildren in Argentina, Chile, and Mexico perform about twostandard deviations below children in Greece, one of the poorest-performing countries in the OECD. In reading competence (basedon the OECD’s Programme for International Student Assessment[PISA] 2001), the average Indonesian student performed at thelevel of a French student at the seventh percentile (World Bank2006). More than 20 percent of firms in many developing coun-tries rate inadequate skills and education of workers as a major orsevere obstacle to their operations (World Bank 2005).

and given for too long, can slow downadjustment. In developing countries, in-formality makes targeting of benefitsdifficult as the unemployed may alsohave jobs in the informal sector, and theregistered unemployed may be middle-rather than low-income workers (Hoek-man and Winters 2005).29 However,unemployment insurance may also be analternative source of credit for self-employment. Individual savings ac-counts or similar types of unemploy-ment insurance may be a better solutionfor developing countries, although thereis a risk that workers have insufficientresources.

• Mandatory severance pay is the mostcommon income support program indeveloping countries, as compliance iscomplaint-driven and an expensivebureaucracy is not required. However, ifoverly large, severance pay may discour-age hiring or reforms as costs becomeunmanageable.

• One-off compensation programs havealso been used in both developing (publicsector downsizing) and developed (re-structuring of declining industries) coun-tries. While supporting relatively well-offworkers (the previous beneficiaries ofrents), they are often seen as politicallynecessary for reform—although experi-ence suggests that they have often not suc-ceeded in attaining their stated goals.30

• Wage insurance gives workers a propor-tion of their former wage for a set periodof time, conditional upon their findingnew employment. This eases adjustment,provides an incentive to take a new,albeit lower-paying, job and (in effect)subsidizes on-the-job retraining—themost effective kind (Kletzer 2001).

Global competition and movement ofworkers argues for separating health care fromemployment status. The possibility that in fu-ture more types of workers could experience

periods of unemployment or more frequentjob change argues for new mechanisms to en-sure that they are not left without access to es-sential health services. Moreover, health carebenefits provided by firms are a burden onglobally contestable jobs and make employerswary of forming long-term relationships withprospective employees (Leamer 2006). In de-veloping countries, extending universal basicmedical care not linked to other aspects of for-mality could help to reach the poorest, butrisks increasing incentives for informality(Arias and others 2005). One option is generalhealth provision, funded by tax revenue ratherthan attached to employment; a more modestalternative is the creation of a health insurancesubsidy for displaced workers, as proposed byKletzer (2001).31

Active labor-market programs can be effec-tive in keeping workers in the labor market andupgrading their skills, but experience has beenmixed and programs need to be designed tosuit the conditions in developing countries.32

While there is considerable experience in theOECD countries,33 less is known about poli-cies in developing countries.34 For the latter,the key issues are the size of the informal sec-tor, limited administrative capacity, and the ab-sence of broader social safety nets. Leakagerisks are higher, as the unemployed may alsohave informal jobs and may not be low-income. Policies need to improve conditions inthe informal sector, but avoid creating addi-tional incentives for informality.

In both developed and developing countriessuccessful interventions are comprehensive,oriented to labor demand, linked to real work-places, and carefully targeted (box 4.9). Inter-ventions are also more effective when theeconomy is growing. But longer-term assess-ments are needed (most cover one to twoyears) and a range of effects—deadweight (im-pact would have been achieved in the absenceof the program), substitution (participantssubstitute for nonparticipants in the labormarket and the employment effect is zero),and displacement (firms with subsidized


130

workers displace those without)—need to betaken into account. Given their mixed record,and the challenges of appropriate design, gov-ernments need to be realistic about what ac-tive labor-market policies can achieve.

Globalization may undermine funding for programs to support laborWhile integrating into the world economyrequires that import taxes be kept low andrelatively uniform, for the least developed


131

Employment services. Generally have a positiveimpact on employment and earnings and are cost-effective. But they are of limited use where struc-tural unemployment is high and labor demand low.May be less effective in developing countrieswhere informality is high and implementationcapacity limited.

Public works. Effective, including for informalworkers, as a short-term safety net in developingcountries but do not improve future labor-marketprospects, especially where a stigma is attached toparticipation. Wages need to be sufficiently low totarget those with low incomes and few jobprospects, and projects should also target poorareas. Can be most redistributive, but requiregovernment expenditure.

Training. Can result in higher employment rates,if not earnings. Programs work best with active em-ployer involvement (on-the-job training), and lim-ited evidence from developing and transitioneconomies suggests better results for women thanmen. But firms are reluctant to train lower-skilledworkers; in the OECD countries the least qualifiedare only a quarter to a third as likely as the highlyqualified to participate in job-related training. Agrowing number of countries fund enterprise-basedtraining via compulsory levies (usually 1 percent ofpayroll), with reimbursement based on training pro-vided in some cases (examples include Singaporeand Mauritius).

Retraining after mass layoffs. No positive impact,except with a comprehensive package of employmentservices, and expensive. Workers are often geograph-ically concentrated with industry-specific skills.Best results have been achieved with longer programsthat include some worker contribution to costs.

Box 4.9 Overview of the impact of active labor-market programs

Training for youth. Less successful than earlier in-vestments in the education system. Some success inLatin America with programs that integrate trainingwith remedial education, job search assistance, andsocial services. For example, the “Jovenes” programsof Argentina, Chile, Peru and Uruguay are targetedat disadvantaged youth. They combine training andwork experience with other services, include the pri-vate sector, and are financed by tripartite levy-grantschemes or the government. They have substantialand positive impacts on employment and earnings,but can be small scale, may not be cost-effective, andparticipants can displace other workers.

Employment subsidies. Mostly for disadvantagedgroups, although some countries (Belgium, France, theNetherlands) provide for all low-paid work. A signifi-cant share of overall active labor-market program(ALMP) spending in several OECD countries. Mostdo not have a positive impact and have substantialdeadweight (workers would have been employedwithout the subsidy) and substitution (worker dis-places a nonsubsidized worker) costs.

Microenterprise development and self-employmentassistance. Some evidence of positive impacts forolder and better-educated workers, but take-up islow and business failure rate is high.

Sources: Betcherman, Olivas, and Dar 2004; World Bank 2005;Rama 2003; Arias and others 2005; OECD 2005a, 2005d;Heckman and Pagés 2000.

Note: Betcherman, Olivas, and Dar (2004) build on an earlierWorld Bank study of 72 scientific (that is, using a controlgroup) evaluations of ALMPs by Dar and Tzannatos (1999) byadding 87 new studies, 39 of which cover programs in develop-ing and transition economies. Similar conclusions on a numberof points have been made in OECD reviews (see Martin 2000).

countries they are a key source of revenue rel-ative to value-added tax (VAT) and sales taxes(figure 4.6). High-income countries are able torecover revenues lost from trade liberalizationfrom other sources: on average, middle-income countries recover 45–60 percent of losttariff revenues while least developed countriesrecover less than 30 percent (Baunsgaard andKeen 2005).

But many countries collect far less in tariffrevenue than the applied tax rates would sug-gest, owing to the widespread (discretionary)granting of exemptions. Exemptions make thetax regime opaque and difficult to administerand can lead to a distorted incentive structurethat discriminates against small firms with lessinfluence. Further, there is little evidence thatexemptions have a significant impact on in-vestment, their primary justification. Manycountries could substantially reduce appliedtariffs while maintaining or even increasingrevenue if exemptions were removed and col-lection improved. However, it is still necessaryto address the development challenge high-lighted in figure 4.6 of moving from easy-to-collect trade taxes to harder-to-collectconsumption and income taxes.35 Simplyimplementing a VAT is not sufficient; a highdegree of collection efficiency (the ratio ofactual to potential revenues) is needed.36

While feelings of greater economic insecu-rity among workers may lead to greater de-mands for social insurance, it has been arguedthat globalization is limiting the capacity ofgovernments to fund such protection—and tosupport the productivity-enhancing measuresdiscussed above. The fear is that globalizationand greater mobility of capital and wealthyworkers will undermine the tax base, as thesefactors move to the lowest tax locations, com-promising social welfare programs for thosebearing the burden of adjustment. There is littleevidence of this process in developed countries(OECD 2000b), however, and the preferencesof wealthy individuals for education, health,law and order, and social welfare suggest thatfears of large numbers of skilled workers emi-grating to low tax havens are unlikely to be jus-tified. Moreover, much of the knowledge ofthese workers is gained and creates value frominteractions and synergies with clusters of othersimilar workers. As Leamer (2006) says, itmatters who your neighbors are.

The international community—workingtogether—can help realize the potential of globalization The rise of China, India, and other emergingeconomies amounts to a huge increase in thesupply of unskilled labor on a global scale.This could heighten the existing—and grow-ing—inequality between skilled and unskilledworkers in both developed and developingcountries that has resulted from a mix of tech-nology and trade effects. There are fears thatthere will be no space for other countries, par-ticularly developing countries, to compete inlow-wage exports or in attracting global capi-tal flows. In addition, because China andIndia are rapidly upgrading the skills of theirworkforce, and services activities along thevalue chain are being globally sourced, skilledworkers everywhere are increasingly facingcompetitive pressure. There are mitigatingforces, however; China and India offer hugemarkets for the exports of other countries,their own wages are bound to grow rapidly,and developing the full range of institutions


132

Figure 4.6 Low-income countries dependheavily on import duties for tax revenues

0

1

2

3

4

5

6

7

Low-in

com

e

Lower

-midd

le-

incom

e

Upper

-midd

le-

incom

e

High-in

com

e

Percent of GDP

Source: World Bank staff calculations.

VAT, sales, orturnover taxes

Importduties

needed to underpin a modern, dynamic mar-ket economy will take time.

This new global climate poses challenges—but also offers considerable opportunities—forall countries. The countries best placed to ad-dress the challenges will be those best able toseize the opportunities and generate the newsources of growth and wealth needed to fi-nance additional investments in social safetynets and education. In many developing coun-tries, domestic reforms to reduce rigidities inthe labor market and improve the climate forbusiness and innovation will be critical. Allcountries will need better mechanisms to cush-ion adjustment costs and distribute the benefitsof growth to offset the tendencies toward in-equality and volatility. But collective action bythe international community will also beneeded in two important respects. First, as thepressures of globalization increase the calls forbeggar-thy-neighbor protectionism, the inter-national community will need to band togetherto preserve and extend the open markets thathave underpinned recent advances in growthand poverty reduction. In the absence of openglobal markets, many of the new opportunitiesfrom the coming globalization will disappear.Second, the international community needs toprovide the financial and technical support—the “aid for trade”—to enable the poorestcountries to overcome the infrastructure andcapacity constraints that limit their ability totake advantage of new trade opportunities.

Notes1. Trade among developing countries is also grow-

ing significantly; South-South trade now constitutesone-quarter of developing-country exports, and thistrade is growing 50 percent faster than world trade.

2. A recent U.S. study (Brofenbrenner and Luce2004) concludes that the Bureau of Labor Statistics(BLS) grossly underestimates the total number of jobslost to global production shifts. While the BLS re-ported 4,633 private sector workers in establishmentswith 50 or more employees who lost their jobs be-cause of global outsourcing from January to March2004, the authors, drawing on media reports, findevidence of a minimum of 25,000 jobs lost over thatperiod. Moves were often to several destinations

simultaneously and most were in manufacturing, al-though there was a significant increase in shifts ofwhite-collar-services jobs to India.

3. Ghose (2003) identifies a group of manufacturing-exporting developing countries that have shown im-pressive growth in employment in the sector, fromabout 50.9 million in 1980 to 82.8 million in 1997 (orfrom 79 to 88.7 percent of total employment over thesame period). In this group are China, the ArabRepublic of Egypt, India, Indonesia, Israel, theRepublic of Korea, Malaysia, Mauritius, Morocco, thePhilippines, Singapore, Sri Lanka, Taiwan (China),Thailand, and Turkey. However, other manufacturing-exporting countries have witnessed declines of 13.5 to10.6 million (21 to 11.3 percent) over the same period—this group consists of Argentina, Brazil, Hong Kong(China), Malta, Mexico, Pakistan, and South Africa.Note that Mexico’s figures do not include themaquiladora; if they did, the country would haveappeared in the first group.

4. A further question is the distribution of gains be-tween labor and capital. However, this issue hasproven difficult to analyze and beyond the immediatescope of this chapter, which focuses on the distributionwithin labor markets between skilled and unskilledworkers.

5. Whether the impacts of greater openness operatemore or less through wages as opposed to employmentdepends on labor-market institutions, the efficiency ofcapital markets, and social policies. Hence in the UnitedStates, the more flexible labor market and more effi-cient financial sector mean that wages bear a greatershare of shocks than in the European Union. In devel-oping countries, it also appears that wage responses aregreater than employment impacts, suggestive both oflabor-market rigidities and industry rents engenderedby trade policy (Hoekman and Winters 2005).

6. For full-time workers in the United States be-tween 1979 and 1995, real wages of those with 12 yearsof education fell by 13.4 percent, and real wages of thosewith less than 12 years of education fell by 20.2 percent.During the same period, real wages of workers with 16or more years of education rose by 3.4 percent so thatthe wage gap between these groups grew significantly(Feenstra and Hanson 2003).

7. Some of the overall increases in the skill pre-mium could also be related to the artificially low pricesfor skilled labor prior to opening. Modest increases arealso found in China and Vietnam (World Bank 2002).

8. Economic literature has mainly focused on theimpact of globalization on labor demand elasticities. Au-thors such as Rodrik (1997) argue that globalizationhas led to an increase in the labor demand elasticity, withthe result that changes in product prices now havemagnified impacts on wages and employment. The


133

empirical support for this link is mixed, although esti-mating elasticities is prone to difficulties. Rodrik (1997)finds that the interaction between trade openness andvariation in a country’s terms of trade is positively linkedwith volatility of growth and government expenditures.The latter, Rodrik argues, reflects the increasing demandfor social protection as globalization increases insecu-rity. On the other hand, Iversen and Cusack (2000)argue that what is required is to show that volatilityfrom international markets is greater than that in do-mestic markets; they find that in developed countriesthere is no correlation between trade openness and out-put, earnings, or employment volatility. Others havetried to link trade liberalization to changes in labor de-mand elasticities. Slaughter (2001) finds that labour de-mand elasticities for low-skilled workers in the UnitedStates have increased over time but with no clear link totrade variables. Fajnzylber and Maloney (2005), for aset of Latin American countries, and Krishna, Mitra,and Chinoy (2001) for Turkey do not find strong sup-port for this link. On the other hand, Hasan, Mitra, andRamaswamy (2003) find a strong link between trade re-form and wage and employment volatility in India.Nevertheless, even though a positive link between glob-alization and observed measures of volatility has notbeen found, globalization may still have contributed togreater risks and to heightened economic insecurity(Scheve and Slaughter 2002).

9. Data for 19 developed and developingeconomies suggest that flexible hiring and firing rulesare positively associated with higher rates of entry ofnew firms, which are often better at harnessing newtechnologies (World Bank 2005).

10. Alternatively, of course, it could be argued thatstrict employment protection is a response to thehigher level of anxiety of workers in these countries. Ifthat is the case, however, one would also have toconclude that protection has not been very effective inreducing that anxiety.

11. Skilled workers are considered to be those withsome secondary education, plus those with secondaryeducation or above. This selection does not take intoaccount the quality of the education received orcomparability among countries.

12. This is notwithstanding the fact that trade lib-eralization in India and especially China has advancedgreatly over the last 15 years, in China driven in partby World Trade Organization (WTO) accession and inIndia by unilateral reforms.

13. It should be noted that Hong Kong (China) isthe world’s third-largest outward investor, with flowsof about $40 billion in 2004 (UNCTAD 2005).

14. There is controversy around the exact numberof graduates from Chinese and Indian institutions. Forexample, a recent study by Duke University indicated

that engineering graduates from Chinese universitiesnumbered only 351,000 per year as opposed to previ-ous estimates of over 600,000. Note that the new num-ber is still two and a half times as many graduates as inthe United States; however, China’s population is fourtimes as large.

15. It is interesting to note that the Soviet Unionovertook the United States in the number of researchworkers (Nolting and Feshbach 1980) but did not suc-ceed in achieving strong and sustained innovation-driven growth.

16. Migration of skilled workers can be positive forthe country of origin when those workers come back. InMorocco, high-skilled labor has started to return,bringing substantial know-how and technologicalknowledge into the country, increasing productivity andboosting innovation in terms of improved businesspractices.

17. The value of a relationship-specific investmentis significantly higher within a buyer-seller relationshipthan outside it. An example is where suppliers or sub-contractors to a car producer make investments in de-sign modifications that improve the fit or ease of as-sembly with other parts but which are not relevant tothe production process of other car makers. Suchinvestments tend to be associated with longer-termcontractual commitments between producers andtheir suppliers and less repeated bargaining (Joskow(1987)). Spencer and Qui (2001) find that such rela-tionships in the Japanese car industry tend to limit therange of imports to less important parts and that it ispossible that no parts are imported despite lowerproduction costs overseas.

18. Looking at U.S. imports from 10 major devel-oping countries (which together accounted for 26.5percent of U.S. imports at the time of the cited study),Aggarwal (1995) noted that sectors with egregiouslabor conditions were not a primary share of thesecountries’ exports; that standards were often lower inless export-oriented or nontraded sectors; and that,within the export-oriented sector, labor conditions infirms more involved in exporting were either similar toor better than those in other firms. Raynauld and Vidal(1998) showed that, since 1980, countries with lowstandards had not increased their share of global ex-ports and two-thirds of 39 countries with low laborstandards had seen their international competitiveness(as measured by unit labor costs) stagnate or decline(decline reflects either a decline in labor productivityrelative to the nominal cost of labor or a rise in thenominal cost of labor relative to its productivity) while14 of the 18 high-standards countries had increasedtheir international competitiveness.

19. “Outsourcing” or “offshoring” haveboth been used to refer to the global sourcing of


134

services—technically “outsourcing” refers to the sourc-ing of an activity outside a company (such as the con-tracting out of billing services), which can also take placewithin the domestic market, while “offshoring” is themovement of production of a service outside a country.For firms, offshoring need not be outsourcing when theactivity stays within a foreign affiliate; from the perspec-tive of national labor markets, it is the movement ofproduction to another territory that is the focus of inter-est. Strictly speaking, not all FDI is offshoring, as in caseswhere a foreign affiliate is built to service the localmarket in the host country (Kirkegaard 2005).

20. Note that balance of payments (BOP) statisticsimperfectly measure the full extent of global sourcingof services because of classification and data limita-tions; figures should be taken as an underestimate.

21. Atkinson (2004) notes that a concerted effortby the Chinese government to expand acquisition ofEnglish could see China moving beyond non-language-based services to other business services.

22. Care needs to be taken with comparisons be-tween outsourcing and inward FDI, as the foreign es-tablishment may be created primarily to serve the do-mestic market (see note 19). That said, the comparisonmay be more relevant in terms of the contribution offoreign companies to job creation within the UnitedStates, to offset the movement of jobs overseas.

23. For every $1 of call-center work offshored byU.S. firms, an estimated $1.43 is reinvested in the U.S.economy; the amounts are $1.33 and $1.42 for infor-mation technology services and high-end knowledgeservices (such as equity research, tax preparation, andrisk management), respectively. The net benefit to theU.S. economy of shifting $1 previously spent in theUnited States to India could be as high as 12–14 centsper dollar (McKinsey Quarterly, October 2003).

24. The absence of international standards formany service sector activities reinforces the importanceof reputation in attracting new clients.

25. Nicolás Eyzaguirre, former Minister of Financefrom Chile, comparing the experiences of Chile andArgentina, in a presentation to the Mauritius HighLevel seminar in September 2006.

26. Rates of return can also be influenced by thelack of competition. If incumbents are able to extractlarge rents that are not endangered, the incentive toinnovate is severely restrained because the returnswill replace some of the rents they are actuallycollecting, reducing the net value of innovation. Evi-dence indicates that monopolies are not particularlyinnovative and that small firms stimulate innovation(Geroski 1990). Opening markets by reducing exter-nal barriers and creating better regulatory frame-works for natural monopolies is hence likely to raiserates of innovation.

27. However, IPR protection must be balancedagainst the need to avoid stifling competition.

28. Education can also promote access to technol-ogy and development of new sectors: in 21 countries inLatin America and the Caribbean, an increase of fiveyears in the average level of education in those above15 was associated with an increase in FDI of 3 percentof GDP (de Ferranti and others 2002).

29. While the informal sector can also be a way ofmanaging risk, there are limits to its role as a safetynet, as it often generates most of the flows into unem-ployment (about 60 percent in Argentina, Brazil, andMexico) and is ineffective in cases of multiple/covariateshocks (Arias and others 2005).

30. In the United States, the structure of the politi-cal system (including, for example, passage of TradePromotion Authority in Congress) and nature of pres-sures have led to the development of trade-specificadjustment measures (Brenton 2006). Under the U.S.Trade Adjustment Assistance (TAA) program, qualifiedworkers can receive an additional 52 weeks of unem-ployment insurance provided they are enrolled in an ap-proved training program; a similar program was createdfor the North American Free Trade Agreement(NAFTA) in 1993. TAA and NAFTA payments areabout $300 million annually (Kletzer 2001). The Euro-pean Union is also now considering a GlobalizationFund of half a billion euros to help retrain and relocate35,000–50,000 workers a year whose jobs are lost toglobal sourcing and trade. Money would be available inthe case of layoffs of at least 1,000 people in regionswith a population of at least 800,000 where the unem-ployment rate was already higher than the European ornational average; or where several companies in a sectorlaid off at least 1,000 workers over six months in regionswith a population of up to three million and where thejob losses added up to 1 percent of total employees inthat sector. While the fund will cover a relatively smallnumber of workers, it is seen as important in resistinggrowing calls for protection (Kanter 2006). It is notclear, however, that there is an equity argument for dis-tinguishing between trade-affected and other workers,such as those affected by technological change. In somecircumstances—such as mass layoffs—trade-specific as-sistance may be more cost-effective. However, it shouldbe used sparingly, aimed at orderly adjustment, be time-limited, and include both services and manufacturingworkers as well as workers who have lost their jobs fromboth import and export competition (OECD 2005d).

31. Under Kletzer’s proposal for the United States,all full-time displaced workers would be eligible to re-ceive a health insurance subsidy for up to six months,or until they found a new job, whichever is earlier.

32. ALMPs include employment services, training,public works (which offer short-term employment on


135

community projects in sectors such as construction,rural development, and community services), wage andemployment subsidies, and self-employment assistance.

33. Over 1990–2002, average national expendi-ture on ALMPs in OECD countries remained relativelyconstant at about 0.75 percent of gross domestic prod-uct (GDP). This average masks wide differences, how-ever, with some European countries spending over1 percent of GDP while the United States, Japan,Korea, and the United Kingdom spent under 0.4 per-cent. Training accounted for the bulk of spending(36 percent), followed by public employment services(24.5 percent) and job subsidies (19.5 percent). Transi-tion economies show similar (but lower) patterns ofspending (Betcherman, Olivas, and Dar 2004).

34. There is some evidence that Latin Americancountries have been investing significantly in youthtraining and public works programs, but in Africathere is very little active programming on any signifi-cant scale (Betcherman, Olivas, and Dar 2004).

35. For many resource-rich developing countriesfacing rising demand from China and India andhigher world prices, a crucial opportunity that has tobe addressed is translating higher revenues into in-vestments in social and educational programs that en-hance competitiveness and support diversification. Forthese countries, the priority for domestic reform mustbe to address governance and corruption issues asso-ciated with distributing the revenues from resources.This must be supported by the global communitythrough increased efforts to discipline the activities offirms and governments in countries demanding theseresources.

36. Countries with smaller agricultural and largerurban sectors, with strong political institutions, withhigher per capita incomes, and that are more open totrade tend to have higher collection efficiencies forVAT (Aizenman and Jinjarak 2006).

ReferencesAggarwal, M. 1995. “International Trade, Labor Stan-

dards and Labor Market Conditions: An Evalua-tion of the Linkages.” U.S. International TradeCommission, Office of Economics Working PaperNo. 95-06-C, Washington, DC.

Aizenman, Joshua, and Yothin Jinjarak. 2006. “Glob-alization and Developing Countries—A ShrinkingTax Base?” NBER Working Paper No. 11933,National Bureau of Economic Research, Cam-bridge, MA.

Amiti, Mary, and Shang-Jin Wei. 2004. “Fear of Out-sourcing: Is It Justified?” NBER Working Paper No.10808, National Bureau of Economic Research,Cambridge, MA.

Anderton, Robert, and Paul Brenton. 1999. “Out-sourcing and Low-Skilled Workers in the UK.”Bulletin of Economic Research 51: 267–86.

Anderton, Robert, Paul Brenton, and John Whalley.2006. Globalisation and the Labour Market:Trade, Technology and Less-Skilled Workers inEurope and the United States. London and NewYork: Routledge.

Arias, Omar, Andreas Blom, Mariano Bosch, WendyCunningham, Ariel Fiszbein, Gladys LopezAcevedo, William Maloney, Jaime Saavedra,Carolina Sanchez-Paramo, Mauricio Santamaria,and Lucas Siga. 2005. “Pending Issues in Protec-tion, Productivity Growth, and Poverty Reduc-tion.” World Bank Policy Research WorkingPaper No. 3799, Washington, DC. December.

Atkinson, Robert. 2004. “Understanding the OffshoringChallenge.” Progressive Policy Institute PolicyReport, May, available at www.ppionline.org.

Bates, Jenny. 2000. “International Trade and LaborStandards.” Progressive Policy Institute (PPI)Policy Report, available at www.ppionline.org.

Baunsgaard, Thomas, and Michael Keen. 2005. “TaxRevenue and (or?) Trade Liberalization.” Interna-tional Monetary Fund (IMF) Working PaperNo. 05/112, Washington, DC.

Bergin, Paul, Robert Feenstra, and Gordon Hanson.2006. “Outsourcing and Volatility.” University ofCalifornia.

Betcherman, Gordon, Karina Olivas, and Amit Dar.2004. “Impacts of Active Labor Market Pro-grams: New Evidence from Evaluations withParticular Attention to Developing and TransitionEconomies.” Social Protection Discussion PaperSeries, World Bank, Washington, DC. January.

Bhagwati, Jagdish, Arvind Panagariya, and T. N.Srinivasan. 2004. “The Muddles over Outsourc-ing.” Journal of Economic Perspectives 18: 93–114.

Bhorat, Haroon, and Paul Lundall. 2004. Employ-ment and Labour Market Effects of Globaliza-tion: Selected Issues for Policy Management.Employment Strategy Papers No. 2004/3, ILO,Geneva.

Blinder, Alan. 2006. “Offshoring: The Next IndustrialRevolution?” Foreign Affairs 85(2): 113–28.

Borjas, George. 2003. “The Labor Demand Curve IsDownward Sloping: Reexamining the Impact ofImmigration on the Labor Market.” QuarterlyJournal of Economics 118(4): 1335–74.

Borjas, George, Richard Freeman, and Lawrence Katz.1997. “How Much Do Immigration and Trade Af-fect Labor Market Outcomes?” Brookings Paperson Economic Activity 1: 1–90, Washington, DC.

Bravo-Ortega, Claudio, and Daniel Lederman. 2006.“The Effect of Chinese and Indian Patenting


136

Activity on Latin America’s Patent Counts: Esti-mating the Effects of Cumulative Learning andContemporaneous Patent Tournaments.” Back-ground paper for the Office of the Chief Econo-mist for Latin America and the Caribbean Re-gional Study: “Latin America and the CaribbeanRespond to the Growth of China and India.”World Bank, Washington, DC.

Brenton, Paul. 2006. “Adjusting to Globalisation: Pol-icy Responses in Europe and the US.” In Global-isation and the Labour Market: Trade, Technol-ogy and Less-Skilled Workers in Europe and theUnited States, ed. Robert Anderton, Paul Bren-ton, and John Whalley. London and New York:Routledge.

Brofenbrenner, Kate, and Stephanie Luce. 2004. “TheChanging Nature of Corporate Global Restruc-turing: The Impact of Production Shifts on Jobs inthe US, China and Around the Globe.” Washing-ton, DC: US-China Economic and Security Re-view Commission.

Brown, D., A. Deardorff, and R. Stern. 2002. “The Effectsof Multinational Production on Wages and WorkingConditions in Developing Countries.” In Challengesto Globalization, ed. Robert Baldwin and L. AlanWinters. Chicago: University of Chicago Press.

Card, David. 2005. “Is the New Immigration Really SoBad?” NBER Working Paper No. 11547, NationalBureau of Economic Research, Cambridge, MA.

Cline, William. 2006. “Exports of Manufactures andEconomic Growth: The Fallacy of CompositionRevisited.” Institute for International Economics,Washington, DC.

Cortes, Patricia. 2005. “The Effect of Low-skilledImmigration on US Prices: Evidence from CPIData.” Massachusetts Institute of Technology,Cambridge, MA.

Cravino, Javier, Daniel Lederman, and LarceloOlarreaga. 2006. “Substitution between ForeignCapital in China, India and the Rest of the World:Much Ado about Nothing?” Background paper forthe Office of the Chief Economist for Latin Americaand the Caribbean Regional Study, “Latin Americaand the Caribbean Respond to the Growth of Chinaand India,” World Bank, Washington, DC.

Criscuolo, Chiara, Jonathan Haskel, and MatthewSlaughter. 2004. “Why Are Some Firms MoreInnovative? Knowledge Inputs, KnowledgeStocks and the Role of Global Engagement.”University College London.

Dar, Amit, and P. Zafiris Tzannatos. 1999. “ActiveLabor Market Programs: A Review of theEvidence from Evaluations.” Social ProtectionDiscussion Paper No. 9901, World Bank,Washington, DC. January.

de Ferranti, David, Guillermo Perry, Indermit Gill, andLuis Servén. 2000. Securing Our Future in aGlobal Economy. Washington, DC: Latin Ameri-can and Caribbean Studies, World Bank.

de Ferranti, David, Guillermo Perry, Daniel Lederman,and William Maloney. 2002. From Natural Re-sources to the Knowledge Economy: Trade andJob Quality. Washington, DC: Latin Americanand Caribbean Studies, World Bank.

de Mello, Luiz, and Mombert Hoppe. 2005. “Educa-tion Attainment in Brazil: The Experience ofFUNDEF.” OECD Economics Department Work-ing Papers No. 424, OECD, Paris.

Dehejia, Vivuk, and Yiagadeesen Samy. 2002. “Tradeand Labour Standards—Theory, New EmpiricalEvidence and Policy Implications.” CESifo Work-ing Paper No. 830, December.

Edwards, L. 1999. “Trade Liberalization, StructuralChange and Occupational Employment in SouthAfrica.” Paper presented at the Trade and Indus-trial Policy Secretariat (TIPS) Annual Forum,Muldersdrift, September 19–22.

Eifert, Ben, Alan Gelb, and Vijaya Ramachandran.2005. “The Business Environment and Compara-tive Advantage in Africa: Evidence from Invest-ment Climate Data.” Paper presented at the17th Annual World Bank Conference on Develop-ment Economics (ABCDE), January 27, Dakar,Senegal.

Fajnzylber, Pablo, and William F. Maloney. 2005.“Labor Demand and Trade Reform in LatinAmerica.” Journal of International Economics66: 423–46.

Feenstra, Robert, and Gordon Hanson. 2003. “GlobalProduction Sharing and Rising Inequality: ASurvey of Trade and Wages.” In Handbook ofInternational Trade, ed. E. K. Choi and J. Harrigan.Blackwell Publishing.

Forrester Research. 2002. “3.3 Million US ServicesJobs to Go Offshore.” Cambridge, MA.

Fox, L., G. Betcherman, V. Chandra, B. Eifert, and A.Van Adams. 2004. “Realizing the Potential of theLabor Force in Africa: Barriers and Opportuni-ties.” World Bank, Washington, DC.

Freeman, Richard. 2005. “The Great Doubling: Laborin the New Global Economy.” Harvard Univer-sity, Cambridge, MA.

______. 2006. “Labor Market Imbalance: Shortages,or Surpluses, or Fish Stories?” Paper presentedat Boston Federal Reserve Economic Confer-ence, “Global Imbalances—As Giants Evolve,”Chatham, MA.

Friedman, Thomas L. 2005. The World Is Flat: A BriefHistory of the Twenty-First Century. New York:Farrar, Straus and Giroux.


137

Geroski, P. A. 1990. “Innovation, TechnologicalOpportunity, and Market Structure.” OxfordEconomic Papers 42(3): 586–602.

Ghose, Ajit K. 2003. Jobs and Incomes in a Globaliz-ing World. Geneva: International Labour Office.

Grossman, G., and E. Rossi-Hansberg. 2006. “The Riseof Offshoring: It’s Not Wine for Cloth Anymore.”Princeton University, Princeton, New Jersey.

Hall, Robert, and Charles Jones. 1999. “Why Do SomeCountries Produce So Much More Output perWorker Than Others?” Quarterly Journal ofEconomics 114: 83–116.

Hasan, Rana, Devashish Mitra, and K. V. Ramaswamy.2003. “Trade Reforms, Labor Regulations, andLabor-Demand Elasticities: Empirical Evidencefrom India.” East-West Center Working PapersNo. 59. Honolulu, Hawaii.

Heckman, James, and Carmen Pagés. 2000. “The Costof Job Security Regulation: Evidence from LatinAmerican Labor Markets.” NBER Working PaperNo. 7773, National Bureau of Economic Re-search, Cambridge, MA.

Helpman, Elhanan. 2004. The Mystery of EconomicGrowth. Cambridge, MA: Harvard UniversityPress.

Hoekman, Bernard, and Beata S. Javorcik. 2006.“Lessons from Empirical Research on Interna-tional Technology Diffusion through Tradeand Foreign Direct Investment.” In Global Inte-gration and Technology Transfer, ed. Hoek-man and Javorcik. Washington, DC: World Bank.

Hoekman, Bernard, and L. Alan Winters. 2005.“Trade and Employment: Stylized Facts andResearch Findings.” Mimeo.

ILO (International Labour Office). 1998. Labor andSocial Issues Relating to Export ProcessingZones. Geneva: ILO.

______. 2004. A Fair Globalization: Creating Oppor-tunities for All: Report of the World Commissionon the Social Dimension of Globalization.Geneva: ILO.

______. 2006. Competing for Global Talent. Geneva:ILO.

Iversen, Torben, and Thomas Cusack. 2000. “TheCauses of Welfare State Expansion.” WorldPolitics 52: 313–49.

Jaumotte, Florence, and Nigel Pain. 2005. “AnOverview of Public Policies to Support Innova-tion.” OECD Economics Department WorkingPapers No. 456, OECD, Paris.

Jensen, J. Bradford, and Lori Kletzer. 2005. “TradableServices: Understanding the Scope and Impact ofServices Outsourcing.” IIE Working Paper SeriesNo. WP05-9, Institute for International Econom-ics, Washington, DC.

Joskow, P. L. 1987. “Contract Duration andRelationship-Specific Investments: EmpiricalEvidence from Coal Markets.” American Eco-nomic Review 77: 168–85.

Kanter, James. 2006. “EU Fund to Ease GlobalizationPain.” Reprinted from The International HeraldTribune, March 2, 2006, available at Yale GlobalOnline (http://yaleglobal.edu).

Katz, Larry F., and David H. Autor. 1999. “Changes inthe Wage Structure and Earnings Inequality.” InHandbook of Labor Economics, Vol. 3A, ed.Orley C Ashenfelter and David Card. Amsterdam:Elsevier.

Kirkegaard, Jacob Funk. 2005. “Outsourcing andOffshoring: Pushing the European Model Overthe Hill, Rather than Off the Cliff!” IIE WorkingPaper Series No. WP05-1, Institute for Interna-tional Economics, Washington, DC.

Kletzer, Lori. 2001. “A Prescription to Relieve WorkerAnxiety.” Institute for International EconomicsPolicy Brief No. 01-2, Institute for InternationalEconomics, Washington, DC.

Krishna, Pravin, Devashish Mitra, and Sajjid Chinoy.2001. ‘Trade Liberalization and Labor DemandElasticities: Evidence from Turkey.” Journal ofInternational Economics 55: 391–409.

Krugman, Paul. 1995. “Technology, Trade and FactorPrices.” NBER Working Paper No. 5355,National Bureau of Economic Research, Cam-bridge, MA.

Kuhn, P. J., ed. 2002. “Losing Work, Moving On.” W.E. Upjohn Institute for Employment Research,Kalamazoo, MI.

Lardy, Nicholas. 2004. “Do China’s Abusive LaborPractices Encourage Outsourcing and DriveDown American Wages?” Senate DemocraticPolicy Committee Hearing, March 29. Availableat www.iie.org.

Leamer, E. 2006. “A Flat World, A Level Playing Field,a Small World After All, or None of the Above?”Forthcoming in Journal of Economic Literature.

Lederman, Daniel, Marcelo Olarreaga, and IsidroSoloaga. 2006. “The Growth of China and India inWorld Markets: Opportunity or Threat for LatinAmerican Exporters?” Background paper for theOffice of the Chief Economist for Latin Americaand the Caribbean Regional Study, “Latin Amer-ica and the Caribbean Respond to the Growth ofChina and India.” World Bank, Washington, DC.

Majid, Normaan. 2004. “What Is the Effect of TradeOpenness on Wages?” Employment StrategyPapers No. 2004/18, Employment Analysis Unit,ILO, Geneva.

Martin, J. 2000. “What Works among Active LabourMarket Policies: Evidence from OECD Countries’


138


139

Experiences.” OECD Economic Studies, No. 30.OECD, Paris.

Moran, Theodore. 2004. “Trade Agreements andLabor Standards.” The Brookings Institution Pol-icy Brief No. 133, available at www.brookings.edu.

Nolting, Louvan, and Murray Feshbach. 1980. “R andD Employment in the USSR.” Science 207 (Feb. 1):493–503.

Nunn, Nathan. 2005. “Relationship Specificity, In-complete Contracts and the Pattern of Trade.”University of British Columbia. Forthcoming inQuarterly Journal of Economics.

OECD. 1996. Trade, Employment and Labour Stan-dards: A Study of Core Workers’ Rights and In-ternational Trade. Paris: OECD.

______. 2000a. International Trade and Core LabourStandards. Paris: OECD.

______. 2000b. Policy Competition and Foreign DirectInvestment: A Study of Competition among Gov-ernments to Attract FDI. Paris: OECD Develop-ment Centre Studies.

______. 2004. OECD Employment Outlook. Paris:OECD.

______. 2005a. “The Impact of Structural Policies onTrade-Related Adjustment and the Shift to Ser-vices.” Economics Department Working PapersNo. 427, OECD, Paris.

______. 2005b. Trade and Structural Adjustment: Em-bracing Globalization. Paris: OECD.

______. 2005c. “The Share of Employment PotentiallyAffected by Offshoring—An Empirical Investiga-tion.” Working Party on the Information Econ-omy, DSTI/ICCP/IE(2005)8/FINAL.

______. 2005d. OECD Employment Outlook. Paris:OECD.

Ottaviano, Gianmarco, and Giovanni Peri. 2005. “Re-thinking the Gains from Immigration: Theory andEvidence from the US.” NBER Working PaperNo. 11672, National Bureau of EconomicResearch, Cambridge, MA.

Oxfam. 2004. “Stitched Up: How Rich-CountryProtectionism in Textiles and ClothingTrade Prevents Poverty Alleviation.” BriefingPaper No. 60, Oxfam, U.K. Available athttp://www.oxfam.org.uk/what_we_do/issues/trade/downloads/bp60_textiles.pdf.

Perry, Guillermo, and Marcelo Olarreaga. 2006.“Trade Liberalization, Inequality and PovertyReduction in Latin America.” Paper presented atthe Annual World Bank Conference on Develop-ment Economics (ABCDE) 2006.

Puga, Diego, and Daniel Trefler. 2005. “Wake Up andSmell the Ginseng: The Rise of Incremental Inno-vation in Low-Wage Countries.” Department of

Economics Working Papers tepica-193, Univer-sity of Toronto, Ontario.

Rama, Martin. 2003. Globalization and Workers inDeveloping Countries. World Bank Policy Re-search Working Paper No. 2958, Washington,DC. January.

Raynauld, André, and Jean-Pierre Vidal. 1998. LaborStandards and International Competitiveness: AComparative Analysis of Developing and Indus-trialized Countries. Cheltenham, U.K. andNorthampton, MA: Edward Elgar.

Reinecke, G., and R. Torres. 2001. Studies in the SocialDimension of Globalization: Chile. Geneva: In-ternational Labour Office.

Robbins, Donald. 1997. “Trade and Wages in Colom-bia.” Estudios de Economia 24(1): 47–83.

Rodrik, Dani. 1996. “Labor Standards in InternationalTrade: Do They Matter and What Do We Doabout Them?” In Emerging Agenda for GlobalTrade: High Stakes for Developing Countries,Overseas Development Council Essay No. 20, ed.R. Z. Lawrence, D. Rodrik, and J. Whalley.Washington, DC: John Hopkins University Press.

______. 1997. Has Globalization Gone Too Far?Washington, DC: Institute for InternationalEconomics.

Sapir, André. 2005. “Globalization and the Reform ofEuropean Social Models.” Background documentfor the presentation at ECOFIN Informal Meet-ing in Manchester, U.K., September 9.

Scheve, Kenneth, and Matthew Slaughter. 2002.“Economic Insecurity and the Globalization ofProduction.” NBER Working Paper No. 9339,National Bureau of Economic Research,Cambridge, MA.

Slaughter, Matthew J. 2001. “International Trade andLabor-Demand Elasticities.” Journal of Interna-tional Economics 54: 27–56.

Spencer, Barbara, and Larry Qui. 2001. “Keiretsu andRelationship-Specific Investment: A Barrier toTrade?” International Economic Review 42:871–901.

Stern, Robert M. 2003. “Labour Standards and TradeAgreements.” RSIE Discussion Paper No. 496,available at www.spp.umich.edu.

Stiglitz, J. 2000. “Democratic Development as theFruits of Labor.” Keynote Address, IndustrialRelations Research Associations, Boston, January.

Trefler, Daniel. 2005. “Offshoring: Threats and Op-portunities.” Paper presented at the BrookingsTrade Forum, May, Washington, DC.

UNCTAD (United Nations Conference of Trade andDevelopment). 2004. India’s Outward FDI: AGiant Awakening? UNCTAD/DITE/IIAB/2004/1.New York and Geneva: United Nations.

______. 2005. World Investment Report: Transna-tional Corporations and the Internationalizationof R&D. New York and Geneva: United Nations.

Wood, Adrian. 1997. “Openness and Wage Inequality inDeveloping Countries: The Latin American Chal-lenge to East Asian Conventional Wisdom.” WorldBank Economic Review 11(1): 33–58.

World Bank. 2002. Globalization, Growth, andPoverty: Building an Inclusive World Economy.Washington, DC: World Bank.

______. 2003. Lifelong Learning in the Global Econ-omy: Challenges for Developing Countries.Washington, DC: World Bank.

______. 2005. “Workers and Labor Markets.” InWorld Development Report 2005. Washington,DC: World Bank.

______. 2006. World Development Report: Equityand Development. Washington, DC: World Bank.

Yusuf, Shahid, Kaoru Nabeshima, and Dwight Perkins.2006. “China and India Reshape Global Indus-trial Geography.” In Dancing with Giants: China,India, and the Global Economy, ed. L. AlanWinters and Shahid Yusuf. Washington, DC andSingapore: World Bank and the Institute of PolicyStudies.


140

new pressures in labor markets: integrating large emerging … · 2019-08-20 · increase exports....

Documents