World Bank 42991 Employment Policy Primer March 2008 No. 9 THE EFFECTS OF GLOBALIZATION ON WORKING CONDITIONS IN DEVELOPING COUNTRIES An Analysis Framework and Country Study Results* Globalization--defined as falling barriers to, and direct and indirect effects on the risk of falling or stay- the increase in, trade, migration, and investment across ing in poverty. borders--directly affects workers in both developed As Brown (2007) notes in her literature review, the and developing countries. While most global trade and existing evidence is somewhat inconclusive and there are investment is between the developed countries, global- only a few studies that systematically evaluate the effects ization has increased dramatically in a number of devel- of globalization and working conditions either within or oping countries. across countries. According to the World Development Indicators This note outlines an approach for a systematic database, the ratio of trade in goods to gross domestic cross-country comparison of the relationship between product (GDP) in the low and middle income countries globalization and working conditions. In addition, the has increased between 1990 and 2003; with some coun- results from applying this approach to five countries tries, experiencing especially large changes. (Cambodia, El Salvador, Honduras, Indonesia, and Changes in foreign direct investment (FDI) have Madagascar) are presented. also been significant. For South and East Asia, the ratio of inward FDI to GDP increased between 1990 and An Analysis Framework 2003. As with trade, changes in FDI varied a great deal across countries. These dramatic changes in trade and In studying the impact of globalization on working FDI hold both promise and peril for workers in develop- conditions, it is important to account for each country's ing countries. unique history, institutions, and economic conditions. Understanding the effects of globalization is crit- In particular, defining globalization and working condi- ical for governments concerned about employment, tions in terms of dimensions relevant to each country's working conditions, and, ultimately, poverty reduction. experience is vital for conducting a meaningful analysis. Broadly defined, working conditions includes wages Furthermore, how one defines globalization and work- and other key job characteristics including (but not ing conditions will shape how one thinks about global- necessarily limited to) health and safety, hours, secu- ization's effects on working conditions and, therefore, rity, benefits, and representation. These conditions have the appropriate way to study the link between them. *This note was prepared by Ruwan Jayasuriya, based on a forthcoming study by Drusilla Brown, Raymond Robertson et al, "Globaliza- tion and Working Conditions: Evidence from Five Developing Countries" (working title). The World Bank Employment Policy Primer aims to provide a comprehensive, up-to-date resource on labor market policy issues. The series includes two products: short notes, such as this one, with concise summaries of best practice on various topics; and longer papers with new research results or assessments of the literature and recent experience. Primer papers and notes are available on the labor markets Website at www.worldbank.org/labormar- kets, by contacting the Social Protection Advisory Service at (202) 458-5267, or by email at socialprotection@worldbank.org. T h e E f f e c t s o f G l o b a l i z a t i o n o n Wo r k i n g C o n d i t i o n s i n D e v e l o p i n g C o u n t r i e s Measuring Globalization in conducting a meaningful analysis. Within this con- Globalization has many faces. It deals with the text, providing a description of the existing labor stan- increase in values and volumes of trade in goods and dards along with their evolution, information on private services among countries, which generally result from voluntary initiatives that allow consumers and investors reducing tariffs, quotas and other barriers to trade. The to express their values relating to working conditions, associated FDI aimed at facilitating this process is also a details on trade agreements that have been used to affect key factor. working conditions, and an explanation of any interven- Globalization also deals with the development and tions, regulations and monitoring programs will provide spread of technologies, especially Information and Tele- valuable additional insight. communication Technologies, which allow access to In addition, changes in a country's average working information all over the world, and to connect with conditions can occur as a result of changes in industry production, distribution and knowledge networks, rap- composition or in within-industry changes in working idly increasing the potential for development around conditions over time; in particular, if working condi- the globe. tions differ across industries in a significant manner and Furthermore, it also deals with international migra- are constant over time, movement of workers between tion between developing and developed countries, and industries would change average working conditions. If among themselves, which changes the economic rela- workers moved from industries with poor conditions to tions and structures among and within countries. The industries with better conditions, average working con- spread of democracy and democratic institutions in the ditions would improve (and vice versa). world along with the role of international non-govern- Similarly, exposure to international markets can mental organizations is another facet of interest. increase international attention paid to local working All these faces can affect the behavior and the struc- conditions, and could cause domestic firms to improve ture of labor markets within a country. In particular, their working conditions. The effects of FDI on working depending on which of theses features are more impor- conditions is not clear a priori, but rather depends on tant for a particular country, and the particular (idio- the difference between the standards in foreign firms syncratic) characteristics of the country that determines relative to domestic firms. If firms from countries with how it is prepared (or not) to take advantages of the new better working conditions find it advantageous to keep opportunities and challenges posed by globalization, working conditions somewhere between the domestic will lead to a different impact. level and those in the capital's origin country, this would For example, FDI that produces for the domestic increase the average level of working conditions and market may increase competition for firms causing them may induce domestic firms to improve working condi- to lower costs by making working conditions worse. In tions in order to keep and attract workers. In order to contrast, FDI that produces for export may increase the analyze these kinds of changes, it is important to have demand for workers, which may cause firms to improve information on how working conditions change over working conditions as a way to attract or retain workers. time; ideally at the firm-level. In addition to the nature of changes in globalization, Working conditions can be categorized into wages the speed with which liberalization is implemented (i.e., and non-wage working conditions (i.e., health and safe- gradual phase-in vs. sudden) and number of industries ty, hours, security, benefits, union representation, and (many vs. few) impacted are both relevant to understand details about the working environment)1 in studying how globalization affects workers. their relationship with globalization; in particular, indi- Therefore it is important to first characterize the 1 country's experience with globalization that will aid in In addition to the Core Labor Standards established by the Inter- national Labor Organization (ILO), Elliott and Freeman (2003) also selecting the suitable approach to analyze its impact on list dark, crowded, hot, noisy workplaces, no emergency exits or fire working conditions. extinguishers, inadequate or no time to go to the toilet, no canteen or place to eat, abusive supervisors who strike young workers, below- Measuring Working Conditions minimum wage payments, absence of written contracts, compulsory overtime, sexual or other harassment of workers, and late or short Defining working conditions in terms of dimen- wage payments among other concerns about working conditions in sions relevant to each country's experience is important developing countries. 2 T h e E f f e c t s o f G l o b a l i z a t i o n o n Wo r k i n g C o n d i t i o n s i n D e v e l o p i n g C o u n t r i e s vidual-level measures of wage and non-wage conditions odologies can be used in the analyses: (i) event studies, are needed to conduct a meaningful analysis. (ii) price studies, and (iii) quantity studies. The country studies presented in this note use labor Event studies compare working conditions before force survey data to compare industry specific wage and after a clearly identified liberalization event, while and non-wage working conditions of industries directly controlling for other confounding factors that may have affected by globalization with other industries to pro- changed concurrently. vide valuable insight into the impact of globalization on Price studies, based on neoclassical trade theory, working conditions. The labor force survey data used all rely on changes in relative prices to capture the effects have comparable measures of wages, making wages (and of liberalization. A change in the price of a good relative inter-industry wage differentials) a natural choice for to other goods in the economy will affect the conditions working conditions measure that is comparable across of workers (especially wages) within that industry; thus, countries. This data also contains different (not compa- can be used to capture the impact of globalization on rable) measures of non-wage working conditions that working conditions. provide country-specific detail on working conditions Quantity studies link imports, exports, FDI and unique to each country. other quantity measures with working conditions. By Furthermore, it is important to develop an under- using disaggregate industry-level import/export data, standing of the relationship between wages and non- industry-specific effects of trade can be captured while wage working conditions because a negative correlation controlling for other factors that may affect working between the two would suggest that changes in one off- conditions and globalization measures. set the other in a manner that would make conclusions To investigate the effects of globalization on working about a net effect of globalization on working condi- conditions between industries, wage differentials between tions difficult to identify. industries can be estimated. In addition to globalization variables, worker characteristics (such as gender, age, Empirical Application education and other factors) that also affect wages should The link between globalization and working condi- be included in the regression analysis as controls. tions is not clear a priori in the sense that globalization Based on data availability, non-wage working con- holds the potential to both help and hurt workers. ditions (i.e., health and safety, hours, security, benefits, Globalization may increase foreign investment and union representation, and details about the working exports in ways that would increase labor demand. environment) of industries directly affected by glo- On the other hand, globalization may primarily be balization can be compared with other industries to manifested as an increase in imports and loss of pro- develop a more comprehensive understanding of global- tection, forcing firms to attempt to reduce costs by izations impact on working conditions. worsening working conditions. In any case, globaliza- Labor force surveys, which are becoming increas- tion involves adjustment and empirical work can help ingly available, generally contain this data that can be identify the forces that may affect working conditions. used in analyses. Furthermore, different measures of globalization have different implications for working conditions; hence, Country Study Findings having a sound understanding of what measures of globalization and working conditions are used along Data from Cambodia, El Salvador, Honduras, Indo- with a country's experience will aid in conducting a nesia, and Madagascar were used to analyze the link meaningful analysis. between globalization and working conditions. These Within this context, an effective study of the effects five countries have all had significant experiences with of globalization on working conditions should initially liberalization and globalization in the last twenty years. provide a descriptive analysis of the country's experience Together they represent Latin America, South-East Asia, with globalization and working conditions to identify and Africa, and vary significantly in terms of their geog- key variables (i.e., tariff changes, FDI, and etc.), and raphy, size, population, and income. The heterogeneity also assess the degree of competition for domestic firms between these countries creates a useful comparison of brought about by globalization. Three analytical meth- the effects of globalization on working conditions. 3 T h e E f f e c t s o f G l o b a l i z a t i o n o n Wo r k i n g C o n d i t i o n s i n D e v e l o p i n g C o u n t r i e s The methodology described in this note was applied by easy movement of labor between industries, inter- in each of the country studies, where worker-level data industry wage differentials would need to be non- with information on working conditions and industry- existent (or minor) after controlling for demographic level globalization measures were used in the analyses. characteristics. Industry-specific wage premiums that controlled for Thus, it appears that labor markets in export-ori- workercharacteristicswereestimated,andindustriesthat ented sectors which attract FDI are characterized by were particular recipients of FDI and engines of export "good" jobs with high wages and better non-wage work- growth were identified. Although the countries depict ing conditions. In contrast, agriculture, in particular, much heterogeneity, the prevalence of FDI focused in offers "bad" jobs with low wages and poor non-wage the textile/apparel sectors producing for export is a working conditions. common characteristic in at least some period covered The positive correlation of wages and non-wage by each country study. working conditions is particularly informative because Consistent with theory, these countries have expe- wage data are often more complete than data on specific rienced an increase in demand for labor in sectors aspects of working conditions. A robust positive correla- attracting FDI flows and a fall in the employment shares tion between wages and non-wage working conditions of import-competing industries for which FDI inflows may create the presumption that changes in wages indi- are negligible. In fact, evidence that support a decline cate similar changes in working conditions. in the employment share particularly in agriculture and In the case of Cambodia and Honduras, changes a corresponding rise in the employment shares in the in inter-industry wage differentials over time following FDI-intensive/export industries is observed. an episode of increased integration into world markets Data on industry-specific wage differentials and were observable and are suggestive of a pronounced measures of working conditions for Cambodia, El Sal- impact of globalization on their respective workers. vador, Indonesia and Madagascar indicate that labor in The FDI-intensive/export sectors were found to either the FDI-intensive/export industries have generally bet- pay wages significantly above the mean or the wage ter wages and non-wage working conditions than much premium in that industry either increased over time or of the rest of the local economy. In each case, wages changed with patterns of exports and FDI. In Cambodia, and non-wage working conditions in agriculture--the workers in the textile industry earned as much as 35 default industry--were found to be far below the econ- percent above the mean wage, even though these work- omy-wide average. In contrast, wages and non-wage ers are primarily young women. In Honduras, apparel working conditions in the FDI-intensive/export indus- workers also earned a significant wage premium. In El tries were found to be either at or above the economy- Salvador, workers earned low wages, which grew with wide average. FDI and exports. Cambodia is especially unique because it is the Given the statistically and economically significant first country to have quota access specifically tied to differences in wages between industries, the move- non-wage working conditions. Interestingly, Cambo- ment of workers between industries also has important dian apparel workers also receive the highest wage implications for average working conditions. To the premiums of any of the five countries included in the extent that agriculture, which has wages and non-wage analysis. working conditions that are significantly lower, is a This important finding of a positive correlation primary alternative for many domestic workers, the between wages and non-wage working conditions is move from agriculture to the FDI-intensive/exporting at odds with Adam Smith's compensating differentials sectors represents an improvement in overall working argument, which theorizes that jobs in less favorable conditions. non-wage working conditions must be compensated The FDI-intensive/exporting sectors were found to with higher wages. experience rising employment shares and wage premi- The observed positive correlation between wages ums at the same time, which is suggestive of an increase and non-wage working conditions is consistent with in demand for workers that is consistent with theory. the presence of labor market segmentation. For, if As long as labor markets are imperfect (i.e., wages do labor markets were perfect, or at least characterized not immediately equalize across industries) the increase 4 T h e E f f e c t s o f G l o b a l i z a t i o n o n Wo r k i n g C o n d i t i o n s i n D e v e l o p i n g C o u n t r i e s in demand should increase both wage premiums and approach of the Agreement on Textiles and Clothing employment shares. The observed results confirm this (ACT) ending has resulted in a fall in FDI; especially in outcome. Madagascar. Evidence was also found which suggested that in A summary of the key elements related to the coun- industries where FDI exits, employment shares and try studies is presented in Table 1. wage differentials fall. In the countries included in this study, there are two main causes for the fall in FDI. Conclusions First, the Asian crisis resulted in a significant decrease or negative FDI flows for several years. During this cri- The country study results described in this note sup- sis period, employment shares in agriculture increased port the hypothesis that globalization has contributed to and wage premiums in manufacturing fell. Second, the a shift of workers from sectors with low wage and poor Table 1: Summary of Country Study Results Cambodia El Salvador Honduras Indonesia Madagascar Globalization · Evolved from a tightly · Mainly trade · Mainly trade · Moved from an · Mainly trade Experience controlled socialist liberalization, liberalization, import substitution liberalization, economy in 1980 to a export promotion, capital inflows, and strategy in 1970 to a export promotion, free market economy and attractive FDI labor outflows more export-oriented and capital inflows · Rise in FDI and framework · Rise in FDI and economy in the mid- · Success of Export exports in the gar- · Rise in FDI to the exports in the 1980s Processing Zones ment sector garment and non- garment sector · Rise in FDI and (EPZ) in export and · To a lesser extent, tradable sectors · Development of exports in the employment growth rise in construction · Development of marquila sector garment sector and in · With the phase-out related to tourism marquila sector chemicals of the MFA quotas, · Quota access · Key role played by · The Asian crisis led to Madagascar has specifically tied to large outward migra- a significant decrease lost market share non-wage working tion and resulting or negative FDI flows to China and other conditions flows of remittances for several years Asian countries Measures · Wages · Wages · Wages · Wages · Wages of Working · Hours · Hours/Conditions · Social benefits · Facilities, medical, · Hours Conditions · Accidents and · Contracts, social · Unionization safety, transport, and · Benefits and other disabilities security/payments · Formal training overall conditions Analysis · Regression analysis of · Regression analysis of · Regression analysis · Regression analysis of · Regression analysis Methodology wages and accidents wages and remittance of wages and remit- wages and non-wage of wages · Comparison of hours impact tance impact measures · Comparison of worked · Compare hours, con- · Compare benefits · Comparison of non- benefits and other ditions, benefits and conditions wage measures conditions Observed · Garment sector · Garment sector · 20 percent wage · Wages and non-wage · EPZ wages are lower Results wages and working wages and working premium in the measures are higher than in the industrial conditions are above conditions are above garment sector and + correlated in sector but higher the industry average the industry average · Non-wage condi- the garment sector than in the informal · Earnings that are over · Remittances tions similar across and in chemicals sector 50 percent higher reduce labor force sectors · Case studies depict · Phase-out of MFA has than the average wage participation and the · Negligible impact of potential issues progressively reduced · Has relatively few receipt of other non- remittance on labor missed when using EPZ wages and non- accidents wage benefits force participation aggregate data wage measures 5 T h e E f f e c t s o f G l o b a l i z a t i o n o n Wo r k i n g C o n d i t i o n s i n D e v e l o p i n g C o u n t r i e s non-wage working conditions (i.e., agriculture) to sec- References tors with relatively higher wages and better non-wage working conditions. Brown, D.K., 2007. "Globalization: Implications for Employ- One main concern, however, is that the better wages ment, Wages and Labor Standards, Review of Literature," World Bank, Washington D.C. and non-wage working conditions identified in the tex- Cling, J.-P., Razafindrakoto, M., and F. Roubaud, 2007."Export tile/apparel sector may be an artifact of the Multi-fiber Processing Zones in Madagascar: The Impact of the Dis- Agreement (MFA) or the subsequent Agreement on Tex- mantling of Clothing Quotas on Employment and Labor tiles and Clothing (ACT). These agreements allocated Standards," Institut de Recherche pour le Developpe- ment, DIAL, Paris. bilateral production quotas across many countries and Elliott, K.A. and R. Freeman, 2003. Can Labor Standards created economic rents in the form of higher wages and Improve Under Globalization? Institute for International better non-wage working conditions; some of which Economics, Washington D.C. Marcoullier, D. and R. Robertson, 2007. "Globalization on may have been captured by workers. Working Conditions: Case Study of Honduras," World Some results indicate that FDI has progressively Bank, Washington D.C. diminished in some countries with the phase-out of Neak, S. and R. Robertson, 2007. "Globalization on Work- MFA quotas. Preliminary evidence suggests that the fall ing Conditions: Evidence from Cambodia," World Bank, Washington D.C. in FDI and exports may be linked to falling wages and Robertson, R. and A. Trigueros-Arguellos, 2007. "The Effects non-wage working conditions in FDI recipient sectors, of Globalization on Working Conditions: El Salvador which represents an important and emergent line of 1995-2005," World Bank, Washington D.C. inquiry to better understand the consequences of glo- Sitalaksmi, S., Ismalina P., Fitrady A. and R. Robertson, 2007. "Globalization and Working Conditions: Evidence from balization for labor outcomes. Indonesia," World Bank, Washington D.C. The findings, interpretations, and conclusions expressed herein are those of the author(s), and do not necessarily reflect the views of the International Bank for Reconstruction and Development / The World Bank and its affiliated organizations, or those of the Executive Directors of The World Bank or the governments they represent. 6