l b/PS It&1 Policy Research W L.!IMIkruIą T Hs Country Operations Technical Department Latin America and the Caribbean The World Bank November 1992 WPS 1014 Ethnicity, Education, and Earnings in Bolivia and Guatemala George Psacharopoulos Indigenous people will earn more if they get more schooling. Policy RsearchWod&ing peua disseminate the rindings of work in plogWSS and encourage the eschange of idcas among Bank ataffi'n$ allothers interestcd in developmentisuies.Thesepapers, distuibuted by theResearchAdvisory Staff,carry thenamesoftheauthors,reflect oly theirviews,andshould be used and cited accordingly.The findings, interpreations,and conclusions are the authors'own.They should not be auributed to the Worid Bank. its Board of Directors, its management, or any of its member countries. Polloy Research Country Operations WPS 1014 This paper-a product of the Latin America and the Caribbean Technical Department-is part of a larger effort to document poverty conditions in the region. Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Liliana Longo, room I4-187, extension 39244 (November 1992, 23 pages). Indigenous groups are often associated with acquire less. Just giving ethnic groups basic poverty and so are low levels of educadon. education is bound to improve their position. Guatemala and Bolivia are the two Latin Ameri- can countries in which the ethnic part of the This finding was supported by both within- population is proportionately greatest. And country and cross-country evidence: Indigenous Bolivia is more "schooled" than Guatemala. So people fare better in Bolivia (where there is more Psacharopoulos tried to determine how levels of education) than in Guatemala (where there is ethnicity and educationi affect the level of worker less). eamnings. One possible (although controversial) His investigation was based on data from intervention is to provide schooling in the child's household surveys in the two countries. He first language. Such an intervention has been found that, other things being equal, indigenous successfully implemented on a small scale in people who acquire more human capital enjoy Guatemala. Bilingual programs also exist, on a greater economic rewards than those who small scale, in Bolivia. ThePolicy ResearchWorking PaperSeries disseminates the indings of work under way in theBankr Anobjectiveof the series is to get these fndings out quiclcly, even if presentations are less than fully polished. The fndings, interpretations, and conclusions in these papers do not necessarily represent ofricial Bank policy. Produced by the Policy Research Dissemination Center Ethnicity, Education, and Earnings in Bolivia and Guatemala by George Psacharopoulos* * The World Bank, Washington, DC 20433. Iam grateful to Shelton Davis and Philip Musgrove for commenting on an earlier draft of this paper. Table of Contents I. Introduction 1 II. Guatemala 2 Sample Description 2 The Determinants of Eanings 4 III. Bolivia 11 Sample Description 11 The Determinants of Earnings 12 R. Comparative Lessons 17 References 22 I. Introduction Indigenous groups are often associated with poverty in a given country, and so is a low level of education. Guatemala and Bolivia are the two Latin American countries with the largest proportion of ethnic population. They also differ in terms of the overall educational attainment of the population -- Bolivia is more "schooled" than Guatemala. Hence it would be of interest to examine the respective roles of ethnicity and education in determining the level of worker earnings. Human capital theory came to its forte in the early 1960s (e.g., see Schultz, 1961 and Becker, 1964). This theory offered an alternative explanation on why people's incomes differ, relative to dominant sociological or other theories at the time. If someone has more years of schooling or training, he/she will receive higher rewards later in life relative to a lesser schooled/trained person as a compensation/return to the past investment in human capital. This theory has been extensively tested over the last three decades, culminating in Mincer's (1974) classic book on the relationship between earnings, education and labor market experience in the United States. This theory has also been applied in more than 60 other countries, essentially using only two explanatory variables in the so-called "earnings generation function" -- the years of schooling of the individual, and his/her years of work experience in the labor market. In this paper, I expand the Mincerian human capital earnings function to include the ethnicity of the wage or salary earner. -2- The data on which this pirticular 'mnvstidation is based come from Guatemala's and Bolivia's household surveys. The two countries were chosen because of the large ethnic population living there. Information on whether a person belongs to an "ethnic" category differs in the two countries because of the way this question was asked by the two respective surveys, as explained below. Anthropologists familiar with the two country settings are confident that the "indigenous" variable used here truly reflects the ethnicity of a particular persons. These are the people who, living in countries whose official language is Spanish, still speak among themselves in another idigma, wear a traditional dress, or identify themselves as being In what follows, sections IH and Im present the situation in each particular country, and section IV provides some lessons drawn from the comparative analysis. II. Guatemala Sample Description. The data come from the Encuesta Nacional Socio-Demografica (ENSD) conducted in 1989 by the Instituto Nacional de Estadistica. The survey covered about 11,000 households in all parts of the country, urban and rural. From these households, we focus here on about 12,000 persons who reported positive earnings from employment. The reason we focused on wage and salary earners is in order to be able to superimpose ethnicity to the well tried human capital theory on the determinants of individual earnings. We are specifically interested in the effect of diffaential educationalattainment on earnings, given ethicity, as a potential policy instrument for boosting individual earnings capacity. Non-employment income, su ch as from interest and rents, relates to previous wealth accumulation and is beyond this investigation. The survey questionnaire included the question: ".dEs indigena?". The answer to this question classified an individual into two mutually exclusive ethnic groups. If affirmative, it gave a value of 1 to the indigenous dummy variable, and 0 otherwise. Being indigenous is a combination of spealing one of the many non-Spanish languages (e.g., Quiche, Cakchiquel or Mam), dressing in a traditional way, or simply belonging to a socio-cultural group that links to pre-colonial ancestors. (See Whetten, 1961). Nearly 30 percent of the sample declared themselves indigenous. The sample representativeness of indigenous people corresponds well to the one obtained from independent demographic censuses of the population. (See Stephen and Wearne, 1987). Table 1 presents the rest of the vanables used in this analysis. Average earnings in the sample were 236 Quetzals per month (about US$87 at the official exchange rate). Indigenous workers earned less than one half the income of non-indigenous ones. And indigenous people are three times as likely to belong to the bottom quintile of the income distribution. The level of educational attainment of the two ethnic groups differs dramatically. The indigenous workers have only 1.6 mean years of schwoling, relative to 5.0 years of schooling -4 - for r .,-indigenous group. In terms of actual levels attained, nearly 60 percent of the indigenous group have not conpleted primary education, whereas only one quarter of the non- indigenous group belong to this categor;. A scant 0.5 percent of the indigenous group have had higher education experience, whereas 6.0 percent of the non-indigenous group do so. (See Nyrop, 1983, for an analysis of the country's appalling state of educational development). The prevalence of self-employed is double among the indigenous group, ance this group is four times less likely to be in the public sector of the economy. There are no major differences between the two groups in terms of age, gender, hours worked and potendial labor market experience (defined in the Mincerian way as age, minus years of schooling, minus compulsory schooling starting age of 7). (See Mincer 1974). The Determinants of Earnings. Mean earnings by educational level are neatly ordered as in any other country in the world. Education helps increase the earnings of all groups in the population, whether indigenous or not. But how much of the raw differences presented in the lower panel of Table 1 survive after controlling for other characteristics of the worker? There exist other investigations of earnings differentials in Guatemala, but none to the best of my knowledge has included the ethnic dimension, e.g., see Sumner (1981), Terrel (1989) and Arends (1992). -5- Table 2 presents a series of alternative specifications of an earnings function fitted to this data set. The first function is the basic Mincerian one (see Mincer 1974) that has been fitted in many other countries, including Latin America (e.g., see Kugler and Psacharopoulos, 1989), LnY =a+bS +cEX+dEX2+elnW where Y stands for the earnings of the individual, S his/her number of years of schooling, EX for the number of years of labor force experience and W the number of hours worked per week. The coefficient on years of schooling (b) in this semi-log specification can be interpreted as the average private rate of return to one extra year of schooling regardless of the level of education it refers to. This is because the partial derivative of the function with respect to S, gln Y b = corresponds to an approximate definition of the rate of return to investment in schooling. (For an elaboration on this point see Psacharopoulos, 1973). The 15.1 percent rate of return for Guatemala is on the high side by World and Latin American standards, and this has to be attributed to the extremely low level of educational attainment of the population (3.9 years of schooling). Like every other investment, education is subject the law of diminishing returns. (See Psacharopoulos, 1989). -6- The second function in Table 2 introduces the indigenous variable, controlling also for gender. Other things being equal, males earn approximately 30 percent more than females, and indigenous people 46 percent less than non-indigenous. The significance for this result is that even if indigenous people had the same educational level as the non-indigenous group, they would still earn about one half that of non-indigenous. Why? Could the ethnic group be discriminated against in the labor market? Or could the ethnic group have been discriminated against before entering the labor market, e.g. by receiving less or inferior quality education relative to the non-indigenous group? Alas, we cannot test such hypotheses with the data in hand. The last two earnings functions in Table 3 are fitted within the two ethnic sub-samples. There are two findings worth noticing. First, the human capital earnings function for the indigenous group has one third the explanatory power of the function for the non-indigenous group. This means that there are several other factors, beyond human capital, that determine the earnings of the indigenous group, relative to the non-indigenous group. Second, and most important for policy purposes, the rate of return of investing in the education of the indigenous group is of the order of 12 percent, which makes it an attractive yield. Although this is two percentage points lower than the rate of return for the non-indigenous group, it is reassuring to know that education does have an effect in boosting their earnings, in spite of any possible prior or current labor market discrimination again.* this group. Given the secular improvement of educational development in all parts of the world, ;t - 7 - is of interest to see how such attainment differs among different age cohorts in Guatemala. Table 3 shows that, indeed, the level of education of both ethnic groups has improved considerably during the last decades. And during such improvement, the relative position of the indigenous group has marked considerable gains as well. Whereas indigenous people born in the 1930s had a fivefold educationial disadvantage relative to non-indigenous, the cohort born in the 1970s has only a disadvantage of a factor of two. In order to take into account such a cohort effect, and in order to answer a more pol.cy- oriented, expansion-at-the-margin question, the second function in Table 2 was fitted again within those aged less than 40, and again for those aged less than 30. (See Table 4). Although the rate of return continues to be of the same order of magnitude as in the analysis using the whole sample (of the order of 13 - 14 percent), the "disadvantage coefficient" of being indigenous drops in the successive cohorts (from 46 percent in the whole sample, to 42 percent among those aged less than 40, and to 38 percent among those less than 30). Although the absolute disadvantage of the earnings position of the indigenous group is still substantial, this finding should be very welcome from the policy viewpoint, in the sense that education helps this minority group to improve its relative as well as absolute economic status. -8 - Table 1. Mean Characteristics by Ethnic Group, Guatemala 1989 Characteristic Indigenous Non-indigenous Entire Sample Educational Attainment (%) 1/ - No education 59 24 35 - Primary 38 50 47 - Secondary 3 20 15 - Higher 0.5 6.0 4.3 Years of Schooling 1.6 5.0 3.9 Labor Earnings (Q/month) 121 287 236 Bottom 20% of Inc. Distr.(%) 38 12 20 Male (%) 80 71 74 Age 38 36 37 Experience (Years) 29 24 26 Hours Worked/Week 46 46 46 Economic Sector (%) - Private 36 55 49 - Public 3 12 9 - Self-employed 60 32 41 Earnings (Quetzal/month) - No Education 95 154 124 - Primary 144 230 209 - Secondary 276 402 394 - Higher 543 898 887 - Overall 121 287 236 Number of Observations 3,843 8,658 12,501 1/ Includes dropouts of the respective level. - 9 - Table 2. Human Capital Earnings Functions by Ethnicity, Guatemala Variable Entire Sample Indigenous Non-indigenous Basic Extended Years of Schooling .151 .133 .116 .137 Experience .037 .035 .018 .040 Experience2 -.0004 -.0004 -.0002 -.0005 Hou.s Worked (log) .475 .354 .355 .358 Indigenous -.463 Male .307 .283 .321 Constant 1.999 2.481 2.331 2.370 R2 .338 .387 .124 .369 N 12,141 12,141 3,776 8,365 Notes: Dependent variable is the natural Logarithm of earnings from employment. All coefficients are statistically significant at the 1 % level or better. Table 3. Educational Attainment Among Ethnic Groups by Age, Guatemala Non-indigenous Years of Schooling Education Advantage Age Group Non-indigenous Indigenous Ratio 16 - 20 5.07 2.67 1.90 21 - 30 6.33 2.38 2.66 31 - 40 5.42 1.51 3.59 41 - 50 4.03 .89 4.53 51 + 2.88 .52 5.54 ^ 10- Table 4. Rate of Return and Indigenous Earnings Disadvantage by Age Group, Guatemala Age Group Rate of Return m Indigenous ' (percent) Earning Disadvantage (percent) 40 or less 13.5 - 42 30 or less 13.7 -38 Notes: 1/ Coefficient on years of schooling in the function reported in Table 2, times 100. 2/ Coefficient on the "indigenous" dummy reported in the second function in Table 2, times 100. - 11 - II. Bolivia Sample description. The data are based on the 1989 Encuesta Integrada de Hogares (EIE) conducted by the Instituto Nacional de Estadistica. This survey covers only urban centers with population of more than 10,000. The sample covered 22,100 individuals aged 16+, out of which 10,743 had positive earnings from employment. It should be noted that this survey is urban, whereas the one for Guatemala includes both urban and rural areas. The ethnicity or indigenous variable was also defined in a different way in Bolivia relative to Guatemala. The survey asked the question "G Que idioma(s) habla habitualmente?". ("What language do you usually speak?"). The responses were coded under the ten possibilities listed in Table 5, along with theh frequencies. For the purposes of this exercise we defined as ethnic or indigenous all those who usually spoke one or more of the vernacular languages, even if they also spoke Spanish. This grouping classified as ethnic 34 percent of the population at large (Table 5), or 38 percent of those who have positive earnings from employment (Table 6). These statistics match well independent accounts of the proportion of the ethnic group in Bolivia today. Among the workers, the indigenous groups lags 3.4 years of schooling relative to the non- indigenous group. The overall level of schooling of both groups, however, is much above that of Guatemala. Mean earnings in the sample is 305 Bolivares or about US$104 per month at the - 12 - official exchange rate. The earnings differential between the two groups is again substantial (See lower panel of Table 6). The indigenous workers earn 60 percent less than the non-indigenous group. (For an analysis of the gender earnings gap in Bolivia, using the same survey, see Scott, 1992). The Determinants of Earnings. Table 7 presents the same human capital earnings functions as those fitted earlier for Guatemala. The results show, again, a statistically significant earnings gap of the indigenous versus non-indigenous group, although this gap (approximately 23 percent), is much lower than that found in Guatemala. The retums to ducation to the indigenous group (6.2 percent) are 2.4 percentage points lower relative to the non-indigenous group (8.4 percent). Given the higher level of educational attainment of recent cohorts (Table 8), the eamings functions were run again for those aged less than 40 or less than 30 years old. The indigenous group's earnings disadvantage is much less for the younger cohort -- of the order of 16 percent (Table 9). - 13 - Table 5. The Distribution of Spoken Languages in Urban Bolivia, 1989 Code Language N % 1 Castellano 14191 64.2 2 Quechua 196 .9 3 Aymara 182 .8 4 Castellano y Quechua 3445 15.6 5 Castellano y Aymard 3409 15.4 6 Quechua y Aymara 10 .0 7 Castellano, Quechua y Aymara 341 1.5 8 Castellano y Guaranf 12 .1 9 Castellano y otro nativo 14 .1 10 Castellano y otro no nativo 296 1.3 Indigenous (codes 2 to 8) 7,599 34.4 Non-Indigenous (codes 1, 9, and 10) 14,501 65.6 Overall 225,100 100 - 14 - Table 6. Characteristics of Workers by Ethnic Group, Bolivia, 1989 Characteristic Indigenous Non-indigenous Elntire Sample Educational Attainment(%)1 - No education 12 2 6 - Primary 55 35 42 - Secondary 22 36 31 - Higher 11 27 21 Years of Schooling 6.5 9.9 8.6 Labor Eantings (B/month) 305 493 422 Bottom 20% of Inc. Distr (%) 27 16 20 Male (%) 58 60 59 Age 39 35 37 *Experience (Years) 26 19 22 Hours Worked/Week 42 42 42 Economic Sector (%) - Salaried 45 62 56 - Self-employed 55 38 44 Earnings (Bolivares/month) - No Education 205 273 222 - Primary 282 372 328 - Secondary 329 447 414 - Higher 470 724 673 - Overall 305 493 422 Number of Observations 4,057 6,687 10,743 1/ Includes dropouts of the-respective level. - 15 - Table 7. Human Capital Earnings Functions by Ethnicity, Bolivia, 1989 Variable EntiLr Sample Indigenous Non-indigenous Basic Extended Years of Schooling .092 .076 .062 .084 Experience .048 .046 .034 .052 Experience2 -.0005 -.0005 -.0003 -.0006 Hours Worked (log) .389 .277 .230 .311 Indigenous -.234 Male .364 .370 .369 Constant 2.672 3.124 3.315 2.861 R2 .210 .257 .188 .252 N 10,632 10,632 4,033 6,599 Notes: Dependent variable is the natural logarithm of earnings from employment. All coefficients are statistically significant at the 1 % level or better. - 16 - Table 8. Educational Attainment Among Ethnic Groups by Age, Bolivia Non-indigenous Years of Schooling Education Advantage Age Group Non-indigenous Indigenous Ratio 16 - 20 8.7 6.2 1.40 21 - 30 10.7 8.2 1.30 31 -40 10.5 7.0 1.50 41 - 50 9.2 5.7 1.61 51 + 8.1 4.3 1.88 Table 9. Rate of Return and Indigenous Earnings Disadvantage by Age Group, Bolivia Age Group Rate of Return m' Indigenous 2' (percent) Earnings Disadvantage (percent) 40 or less 8.6 - 22 30 or less 9.0 - 16 Notes: 1/ Coefficient on years of schooling in the function reported in Table 7, times 100. 2/ Coefficient on the "indigenous" dummy reported in the second function in Table 7, times 100. - 17 - IV. CQmpartiv Lessons What does this comparative exercise teach us? What interventions would improve the economic position of a minority ethnic group in a pluralistic society? Because of the differential composition of the sample in the two countries, the Guatemalan estimates were repeated using only the urban sub-sample. Among urban residents, the earnings advantage of the non-indigenous group was 97 percent, versus 137 percent for the country as a whole. On the other hand, there is a major difference in the two countries in where the indigenous people live. In Guatemala, only 13 percent of the indigenous people live in urban areas, versus 34 percent in Bolivia. Hence, the comparison of urban Bolivia to Guatemala as a whole might be more appropriate than resticting the analysis to urban areas in the two countries. Figure 1 shows the contrast between the two countries in terms of educational development and earnings, by ethnicity. The much lower educational attainment of the ethnic group in Guatemala relative to Bolivia, translates into a much sharper earnings differential in favor of the non-ethnic in Guatemala relative to Bolivia. Figure 2 depicts essentially the same relationship, but cast in terms of the returns to education. The two lines assume a continuity in terms of educational development within ethnic - 18 - groups across the two countries. The negative slope of both line3 implies that, for any ethnic group, the higher its level of educational attainment, the lower the returns to education the group enjoys. This is just a further documentation of the law of diminishing returns. What is of importance for the purpose of this analysis, however, is that the line referring to the indigenous group lies well below the line for the non-indigenous group. What this means is that in both countries, even if the two groups had the same level of educational attainment, the indigenous group would enjoy a several percentage points lower rate of return on their schooling investment relative to the non-indigenous group. The results presented here should be qualified by the fact the data do not permit to establish how much of the inferior economic position of the ethnic people in both countries is due to discrimination, either before or after joining the labor market. What we could say at most in this respect, is that the earnings differential between the two groups that remains unexplained, after controlling for human capital endowments, represents an upper bound of labor market discrimination. But there must be several other factors determining the earnings of these two very different groups, that we cannot be sure what is the degree of pure discrimination in the labor market. In spite these caveats, our investigation has shown that, other things equal (including culture) those indigenous people who acquire more Luman capital enjoy relatively higher economic rewards relative to their counterparts who have less human capital. Hence, the sheer provision of education to such groups, mainly in the form of basic education, is bound to - 19 - improve their position. The above intervention is supported not only from within-country evidence, where the more educated, even if indigenous, earn more than the less educated. It is also supported by the cross-country evidence that when a country has a higher level of educational attainment on average (Bolivia), the indigenous people fare much better relative to a country will a lesser educational attainment (Guatemala). Another possible intervention is the provision of schooling in the child's first language. This is a highly controversial issue, given the fact the child will have later to switch to Spanish for further education. Yet such intervention has been successfully implemented on a small scale in rural Guatemala. (See Chesterfield and Steely, 1986). Bilingual programs also exist in Bolivia, although at a small scale. The economic position of the indigenous people can be addressed from the perspective of several disciplines. Here we have used an economic perspective, in particular human capital theory, extended to accommodate the indigenous attribute of the wage or salary earner. In summary, from the above analysis we conclude that in a country with a high proportion of ethnic population, those who are indigenous have much lower levels of educational attainment and receive less rewards in the labor market relative to the non-indigenous group. We also conclude that education is a good investment for both groups, i.e. including the indigenous people. - 20- Eigure1 Education and Earnings by Ethnicity Mean years of schooling 12 - Indigenous E Non-indigenous 99 10 . 2 0 Guatemala Bolivia Mean earnings Index (Indigenous * 100) 300 i Indigenous E_ ]on-Indigenous 237 200 100 o 100 . 0G Guatemala Bolivia -21- Figure 2 The Returns to Education by Ethnicity and Level of Educational Development Rate of return (%/6) .16 14 G 12 G 10- 8 B B 6 2 4 6 8 10 Years of schooling Indigenous _ _mmmmmm Non-Indigenous G Guatemala observation B Bolivia observation - 22 - REFERENCES Arends, M., "Female Labor Force Participation and Earnings in Guatemala," Chapter 17, in G. Psacharopoulos and Z. Tzannatos, eds., Women's Employment and Pay in Latin Ameria, The World Bank, 1992. Becker, G.S., Human Cial. National Bureau of Economic Research, 1964. (Second edition, 1975). Chesterfield, Ray, and H. Ned Seeye, Process Evaluation of the Programa Nacional de Educacion Bilingue (PRONEBI). Final Report for USAID/Guatemala and PRONEBI. Washington, D.C.: United Agency for International Development, 1986. Guerra Borges, A., "Compendio de Geograffa Econ6mica y Humana de Guatemala". Editorial Universitaria de Guatemala, 1983. Kugler, B., Psacharopoulos, G., "Earnings and Education in Argentina: an Analysis of the 1985 Buenos Aires Household Survey", Economics of Education Review, Vol. 8, No. 4, 1989: 353-365. Mincer, J., Schooling. Experience. and Earnings. New York: Colombia University Press, 1974. Nyrop, R., Guatemala: A Country Study, American University, 1983. Psacharopoulos, G., Returns to Education: An International Comparson. Elsevier - Jossey Bass, 1973. Psacharopoulos, G., "Time Trends of the Retums to Education: Cross-National Evidence", Economics of Education Review, vol. 8, no. 3. 1989: 225-231. Schultz. T.W., "Investment in Human Capital," American Economic Review, 51 (1), March, 1961. Scott, K., "Women in the Labor Force in Bolivia: Participation and Earnings," Chapter 7 in 0. Psacharopoulos and Z. Tzannatos, eds. Women's Employment and Pay in Latin Amenca, The World Bank 1992. Stephen, David, and Wearne, Phillip, Central American's Indians with a summary by Rodolfo Stavenhagen. London: Minority Rights Group, 1984. Sumner, Daniel, "Wage Functions and Occupational Selection in a Rural Less Developed Country Setting," Review of Economics and Statistics, vol. 63, 1981:513-519. - 23 - Terreil, Katherine. -An Analysis of Wage Structure in Guatemala City," the Joumalf D.evelping Areas, vol. 23, April 1989: 405-423. Whetten, N., Guatemala: The Land and the People. New Haven: Yale University Press, 1961. Pollcy Research Working Paper Series Contact Title Author Date for paper WPS992 Regional Integration in Sub-Saharan Faezeh Foroutan October 1992 S. Fallon Africa: Experience and Prospects 37947 WPS993 An Economic Analysis of Capital S. Ibi Ajayi October 1992 N. Lopez Flight from Nigeria 34555 WPS994 Textiles and Apparel in NAFTA: Geoffrey Bannister October 1992 A. Daruwala A Case of Constrained Liberalization Patrick Low 33713 WPS995 Recent Experience with Commercial Stijn Claessens October 1992 Rose Vo Bank Debt Reduction Ishac Diwan 33722 Eduardo Fernandez-Arias WPS996 Strategic Management of Population Michael H. Bernhart October 1992 0. Nadora Programs 31091 WPS997 How Financial Liberalization in John R. Harris October 1992 W. Pitayatonakarn Indonesia Affected Firms' Capital Fabio Schiantarelli 37664 Structure and Investment Decisions Miranda G. Siregar WPS998 What Determines Demand for Freight Esra Bennathan October 1992 B. Gregory Transport? Julie Fraser 33744 Louis S. Thompson WPS999 Stopping Three Big Inflations Miguel A. Kiguel October 1992 R. Luz (Argentina, Brazil, and Peru) Nissan Liviatan 34303 WPS1000 Why Structural Adjustment Has Not Ibrahim A. Elbadawi October 1992 A. Maranon Succeeded in Sub-Saharan Africa Dhaneshwar Ghura 39074 Gilbert Uwujaren WPS1001 Have World Bank-Supported Ibrahim A. Elbadawi October 1992 A. Maranon Adjustment Programs Improved Economic 39074 Performance in Sub-Saharan Africa? WPS1002 World Fossil Fuel Subsidies and Bjorn Larsen October 1992 WDR Office Global Carbon Emissions Anwar Shah 31393 WPS1 003 Rent-Sharing in the Multi-Fibre Kala Krishna October 1992 M. T. Sanchez Arrangement: Evidence from U.S.- Ling Hui Tan 33731 Hong Kong Trade in Apparel WPS1004 Family Planning Programs in Sub- Regina McNamara October 1992 0. Nadora Saharan Africa: Case Studies from Therese McGinn 31091 Ghana, Rwanda, and the Sudan Donald Lauro John Ross WPS1 005 An Approach to the Economic Laszlo Lovei October 1992 M. Dhokai Analysis of Water Supply Projects 33970 Policy Research Working Paper Series Contact Title Author Date for paper WPS1 006 Preparing Multiyear Railway Jorge M. Rebelo October 1992 A. Turner Investment Plans: A Market-Oriented 30933 Approach WPS1007 Global Estimates and Projections Rodolfo A. Bulatao C tober 1992 0. Nadora of Mortality by Cause, 1970-2015 Patience W. Stephens 31091 WPS1 008 Do the Poor Insure? A Synthesis of Harold Alderman October 1992 C. Spooner the Literature on Risk and Christina H. Paxson 32116 Consumption in Developing Countries WPS1 009 Labor and Women's Nutrition: Paul A. Higgins October 1992 C. Spooner A Study of Energy Expenditure, Harold Alderman 32116 Fertility, and Nutritional Status in Ghana WPS1 010 Competition and Efficiency in Dimitri Vittas October 1992 W. Pitayatonakarn Hungarian Banking Craig Neal 37664 WPS1 011 How Tax Incentives Affect Decisions Robin Boadway November 1992 C. Jones to Invest in Developing Countries Anwar Shah 37754 WPS1012 The Brady Plan, the 1989 Mexican Haluk Unal November 1992 W. Patrawimolpon Debt Reduction Agreement, and Bank Asli Demirg0c-Kunt 37664 Stock Returns in the United States Kwok-Wai Leung and Japan WPS1 013 The Impact of Mexico's Retraining Ana Revenga November 1992 D. Young Program on Employment and Wages Michelle Riboud 30932 Hong Tan WPs1014 Ethnicity, Education, and Earnings George Psacharopoulos November 1992 L.Longo in Bolivia and Guatemala 39244 WPS1015 Benefit Incidence Analysis in Thomas M. Selden November 1992 C. Jones Developing Countries Michael J. Wasylenko 37754