Trade and Education Divergence A Gender Perspective

Document Sample
Trade and Education Divergence A Gender Perspective Powered By Docstoc
					          Trade and Education Divergence: A gender Perspective

                                      Ramya Vijaya

                                Richard Stockton College


         This paper investigates the impact of trade flows on gender differentiated
educational attainment data. We find that in developing economies there is a persistent
negative relationship between changes in trade volumes and women’s enrollment in
secondary education. These results can be understood by placing in a gendered context,
the prediction from traditional trade theory that countries specializing in low-skilled
production for trade purposes can experience a reduction in incentives to invest in
education. Since low-skilled export production has relied predominantly on female labor,
the trade education link would be particularly significant for female enrollment rates.
These findings we believe are an important extension to the literature that has questioned
the view that trade and the resultant increases in income and employment would have an
unqualified positive impact on women’s well being.


       This paper presents a cross-country empirical evaluation of the impact of

increased trade volumes on women’s participation in higher education. Neo-classical

trade theory suggests that trade-related employment patterns can create changes in

incentives to acquire education depending on a country’s initial comparative advantage.

Such changes can lead to a divergence in skill investments between capital-intensive

developed and labor-intensive developing economies. This paper presents a gender

differentiated analysis of the divergence hypothesis that incorporates the insights from

feminist economics literature. Based on observation that major exporters of manufactured

goods have relied on low-wage labor of women workers, this paper considers the

implication of trade-related employment opportunities for gender skill differentials. We

find a negative relationship between trade participation female enrollment in secondary

education for the group of low and middle income countries.

       We believe that this link between trade and education is important given the

current trends in participation in education. As one of the United Nations Millennium

development goals, women’s participation in education has come under increasing policy

focus. Studies that have evaluated the progress towards the goal of gender equality in

education note that gender gaps in enrollment in secondary education have been closing

(UNIFEM 2002). However they also note that there are considerable shortfalls in the

levels of enrollment. Less than 20 percent of the countries have achieved enrollment rates

of 90 percent or more for secondary education for girls. In fact the average enrollment

rate in secondary education for women in low and middle-income countries is 64 percent

(UNESCO Education Database Data for 2002-2003).

       Moreover, the studies also note that while the narrowing of the gender gap in

education is a significant accomplishment it falls short of being a true indicator of the

empowerment of women for two reasons. The first as noted above is the low levels of

enrollment. The second more important consideration is the links between education and

the availability of economic opportunities to apply the education. This latter criterion not

only provides a more accurate assessment of the empowerment of women, it also

provides an indication of the sustainability of the efforts. That is higher enrollments are

more likely to continue when there are economic opportunities that reward the investment

in education. Since the data indicates that a large number of women depend on trade-

related employment opportunities, the theoretical prediction of trade impacting

educational incentives provides an important context for examining the dynamic links

between education and economic opportunity for women.

       The following section expands on the theoretical relationship between trade and

incentives to invest in education.

Trade and Education:

       According to the traditional Heckhsher-Ohlin trade theory, trade increases the

demand for and hence the production of goods produced by the country’s abundant

resources. Since the abundant resource in many developing economies is unskilled labor,

trade is expected to lead to an expansion of production in low-skilled labor-intensive

industries in these countries. Ronald Findlay (1995) extends the H-O model to show that

countries specializing in the production and export of labor-intensive goods will tend to

see a decline in the incentive to invest in education due to the immediate expansion in

low-skilled employment. In the long run this can lead to a tendency to “deccumulate”

rather than accumulate human capital in these countries.

       The developed economies on the other hand start out with a relatively higher

endowment of capital and skilled labor and therefore tend to specialize in high-skilled

production. This expands the demand for high-skilled labor and therefore provides an

incentive to invest further in education and skill acquisition.

       This demand-induced divergence in incentives therefore predicts increasing

difference between patterns of investment in education between the developed and

developing countries. Consequently the developing countries are never able to ‘catch up’

or converge to the skill levels of the richer countries.

       Empirical evidence for the divergence hypothesis is mixed. Adrian Wood and

Cristóbal Ridao-Cano (1997) examine data for 90 countries during the period 1960-90 to

see if incentives to invest in education are indeed impacted by increased trade openness.

The findings indicate that greater trade openness does indeed have a negative impact on

school enrollment rates in developing countries and a positive impact on enrollments

rates in developed countries as predicted by the H-O theory. However Wood and Ridao-

Cano also suggest that the above negative impact can be countered by other trade-related

mechanisms. On such mechanism they note, is the income effect of trade. That is as trade

increases overall output and income growth in the economy, people are more able to

spend on human capital investments.

       In addition increased trade links are also expected to provide developing countries

with the opportunity to come in contact with and learn from the more advanced

technologies of the developed word, therefore narrowing the knowledge gap between the

low-skilled and high-skilled economies (Jeffery D. Sachs and Andrew M. Warner 1995;

David T. Coe, Elhanan Helpman and Alexander Hoffmaister, 1997).

       While the evidence is inconclusive, there is another more important reason for

revisiting the debate. The above theoretical framework and the empirical studies assume

homogenous demand for both male and female labor. Evidence from feminist economic

literature indicates that the impact of trade has not been gender neutral. Depending on the

level of economic development there are varying but distinct gender patterns in trade-

related employment changes in most countries. For instance developing countries that

adopted a trade-oriented growth strategy tended to draw substantially on female labor to

expand the export-oriented manufacturing sector. This trend often referred to as the

feminization of labor has been noted by several studies such as Guy Standing (1989,

1999), Nilufer Cagatay and Sule Ozler (1995) and Susan Joekes and Ann Weston (1994).

Rekha Mehra and Sarah Gammage (1999) also cite evidence for this trend from various

developing countries such as Dominican Republic, Sri Lanka and Tunisia.

       In these countries therefore the gender trend in trade-related employment could

also imply a gender pattern to the changes in incentive to invest in education. That is

since the trade led expansion in low-skilled production largely relies on recruiting low-

skilled female workers, it is primarily women’s incentive to invest in education that

might change.

       Moreover interactions between employment expansion and women’s

commitments to household production might also imply a greater potential for negative

changes in women’s rather than men’s incentives to invest in education. Several studies

have evaluated the impact of time poverty as paid employment expansion leads to a

double burden on women’s time, particularly in developing economies (William Darity

1995, Lourdes Benería and Shelley Feldman 1992). For example in the context of sub-

Saharan Africa Darity (1995) presents a two-sector framework for the economy – a paid

cash-crop sector and an unpaid household or subsistence sector. As more women enter

the cash crop sector, there is much less time available for the household or subsistence

sector. Extending the model, the study observes that in low-income economies where

women shoulder the time burden, an export boom might lead to nutritional deprivation

for women as subsistence production shrinks. A similar argument could be made that the

time burden increases the possibility that with an export boom women might be more

likely to withdraw from education investments.

       On the other hand there is still the income effect to contend with. As mentioned

earlier, it has been suggested that overall increases in income due to trade can

compensate for the negative incentive effect. However evidence from some studies

indicates that the income effect may also not be gender-neutral. For example in the case

of Taiwan, Susan Greenhalgh (1985) observed that families tended to use the earnings of

girls employed in low-skilled export-oriented sectors to fund higher education for male

siblings, thereby continuing the gender differentials in education. Similarly increased

employment of mothers might also shift the burden of household production on to

daughters, therefore leaving less time for education participation.

       There are also questions regarding the long-term sustainability of the gains from

trade-related employment expansions for women. In many developing economies the

initial trend of feminization began to reverse as countries transitioned into higher skilled,

capital-intensive exports. Evidence for this defeminization trend from a range of

countries is presented for example in the 1999 UN world survey of women in

development and Berik (2000).

       The seeds of this defeminization it seems were present even in the initial trend of

feminization. Seguino (2000) argues that even as countries adopted the female-oriented

manufacturing export growth strategy, the earnings from the expanding export sector

were often used to import more capital-intensive technologies to stimulate further growth.

However the sectors that applied the new capital-intensive production processes mostly

employed men while women were stuck in the low-skilled export sector. Therefore even

with defeminization of overall manufacturing (with rising sophistication of exports)

women’s opportunities in manufacturing and service sectors may be restricted to low skill

employment, thus we might still expect a gender skill divergence due to trade.

       In summary, the H-O theory suggests declining incentives to invest in higher

education in response to an expansion of low-skilled employment. However as observed

before, if the expansion in low-skilled employment is primary targeted toward women it

is specifically women’s incentive to invest in education that might be impacted.

Moreover to the extent that women face the time poverty issue while engaging in both

paid and unpaid work, time to invest in education becomes more of an issue for women.

In addition, to the extent that higher skilled capital-intensive sectors are closed off to

women workers and they continue to be crowded into the low-skill manufacturing

exports (Seguino 2000), the impact of the H-O disincentive is further reinforced.

       These varied impacts therefore support a re-examination of the evidence for the

H-O based incentive effect from a gender perspective. We therefore conduct an empirical

evaluation on gender differentiated education data. In the following section we discuss

the empirical model.

Empirical Specification:

In order to examine the impact of trade on the incentive to acquire higher skills, Wood

and Ridao Cano (1999) use data on school enrollments rates as the measure of the

decision to invest in skills. Here we are interested in extending the model to investigate

the gender-differentiated impact. We therefore use gender differentiated school

enrollment data.

       The independent variable that is of primary interest is the extent of trade

expansion in the countries. Again following Wood and Ridao-Cano, total trade (that is

the volume of imports and exports) as a percentage of GDP is used as the primary

indicator of the economy’s trade participation.

       Besides trade related employment, other economic factors such as overall living

standards, availability of other non-trade opportunities and the amount of resources

allocated to education would also influence enrollment rates across countries. Higher

living standards and stronger economies can be expected to increase people’s abilities to

afford investments in education. Similarly greater resources for education can make

education more accessible to greater sections of the population and therefore increase

enrollment rates. In order to account for some of the economic influences, the empirical

estimation includes the per capita Gross Domestic Product (GDP) and the number of

secondary education teachers as explanatory variables. The per capita GDP variable is

used alternatively as a measure of overall living standards and economic performance.

The number of teachers is intended to measure the resource allocation towards education.

       Since we are interested in the change in education patterns, it is more appropriate

to estimate the equation with the rate of change in enrollment rates rather than the current

enrollment rate. All the variables in the equation therefore represent rates of change. The

estimating equation therefore is:

       ERit = 0i + 1 TRit + 2 Percapit + 3 Teacherit + eit

       Where ER is the rate of change in school enrollment for each year. Separate

equations are estimated for the male and female enrollment rates. TR is the change in

ratio of trade to GDP, Percap the change in per capita GDP and Teacher, the change in

the number of secondary teachers employed. 0i is the country specific fixed effect, i and

t are the country and time indices respectively and eit is the error term.

Data Description:

       The gender trends and implications of trade differ across groups of countries

depending on the stages of development and economic structure. We therefore expect the

impact of trade on education to also differ for different groups of countries. For instance

the feminization of labor hypothesis has focused considerably on countries that adopted

the manufacturing export oriented approach to growth. Seguino (2000) refers to this

group of countries as semi-industrialized since a large share of the exports is produced in

relatively low-skilled manufacturing industries such as textiles, apparel and electronics.

Therefore they are different from the developed countries which focus on high-skilled

production and also different from countries that rely mostly on agricultural exports.

Since it is this low-skill manufacturing export sector that has been at the forefront of

increasing female employment, it is particularly in these countries that we might expect

to find the trade induced changes in education incentives for women.

        On the other hand agricultural exports could also lead to reduced time for

education due to the time poverty issue that women are faced with. In our empirical

analysis therefore we attempt to accommodate these differences by estimating the model

for different groups of countries and comparing the results. First however we start by

collecting available data for all countries. The years for our panel data are from 1980-

2000. The time period was chosen with a view to parallel the growth in importance of

export orientation as a growth strategy while accommodating the availability of

consistent yearly data.

        The data comes from the World Bank World Development Indicators (WDI,

2003). The school enrollment data is the gross enrollment in secondary education from

the WDI. We use the secondary enrollment here since as we mention we are more

interested in education higher than the primary level which is more relevant for job skills.

        The use of enrollment data suffers from several disadvantages. The enrollment

data does not provide a clear picture of actual attendance or completion of relevant grade.

Moreover it also does not provide any indication of school quality and differences in

programs of study between male and female students. Data on schooling also leaves out

crucial information on skills training acquired at work. However quantitative assessments

for such a comprehensive notion of skills are not available on a consistent cross-country

basis. Moreover there are also limitations on the availability of gender-disaggregated

data. Therefore, following Wood and Ridao-Cano the secondary school enrollment rates,

which are available on a gender-disaggregated basis, continue to be the dependent

variables in this study.

       We also limit this analysis to secondary rather then tertiary enrollments for a few

reasons. First the availability of yearly data for the longer time horizon we are interested

in here is sparse in the case of tertiary enrollment and would result in many countries

dropping out of the analysis. Moreover we also believe that summary tertiary education

data is less comparable across countries and gender due to differences in specializations.

Secondary education on the other hand would tend to represent more generalized skills

that can be compared.

       The independent variable - GDP per capita, number of secondary school teachers

and trade as a percentage of GDP, are also from the WDI. Besides these economic

influences, non-economic factors such as cultural and social norms towards education,

particularly female education can also be expected to have a substantial impact on the

school participation rates across countries. However these influences tend to be very

specific to each country and consistent quantifiable cross-country data is not readily

available. However this study takes advantage of the panel data format where a fixed

effects model allows a country specific constant term to account for some of these social

and cultural influences.

Clarifying Causality and Other Estimation Issues:

       The direction of causality in the relationship between trade and female education

is also an important issue that the empirical evaluation has to address. The primary

hypothesis tested here is that trade expansion might lower female incentives to invest in

education since such expansions have relied considerably on increased employment of

women in low-skilled work particularly in developing economies. Therefore trade has the

potential to reinforce traditional biases and negatively impact women’s education

participation rates. However the gender pattern in trade-related employment may in fact

be caused by the fact that women are perceived to invest less in skill-acquisition.

Consequently the causality is not clear. That is a pattern of low female participation in

higher education exists which causes a certain pattern of trade-related employment that in

turn deepens the pattern of low female participation in education.

       In order to account for this cyclical nature of causality and isolate the impact of

trade on the current trend in female education participation, the instrumental variables

approach is adopted for the empirical estimation. We use the five year lagged value of the

rate of change in trade as an instrument for the trade variable since past changes in trade

are unlikely to be influenced by current rates of education participation. The five year

lags also is more realistic given that changes in trade volumes will likely have a more

gradual rather than immediate impact on enrollments. The two-stage least square

estimation method is used to estimate the regression with the instrumental variable.

Estimation Results:

       The initial estimation is conducted without any restrictions, including available

data for all countries. The results in Table 1 indicate that the amount of resources devoted

to education has the strongest impact on enrollment rates. The teacher variable has a

statistically significant influence on both male and female enrollments. A one percent

increase in the rate of change in the number of secondary school teachers causes a

0.12(female) to 0.13(male) per cent increase in the rate of change in enrollment rates. The

per capita GDP also has a significant positive impact on male enrollment rates. The trade

variable is not a statistically significant influence here. However this group includes all

countries, ranging from high income developed economies to low income developing

economies. We would therefore not expect a single pattern for the influence of trade on

education to emerge here. Since many of the gender and trade influences discussed earlier

are more applicable to developing economies, we re-estimate the equations with a

particular focus on such economies.

Table 1: Instrumental Variable Regression- All Countries (standard errors in parentheses)

Variable                 Female Enrollment                  Male Enrollment

Trade                    -0.006                             -0.022
                          (0.010)                           (0.017)

Percapita                 0.020                             0.061*
                         (0.032)                            (0.038)

Teachers                  0.123***                          0.139***
                         (0.019)                            (0.022)

N                          663                              663
R-squared                 0.09                              0.08
Notes: *** Statistically significant at 99%; ** statistically significant at 95%; *statistically
significant at 90%.

        We begin by selecting countries that are classified as low and middle income

countries in the World Development Indicators. Table 2 presents the results for this group

of countries subject to data availability. The list of countries and the descriptive statistics

is provided in tables 6 and 7. For this group, the teacher variable continues to be a strong

influence on enrollments, highlighting the importance of sustained resource allocations to

education. In the female enrollment equation, the trade variable is now statistically

significant and negative. In the male enrollment equation the trade variable is not

statistically significant. This provides support to the main hypothesis regarding the

gendered impact of trade on education discussed in this paper.

Table 2: 2-Stage Instrumental Variable Regression- Middle and Low Income Countries (standard
errors in parentheses)

Variable                 Female Enrollment                  Male Enrollment

Trade                    -0.030*                            -0.024
                          (0.017)                           (0.020)

Percapita                -0.014                              0.018
                          (0.048)                           (0.056)

Teachers                 0.212***                           0.235***
                         (0.024)                            (0.031)

N                        503                                503

R-squared                 0.15                              0.13
Notes: *** Statistically significant at 99%; ** statistically significant at 95%; *statistically
significant at 90%.

        To see if the results are driven by any particular group of countries and to test for

robustness, the next set of regressions groups the countries according to different

criterion. We begin by excluding from the middle and low income group, countries that

are primarily oil exporters since these countries might have a pattern of relatively high

per capita income combined with low human capital and gender equality. Our criterion

for oil exporters is a country where oil constitutes more than 90 percent of total exports.

The results in table 3 are similar to table 2. The trade variable continues to be negative

and significant in the female enrollment equation and not significant in the male

enrollment equation. The economic impact of the variable is also similar- about a 0.03

percent decline in the rate of change of female enrollment with a 1 percent increase in the

rate of change in trade. Here the GDP per capita has a statistically significant positive

impact on male enrollments.

Table 3: Instrumental Variable Regression- Middle and Low Income Non-Oil Countries (standard
errors in parentheses)

Variable                 Female Enrollment                  Male Enrollment

Trade                    -0.027*                            -0.001
                          (0.012)                           (0.020)

Percapita                 0.028                             0.088*
                         (0.051)                            (0.053)

Teachers                   0.215***                         0.224***
                          (0.028)                           (0.030)
N                         442                               442
R-squared                 0.15                              0.14
Notes: *** Statistically significant at 99%; ** statistically significant at 95%; *statistically
significant at 90%.

        In the next set of regressions, the focus is on predominantly food exporting

countries. The countries are selected from the low and middle income group based on the

food exports to total exports ratio. The ratio for the selected countries is one standard

deviation above the mean ratio for the entire group. For such countries the Darity model

(1995) indicates a link between trade and time poverty for women which might lead to a

negative impact on education. The results report a negative relationship between trade

and female enrollments. The trade variable is negative and significant only in the female

enrollment equation (table 4). The teacher variable is once again positive

and statistically significant for both the male and female enrollments.

Table 4: 2-Stage Instrumental Variable Regression - Low and Middle Income Food Exporting
Countries (standard errors in parentheses)

Variable                 Female Enrollment                  Male Enrollment

Trade                    -0.043*                            -0.010
                          (0.027)                           (0.035)

Percapita                - 0.001                            0.082
                         (0.077)                            (0.102)

Teachers                  0.206***                          0.272***
                         (0.046)                            (0.062)

N                         137                               137
R-squared                 0.19                              0.16
Notes: *** Statistically significant at 99%; ** statistically significant at 95%; *statistically
significant at 90%.

        The final set of regressions attempts to focus on ‘semi-industrialized’ countries

that concentrate on manufacturing exports. We selected countries from the middle and

low income group where the ratio of manufacturing to total exports is one standard

deviation above the mean manufacturing to exports ratio for the whole group (country list

in table 7). The expansion in low-skilled trade-related employment is associated with

expanding manufacturing exports in many developing economies (Seguino 2000). It is

therefore for this group of countries that we would expect the strongest support for the

negative impact of trade on education incentives. In table 4 the trade variable is negative

and significant in both the male and female enrollment equations. A 1 percent increase in

the rate of change in trade causes a 0.8 percent decline in the rate of change of

educational enrollments (male and female).

        While these results indicate support for the H-O hypothesis, they do not suggest a

gender divergence like the earlier groups since there is a negative impact for both male

and female enrollments. This however adds another important gender dimension to the

H-O hypotheses. The decline in the gender gap for labor-intensive economies might be

the result of a downward convergence, where male enrollments are declining and

stagnating. This would fit with the observed pattern of continued low enrollment rates for

both men and women even while gender gaps are narrowing.

Table 5: Instrumental Variable Regression -Low and Middle Income Manufacturing Exporting
Countries (standard errors in parentheses)

Variable                 Female Enrollment                  Male Enrollment

Trade                    -0.084**                           -0.088**
                          (0.043)                           (0.044)

Percapita                 0.001                             0.007
                         (0.02)                             (0.095)

Teachers                  0.41***                           0.364***
                         (0.055)                            (0.057)

N                         195                               195
R-squared                 0.30                              0.22
Notes: *** Statistically significant at 99%; ** statistically significant at 95%; *statistically
significant at 90%.


        For all the above estimations the teacher variable has the strongest influence on

enrollment rates. The positive influence of this variable is considerably more powerful

that the negative impact of the trade variable in all the cases. This is not surprising given

that enrollment rates have been increasing in most regions of the world indicating that the

positive influences on enrollments have been stronger. Resource allocation towards

education has also been receiving more attention as evidenced by the inclusion of

primary and secondary education targets in the Millennium Development Goals. This

positive trend will hopefully continue. However this analysis highlights the importance of

understanding the obstacles that might be posed by unsuitable economic environments.

        The empirical analysis presented here indicates that increased trade has in fact

created an environment that might diminish the success of efforts to encourage education

enrollments in some groups of countries. The negative impact of trade changes on

women’s educational attainment is particularly persistent across different groups of

developing economies. Feminist economic studies have increasingly questioned the view

that income and employment gains from trade would have an unqualified positive impact

on women’s well being. Such studies have previously focused on the issue of time

poverty and on the persistence of gender wage gaps that are often supported by the nature

of trade-related employment. The links between trade and education adds a crucial

extension to this literature.

Table 6: Country Lists

Middle and Low-Income (ML)              ML-Manufacturing     ML-Food

Algeria                  Panama         Argentina            Belize
Argentina                Peru           Bangladesh           Cameroon
Bahrain                  Philippine     Barbados             Colombia
Bangladesh               Saudi Arabia   Brazil               Costa Rica
Barbados                 Senegal        China                Cote d’Ivoire
Belize                   Sri Lanka      Dominican Republic   Ecuador
Bolivia                  Swaziland      El Salvador          Ethiopia
Botswana                 Syria          Indonesia            Fiji
Brazil                   Tanzania       Jamaica              Gambia
Cameroon                 Thailand       Jordan               Ghana
Chile                    Togo           Malaysia             Guatemala
China                    Tunisia        Malta                Guyana
Colombia                 Turkey         Mexico               Honduras
Costa Rica               Uganda         Morocco              Kenya
Cote d’Ivoire            Venezuela      Pakistan             Madagascar
Dominican Republic       Zambia         Philippines          Malawi
Ecuador                  Zimbabwe       Sri Lanka            Mozambique
Egypt                                   Thailand             Nicaragua
El Salvador                             Tunisia              Panama
Ethiopia                                Turkey               Senegal
Fiji                                    Zimbabwe             Tanzania


Table 7: Descriptive Statistics: Mean and Standard Deviation (in parenthesis)

Variable                    All Countries               Middle & Low Income
Female Enrollment               51.60                          40.25
                                (32.73)                        (24.07)

Male Enrollment                 54.73                            44.64
                                (28.71)                          (21.36)

Trade                           36.50                            70.59
                                (50.74)                          (40.20)

GDP per capita                  5190.50                          2938.74
                                (4904.12)                        (2288.3)

Teachers                        141706                           113932
                                (398838.3)                       (417997.9)
Rates of Change
Female Enrollment               0.030                            0.034
                                (0.031)                          (0.032)

Male Enrollment                 0.022                            0.024
                                (0.026)                          (0.028)

Trade                          -0.009                            0.0008
                                (0.041)                          (0.023)

GDP Per capita                  0.050                            0.047
                                (0.023)                          (0.020)

Teachers                        0.037                            0.048
                                (0.041)                          (0.040)

Number of Countries             85                               63


Beneria, Lourdes, and Shelley Feldman. (1992) Unequal Burden: economic crises,
persistent poverty, and women's work. Boulder, CO: Westview Press.

Berik, Gunseli. (2000) “Mature Export-Led Growth and Gender Wage Inequality in
Taiwan.” Feminist Economics 6 (3): 1-26.

Blecker, Robert and Stephanie Seguino. (2002) “Macroeconomic Effects of Reducing
Gender Wage Inequality in an Export-oriented, Semi-industrialized Economy.” Review of
Development Economics, 6 (1): 103-119.

Bowles, Samuel. (1991) “What Markets Can and Cannot do.” Challenge July-August.

Cagatay, Nilufer and Sule Ozler. (1995) “Feminization of the labor force: The effects of
long-term development and structural adjustment.”. World Development 23(11): 1883-

Coe, David, Elhanan Helpman, and Alexander W Hoffmaister. (1997) “North-South
R&D Spillovers.” Economic Journal 107: 134-59.

Darity, William. (1995) “The Formal Structure of A Gender-Segregated Low-Income
Economy.” World Development 23(11): 1963-1968.

Eric Edmonds & Nina Pavcnik. (2002) “Does Globalization Increase Child Labor?
Evidence from Vietnam.” National Bureau of Economic Research Working Paper 8760.

Elson, Diane. (1999) “Labor Markets as Gendered Institutions: Equality, Efficiency and
Empowerment Issues.” Word Development 27(3): 611-627.

Findlay, Ronald. (1995) Factor Proportions, Trade and Growth. Cambridge. MA:MIT

Greenhalgh, Susan. (1985) “Sexual Stratification: The other Side of ‘Growth with
Equity’ in East Asia.” Population and Development Review 11(2): 265-314.

Grown, Caren, Diane Elson, and Nilufer Cagatay. (2000) “Introduction.” World
Development 28(7): 1145-1155.

Hanson, Gordon and Ann Harrison. (1999) “Trade Liberalization and Wage Inequality in
Mexico.” Industrial and Labor Relations Review 52(2): 271-88.

Joekes, Susan P. and Ann Weston. (1994) Women and the New Trade Agenda. New

Kabeer, Naila. (2001) The Power to Choose: Bangladeshi Women Workers and Labour
Market Decisions. Dhaka. The University Press Limited.

Leamer, Edward. (1998) “In Search of the Stolper-Samuelson Linkages between
International Trade and Lower Wages.” in S. Collins ed. Imports, Exports and the
American Worker. Washington DC: Brookings Institution.

Mehra, Rekha, and Sarah Gammage. (1999) “Trends, Countertrends, and Gaps in
Women's Employment.” Word Development 27(3): 533-550.

Mayer, Jorg. (2000) Globalization, Technology Transfer and Skill Accumulation in Low-
Income Countries. United Nations Conference on Trade and Development, Washington:
United Nations.

Ozler, Sule. (2000) “Export Orientation and Female Share of Employment: Evidence
from Turkey.” Word Development 28(7): 1239-1248.

Parker, Susan Wendy, And Carla Pederzini. (2000) “Gender Differences in Education in
Mexico.” World Bank Departmental Working Paper no. 21023.

Sachs, Jeffrey.D, and Andrew Warner. (1995) “Economic Reform and the Process of
Global Integration.” Brookings Papers on Economic Activity 1:1-118.

Seguino, Stephanie. (2000) “Gender Inequality and Economic Growth: A Cross-Country
Analysis.” Word Development 28(7): 1211-1228.

Standing, Guy. (1989) “Global Feminization through Flexible Labor.” Word
Development 17(7): 1077-1095.

Standing, Guy. (1999) “Global Feminization through Flexible Labor: A Theme
Revisited.” Word Development 27(3): 583-602.

Wood, Adrian, and Cristobal Ridao-Cano. (1999) “Skill, Trade and International
Inequality.” Oxford Economic Papers 51: 89-119.

World Survey on the Role of Women in Development: Globalization, Gender and Work.
(1999) New York. United Nations Press.

Progress of the World’s Women. (2002) United Nations Development Fund for Women.


Shared By:
leader6 leader6