Across developing countries, there is considerable under-investment in children's human capital; it is reflected in low immunization rates, child malnutrition, high drop-out rates, etc. Because of the (both individual and aggregate) long-term effects of human capital investment during childhood, governments across the globe have designed and implemented policies to encourage parents to invest more in the health and education of their children (numerous conditional cash transfer programs across countries are some examples). However, there is little hope for designing the most effective policies without an understanding of what underlies the decisions made by parents.
Many studies have looked at the effects of parental characteristics and household environment (parents’ education, household’s financial resources, and number and composition of siblings) on investment in children in developing countries (Thomas et al. 1991 , Duflo 2000 , Sawada and Lokshin 2009 ). However, we know much less about the role of the state of the labor market on parental investment decisions. My job market paper  examines the effects of changes in labor market opportunities in Mexico for women on the bargaining power of women within households and, ultimately, on investment in children's health.
The empirical strategy takes advantage of the segregated nature of the labor market for older versus younger women in the Mexican manufacturing sector and the geographic heterogeneity in demand across municipalities for younger versus older female labor within the Mexican export manufacturing sector between 2002-2005. A feature of many export industries across developing countries, including Mexico, has been the employment of female labor. However, importantly, there are sizeable and persistent differences in the share of older female labor out of total female labor across different industries. In some of these industries, such as manufacturing of rechargeable batteries, automobiles' electric systems, and household appliances, female labor older than the age of 24, who are mostly married, constitutes around 50 percent of the total female labor, while in others like manufacturing of vessels, cars, trucks, and motorcycles this ratio increases to almost 90 percent. An increase in labor demand in industries with strong preferences for hiring young, single, women, hardly affects the employability and the wage rate of older, married, women, and therefore their bargaining power.
To estimate the changes in demand for different gender-age categories of labor in each municipality, the paper uses two different methodologies. First, following Bartik (1991) , Blanchard and Katz (1992) , Bound and Holzer (2000) , and Autor and Duggan (2003) , and using data from the Social Security Institute of Mexico (IMSS), that includes monthly employment data from all formal private-sector establishment, I construct demand indices that capture exogenous shifts in local labor demand for different gender-age categories: women in the age category of 15-24 years, women who are 25 years of age or older, and men. The demand index for each gender-age-municipality cell is constructed based on the nationwide changes in employment of that gender-age category in different industries, weighted by the local labor market-specific shares of employment in each industry. The second methodology utilizes the differential effects of increases in Chinese exports to the U.S. following China's entry into the WTO in 2001 across industries in Mexican export manufacturing sector. I find similar findings using the two methodologies.
To examine changes in women's bargaining power within households, the paper uses data from two waves of the Mexican Family Life Survey (MxFLS 2002 and 2005). MxFLS has data on whether a husband or wife (or both) makes the decision in 12 different categories. I make use of this data to construct a qualitative measure of both spouses' bargaining power; the share of decisions made by each partner. This measure (along with the expenditure share of different goods that could be indicative of women’s bargaining power) is used to reveal whose preferences are reflected to a greater degree in household decisions and an increase in this variable is interpreted for each spouse as an increase in her/his bargaining power.
The paper finds that Increases in labor demand for older women (who are mostly married) increase women's bargaining power within households. The magnitudes suggest that a 10 percent growth in demand for older women in the export sector translates into a 13 percent increase in the share of decisions made by the wife in the household. Importantly, consistent with the theory of bargaining power, this effect is not limited to working women.
The paper then analyzes the effect of changes in demand for different gender-age categories of labor on investment in children’s health; the paper looks at the expenditure share of health services within households as well as two measures of health investment for girls and boys. One is the reported health condition of the child, and the other is whether a child older than the age of 4 has completed the preventive vaccination course. The paper finds that, in a typical Mexican municipality, a 10 percent increase in labor demand for older women in the export sector increases the chance of a daughter being in good health by 10 percent, and a daughter completing the vaccination course by 14 percent. I do not find any evidence that an increase in labor demand for older women changes investment in boys’ health. These results are consistent with evidence from the previous literature that suggest that mothers value daughters relatively more than fathers do (Thomas 1990 ; Thomas 1994 ; Behrman 1993 ; Strauss and Thomas 1995 ). However, it could also be the case that mothers are more equity minded and gender gaps in children’s health decline when their bargaining power increases.
Finally, the paper finds that changes in demand for younger women (who are mostly single) do NOT affect investment in girls' health. This finding is important; families may invest more in girls because the returns to that investment, in terms of longer-run labor market success, have increased. If households invested in daughters' health as the result of better employment prospects for them (as opposed to increased bargaining power of mothers), one would expect a positive demand shock for younger women to result in a positive and significant effect on investment in girls' health as well; this does not play out in the data.
Overall, the paper contributes to our understanding of the gender gap in human capital that has been of considerable concern in developing countries (Sen and Sengupta 1983 ; Das Gupta 1987 ; Baird, Friedman, Schady 2011  provide evidence). The findings here suggest that labor market opportunities might affect parental decisions to invest in children's health, and these decisions differ depending on who within the household experiences increased labor market opportunities. More specifically, more opportunities for women could result in disproportional investment in girls' health if mothers face more and better options to participate in the labor market.
Kaveh Majlesi  is a Ph.D. candidate at the University of Texas at Austin and is on the job market.