Three essays on labor economics

This dissertation addresses three broad issues within the fields of labor economics: family economics, wage compensation, and household decision making.;Chapter 2 studies the reasons for low levels of child support paid by non-custodial divorced fathers. When mothers remarry, part of their financial burden of raising their children shifts from fathers to stepfathers, providing an incentive for the father to encourage mother's remarriage through the use of strategically picked payment level. Assuming father's income is at least partially unobservable by mother after divorce, I first formulate a theoretical model of the father's optimal transfer and the mother's remarriage decision in a two-period dynamic model. The equilibrium closed form solution reveals that the father's optimal transfer is indeed lower when his income is partly unobservable to the mother than when it is fully observable. The issue is further investigated empirically, using data from the Panel Study of Income Dynamics (PSID) in the U.S. With income volatility as proxies for degree of unobservability, I examine the relationship between father's payment and his income unobservability, and find the more volatile his income is the lower he pays. In terms of the magnitude, associated with 1% increase in the income variation/level ratio, the annual reduction in child support transfer is estimated to be S6.6-32.3 in amount or 0.9-1.96% in likelihood, depending on the measure of income volatility used. Since alternative hypotheses are ruled out by observing no relationship between income volatility and additional controls, this provides evidence on the strategic explanation of inadequate payment.;Chapter 3 estimates the value of occupational risk to life in a time series context. Conventional estimates of this type are generally based on linear extrapolation of wage-risk premiums, assuming linear risk pattern and exogenous labor supply within a static context. For what it is worth, all these restrictions are not necessarily true and hence relaxed in this chapter. Using job information currently and 5 years ago of Census 1970, I find job risk has a long term effect on both employment probability and earned income if employed. In particular, previous risky job workers are less likely to remain employed in the future, but once employed, they tend to earn more income than their safe counterparts. With the latter effect almost 7 times as large as the former, the net impact on future expected labor income ends up being significantly positive, causing conventional static estimates to be downward biased. In terms of value of life, evaluated at the 1st percentile risk level of 0.3 per 100,000 workers, value of life estimated within a time series context is ;Chapter 4 examines the determinants of women's decision making power and their effects on the use of reproductive health care services in Pakistan. Based on a newly available nationally representative living standard survey, the most important determinant of women's decision making power identified is education. Logit model estimation further reveals that women's decision making power is significantly positive correlated with most reproductive health service uptake indicators, while husbands/male household heads' has the opposite effect. This finding rejects common preference model in support of bargaining theories, and provides theoretical foundation for women empowering policies. Particularly important aspects of women's decision making power to health services uptake identified through factor analysis are personal development and family planning, which might be able to explain the lack of consistent evidence in the literature. In view of men's decisive role, policy implication on men's part is also emphasized for the first time.;Keywords: Child Support, Remarriage, Risk, Value of Life, Reproductive Health, Women Empowerment...