The household production literature emphasizes that technical or biological processes condition input selection by households in their production activities, along with prices and income. Exogenous variations in health, to the extent that they are perceived by individuals (heterogeneity), lead to correlations between inputs and health outcomes that cannot be used to derive causal conclusions. Therefore, estimates of health technology must be obtained from a behavioral model in which health inputs are themselves choices. Consistent estimates are reported of the effect of endogenous inputs, such as medical care, smoking, and fertility, on birth weight and fetal growth in the presence of health heterogeneity.

Although estimates of the fertility-labor supply relationship abound, a full appreciation of the interpretation of such estimates has been lacking, regardless of the empirical strategy employed. This paper attempts to elucidate, within the context of a life-cycle decision-making process, the information contained in the estimated association between fertility and labor supply as calculated from "single" and "simultaneous-equations" estimation techniques. We also present a statistical methodology based upon the occurrence of twins in the first pregnancy and provide estimates, using that methodology, of the extent to which women's life-cycle labor supply decisions respond to exogenous (and, in this case, unanticipated) extra children.

In the last fifteen years, our understanding of the behavior of households in both highand low-income countries has increased significantly. Econometric studies contributing to this body of knowledge, concerned with the determinants of such behavior as labor supply, fertility, health and food consumption, however, have generally taken the structure of the household as exogenously given. While some progress has been made concerning how changes in the legal structure alter patterns of family formation and breakup, the wide variety of family organizations observed across countries of the world or the evolution of family structure within countries cannot be readily explained by existing models. While some economists have suggested that organizations based on kinship can be understood in terms of transaction economies (Yoram Ben-Porath, 1980; Robert Pollak, 1985), there are now few precise implications of this approach, and little or no evidence of its usefulness for predicting or explaining the existence of any particular family structure in specific settings characterized by their natural endowments and/or legal structure. In this paper I discuss some recent studies of family and household organization in one specific context, rural India, that have sought to formulate a model of household structure based on the need for individuals in lowincome, private information settings to protect themselves against intertemporal fluctuations in resources arising from the natural vagaries of water supply. I also present some new evidence on the relationship between family arrangements and measured risk characteristics of the agricultural environment in these settings. While many economists have highlighted risk and information consideration in the study of such formal rural institutions as sharecropping, permanent servitude, and contractual interlinking, studies of these individual contractural arrangements have ignored the family itself as a risk-mitigating institution. Moreover, much of this literature has been solely concerned with ex ante measures to reduce risk-contractual instruments or production factors that reduce the effects of variability in the exogenous production input rainfall on income. Mechanisms serving to preserve consumption stability ex post in the face of variability in realized income have been neglected.

Payoffs from Panels in Low-Income Countries: Economic Development and Economic Mobility by Mark R. Rosenzweig. Published in volume 93, issue 2, pages 112-117 of American Economic Review, May 2003

The roles of the entitlements of the AFDC program and marital prospects in the fertility and marriage choices of young women are assessed in the context of a model incorporating heritable endowment heterogeneity, assortative mating, concern for child quality, and potential parental and public support alternatives. Estimates based on data describing the fertility and marital experience up to age 23 of the eight birth cohorts of women in the NLSY provide evidence that higher AFDC benefit levels and lower marital prospects induce young women to choose to have a child outside of marriage.

This paper presents evidence from empirical studies that test hypotheses derived from models of household behavior pertaining to the interrelationships among population growth, human capital, and economic development. These studies have exploited quasi-natural experiments embodied in the cross-area variability in the wage rates of children in a number of low-income countries, the intercouple variation in the biological propensity to conceive, and the geographically selective introduction of new high-yielding seed varieties in India in the period 1961-71. The different varieties of evidence support the hypotheses that alterations in the returns to human capital associated with exogenous technical change lead simultaneously to increases in human capital investments and to reductions in fertility and that the costliness of fertility control is a significant but modest factor in inhibiting human capital investments.

A multiperiod household model of fertility behavior applicable to rural-agricultural settings in which the pecuniary as well as the psychic returns from children are significant is formulated. Implications regarding the effects of variables associated with the market for agricultural labor on the demand for farm children, including technical change, agricultural wage rates, farm value, and nonfarm employment opportunities, derived from the model are tested on U.S. aggregate data covering the period 1939-60. The empirical results obtained appear to support the theoretical model and suggest the importance of the reduction in the value of children as productive assets in agriculture as a factor in the postwar decline in the U.S. farm birth rate.