In this proposed multi-level and longitudinal household data from a large national probability sample of over 4500 rural households in India covering the period 1968-1982 and new data covering the period through 1999, merged with supplementary area-specific data from governmental sources and from satellites, will be used to reexamine comprehensively the effects of economic growth on the distribution of incomes across households and individuals, on economic mobility and on the formation and dissolution of households. A unique and critical feature of the merged data is that they document a period of time in India which conforms closely to a natural experiment in which the source of economic development - the "green revolution" - was exogenous to the population, was measurable, and, due to natural conditions, was geographically selective. It is thus possible to document how economic development, propelled by agricultural productivity growth, differentially affected human capital returns and investments and other household behavior.
The underlying theme of this work is that the evaluation of economic change on individual welfare must importantly account for how individuals are grouped into households and how household interact through transfers, markets, and local political institutions. The particular aims of the project include (a) the achievement of a better understanding of household behavior and intra-household distribution, (b) an assessment of the contributions of interhousehold transfers to income security and investment in a dynamic context, (c) estimation of the determinants of household break-up and formation, with particular focus on the impact of technical change, (d) an evaluation of the role of dowry in affecting investments in gender-specific human capital and the distribution of resources across households, and (e) the evaluation of the role of local governments in allocating land resources and other public goods with attention to both distributive and efficiency criteria and with particular attention to forest resources situated on common lands.
These new inquiries will be based on the availability of a new survey round of the panel data used in the prior project work. The new data not only expand the time frame for analysis another 17 years, for a total span of temporal coverage of more than 30 years, but provide new information particularly suited to analyses of households in a dynamic context. In particular, the new data provide information more consistent with new frameworks describing family and household relations and will include split-off households in the panel for the first time as well as retrospective information on the marriages of all family members and household division histories. These novel and unique household data will in particular permit a detailed exploration of family and household relations and how these and the outcomes of family decisions are affected over the long as well as the short-run by economic change. The new data will also provide retrospective data on the governance structures of the villages, enabling the construction of a village-level data set of approximately 250 villages over 30 years with which to explore whether and how village governance affects the distribution of resources across households, common land use and the exploitation of forest resources. The availability of these unique data spanning a long time-period yet providing detail at the individual, household level and family level will thus provide an improved empirical basis for assessing theories of the distributional effects of growth and development as well as of household behavior.
