Demand systems estimation increasingly makes use of household-level microdata, mainly to measure the effects of demographic variables. Data based on these household-expenditure surveys present a major estimation problem. For any given household, many of the goods have zero consumption, implying a censored dependent variable. Techniques which do not take this censored dependent variable into account will yield biased results. We utilize a censored regression approach that is computationally simple, consistent, and asymptotically efficient. The results are then presented and compared with those obtained using an uncensored technique.

/pdf/demand-systems-estimation-with-microdata-a-censored-4by0b6q6ud.pdf

Demand Systems Estimation With Microdata: A Censored Regression Approach

This paper places the debate over using consumption or income in studies of inequality growth in a formal intertemporal setting. It highlights the importance of permanent and transitory income uncertainty in the evaluation of growth in consumption inequality. We derive conditions under which the growth of variances and covariances of income and consumption can be used to separately identify the growth in the variance of permanent and transitory income shocks. Household data from Britain for the period 1968-1992 are used to show a strong growth in transitory inequality toward the end of this period, while younger cohorts are shown to face significantly higher levels of permanent inequality.

/pdf/consumption-inequality-and-income-uncertainty-34fhv7x4xv.pdf

Consumption Inequality and Income Uncertainty

Book description:
This book is an in-depth discussion of rising inequalities in the western world. It explores the extent to which rising inequalities are the mechanical consequence of changes in economic fundamentals (such as changes in technological or demographic parameters), and to what extent they are the contingent consequences of country-specific and time-specific changes in institutions.

Both the 'fundamentalist' view and the 'institutionalist' view have some relevance. For instance, the decline of traditional manufacturing employment since the 1970s has been associated in every developed country with a rise of labor-market inequality (the inequality of labor earnings within the working-age population has gone up in all countries), which lends support to the fundamentalist view. But, on the other hand, everybody agrees that institutional differences (minimum wage, collective bargaining, tax and transfer policy, etc.) between Continental European countries and Anglo-Saxon countries explain why disposable income inequality trajectories have been so different in those two groups of countries during the 1980s-90s, which lends support to the institutionalist view.

The chapters in this volume show the strength of both views. Through empirical evidence and new theoretical insights the contributors argue that institutions always play a crucial role in shaping inequalities, and sometimes preventing them, but that inequalities across age, sex, and skills often recur. From Sweden to Spain and Portugal, from Italy to Japan and the USA, the volume explores the diversity of the interplay between market forces and institutions.

http://sites.psu.edu/scottcolby/wp-content/uploads/sites/13885/2014/07/Blundell1987_Bivariate-alternatives-to-the-tobit-model.pdf

Bivariate alternatives to the tobit model

This survey is designed to emphasize the relationship between empirical a nd theoretical considerations in the analysis of consumer behavior. A lthough it focuses primarily on analysis at the individual or micro l evel, it tries, where possible, to draw implications for the estimati on of aggregate relationships. The importance of micro-level analysis stems from the rapid expansion in available panel and survey data se ts. This development has generated its own theoretical and empirical issues, many of which are raised in this survey. Copyright 1988 by Royal Economic Society.

M.J. Keen

Papers

Estimating consumption from expenditure data