Outcome Tests and Screening Errors

Outcome Tests and Screening Errors

Hamish Low (Federal Reserve Bank of Chicago)

You may follow the seminar at the following link.

ABSTRACT:

This paper clarifies the conditions needed to test for discrimination, and shows how to test whether observed disparate treatment is taste-based or statistical. We use the context of applications for disability insurance where an authority (the SSA) decides whether to award benefits, possibly with a taste for discrimination against specific groups. We stress the need to specify the objective of the authority, to characterize their information set, and the importance of independent data on an individual's type. We focus on an authority with an objective to avoid Type I and Type II errors. First, we show that outcome tests \`a la Becker may fail to detect discrimination when the distribution of estimated eligibility differs for people who are truly eligible from those who are not -- even in circumstances in which the distribution of ``ground truth'' is independent of group affiliation. Second, we show that self-reported data on eligibility (specifically, self-reported disability) and detailed data on the information set of the authority can distinguish between taste-based discrimination and statistical discrimination. We find strong evidence of taste-based discrimination against women in the award of disability benefits. 

BIO:

Hamish Low is a senior economist and economic advisor in the microeconomics team in the research department at the Federal Reserve Bank of Chicago. Prior to joining the Fed, Hamish was the James Meade Professor of Economics at the University of Oxford and Nuffield College, and before that, professor of economics at the University of Cambridge and Trinity College, Cambridge. Hamish is also a research fellow at the Institute for Fiscal Studies and at the Centre for Economic Policy Research. Hamish Low holds a B.A. in philosophy, politics and economics from the University of Oxford, and a Ph.D. in economics from University College London. His research agenda sits within the broad fields of public economics and labor economics, and is focused on understanding the risks that individuals and families face and the choices that they make (or don't) to insure themselves. This includes risks that are primarily economic and those due to health shocks and changes in family structures. A key part of this agenda has been to understand the role of government in mitigating the uncertainty. He has published widely in the top outlets in economics including the American Economic Review, Econometrica, and the Journal of Political Economy.