TY - JOUR AU - Barsky,Robert B. AU - Kimball,Miles S. AU - Juster,F. Thomas AU - Shapiro,Matthew D. TI - Preference Parameters and Behavioral Heterogeneity: An Experimental Approach in the Health and Retirement Survey JF - National Bureau of Economic Research Working Paper Series VL - No. 5213 PY - 1997 Y2 - October 1997 UR - http://www.nber.org/papers/w5213 L1 - http://www.nber.org/papers/w5213.pdf N1 - Author contact info: Robert B. Barsky Department of Economics University of Michigan Ann Arbor, MI 48109-1220 Tel: 734/764-9476 Fax: 734/764-2769 E-Mail: barsky@umich.edu Miles S. Kimball Department of Economics University of Michigan Ann Arbor, MI 48109-1220 Tel: 734/764-2375 Fax: 734/764-2769 E-Mail: mkimball@umich.edu Thomas Juster Program Dir, Survey Research Ctr Institute for Social Research University of Michigan Ann Arbor, MI 48104 E-Mail: N/A user is deceased Matthew D. Shapiro Department of Economics University of Michigan 611 Tappan St Ann Arbor, MI 48109-1220 Tel: 734/764-5419 Fax: 734 764-2769 E-Mail: shapiro@umich.edu AB - Individuals' preferences underlying most economic behavior are likely to display substantial heterogeneity. This paper reports on direct measures of preference parameters relating to risk tolerance, time preference, and intertemporal substitution. These experimental measures are based on survey respondents' choices in hypothetical situations. The questions are constructed with as little departure from the theorist's concept of the underlying parameter as possible. The individual measures of preference parameters display substantial heterogeneity. The majority of respondents fall into the least risk-tolerant group, but a substantial minority display higher risk tolerance. The individual measures of intertemporal substitution and time preference also display substantial heterogeneity. The mean risk tolerance is 0.25; the mean elasticity of intertemporal substitution is 0.2. Estimated risk tolerance and the elasticity of intertemporal substitution are essentially uncorrelated across individuals. Because the risk tolerance measure is obtained as part of the main questionnaire of a large survey, it can be related to a number of economic behaviors. Measured risk tolerance is positively related to a number of risky behaviors, including smoking, drinking, failing to have insurance, and holding stocks rather than Treasury bills. Although measured risk tolerance explains only a small fraction of the variation of the studied behaviors, these estimates provide evidence about the validity and usefulness of the measures of preference parameters. ER -