CFAES Give Today

Department of Agricultural, Environmental, and Development Economics


Health information, Human Capital, And Risky Behavior: New (improved) estimates of how information affects behavior

Feb 17, 2017, 10:30am - 12:00pm

Dean Lillard, Associate Professor in the Department of Human Sciences at The Ohio State University, will present "Health information, Human Capital, And Risky Behavior: New (improved) estimates of how information affects behavior" on Friday, February 17 from 10:30am to 12:00pm in room 250A, Agricultural Administration.

In economic models of behavior, information plays an important role. Sometimes the role lurks in the background - when usually the model assumes that individuals have some level of knowledge (full or partial). But even when models directly incorporate variation in information as a factor that explains variation in behaviors, many studies specify relatively simple empirical models that only partly capture how responses to information vary with levels of human capital. In this paper I develop empirical measures of the stock of information people have about the risks of smoking. I use plausibly exogenous variation in exposure to that information to first explore whether and how the risky behavior changes when people are exposed to more information. I then explore how those responses vary across people with different levels of human capital. I use several sources of individual survey data. I use data that track smoking behavior of individuals over their whole life-course - from birth until the date of the interview. The data include individuals who started smoking as early as 1930 - well before scientific evidence had accumulated. I measure the information stock as the sum of potential exposure to magazine articles about smoking. (Potential) exposure varies across individuals, geographic space, and time. I use those data to explore whether and how individuals changed their behavior as they accumulated information and as new information was revealed and whether and how responses to information varies in ways one would predict across people who possess different levels of human capital.