The Production of Published Research By U.S. Health Economists
Posted: 22 Jun 2007
Date Written: June 2007
Abstract
A 2005 survey reveals that U.S. health economists spend approximately 55 percent of their professional time on research. Assistant, associate and full professors, on average, reported publishing 8.8, 17.0 and 22.1 peer reviewed papers, respectively, over the past five years. Non-academic health economists, on average, published 8.2 papers in the same time period.
This paper examines the research productivity of U.S. health economists, measured as the number of peer-reviewed papers published in the last five years, both overall and in specific types of journals. We focus on productivity over the past five years rather than over the entire career in order to minimize the impact of changes over time in research environment.
We model the production of published research as a function of labor inputs, human capital, and location-specific capital. Labor inputs include the average number of hours per week spent on research and spent on teaching (which may have synergies with research). Human capital includes the field of one's doctorate, the rank of the department that granted the doctorate, and number of years of work experience since obtaining the doctorate. Location-specific capital includes the number of health economist colleagues and the type of school in which the faculty member is employed. An analogous model with appropriate modifications is estimated for non-academic health economists.
The data are drawn from the 2005 survey of health economists conducted under the auspices of the International Health Economics Association and AcademyHealth. The fact that these data were collected with the cooperation of AcademyHealth, and that health economists are the individuals studied, makes this paper appropriate for the Interest Group Business Meeting portion of the agenda. This online survey collected information on demographic and professional characteristics, publications, incomes and perceptions of professional and policy issues. (See Cawley and Morrisey (JHE 2007) for an analysis of the salary data.) Overall the survey had a response rate of 32 percent. For this analysis the survey yields 129 and 72 useable observations of doctorally trained academic and non-academic health economists, respectively.
The empirical models are translog production functions which allow for nonlinearity in the direct effects and interactions for each type of labor input, human capital, and location-specific capital. Seemingly unrelated translog regression models are estimated to explore the effects of the various inputs on publications in five types of journals: general-interest economics, health economics, health services research, health policy and medical journals.
Keywords: health economists, research, production function
JEL Classification: I1
Suggested Citation: Suggested Citation