Christiana E. Hilmer

On the Search for Talent in Academic Labor Markets: Easily Observable Later‐Graduate Study Outcomes as Predictors of Early‐Career Publishing, Placement, and Tenure

Three easily observed later‐graduate study outcomes, working with a prominent advisor, publishing while in graduate school, and coauthoring with the advisor are demonstrated to be significant determinants of early‐career publishing success for a sample of nearly 3,000 Ph.D. recipients from more than 100 domestic programs between 1990 and 1993. Out‐of‐sample predictions for 681 Ph.D. recipients from the class of 1994 indicate that these easily observed later‐graduate study signals dramatically improve our ability to forecast which students become prolific and which students fail to publish more than one article. Analyzing initial domestic tenure‐track economics placements indicates that hiring departments placed far more emphasis on Ph.D. program pedigree, leading to an initial tenure rate of less than 50% but eventually nearly three‐fourths currently hold tenured positions in the United States.

Dissertation advisors and initial job placements for economics PhD recipients

Using a unique data set that matches economics PhD recipients with their advisors, we find that after controlling for programme quality, the relative standing of a students dissertation advisor has a significant impact on his or her initial job placement.

Estimating Indirect Production Functions with a More General Specification: An Application of the Lewbel Model

Whereas consumer theory employs several different empirical specifications for estimating indirect utility functions, producer theory has relied on the Translog specification to estimate the indirect production function. In this paper, we apply Lewbel’s more general functional specification and investigate its implications for the estimation of indirect production functions in productivity analysis. An attractive feature of the Lewbel model is that it nests both the Translog and the almost ideal supply system, offering a method to assess the empirical validity of all three specifications. Aggregate U.S. production data are used to examine the performance of the three models in an empirical application.

How does a 48-hour test-taking window affect performance? Results from a testing centre

This note examines the effect that different self-selected test-taking strategies have on a students performance in an introductory economics course. Due to the policies of a centralized testing centre, students are allowed to choose when in a 48-hour period they take each exam as well as how long they work on the exam. The results suggest that the relationship between completion time and exam score follows an inverted U-shape while the relationship between relative order and exam score follows a U-shape. In other words, students who turn in their exams in a relatively short or a relatively long time tend to perform worse while students who choose to take their exams relatively early or relatively late tend to perform better.

So You Want to Hire a Future Star? Student-Advisor Matching and Early Career Productivity for Economics Ph.D.S

This paper examines a unique data set containing information on a Ph.D. recipients dissertation advisor, graduate program, and early career research productivity. We ask whether the match between an economics Ph.D. student and his or her dissertation advisor provides additional information as to the students likelihood for early career success beyond the program from which the student graduated. Comparing the distributions of research productivity across reputation ranks suggests that substantial overlap exists, with top students from lower ranked programs outperforming a significant percentage of students from top ranked programs. It is further observed that the high performing students from less respected programs tend to work with advisors ranked higher among Coupes (2003) top 1000 global economists than the less productive students from elite programs. Regressions that control for both program reputation and advisor rank confirm that, all else equal, students working with ranked advisors are significantly more likely to publish in their early careers, especially in top 5 and top 36 journals, than students working with unranked advisors. Moreover, adding controls for advisor rank diminishes the estimated impact of the students Ph.D. program. Together these facts suggest that the student-advisor match provides an important signal of the students likelihood for early career publishing success.