Research
Published Papers
"Are Students Time-Constrained? Course Load, GPA, and Failure," with Alexander Amaya (JPubE) -- Given the simultaneous rise in time-to-graduation and college GPA, it may be that students reduce their course load to improve their performance. Yet, evidence to date only shows increased course loads increase GPA. We provide a mathematical model showing many unobservable factors -- beyond student ability -- can generate a positive relationship between course load and GPA unless researchers control student schedules. West Point regularly implements the ideal experiment by randomly modifying student schedules with additional training courses. Using 19 years of administrative data, we provide the first causal evidence that taking more courses reduces GPA and increases course failure rates, sometimes substantially.
“Enacting the Rubric: Teacher Improvements in Windows of High-stakes Observation,” with Emily Wiseman (2021, Education Finance and Policy) – A key purpose of high-stakes in-class observations is to improve teacher practice, but little evidence exists on whether teachers respond as expected. We estimate how teachers adjust their practice when anticipating an unannounced, high-stakes in-class observation in Washington D.C. We use the probability of receiving an evaluation as a treatment variable to identify how teachers make preparations for an impending evaluation, exploiting the random timing of evaluations from both principals and district employees. We find that as the probability of an evaluation increases 10 percentage points, teachers improve their evaluation score by a statistically significant 0.06 standard deviations. We also find that teachers that are not facing dismissal threat do not make preparations for their fifth and final evaluation of the year. This suggests two takeaways: (1) teachers are aware of how to improve their practice for evaluations and (2) they respond to high-stakes in-class observations. In conjunction with other evidence that these preparations improve student outcomes, we conclude that high-stakes evaluations are an effective tool for improving teacher practice.
“The Changing Occupational Distribution by College Major,” with Michael Ransom. 2017. Research in Labor Economics, Vol. 45 - In this paper, we examine the occupational distribution of individuals who hold bachelor degrees in particular fields in the United States using data from the various waves of the National Survey of College Graduates. We propose and calculate indices that describe two related aspects of the occupational distribution by major field of study: distinctiveness (how dissimilar are the occupations of a particular major when compared with all other majors) and variety (how varied are the occupations among those who hold that particular major). We discuss theoretical properties of these indices and statistical properties of their estimates. We show that the occupational variety has increased since 1993 for most major fields of study, particularly between the 1993 and 2003 waves of the survey. We explore reasons for this broadening of the occupation distribution. We find that this has not led to an increase in reported mismatch between degree and occupation.
Working Papers (copy available upon request)
“Multi-tasking with Production Uncertainty: A Real-Effort Laboratory Experiment,” with Michael Kofoed (R&R Economic Inquiry) – Policy advocates often push for market-like performance incentives for publicly provided services such as in education or healthcare (as with Medicare in the U.S.). The evidence supporting output-based, performance incentives for such jobs is mixed at best. One possible explanation is that the production process for these jobs is characterized by production uncertainty, mathematically represented as uncertainty in the marginal product of inputs. In a multi-tasking principal-agent model, I show how production uncertainty can make output-based incentives induce inefficient behavior. The model shows employees will inefficiently favor inputs with lower production uncertainty and reduce overall effort. In such cases, input-based incentives may prove more effective. I conduct a laboratory experiment to test these predictions. The results confirm the two predictions of the model: participants shift away from their efficient input type as production uncertainty grows for that type, and production uncertainty induces an overall reduction in effort.
“Does Monitoring Change Teacher Pedagogy and Student Outcomes?” (R&R JOLE) –Despite limited real-world evidence on how monitoring affects employee time allocation, teacher performance pay programs often use monitoring in the form of unannounced in-class observations. Such programs have had mixed results, and it remains unknown which aspects make them effective. To address this gap, I use exogenous variation in the presence and intensity of monitoring in D.C. Public schools to show it significantly improves student test scores and reduces suspensions. As monitoring intensifies, teachers use more individualized teaching and emphasize higher-level learning. This is novel evidence validates monitoring as a potential tool for enhancing teacher pedagogy and employee performance broadly.
"How do Veterans Fare as Entrepreneurs? A Descriptive Analysis of Vetrepreneurs," with William Skimmyhorn (R&R International Small Business Journal) - The large literature on the correlates of entrepreneurship pays relatively little attention to America’s military Veterans. We leverage two national surveys with rich demographic and behavioral data to estimate how military experience predicts entrepreneurship and financial success. After accounting for demographic differences, Veterans are less likely to become entrepreneurs than their peers. Veteran entrepreneurs also earn less and have lower financial satisfaction. These results run contrary to the common narrative that the average military experience imparts human capital that prepares Veterans for entrepreneurship. We also find that female, Black and higher educated Veterans are relatively more likely to pursue entrepreneurship.
"More than Sheepskin: A Natural Experiment on College and Earnings," with Tim Simmons-Justicz and Joseph Price (Under Review) - While the causal effect of college on earnings is well established, there remains little causal evidence distinguishing human capital gains from the signal a college degree sends. We use an unexpected and sudden reduction in the length of time Army officers remained in school to identify the returns of an additional year of education. These early graduates were awarded a Bachelor of Science – which is typically a four year degree – after completing only two or three years of school. We find an additional year of college leads to a 10% increase in earnings 20 years post-graduation.
"Army Enlistment and Free Community College," with Michael Brown, Celeste Carruthers, Michael Kofoed, and Jenna Kramer - We examine how "free community college" programs affect U.S. Army enlistment, taking advantage of the roll-out of the Tennessee Promise model from one county to the rest of the state. We find that availability of this particular form of tuition guarantee reduced enlistments by four recruits in average county per year or about a 25 percent reduction. We find that this reduction is driven by poorer, white students.
"Jack of All Trades or Master of One: How General and Specific Human Capital Endure the Recession" - The workforce in the U.S. is more college educated than ever before. College graduates replaced the majority of jobs lost during the Great Recession, yet the college wage premium remains. I find that employers respond to the increased supply of college graduates by changing the skills required of employees. In a task-based model, I demonstrate how this skillset shift favors general human capital over specific. The theoretical innovation is to classify general human capital as a risk-reducing option for employers. The empirical innovation is to measure general and specific human capital within a major using a major's information content about occupational choice and skills. This measure has a direct interpretation in the theoretical model. Using machine learning, I present evidence that highly content-specific college majors experienced greater wage losses and have been slower to recover after the Great Recession.
"The Cost of Bad Timing: The Effect of Military Exit Timing on Veteran Educational Attainment," with Carl Wojtaszek and Michael Kofoed - Successful job transitions often hinge on a worker’s access to education and re-training. Yet factors that impact this important population’s education investment decision have been largely unexplored. We use a two-stage least-squares approach to investigate the effect of military exit timing on educational attainment during the 1990s and then again in the 2010s after passage of the Post 9-11 GI Bill. Using military records, we find that longer wait times between a soldier’s exit and a semester start date reduces the likelihood of enrollment in the 1990s, but the reverse effect in the 2010s. We explore the differences between time periods by looking at the increased use of online education and for-profit schools.
Work in Progress
"On-the-job Training and Educational Attainment," with Jesse Bruhn, Jake Fabian, Matthew Gudgeon, Luke Gallagher, and Adam Isen - The specific job options of a newly enlisted soldier depend on passing strict cutoff scores on multiple dimensions of the Armed Forces Qualification Test (AFQT). With a novel multiple regression discontinuity approach, we use these many cutoffs to measure the causal effect of job experience on later life outcomes, such as educational attainment and health.
"Can Diversity Improve Equity? Achieving Racial Parity in Leader Assessment of Minority Team Members," with Romaine Campbell and supported by the Upjohn Institute Early Career Research Grant - Cadets at West Point are randomized into their first platoons as part of their initial training. This randomization leads to differences in the racial composition of a Cadet's platoon. We track Cadets throughout their time at West Point and five years into their career as officers to assess whether early diversity in their training platoon leads to longer-term changes in their team performance and the ratings of their team mates from under-represented minorities.
"Performance Pay for Army Recruiters: A National RCT" - Using the results of a national randomized controlled trial, I examine the effects of performance pay for Army recruiters. The RCT provides two treatment branches to assess whether incentives should reward recruiters for simply the number of new recruits or for the quality of those recruits as measure by their success and persistence within the Army.
Leadership Economics - an undergraduate textbook on public choice, information economics, personnel economics, and a microeconomic model of soft human capital. 200+ pages completed; 11 chapters complete with problem sets and case studies.