Wilbert van der Klaauw

Identification and Estimation of Treatment Effects with a Regression-Discontinuity Design

Ž. THE REGRESSION DISCONTINUITY RD data design is a quasi-experimental design with the defining characteristic that the probability of receiving treatment changes discontinuously as a function of one or more underlying variables. This data design arises frequently in economic and other applications but is only infrequently exploited as a source of identifying information in evaluating effects of a treatment. In the first application and discussion of the RD method, Thistlethwaite and Campbell Ž. 1960 study the effect of student scholarships on career aspirations, using the fact that awards are only made if a test score exceeds a threshold. More recently, Van der Klaauw Ž. 1997 estimates the effect of financial aid offers on students’ decisions to attend a particular college, taking into account administrative rules that set the aid amount partly on the basis of a discontinuous function of the students’ grade point average and SAT Ž. score. Angrist and Lavy 1999 estimate the effect of class size on student test scores, taking advantage of a rule stipulating that another classroom be added when the average Ž. class size exceeds a threshold level. Finally, Black 1999 uses an RD approach to estimate parents’ willingness to pay for higher quality schools by comparing housing prices near geographic school attendance boundaries. Regression discontinuity methods have potentially broad applicability in economic research, because geographic boundaries or rules governing programs often create discontinuities in the treatment assignment mechanism that can be exploited under the method. Although there have been several discussions and applications of RD methods in the literature, important questions still remain concerning sources of identification and ways of estimating treatment effects under minimal parametric restrictions. Here, we show that identifying conditions invoked in previous applications of RD methods are often overly strong and that treatment effects can be nonparametrically identified under an RD design by a weak functional form restriction. The restriction is unusual in that it requires imposing continuity assumptions in order to take advantage of the known discontinuity in the treatment assignment mechanism. We also propose a way of nonparametrically estimating treatment effects and offer an interpretation of the Wald estimator as an RD estimator.

Leaving Teaching in the UK: A Duration Analysis

In this paper, the authors analyze the decision by teachers to leave the profession. Their results affirm the importance of relative earnings in the tenure and turnover decisions of teachers. The econometric modeling approach used yields important insights into the appropriateness of adopting a flexible, semiparametric specification of the duration dependence structure and of the unobserved heterogeneity distribution in duration models. Copyright 1995 by Royal Economic Society.

The Turnover of Teachers: A Competing Risks Explanation

In this paper, we analyze the decision by teachers to leave the profession in a dependent competing risks framework. The econometric model allows for a flexible, semiparametric specification of the duration-dependence structure and of the unobserved heterogeneity distribution in each exit-specific hazard function. Our results obtained for a large sample of UK teachers affirm the importance of teacher salaries and opportunity wages in the turnover decision of teachers and illustrate the insight gained from differentiating between multiple destinations or exit types.

Female Labour Supply and Marital Status Decisions: A Life-Cycle Model

This paper studies the interdependence between and the determinants of life-cycle marital status and labour force participation decisions of women. A dynamic utility maximization model is presented and estimated using longitudinal data on women from the Panel Study of Income Dynamics. The MLE method employed, involves solving a dynamic programming problem. Further, a minimum distance estimator is proposed which allows for the incorporation of wage data in a computationally simple way. The estimates are used to predict changes in the life-cycle patterns of employment, marriage and divorce due to differences in education, race, the females earnings and her (potential) husbands earnings. The estimation results indicate that the utility gains to marriage are decreasing in the females wage rate and increasing in her (potential) husbands earnings, while the opposite is found for gains to working. Ignoring the endogeneity of marital status decisions is shown to lead to an underestimation of own and husbands wage effects on female labour supply.

Regression-Discontinuity Analysis: A Survey of Recent Developments in Economics

This paper provides a discussion of recent developments related to the applicability of the regression discontinuity design in economics. It reviews econometric issues, such as identification and estimation methods, as well as a number of sensitivity and validity tests of importance in empirical application.

Evaluating the Cost of Conscription in The Netherlands

In this article we investigate the effect of military service in the Netherlands on future earnings. Estimating the cost or benefit of military service is complicated by the complex selection that determines who eventually serves in the military: On the one hand, potential conscripts have to pass medical and psychological examinations before entering the military, and on the other hand numerous (temporary) exemptions exist that can be manipulated by young men to avoid military service. We use substantial, policy-induced variation in aggregate military enrollment rates to deal with these selection issues. We find that approximately 10 years after serving in the military former conscripts have earnings that are on average 5% lower than the earnings of members of their birth cohort who did not serve in the military. These findings are shown to be robust against a variety of specifications.

An Introduction to the FRBNY Consumer Credit Panel

In this paper, we introduce the FRBNY Consumer Credit Panel, a new longitudinal database with detailed information on consumer debt and credit. The panel uses a unique sample design and information derived from consumer credit reports to track individuals’ and households’ access to and use of credit at a quarterly frequency. In any given quarter ranging from the first quarter of 1999 to the present, the panel can be used to compute nationally representative estimates of the levels and changes in various aspects of individual and household liabilities. In addition to describing the sample design, the use of sample weights, and the credit report information included in the database, we provide some comparisons of population statistics and consumer debt estimates derived from our panel with those based on data from the American Community Survey and the Flow of Funds Accounts of the United States.

Real Estate Investors, the Leverage Cycle, and the Housing Market Crisis

We explore a mostly undocumented but important dimension of the housing market crisis: the role played by real estate investors. Using unique credit-report data, we document large increases in the share of purchases, and subsequently delinquencies, by real estate investors. In states that experienced the largest housing booms and busts, at the peak of the market almost half of purchase mortgage originations were associated with investors. In part by apparently misreporting their intentions to occupy the property, investors took on more leverage, contributing to higher rates of default. Our findings have important implications for policies designed to address the consequences and recurrence of housing market bubbles.

Land Use Regulation and Welfare

We evaluate the effect of land use regulation on the value of land and on welfare. Our estimates are based on a decomposition of the effects of regulation into three components: an own‐lot effect, which reflects the cost of regulatory constraints to the owner of a parcel; an external effect, which reflects the value of regulatory constraints on ones neighbors; a supply effect, which reflects the effect of regulated scarcity of developable land. Using this decomposition, we arrive at a novel strategy for estimating a plausibly causal effect of land use regulation on land value and welfare. This strategy exploits cross‐border changes in development, prices, and regulation in regions near municipal borders. Our estimates suggest large negative effects of regulation on the value of land and welfare in these regions.

What Students Remember and Say about College Economics Years Later

In his presidential address to the American Economic Association, George Stigler (1963) offered the provocative hypothesis that students would retain very little knowledge from principles courses in economics five years or more after taking the courses. The few empirical studies that have been published on this topic generally found no or small lasting effects, at least for those who took fewer than four courses (see e.g., G. L. Bach and Phillip Saunders, 1965; Gerald J. Lynch, 1990). That raises even broader questions about the long-term effects of studying economics in college, in terms of individuals’ behavior as consumers, workers, and voters, which we are now beginning to investigate using both survey and transcript data. We have two major goals in this study. First, we want to learn how students perceive their classroom experience in economics courses years after leaving school, both in absolute terms and compared to other courses they took. We drew samples of economics, business, and other majors, who attended our four universities in 1976, 1986, and 1996, and asked which topics regularly covered in economics courses they now viewed as being most (and least) important. We asked whether they now viewed the economics courses they took as interesting, important, too difficult, or too abstract. We also compared their perceptions of teaching methods and grading rigor in economics courses to those developed in other courses. Our second major goal represents an empirical test of the common claim that economics is a unique “way of thinking.” If it is, we might reasonably expect people with more training in economics to have different views on policy issues, and to make different decisions as consumers, workers, savers, investors, and voters. We collected survey data on many of these choices, and by matching those responses with transcript data, we plan to investigate whether there are observable behavioral responses associated with being an economics major, or simply taking some minimum number of economics courses, compared to students who took fewer courses or none at all. There are other uses for these data. They can likely be used to study choice of major and course-taking behavior as well as common laboreconomics issues such as human capital versus screening in the labor market. Certainly the data can provide insight into curriculum development for economics departments, and business schools as well. There is, obviously, the potential for response biases in the analysis of these data, but we expect that having transcript data for respondents and nonrespondents will allow us to deal with these issues econometrically. This first, brief report from this ongoing project addresses only the first goal of the project, specifically, how our former students evaluate their experience with economics courses and instructors. What economic courses and concepts do they now believe are most important? Do they wish they had taken more, or less, coursework in economics? What do they think now about various economic issues? And * Allgood: Department of Economics, University of Nebraska, Lincoln, NE 68588 (e-mail: sallgood@unlnotes.unl. edu); Bosshardt: Department of Economics, Florida Atlantic University, Boca Raton, FL, 33431 (e-mail: wbosshar@fau.edu); van der Klaauw: Department of Economics, University of North Carolina, Chapel Hill, NC, 27599 (e-mail: vanderkl@email.unc.edu); Watts: Department of Economics, Purdue University, West Lafayette, IN 47907, and while this paper was written, the German International School of Management and Administration, Hanover, Germany (e-mail: mwatts@mgmt.purdue.edu). We thank the Board of the Calvin K. Kazanjian Economics Foundation for the grant that made this work possible, and the AEA Committee for Economic Education for bringing us together to write the proposal, as described in Michael Salemi et al. (2002). Mike Salemi provided helpful comments on an earlier draft of this paper. April Fidler provided major assistance in project coordination, administration, and data entry. Georg Schaur worked extensively with data organization and preliminary tabulations.