Hidehiko Ichimura

Interests

No activities entered.

Courses

Scholarly Contributions

Chapters

• Ichimura, H., & Linton, O. (2005). Asymptotic expansions for some semiparametric program evaluation estimators. In Identification and inference in econometric models: essays in honor of Thomas J. Rothenberg(pp 149-170). Cambridge University Press. doi:10.1017/cbo9780511614491.009
  More info

  We investigate the performance of a class of semiparametric estimators of the treatment effect via asymptotic expansions. We derive approximations to the first two moments of the estimator that are valid to “second order.” We use these approximations to define a method of bandwidth selection. We also propose a degrees of freedom–like bias correction that improves the secondorder properties of the estimator but without requiring estimation of higher-order derivatives of the unknown propensity score. We provide some numerical calibrations of the results.

Journals/Publications

• Fukai, T., Ichimura, H., Kitao, S., & Mikoshiba, M. (2025). Medical expenditures over the life-cycle: persistent risks and insurance. Japanese Economic Review, 76(Issue 2). doi:10.1007/s42973-025-00190-z
  More info

  This paper builds a life-cycle model of single and married households and evaluates the roles of the national health insurance system. We use the administrative data on nationwide health insurance claims in Japan to analyze medical expenditure risks and calibrate the model with the stochastic process that varies by age and gender. Economic and welfare effects of health insurance reform depend on household income levels and generosity of the welfare program. Without health insurance, high-income households turn to self-insurance, significantly increasing aggregate savings. Low-income households, especially low-skilled single men and women, reduce savings and many of them become welfare recipients. Raising copayment rates for the elderly increases household savings, but depletes wealth of low-income households and leads to a rise in the number of welfare recipients.
• Graham, B., Ichimura, H., Jansson, M., & Khan, S. (2025). Introduction to the Annals Issue in Honor of James Powell. Journal of Econometrics. doi:10.1016/j.jeconom.2025.106051
• Fukai, T., Fukuda, S., Ichimura, H., Nakata, D., Sato, I., & Terada, K. (2024). National Transfer Accounts (NTA) in Japan: 1984−2014. Japanese Economic Review, 75(Issue 4). doi:10.1007/s42973-024-00175-4
  More info

  We developed standardized methods that allow for computing National Transfer Accounts (NTA) statistics consistently from 1984 to 2014 and applied this uniform approach to analyze the evolution of three NTA accounts. By detailing our computation methods, we ensure that the results can be replicated, providing a consistent basis for discussion using NTA. Our analysis revealed that the 23–39 age group experienced the slowest consumption growth since 2004, despite having higher labor income growth compared to other age groups. This discrepancy is attributed to a greater public transfer burden on this age group. This finding offers a new perspective on the potential causes of declining marriage and fertility rates in Japan. Additionally, we discuss potential improvements to the NTA estimation methods and framework.
• Ichimura, H., Lei, X., Lee, C., Lee, J., Park, A., & Sawada, Y. (2024). Wellbeing of the older individuals in East Asia. Japanese Economic Review, 75(Issue 4). doi:10.1007/s42973-024-00168-3
  More info

  Rapid demographic transition in East Asia has resulted in “super” aging. Because of steadily decreasing fertility and increasing life expectancy, the proportion of older individuals in the population and the old-age dependency ratio are rising across all East Asian countries, particularly China, the Republic of Korea, and Japan. This study empirically investigated the well-being of older individuals in these three countries using comparable micro-level data from the China Health and Retirement Longitudinal Study, Korean Longitudinal Study on Aging, and Japanese Study of Aging and Retirement. Specifically, we examined the depressive symptoms scale as a measure of well-being and estimated the impact of four broad categories: demographic; economic; family-social; and health. The decomposition-and-simulation analyses reveal that although differences in the characteristics of older individuals in the three countries among countries explain many differences in mean depression rates, there remain significant differences across countries, which cannot be explained. Even after considering multiple factors, the study found that older individuals in Korea were more likely to be depressed than those in China or Japan.
• Ichimura, H., Newey, W. K., Chernozhukov, V., Escanciano, J. C., & Robins, J. M. (2022). Locally Robust Semiparametric Estimation. Econometrica.
• Ichimura, H., & Newey, W. (2022). The influence function of semiparametric estimators. Quantitative Economics, 13(1), 29–61. doi:https://doi.org/10.3982/QE826
• Ichimura, H., Fukai, T., & Kawata, K. (2021). Describing the impacts of COVID-19 on the labor market in Japan until June 2020. The Japanese Economic Review, 72(3), 439-482. doi:10.1007/s42973-021-00081-z
• , H. I., & , W. K. (2018). The Influence Function of Semiparametric Estimators.
  More info

  Often semiparametric estimators are asymptotically equivalent to a sampleaverage. The object being averaged is referred to as the influence function.The influence function is useful in formulating primitive regularity conditionsfor asymptotic normality, in efficiency comparions, for bias reduction, and foranalyzing robustness. We show that the influence function of a semiparametricestimator can be calculated as the limit of the Gateaux derivative of aparameter with respect to a smooth deviation as the deviation approaches apoint mass. We also consider high level and primitive regularity conditions forvalidity of the influence function calculation. The conditions involve Frechetdifferentiability, nonparametric convergence rates, stochastic equicontinuity,and small bias conditions. We apply these results to examples.[Journal_ref: ]
• , T. F., , H. I., & , K. K. (2018). Quantifying Health Shocks Over the Life Cycle.
  More info

  We first show (1) the importance of investigating health expenditure processusing the order two Markov chain model, rather than the standard order onemodel, which is widely used in the literature. Markov chain of order two is theminimal framework that is capable of distinguishing those who experience acertain health expenditure level for the first time from those who have beenexperiencing that or other levels for some time. In addition, using the modelwe show (2) that the probability of encountering a health shock first de-creases until around age 10, and then increases with age, particularly, afterage 40, (3) that health shock distributions among different age groups do notdiffer until their percentiles reach the median range, but that above themedian the health shock distributions of older age groups gradually start tofirst-order dominate those of younger groups, and (4) that the persistency ofhealth shocks also shows a U-shape in relation to age.[Journal_ref: ]
• , V. C., , J. C., , H. I., , W. K., & , J. M. (2018). Locally Robust Semiparametric Estimation.
  More info

  We give a general construction of debiased/locally robust/orthogonal (LR)moment functions for GMM, where the derivative with respect to first stepnonparametric estimation is zero and equivalently first step estimation has noeffect on the influence function. This construction consists of adding anestimator of the influence function adjustment term for first stepnonparametric estimation to identifying or original moment conditions. We alsogive numerical methods for estimating LR moment functions that do not requirean explicit formula for the adjustment term. LR moment conditions have reduced bias and so are important when the firststep is machine learning. We derive LR moment conditions for dynamic discretechoice based on first step machine learning estimators of conditional choiceprobabilities. We provide simple and general asymptotic theory for LRestimators based on sample splitting. This theory uses the additivedecomposition of LR moment conditions into an identifying condition and a firststep influence adjustment. Our conditions require only mean square consistencyand a few (generally either one or two) readily interpretable rate conditions. LR moment functions have the advantage of being less sensitive to first stepestimation. Some LR moment functions are also doubly robust meaning they holdif one first step is incorrect. We give novel classes of doubly robust momentfunctions and characterize double robustness. For doubly robust estimators ourasymptotic theory only requires one rate condition.[Journal_ref: ]
• Ichimura, H., & Arai, Y. (2018). Simultaneous Selection of Optimal Bandwidths for the Sharp Regression Discontinuity Estimator. Quatitative Economics, 9(1), 441-482. doi:https://doi.org/10.3982/QE590
• Ichimura, H., & Lee, S. (2018). Corrigendum to “Characterization of the asymptotic distribution of semiparametric M-estimators” [J. Econometrics 159 (2) (2010) 252–266](S0304407610001302)(10.1016/j.jeconom.2010.05.005). Journal of Econometrics, 202(Issue 2). doi:10.1016/j.jeconom.2017.07.003
  More info

  This note provides correction to Ichimura and Lee (2010).
• Ichimura, H., Ahn, H., Powell, J. L., & Ruud, P. A. (2018). Simple Estimators for Invertible Index Models. Journal of Business and Economic Statistics, 36(1), 1-10. doi:https://doi.org/10.1080/07350015.2017.1379405
• Arai, Y., & Ichimura, H. (2016). Optimal bandwidth selection for the fuzzy regression discontinuity estimator. Economics Letters, 141(Issue). doi:10.1016/j.econlet.2016.01.024
  More info

  A new bandwidth selection method for the fuzzy regression discontinuity estimator is proposed. The method chooses two bandwidths simultaneously, one for each side of the cut-off point by using a criterion based on the estimated asymptotic mean square error taking into account a second-order bias term. A simulation study demonstrates the usefulness of the proposed method.
• Ichimura, H., Sawada, Y., & Shimizutani, S. (2016). Conference on Economics of Ageing in Japan and other Societies: Introduction. Japanese Economic Review, 67(Issue 2). doi:10.1111/jere.12105
• Arai, Y., Ichimura, H., & Kawaguchi, D. (2015). The educational upgrading of Japanese youth, 1982-2007: Are all Japanese youth ready for structural reforms?. Journal of the Japanese and International Economies, 37(Issue). doi:10.1016/j.jjie.2015.04.002
  More info

  Are all Japanese youth ready for the structural reforms proposed as a supply-side policy of Abenomics? To answer this question, we assess how well Japanese youth have coped with the labor market's long-term structural changes, induced primarily by deepening interdependence with emerging economies and rapid technological progress over the last three decades. We examine the role of educational upgrading on the labor-market outcomes of youth between the ages of 25 and 29, using six waves of micro data from the Employment Status Survey spanning from 1982 to 2007. The analysis demonstrates that the demand growth for skilled labor relative to unskilled labor has been met by the educational upgrading of youth through the expansion of tertiary education, including education in vocational schools. Youth left behind the trend of educational upgrading, however, have suffered significantly from decreasing employment opportunities and deteriorated working conditions.
• Costa Dias, M., Ichimura, H., & van den Berg, G. J. (2013). Treatment evaluation with selective participation and ineligibles. Journal of the American Statistical Association, 108(Issue 502). doi:10.1080/01621459.2013.795447
  More info

  Matching methods for treatment evaluation based on a conditional independence assumption do not balance selective unobserved differences between treated and nontreated. We derive a simple correction term if there is an instrument that shifts the treatment probability to zero in specific cases. Policies with eligibility restrictions, where treatment is impossible if some variable exceeds a certain value, provide a natural application. In an empirical analysis, we exploit the age eligibility restriction in the Swedish Youth Practice subsidized work program for young unemployed, where compliance is imperfect among the young. Adjusting the matching estimator for selectivity changes the results toward making subsidized work detrimental in moving individuals into employment. © 2013 American Statistical Association.
• Altonji, J. G., Ichimura, H., & Otsu, T. (2012). Estimating Derivatives in Nonseparable Models With Limited Dependent Variables. Econometrica, 80(Issue 4). doi:10.3982/ecta8004
  More info

  We present a simple way to estimate the effects of changes in a vector of observable variables X on a limited dependent variable Y when Y is a general nonseparable function of X and unobservables, and X is independent of the unobservables. We treat models in which Y is censored from above, below, or both. The basic idea is to first estimate the derivative of the conditional mean of Y given X at x with respect to x on the uncensored sample without correcting for the effect of x on the censored population. We then correct the derivative for the effects of the selection bias. We discuss nonparametric and semiparametric estimators for the derivative. We also discuss the cases of discrete regressors and of endogenous regressors in both cross section and panel data contexts. © 2012 The Econometric Society.
• Ichimura, H. (2012). The 2011 Japanese Economic Association-Nakahara Prize. Japanese Economic Review, 63(Issue 3). doi:10.1111/j.1468-5876.2012.00583.x
• Ichimura, H., & Lee, S. (2010). Characterization of the asymptotic distribution of semiparametric M-estimators. Journal of Econometrics, 159(Issue 2). doi:10.1016/j.jeconom.2010.05.005
  More info

  This paper develops a concrete formula for the asymptotic distribution of two-step, possibly non-smooth semiparametric M-estimators under general misspecification. Our regularity conditions are relatively straightforward to verify and also weaker than those available in the literature. The first-stage nonparametric estimation may depend on finite dimensional parameters. We characterize: (1) conditions under which the first-stage estimation of nonparametric components do not affect the asymptotic distribution, (2) conditions under which the asymptotic distribution is affected by the derivatives of the first-stage nonparametric estimator with respect to the finite-dimensional parameters, and (3) conditions under which one can allow non-smooth objective functions. Our framework is illustrated by applying it to three examples: (1) profiled estimation of a single index quantile regression model, (2) semiparametric least squares estimation under model misspecification, and (3) a smoothed matching estimator. © 2010 Elsevier B.V. All rights reserved.
• Blundell, R., Gosling, A., Ichimura, H., & Meghir, C. (2007). Changes in the distribution of male and female wages accounting for employment composition using bounds. Econometrica, 75(Issue 2). doi:10.1111/j.1468-0262.2006.00750.x
  More info

  This paper examines changes in the distribution of wages using bounds to allow for the impact of nonrandom selection into work. We show that worst case bounds can be informative. However, because employment rates in the United Kingdom are often low, they are not informative about changes in educational or gender wage differentials. Thus we explore ways to tighten these bounds using restrictions motivated from economic theory. With these assumptions, we find convincing evidence of an increase in inequality within education groups, changes in educational differentials, and increases in the relative wages of women.
• Ichimura, H., & Taber, C. (2002). Semiparametric Reduced-form Estimation of Tuition Subsidies. American Economic Review, 92(Issue 2). doi:10.1257/000282802320189410
• Ichimura, H., & Taber, C. (2001). Propensity-score matching with instrumental variables. American Economic Review, 91(Issue 2). doi:10.1257/aer.91.2.119
• Heckman, J. J., Ichimura, H., & Todd, P. (1998). Matching As An Econometric Evaluation Estimator. Review of Economic Studies, 65(Issue 2). doi:10.1111/1467-937x.00044
  More info

  This paper develops the method of matching as an econometric evaluation estimator. A rigorous distribution theory for kernel-based matching is presented. The method of matching is extended to more general conditions than the ones assumed in the statistical literature on the topic. We focus on the method of propensity score matching and show that it is not necessarily better, in the sense of reducing the variance of the resulting estimator, to use the propensity score method even if propensity score is known. We extend the statistical literature on the propensity score by considering the case when it is estimated both parametrically and nonparametrically. We examine the benefits of separability and exclusion restrictions in improving the efficiency of the estimator. Our methods also apply to the econometric selection bias estimator.
• Heckman, J., Ichimura, H., Smith, J., & Todd, P. (1998). Characterizing selection bias using experimental data. Econometrica, 66(Issue 5). doi:10.2307/2999630
  More info

  Semiparametric methods are developed to estimate the bias that arises from using nonexperimental comparison groups to evaluate social programs and to test the identifying assumptions that justify matching, selection models, and the method of difference-in-differences. Using data from an experiment on a prototypical social program and data from nonexperimental comparison groups, we reject the assumptions justifying matching and our extensions of it. The evidence supports the selection bias model and the assumptions that justify a semiparametric version of the method of difference-in-differences. We extend our analysis to consider applications of the methods to ordinary observational data.
• Ichimura, H., & Thompson, T. S. (1998). Maximum likelihood estimation of a binary choice model with random coefficients of unknown distribution. Journal of Econometrics, 86(Issue 2). doi:10.1016/s0304-4076(97)00117-6
  More info

  We consider a binary response model yi = 1{x′ißi + εi ≥ 0} with xi independent of the unobservables (ßi, εi). No finite-dimensional parametric restrictions are imposed on F0, the joint distribution of (ßi, εi). A nonparametric maximum likelihood estimator for F0 is shown to be consistent. We analyze some conditions under which F0 is or is not identified. In particular, we show that if the support of F0 is a subset of any half of the unit hypersphere, then F0 is identified relative to all distributions on the unit hypersphere. We also provide some Monte Carlo evidence on the small sample performance of our estimator. © 1998 Elsevier Science S.A. All rights reserved.
• Kamiya, K., & Ichimura, H. (1998). A revealed preference theory for non-expected utility on "certain × uncertain" consumption pairs. Japanese Economic Review, 49(Issue 1). doi:10.1111/1468-5876.00071
  More info

  In this paper we consider a two-period decision problem, where the feasible set is the set of "certain × uncertain" consumption pairs. That is, the decision-maker chooses (x, m) in a feasible set, where x is a certain first-period consumption and m is a random second-period consumption, a Borel probability measure on the set of real numbers. The purpose of this paper is to present revealed preference theory for non-expected utility on "certain × uncertain" consumption pairs. We present necessary and sufficient conditions for the data to be consistent with some non-expected utility functions.
• Heckman, J. J., Ichimura, H., & Todd, P. E. (1997). Matching As An Econometric Evaluation Estimator: Evidence from Evaluating a Job Training Programme. Review of Economic Studies, 64(Issue 4). doi:10.2307/2971733
  More info

  This paper considers whether it is possible to devise a nonexperimental procedure for evaluating a prototypical job training programme. Using rich nonexperimental data, we examine the performance of a two-stage evaluation methodology that (a) estimates the probability that a person participates in a programme and (b) uses the estimated probability in extensions of the classical method of matching. We decompose the conventional measure of programme evaluation bias into several components and find that bias due to selection on unobservables, commonly called selection bias in econometrics, is empirically less important than other components, although it is still a sizeable fraction of the estimated programme impact. Matching methods applied to comparison groups located in the same labour markets as participants and administered the same questionnaire eliminate much of the bias as conventionally measured, but the remaining bias is a considerable fraction of experimentally-determined programme impact estimates. We test and reject the identifying assumptions that justify the classical method of matching. We present a nonparametric conditional difference-in-differences extension of the method of matching that is consistent with the classical index-sufficient sample selection model and is not rejected by our tests of identifying assumptions. This estimator is effective in eliminating bias, especially when it is due to temporally-invariant omitted variables.
• Heckman, J. J., Ichimura, H., Smith, J., & Todd, P. (1996). Sources of selection bias in evaluating social programs: An interpretation of conventional measures and evidence on the effectiveness of matching as a program evaluation method. Proceedings of the National Academy of Sciences of the United States of America, 93(Issue 23). doi:10.1073/pnas.93.23.13416
  More info

  This paper decomposes the conventional measure of selection bias in observational studies into three components. The first two components are due to differences in the distributions of characteristics between participant and nonparticipant (comparison) group members: the first arises from differences in the supports, and the second from differences in densities over the region of common support. The third component arises from selection bias precisely defined. Using data from a recent social experiment, we find that the component due to selection bias, precisely defined, is smaller than the first two components. However, selection bias still represents a substantial fraction of the experimental impact estimate. The empirical performance of matching methods of program evaluation is also examined. We find that matching based on the propensity score eliminates some but not all of the measured selection bias, with the remaining bias still a substantial fraction of the estimated impact. We find that the support of the distribution of propensity scores for the comparison group is typically only a small portion of the support for the participant group. For values outside the common support, it is impossible to reliably estimate the effect of program participation using matching methods. If the impact of participation depends on the propensity score, as we find in our data, the failure of the common support condition severely limits matching compared with random assignment as an evaluation estimator.
• Ichimura, H. (1993). Semiparametric least squares (SLS) and weighted SLS estimation of single-index models. Journal of Econometrics, 58(Issue 1-2). doi:10.1016/0304-4076(93)90114-k
  More info

  For the class of single-index models, I construct a semiparametric estimator of coefficients up to a multiplicative constant that exhibits 1 √n-consistency and asymptotic normality. This class of models includes censored and truncated Tobit models, binary choice models, and duration models with unobserved individual heterogeneity and random censoring. I also investigate a weighting scheme that achieves the semiparametric efficiency bound. © 1993.
• Hausman, J. A., Newey, W. K., Ichimura, H., & Powell, J. L. (1991). Identification and estimation of polynomial errors-in-variables models. Journal of Econometrics, 50(Issue 3). doi:10.1016/0304-4076(91)90022-6
  More info

  Methods of estimation of regression coefficients are proposed when the regression function includes a polynomial in a 'true' regressor which is measured with error. Two sources of additional information concerning the unobservable regressor are considered: either an additional indicator of the regressor (itself measured with error) or instrumental variables which characterize the systematic variation in the true regressor. In both cases, estimators are constructed by relating moments involving the unobserved variables to moments of observables; these relations lead to recursion formulae for computation of the regression coefficients and nuisance parameters (e.g., moments of the measurement error). Consistency and asymptotic normality of the estimated coefficients is demonstrated, and consistent estimators of the asymptotic covariant matrices are provided. © 1991.

Proceedings Publications

• Ichimura, H., Konishi, Y., & Nishiyama, Y. (2007). Measuring of firm specific productivities: Evidence from Japanese plant level panel data. In International Congress on Modelling and Simulation - Land, Water and Environmental Management: Integrated Systems for Sustainability, MODSIM07.

Presentations

• Ichimura, H. (2023). Identification of LATE Revisited. Southern Methodist University Econometrics SeminarDepartment of Economics, Southern Methodist University.
• Ichimura, H. (2023). LATE with Covariates. UC San Diego, Econometrics SeminarDepartment of Economics, UC San Diego.
• Ichimura, H. (2021, January). Locally Robust Semiparametric Estimation. Econometrics Seminar at UC San Diego. Department of Economics, UC San Diego: Department of Economics, UC San Diego.
• Ichimura, H. (2020, February). Locally Robust Semiparametric Estimation. Econometrics Seminar at Emory University. Emory University, Atlanta, GA: Department of Economics, Emory University.
• Ichimura, H. (2020, September). Locally Robust Semiparametric Estimation. Econometrics Seminar at Rutgers University. Department of Economics, Rutgers University: Department of Economics, Rutgers University.
• Ichimura, H. (2019, April). General Discussion. Robustness in Economics and Econometrics. University of Chicago: Department of Economics, University of Chicago.
• Ichimura, H. (2019, April). Locally Robust Semiparametric Estimation. Econometrics Seminar at Johns Hopkins University. Baltimore, Maryland: Department of Economics, Johns Hopkins University.
• Ichimura, H. (2019, August). General Discussion. From Theory to Statistics to Empirics: An Econometric Conference in honor of James Heckman. University of Chicago: Department of Economics, University of Chicago.
• Ichimura, H. (2019, February). Locally Robust Semiparametric Estimation. Econometric seminat at Pennsylvania State University. University Park, PA: Department of Economics, Pennsylvania State University.
• Ichimura, H. (2019, January). Locally Robust Semiparametric Estimation. Econometric seminat at University of Texas, Austin. Department of Economics, University of Texas, Austin: Department of Economics, University of Texas, Austin.
• Ichimura, H. (2019, May). General Discussion. Whitney Newey's 65th Birthday Conference. MIT, Cambridge, MA: Department of Economics, MIT.
• Ichimura, H. (2019, October). Presidential Address: What Do We Learn from Panel Data Created UsingJapanese Census?. Japanese Economic Association, Fall meeting. Kobe University, Kobe, Hyogo, Japan: Kobe University.
• Ichimura, H. (2018, December). Quantifying Health Shocks Over the Life Cycle. 9th Annual APRU Research Conference on Population Aging. The Hong Kong University of Science and Technology: The Hong Kong University of Science and Technology.
• Ichimura, H. (2018, November). General discussion. Incomplete Models 2018. Department of Economics, Northwestern University: Northwestern University and Cemmap.
• Ichimura, H. (2018, November). Locally Robust Semiparametric Estimation. Econometric seminar at UCLA. Department of Economics, UCLA: Department of Economics, UCLA.
• Ichimura, H. (2018, October). A Progress Report on Japanese Study on Aging and Retirement. ELSA International Consultants Meeting. Institute of Fiscal Studies: Institute of Fiscal Studies and RAND.
• Ichimura, H. (2018, October). Comments on "The Effect of Education on Late-Life Cognition: A Cross-Country Comparison" By Marco Angrisani, Jinkook Lee, Erik Meijer. Conference on Cross-Country Analysis of Retirement, Health, and Well-Being. University of Southern California: University of Southern California.
• Ichimura, H. (2018, October). Locally Robust Semiparametric Estimation. Econometric seminar at Northwestern University. Department of Economics, Northwestern University: Department of Economics, Northwestern University.
• Ichimura, H. (2018, October). Locally Robust Semiparametric Estimation. Econometric seminar at University of Chicago. Department of Economics, University of Chicago: Department of Economics, University of Chicago.
• Ichimura, H. (2018, September). General discussion. Conference for Joe Altonji's 65th Birthday. Department of Economics, Yale University: Department of Economics, Yale University.
• Ichimura, H. (2018, September). Locally Robust Semiparametric Estimation. Econometric seminar at Georgetown University. Department of Economics, Georgetown University: Department of Economics, Georgetown University.
• Ichimura, H. (2018, September). Locally Robust Semiparametric Estimation. Econometric seminar at University of Maryland. Department of Economics, University of Maryland: Department of Economics, University of Maryland.