Graduate School of Business
655 Knight Way
Stanford, CA 94305
NBER Program Affiliations:
NBER Affiliation: Faculty Research Fellow
Information about this author at RePEc
NBER Working Papers and Publications
|April 2016||Who Wants Affordable Housing in their Backyard? An Equilibrium Analysis of Low Income Property Development|
with Timothy McQuade: w22204
We nonparametrically estimate spillovers of properties financed by the Low Income Housing Tax Credit (LIHTC) onto neighborhood residents by developing a new difference-in-differences style estimator. LIHTC development revitalizes low-income neighborhoods, increasing house prices 6.5%, lowering crime rates, and attracting racially and income diverse populations. LIHTC development in higher income areas causes house price declines of 2.5% and attracts lower income households. Linking these price effects to a hedonic model of preferences, LIHTC developments in low-income areas cause aggregate welfare benefits of $116 million. Affordable housing development acts like a place-based policy and can revitalize low-income communities.
|The Long-term Consequences of Teacher Discretion in Grading of High-stakes Tests|
with Petra Persson: w22207
We examine the long-term consequences of teacher discretion in grading of high-stakes tests. Bunching in Swedish math test score distributions reveal that teachers inflate students who have “a bad test day,” but do not to discriminate based on immigrant status or gender. By developing a new estimator, we show that receiving a higher grade leads to far-reaching educational and earnings benefits. Because grades do not directly raise human capital, these results emphasize that grades can signal to students and teachers within the educational system, and suggest important dynamic complementarities between students’ effort and their perception of their own ability.
|February 2010||Clustering, Spatial Correlations and Randomization Inference|
with Thomas Barrios, Guido W. Imbens, Michal Kolesar: w15760
It is standard practice in empirical work to allow for clustering in the error covariance matrix if the explanatory variables of interest vary at a more aggregate level than the units of observation. Often, however, the structure of the error covariance matrix is more complex, with correlations varying in magnitude within clusters, and not vanishing between clusters. Here we explore the implications of such correlations for the actual and estimated precision of least squares estimators. We show that with equal sized clusters, if the covariate of interest is randomly assigned at the cluster level, only accounting for non-zero covariances at the cluster level, and ignoring correlations between clusters, leads to valid standard errors and confidence intervals. However, in many cases this m...
Published: Thomas Barrios & Rebecca Diamond & Guido W. Imbens & Michal Kolesï¿½r, 2012. "Clustering, Spatial Correlations, and Randomization Inference," Journal of the American Statistical Association, Taylor & Francis Journals, vol. 107(498), pages 578-591, June. citation courtesy of