"The Impact of SNAP Work Requirements" (with Erik Scherpf and Young Jo)
Abstract: "Our paper examines the impact of SNAP work requirements on the labor supply of participants and on overall participation in SNAP. We perform a regression discontinuity analysis of the impact of work requirements for able bodied adults without dependents (ABAWDs) on labor supply and participation, exploiting the fact that the work requirement applies only to individuals under 50 years old. Using a novel dataset containing ABAWD work requirement waiver information merged with SNAP administrative records and American Community Survey (ACS) data, we find the work requirements have no impact on labor force participation and the number of hours worked. We do find that the work requirements reduce participation in SNAP. There is some evidence that those with worse job prospects are especially less likely to participate in SNAP as a result of the work requirements. We find little evidence that ABAWDs respond to the work requirements by claiming disability."
"The Implications of Misreporting for Longitudinal Studies of SNAP" (with Erik Scherpf)
Abstract: "Researchers studying a variety of important economics, nutrition, and health topics use survey data containing information on SNAP participation. In order to study the dynamics of SNAP participation or recognizing possible selection bias in cross-sectional estimators, many researchers use longitudinal estimators to estimate the causal effects of SNAP. However, misreporting of SNAP participation is common in survey datasets, and bias from misreporting can be larger for longitudinal estimators. In an analysis of data combining newly compiled administrative datasets on SNAP participation from nine states and covering the years 2005-2015 with individual records from the CPS ASEC survey, we confirm findings in previous studies of substantial misreporting and find evidence that the misreporting is not done at random. Additionally, we examine bias caused by misreporting in a longitudinal estimators and find severe bias, much greater in magnitude than bias caused by misreporting in cross-sectional estimators. We find that a longitudinal conditional distribution estimator may be an attractive solution for researchers using public use survey datasets."
"The Supplemental Nutrition Assistance Program (SNAP) and the Economy: New Estimates of the SNAP Multiplier" (with Pat Canning)
ERS Economic Research Report
Abstract: "The Supplemental Nutrition Assistance Program (SNAP) is one of the largest safety net programs in the United States—the U.S. Department of Agriculture spent $65.3 billion on the program in fiscal year 2018 and served an average of 40.3 million people per month. By design, SNAP has a countercyclical effect on the wider economy, that is, program enrollment increases when incomes fall and vice versa. The Great Recession of 2007-09 motivated new interest in the impacts of different Federal stimulus tools, including SNAP spending. We examine the countercyclical impacts of SNAP by measuring how SNAP benefits affect gross domestic product, employment, and incomes across the farm economy and for all other industries impacted by SNAP. A review of the literature suggests that SNAP spending during a recession stimulates economic output more than several other fiscal policy tools that have been used to increase economic activity. We estimate multiplier effects of SNAP expansion using a newly compiled Social Accounting Matrix multiplier model and the most recent data available for this purpose. We find that $1 billion in SNAP benefits spent during an economic downturn provides direct added income to the businesses where those benefits are spent and indirect added income to their suppliers and their employees, who in turn spend more and further increase the effect of the initial outlay. This multiplier effect generates an additional $0.5 billion, making the total effect of the $1 billion in SNAP benefits $1.5 billion in gross domestic product, which supports 13,560 new jobs—including $32 million added income going to agricultural industries that support 480 agricultural jobs. "
"Does the Precision and Stability of Value-Added Estimates of Teacher Performance Depend on the Types of Students They Serve?" (with Cassie Guarino and Jeff Wooldridge)
Economics of Education Review
Abstract: "In this paper, we investigate how the precision and year-to-year stability of a teacher’s value-added estimate relate to student characteristics. We find that teachers serving initially higher performing students have more precise value-added estimates and in most cases have higher year-to-year stability levels than teachers with lower performing students. We also decompose the variation in value-added estimates into components that reflect persistent and transitory variation in true teacher performance as well as variation caused by imprecision in the estimates. We find that teachers with lower performing students have less precision in their estimates and more transitory variation in value-added from year to year than other teachers. Our estimates imply that if teachers serving initially lower performing students had levels of precision and transitory variation in their value-added estimates equal to those serving higher performing students, the year-to-year stability in their estimates would actually exceed that of teachers with initially higher performing students."
"Using a Policy Index To Capture Trends and Differences in State Administration of USDA's Supplemental Nutrition Assistance Program" (with Laura Tiehen and David Marquardt)
ERS Economic Research Report
Abstract: "ERS study creates a SNAP Policy Index, a metric that captures differences in States' administration of USDA's Supplemental Nutrition Assistance Program (SNAP)—formerly the Food Stamp Program. Using data from ERS's SNAP Policy Database, the study finds a general trend from 1996 to 2014 toward more accommodative State-level SNAP policies."
"Enrollment without Learning: Teacher Effort, Knowledge, and Skill in Primary Schools in Africa" (with Tessa Bold, Deon Filmer, Gayle Martin, Ezequiel Molina, Christophe Rockmore, Jakob Svensson, and Waly Wane)
Journal of Economic Perspectives
Abstract: "School enrollment has universally increased over the last 25 years in low-income countries. Enrolling in school, however, does not assure that children learn. A large share of children in low-income countries complete their primary education lacking even basic reading, writing, and arithmetic skills. Teacher quality is a key determinant of student learning, but not much is known about teacher quality in low-income countries. This paper discusses an ongoing research program intended to help fill this void. We use data collected through direct observations, unannounced visits, and tests from primary schools in seven sub-Saharan African countries to answer three questions: How much do teachers teach? What do teachers know? How well do teachers teach?"
"Evaluating Specification Tests in the Context of Value-Added Estimation" (with Cassie Guarino, Mark Reckase, and Jeff Wooldridge)
Journal of Research on Educational Effectiveness
Abstract: "We study the properties of two specification tests that have been applied to a variety of estimators in the context of value-added measures (VAMs) of teacher and school quality: the Hausman test for choosing between random and fixed effects and a test for feedback (sometimes called a “falsification test”). We discuss theoretical properties of the tests to serve as background. An extensive simulation study provides important provides further insight to the VAM setting. Unfortunately, while both the Hausman and feedback tests have good power for detecting the kinds of nonrandom assignment that can invalidate VAM estimates, they also reject in situations where estimated VAMs perform very well. Consequently, the tests must be used with extreme caution when student tracking is used to form classrooms."
"A Comparison of Growth Percentile and Value-Added Models of Teacher Performance" (with Cassie Guarino, Mark Reckase, and Jeff Wooldridge)
Statistics and Public Policy
Abstract: "School districts and state departments of education frequently must choose between a variety of methods to estimating teacher quality. This paper examines under what circumstances the decision between estimators of teacher quality is important. We examine estimates derived from growth percentile measures and estimates derived from commonly used value-added estimators. Using simulated data, we examine how well the estimators can rank teachers and avoid misclassification errors under a variety of assignment scenarios of teachers to students. We find that growth percentile measures perform worse than value-added measures that control for prior year student test scores and control for teacher fixed effects when assignment of students to teachers is nonrandom. In addition, using actual data from a large diverse anonymous state, we find evidence that growth percentile measures are less correlated with value-added measures with teacher fixed effects when there is evidence of nonrandom grouping of students in schools. This evidence suggests that the choice between estimators is most consequential under nonrandom assignment of teachers to students, and that value-added measures controlling for teacher fixed effects may be better suited to estimating teacher quality in this case."
"Trends in Breastfeeding Disparities in U.S. Infants by WIC Eligibility and Participation " (with Harry Zhang, Rajan Lamichhane, Patrick McLaughlin, and Mia Wright)
"What Do Teachers Know and Do? Does it Matter? Evidence from Primary Schools in Africa" (with Tessa Bold, Deon Filmer, Gayle Martin, Ezequiel Molina, Christophe Rockmore, Jakob Svensson, and Waly Wane)
Abstract: "School enrollment has universally increased over the past 25 years in low-income countries. However, enrolling in school does not guarantee that children learn. A large share of children in low-income countries learn little, and they complete their primary education lacking even basic reading, writing, and arithmetic skills?the so-called "learning crisis." This paper uses data from nationally representative surveys from seven Sub-Saharan African countries, representing close to 40 percent of the region's total population, to investigate possible answers to this policy failure by quantifying teacher effort, knowledge, and skills. Averaging across countries, the paper finds that students receive two hours and fifty minutes of teaching per day?or just over half the scheduled time. In addition, large shares of teachers do not master the curricula of the students they are teaching; basic pedagogical knowledge is low; and the use of good teaching practices is rare. Exploiting within-student, within-teacher variation, the analysis finds significant and large positive effects of teacher content and pedagogical knowledge on student achievement. These findings point to an urgent need for improvements in education service delivery in Sub-Saharan Africa. They also provide a lens through which the growing experimental and quasi-experimental literature on education in low-income countries can be interpreted and understood, and point to important gaps in knowledge, with implications for future research and policy design."
"Ranking Teachers when Teacher Value-Added is Heterogeneous Across Students"
Abstract: "The typical measure used by researchers and school administrators to evaluate teachers is based on how the students' achievement increases after being exposed to the teacher, or based on the teacher's "value-added''. When teacher value-added is heterogeneous across her students, the typically used measure reflects differences in the average value-added the teacher provides. However, researchers, administrators, and parents may care not just about the average value-added, but also its variance. In this paper, I examine the robustness of typical teacher quality measures to alternate ranking systems factoring in the dispersion of value-added. Encouragingly, ranking systems factoring in the dispersion produce similar rankings as the ranking system based only on the mean. I also examine whether classroom characteristics and teacher experience affect a teacher's value-added variance and find that they explain little of the variation in value-added variances"
"Left with Bias? Quantile Regression with Measurement Error in Left Hand Side Variables"
Abstract: "This paper examines the effect of measurement error in the dependent variable on quantile regression, because unlike OLS regression, even classical measurement error can generate bias. I examine the pattern and size of the bias using both simulation and an empirical example. The simulations indicate that classical error can cause bias and that non-classical measurement error, particularly heteroskedastic measurement error, has the potential to produce substantial bias. Also, the size and direction of the bias depends on the amount of heterogeneity in the effects across quantiles and the regression error distribution. Using restricted access Health and Retirement Study data containing matched IRS W-2 earnings records, I examine whether estimates of the returns to education statistically differ using a precisely measured and mismeasured earnings variable. I find that returns to education are over-stated by roughly 1 percentage point at the median and 75th percentile using earnings reported by survey respondents."