Regression Discontinuity Framework Research Articles

More Harm than Good?

By analyzing a French program that targeted low-achieving and socially disadvantaged junior high schools we provide evidence that school-based compensatory education policies create sorting effects. We use geocoded original data and a regression discontinuity framework to show that the program decreases the individual probability of attending a treated school and symmetrically increases the probability of attending a private school. The effects are driven by pupils from high socioeconomic backgrounds, resulting in an increase in social segregation across schools.

Womb at Work: The Missing Impact of Maternal Employment on Newborn Health

Parental leave policies across the globe have become much more generous than they used to be. This is also true for prenatal maternal leave. While this may be costly in the short run, little is known about the effect of maternal employment during pregnancy on newborn health. In this paper, I exploit three sharp policy changes on the duration of paid parental leave in Austria that strongly affected the share of mothers who work up to the 32nd week of pregnancy. I use administrative data from Austria on the working history of women linked to the full Austrian birth register and coupled with a regression discontinuity framework to identify the effect of prenatal employment on their offspring. Maternal employment during pregnancy with the second child reacts strongly to these policy changes. The share of employed mothers sharply declined in 1990 by 19.1 percentage points, increased in 1996 by 7.2 percentage points and declined again by 6.4 percentage points in 2000. None of these changes in prenatal employment translated into effects on newborn health measured via birth weight, gestational length, and Apgar scores. This result holds true for mothers of different socioeconomic backgrounds and across industries. The effect is precisely estimated, which suggests that prenatal employment prior to the 32nd week of pregnancy does not causally affect the fetus for measures visible at birth.

The Hardships of Long Distance Relationships: Time Zone Proximity and Knowledge Transmission within Multinational Firms

Using a unique dataset on worldwide multinational corporations with precise location of headquarters and affiliates, I present evidence of a trade-off between distance to the headquarters and the knowledge intensity of the foreign subsidiary’s economic activity, emerging from dynamics related to the proximity-concentration hypothesis. This trade-off is strongly diminished the higher the overlap in working hours between the headquarters and its foreign subsidiary. In order to rule out biases arising from confounding factors, I implement a regression discontinuity framework to show that the economic activity of a foreign subsidiary located just across the time zone line that increases the overlap in working hours with its headquarters is, on average, about one percent higher in the knowledge intensity scale. I find no evidence of the knowledge intensity and distance trade-off weakening when a non-stop flight exists between the headquarters and the foreign subsidiary. The findings suggest that lower barriers to real-time communication within the multinational corporation play an important role in the location strategies of multinational corporations.

The Competitive Effects of the Louisiana Scholarship Program On Public School Performance

Given the significant growth rate and geographic expansion of private school choice programs over the past two decades, it is important to examine how traditional public schools respond to the sudden injection of competition for students and resources. This article uses 1) a school fixed effects approach, and 2) a regression discontinuity framework to examine the achievement impacts of the Louisiana Scholarship Program (LSP). This targeted school voucher program has provided public funds for low-income students in low-performing public schools to enroll in participating private schools since the 2012-13 school year. The main findings of the competitive effects analysis reveal neutral to positive impacts that are small in magnitude. Effects are largest for students attending those public schools most affected by the competitive threat. Policy implications are discussed.

Does Dilution Due to Option Exercises Cause Share Repurchases?

Equity-based compensation causes increases in firms' share count and dilutes Earnings Per Share (EPS), which provides firms with an incentive to raise EPS using either share buybacks or earnings management. We employ a regression discontinuity framework to provide evidence of a causal link between equity-based pay and EPS management. Our tests compare firms experiencing dilution from ``just-in-the-money'' option exercises with firms whose options end up narrowly out-of-the-money. We find that firms engage in real- and accruals-based earnings management, but not buybacks, to boost EPS around these plausibly exogenous dilutive events. These effects are stronger when executives' bonuses depend directly on EPS.

The effect of subway expansions on vehicle congestion: Evidence from Beijing

Does public transportation reduce vehicle congestion? Using a regression-discontinuity framework, we examine the effect of six subway openings on short-run congestion in Beijing between 2009 and 2015. We find that vehicle congestion drops sharply immediately after new subway openings. In our central specification, each of the subway openings decreased delay times by an average of 15% in the short run over the city of Beijing.

TAKS-ing students? Texas exit exam effects on human capital formation

High-stakes exit exams are pervasive in the American education system and have the ability to affect students far beyond their earned scores. This paper considers how exit exams in Texas affect student motivational responses and classroom behavior before the end of high school. Employing a regression discontinuity framework, I examine the impact of failing the exam the first time it is administered. Considering behavioral responses to the administration of the Texas Assessment of Knowledge and Skills (TAKS), I study the impact on students’ courses taken, attendance, and disciplinary actions after the exam in the final year of high school. I find that, in line with a model of motivation with heterogeneous effects, some students who fail respond through an increase in the number of courses taken in their senior year, and find a smaller increase in disciplinary infractions.

The Hardships of Long Distance Relationships: Knowledge Transmission and the Ease of Communication within Multinational Firms

Using a unique dataset on worldwide multinational corporations with precise location of headquarters and affiliates, I present evidence of a trade-off between distance to the headquarters and the knowledge intensity of the foreign subsidiary’s economic activity, emerging from dynamics related to the proximity-concentration hypothesis. This trade-off is strongly diminished the higher the overlap in working hours between the headquarters and its foreign subsidiary. In order to rule out biases arising from confounding factors, I implement a regression discontinuity framework to show that the economic activity of a foreign subsidiary located just across the time zone line that increases the overlap in working hours with its headquarters is, on average, about one percent higher in the knowledge intensity scale. I find no evidence of the knowledge intensity and distance trade-off weakening when a non-stop flight exists between the headquarters and the foreign subsidiary. The findings suggest that lower barriers to real-time communication within the multinational corporation play an important role in the location strategies of multinational corporations.

DO PRIVATE TUTORS ENHANCE ENGLISH LANGUAGE ABILITY? REGRESSION DISCONTINUITY EVIDENCE FROM A POLICY EXPERIMENT IN INDIA

ABSTRACTEmpirical evidence has shown that returns to English language ability are substantial in India. Research has also focused on the impact of private tutoring in the context of developing economies to find evidence that remedial teaching leads to better student achievement and higher test scores. In this paper I analyze whether private tutoring helps in developing English language ability. Simple OLS estimates suggest statistically significant effects but estimated coefficients are rather small at 0.2 percentage points. Presence of confounders and selection bias may potentially understate the true effects of private tutoring on English language ability. To address this issue, I exploit a unique policy experiment in India in a regression discontinuity framework to identify cohorts eligible for private tutoring and compare their outcomes to the ineligible. I use this potentially exogenous policy variation as an instrument for private tutoring and find significant increases in English language ability estimated at 16 percentage points which is much larger than the simple OLS effects. Since standard academic curriculum is not directed towards improving foreign language ability, private tutoring as an input in the education production function seems to offer high marginal returns. From a policy perspective, this implies that private tutoring should be encouraged for enhancing English ability even though there are concerns that this provides disincentives for teachers to teach in school.

A Quasi-Experimental Evaluation of the Impact of Public Assistance on Prisoner Recidivism

The Welfare Act of 1996 banned welfare and food stamp eligibility for felony drug offenders and gave states the ability to modify their use of the law. Today, many states are revisiting their use of this ban, searching for ways to decrease the size of their prison populations; however, there are no empirical assessments of how this ban has affected prison populations and recidivism among drug offenders. Moreover, there are no causal investigations whatsoever to demonstrate whether welfare or food stamp benefits impact recidivism at all. This paper provides the first empirical examination of the causal relationship between recidivism and welfare and food stamp benefits Using a survival-based estimation, we estimated the impact of benefits on the recidivism of drug-offending populations using data from the National Corrections Reporting Program. We modeled this impact using a difference-in-difference estimator within a regression discontinuity framework. Results of this analysis are conclusive; we find no evidence that drug offending populations as a group were adversely or positively impacted by the ban overall. Results apply to both male and female populations and are robust to several sensitivity tests. Results also suggest the possibility that impacts significantly vary over time-at-risk, despite a zero net effect. Overall, we show that the initial passage of the drug felony ban had no measurable large-scale impacts on recidivism among male or female drug offenders. We conclude that the state initiatives to remove or modify the ban, regardless of whether they improve lives of individual offenders, will likely have no appreciable impact on prison systems.

Closing Routes to Retirement: How Do People Respond?

We present quasi-experimental evidence on the employment effects of an unprecedented large increase in the early retirement age (ERA). Raising the ERA has the potential to extend contribution periods and to reduce the number of pension beneficiaries at the same time, if employment exits are successfully delayed. However, workers may not be able to work longer or may choose other social support programs as exit routes from employment. We study the effects of the ERA increase on employment and potential program substitution in a regression-discontinuity framework. Germany abolished an important early retirement program for women born after 1951, effectively raising the ERA for women by three years. We analyze the effects of this huge increase on employment, unemployment, disability pensions, and inactivity rates. Our results suggest that the reform increased both employment and unemployment rates of women age 60 and over. However, we do not find evidence for active program substitution from employment into alternative social support programs. Instead employed women remained employed and unemployed women remained unemployed. The results suggest an increase in inequality within the affected cohorts.

SUICIDE AND THE SOCIAL SECURITY EARLY RETIREMENT AGE

Using a regression discontinuity framework, this paper documents a previously unnoticed drop of 7%–8% in the U.S. suicide rate upon reaching age 62 during 1990–2014. This decline is concentrated among men, nearly doubled in size over the most recent decade as the income gap between those just older and younger narrowed, and represents the only trend break among ages 45–79. These findings, along with the observed timing of retirement and benefit claims and research on how income affects suicide, suggest that the likely explanation is Social Security early retirement benefit eligibility rather than retirement per se. (JEL I12, H55)

Closing Routes to Retirement: How Do People Respond?

We present quasi-experimental evidence on the employment effects of an unprecedented large increase in the early retirement age (ERA). Raising the ERA has the potential to extend contribution periods and to reduce the number of pension beneficiaries at the same time, if employment exits are successfully delayed. However, workers may not be able to work longer or may choose other social support programs as exit routes from employment. We study the effects of the ERA increase on employment and potential program substitution in a regression-discontinuity framework. Germany abolished an important early retirement program for women born after 1951, effectively raising the ERA for women by three years. We analyze the effects of this huge increase on employment, unemployment, disability pensions, and inactivity rates. Our results suggest that the reform increased both employment and unemployment rates of women age 60 and over. However, we do not find evidence for active program substitution from employment into alternative social support programs. Instead employed women remained employed and unemployed women remained unemployed. The results suggest an increase in inequality within the affected cohorts.

Parental Investments in Early Life and Child Outcomes Evidence from Swedish Parental Leave Rules

How do parental resources early in life affect children's health and schooling outcomes? We address this question by exploiting the so-called speed premium (SP) in the Swedish parental leave (PL) system. The SP grants mothers higher PL benefits for the subsequent child without the need to re-quality for benefits through pre-birth market work, if the two births occur within a pre-specified interval. The eligibility birth interval for this automatic renewal of PL benefits was set to 24 months in 1980–1985, and to 30 months since 1986. This allow us to use a Regression Discontinuity framework to study how additional parental time and monetary resources impacts both the existing and new child's educational and health outcomes. We find that maternal eligibility to the SP increases the 9th grade GPA, and the likelihood of college attendance of the first-born child, but it does not affect the secondborn. These impacts can be generalized for higher-order parities. The effects are prevalent primarily among children to high-income mothers. Impacts are driven by a combination of a persistent positive income shock, and substitution away from informal care to maternal time.

Cognitive, Socioemotional and Behavioral Returns to College Quality

We exploit the variation in the admissions cutoffs across colleges of a leading Indian university in a regression discontinuity framework to estimate the causal effects of enrolling in a selective college on: cognitive attainment, behavioral preferences, and Big Five personality. We find that enrolling in a selective college improves only females' exam scores. Further, marginally admitted females in selective colleges become less overconfident and less risk averse while males in selective colleges experience a decline in extraversion and conscientiousness. Higher attendance rates among females explain the gender differences in returns to better college and peer environment.

Advising students on their field of study: Evidence from a French University reform

This paper measures the effect of a policy implemented in France in 2009 advising students on their field of study at university. Applicants receive reviews from universities on their chances of graduating, which are determined relative to their numerical grades in high-school. To measure the causal impact of the reviews on the choice of their field of study, we compare students with similar high-school numerical grades but different reviews in a regression discontinuity framework. From a database of first year undergraduate applicants, we estimate that receiving a positive signal in a given field of study has little impact on the probability of registration, while receiving a negative signal in a given field decreases the proportion of students enrolling in this field by 14 percentage points.

The Impact of Secondary Schooling in Kenya

This paper estimates the impacts of secondary school on human capital, occupational choice, and fertility for young adults in Kenya. The probability of admission to government secondary school rises sharply at a score close to the national mean on a standardized 8th grade examination, permitting the estimation of causal effects of schooling in a regression discontinuity framework. The analysis combines administrative test score data with a recent survey of young adults to estimate these impacts. The results show that secondary schooling increases human capital, as measured by performance on cognitive tests included in the survey. For men, there is a drop in the probability of low-skill self-employment, as well as suggestive evidence of a rise in the probability of formal employment. The opportunity to attend secondary school also reduces teen pregnancy among women.

The Impact of Mandated Corporate Social Responsibility: Evidence from India's Companies Act of 2013

Firms’ Corporate Social Responsibility (CSR) activity has become the subject of a large literature in recent years. This paper analyzes CSR activity using quasi-experimental variation created by Section 135 of India’s Companies Act of 2013, which requires (on a “comply-or-explain” basis) that firms satisfying specific size or profit thresholds spend a minimum of 2% of their net profit on CSR. We examine effects along a number of different dimensions including firm value, CSR spending, and other outcomes, as well as exploring broader theoretical implications. Our analysis uses financial statement and stock price data on Indian firms from the Prowess database, along with hand-collected data from firms’ disclosures of CSR activity. By combining a regression discontinuity (RD) framework (based on a nonparametric local polynomial regression approach) with a standard event study, we find a negative and substantial effect on the value of affected firms (relative to unaffected firms) around the crucial event date. This effect seems to be concentrated among firms that are less customer-facing, as indicated by low advertising expenditures. Using a difference-in-difference approach, we find significant increases in CSR activity among firms affected by Section 135, especially in the fraction of firms engaging in CSR spending. The fraction of firms subject to Section 135 that engage in advertising expenditures appears to have declined, consistent with substitution between advertising and CSR. There is no robust evidence of any significant impact on sales or accounting performance, although a modest decline in the return on assets cannot be ruled out. For a subset of large firms, we hand-collect comprehensive CSR data and find that while firms initially spending less than 2% increased their CSR activity, large firms initially spending more than 2% reduced their CSR expenditures after Section 135 came into effect. We explore various explanations for this presumably unintended consequence of Section 135, and also seek to derive some wider implications of this analysis for understanding the role of CSR.

PHP9 - DETERMINING THE FEASIBILITY OF USING A REGRESSION DISCONTINUITY APROACH TO COMPARE EMERGENCY ROOM UTILIZATION BETWEEN MEDICAID FEE-FOR-SERVICE AND COMMERCIALLY OBTAINED COVERAGE IN THE ACA MEDICAID EXPANSION POPULATION

Arkansas implemented a premium assistance strategy to provide individual commercial health insurance for the Medicaid expansion population of the Patient Protection and Affordable Care Act. To mitigate adverse selection, a 12 item patient reported frailty screener, ranging from 0.023 - 1 was used to assign individuals with higher health needs (frailty score > 0.1799) to traditional Medicaid coverage instead of commercial insurance. The objective of this study was to determine the feasibility of using a regression discontinuity (RD) framework to contrast the rates of overall emergency room (ER) use, mental health related ER use and potentially preventable ER use between patients assigned to traditional Medicaid and patients enrolled in commercial insurance. Our study was comprised of ambulatory, non-incarcerated newly enrolled persons between 18 and 64 years of age, who had least 180 days of stable enrollment in commercial insurance or Medicaid in 2014 and completed a frailty screener prior to enrollment. Administrative claims data from Medicaid and two of the commercial insurers providing coverage to the Medicaid expansion population were used to calculate the rates of overall ER, mental health related ER and potentially preventable ER use. To explore the feasibility of using RD, we stratified ER rates by frailty score deciles and estimated a linear regression to test the association between the pre-score (frailty score) and our outcome measure. Our sample was comprised of 17,453 subjects of which 10,894 were above the frailty cut-point. A positive linear trend was observed between overall ER use (β=1,917; p=0.2443), mental health related ER use (β=355; p=0.0999), and potentially preventable ER use (β=979; p=0.3223) and frailty score. All three ER measures showed a positive linear trend association with the frailty score thus supporting future application of the RD framework for inferential analysis.

The Effect of English Language Learner Reclassification on Student ACT Scores, High School Graduation, and Postsecondary Enrollment: Regression Discontinuity Evidence from Wisconsin

AbstractThe recent increase in the number of students classified as English language learners (ELLs) has focused significant attention on reclassification policy, which governs the process by which ELLs move toward, and are deemed to reach, full English proficiency. In this paper, we draw on a data set containing annual individual‐level records for every Wisconsin student ever classified as an ELL between the 2006–07 and 2012–13 school years to estimate the effects of being reclassified at the end of 10th grade—a crucial period on the pathway to postsecondary education—on several measures related to students’ postsecondary attainments. We estimate these effects in a regression discontinuity framework, exploiting Wisconsin's policy rule that automatically reclassifies ELLs who score above a specified cutoff on the state's English language proficiency exam. Our analysis indicates that being reclassified as fully English proficient in 10th grade has a positive effect on students’ ACT scores. It also provides some evidence of a positive effect on high school graduation and the probability of enrolling in a postsecondary institution the fall after graduation. Together, our analyses provide evidence on the effects of a policy directly relevant to the country's fastest growing student population, and we close the paper with a discussion of the implications for research and policy.