Fuzzy Regression Discontinuity Research Articles

Fiscal transfers, local government, and entrepreneurship

AbstractCan local government spending spur entrepreneurial activity? To answer this question, we study a setting where, around multiple pre‐determined and non‐manipulable thresholds, municipalities with lower tax revenues receive direct and different monetary grants from the national budget. Employing a fuzzy regression discontinuity design, we find a positive impact of fiscal transfers on the number of firms, especially sole proprietorships and small firms. The impact is stronger in municipalities where the opposition is more involved in the legislative process or more parties are represented in the municipal council.

Improving public school productivity: Evidence from model schools in India

This paper studies the impact of India’s “model” school program which aimed to provide high quality education to economically disadvantaged students. Model schools combine better infrastructure with more accountability, contract teachers, and lower per-pupil spending than regular public schools. Using a fuzzy Regression Discontinuity Design based on entrance exam cutoffs, I find attending a model school for five years increases test scores in math by 0.38 standard deviations, in science by 0.26 sd, and in social science by 0.26 sd on average. Furthermore, model schools increase the probability of joining pre-university by 11.5 percentage points. The results suggest it is possible to deliver substantial improvement of outcomes in public schools at a slightly lower level of school spending through a package of reforms.

Student Success Coaching Is Best Delivered Early and Often

This research investigates the effects of a success coaching program by examining the timing of the first appointment and the frequency of additional coaching. Using a fuzzy regression discontinuity approach, we provide Local Average Treatment Effects of the program when students make appointments early in the semester. Students with appointments scheduled before discontinuity had an average of two additional meetings compared to students who waited to make their first appointment. We find each coaching session yields an effect of about 0.2–0.3 grade points and approximately 0.8 additional credits earned. This article contributes to understanding of the timing and frequency effects for success coaching and recommends programs aim to have first interactions with students happen before the sixth week of classes and offer four to six appointments per semester for maximum impact.

Building social trust through education

Social trust plays a critical role in shaping the quality of governance. However, we still lack sufficient causal evidence of how social trust can be cultivated through the lens of public policy. In this paper, we, therefore, address this issue by identifying college education as a significant source of social trust. We employ a fuzzy regression discontinuity design based on the massive college enrollment expansion in China that began in 1999. Specifically, using nationwide survey data from the Chinese Household Income Project (CHIP), we find that college education improves both generalized and personalized trust at the individual level. These effects are most pronounced among females and individuals in less developed areas (i.e., Central and Western China). Further analysis demonstrates that college education enhances individuals' social trust by elevating their socioeconomic status, specifically by increasing their chances of obtaining urban Hukou and raising their income. Overall, our findings suggest that in developing countries, social trust can be cultivated by improving access to higher education.

Impact of studying abroad on language skill development: Regression discontinuity evidence from Japanese university students

The importance of English communication skills has increased with globalization, and governments in various countries have encouraged students to go abroad. However, the causal impact of studying abroad has rarely been investigated, particularly in non-European countries. This study adopted a fuzzy regression discontinuity design (RDD) for the flagship scholarship program of the Japanese government for youth studying abroad. We found that scholarships significantly increased the probability of studying abroad by 77–80 points. We measured English proficiency using a multiple-choice test we originally developed and found that studying abroad increased English proficiency by 33–38% (or 1.15–1.35 standard deviation). We also found that studying abroad significantly improved their international posture and perceived communication competence in a foreign language, which are the two traits found to be important determinants of future development in language ability in applied linguistics literature.

Impact of pension income on healthcare utilization of older adults in rural China

ObjectiveIn China, rural residents experience poorer health conditions and a higher disease burden compared to urban residents but have lower healthcare services utilization. Rather than an insurance focus on enhanced healthcare services utilization, we aim to examine that whether an income shock, in the form of China’s New Rural Pension Scheme (NRPS), will affect outpatient, inpatient and discretionary over-the-counter drug utilization by over 60-year-old rural NRPS residents.MethodsProviding a monthly pension of around RMB88 (USD12.97), NRPS covered all rural residents over 60 years old. Fuzzy regression discontinuity design (FRDD) was employed to explore the NRPS causal effect on healthcare services utilization, measured by outpatient and inpatient visits and discretionary over-the-counter drug purchases. The nationwide China Health and Retirement Longitudinal Study (CHARLS) 2018 provided the data.ResultsWithout significant changes in health status and medication needs, 60-plus-year-old NRPS recipients significantly increased the probability of discretionary OTC drug purchases by 33 percentage points. NRPS had no significant effect on the utilization of outpatient and inpatient utilization. The increase in the probability of discretionary OTC drug purchases from the NRPS income shock was concentrated in healthier and low-income rural residents. Robustness tests confirmed that FRDD was a robust estimation method and our results are robust.ConclusionNRPS was an exogenous income shock that significantly increased the probability of discretionary over-the-counter drug purchases among over 60-year-old rural residents, but not the utilization of inpatient or outpatient healthcare services. Income remains an important constraint for rural residents to improve their health. We recommend policymakers consider including commonly used over-the-counter drugs in basic health insurance reimbursements for rural residents; provide health advice for rural residents to make discretionary over-the-counter drug purchases; and to mount an information campaign on over-the-counter drug purchasing in order to increase the health awareness of rural residents.

Can German vocational training combat skill shortages in developing countries? Evidence from dual training system in the Philippines

AbstractThis paper examines the labor market impacts of the Philippine Dual Training System (DTS), a vocational training program that combines firm‐based on‐the‐job training (OJT) and conventional institution‐based training, using a recent survey that tracked graduates from DTS and regular institution‐based programs. The estimation results using fuzzy regression discontinuity design show a significantly positive impact on the most recent monthly earnings. The impact significantly increases with the OJT intensity, measured by the number of hours a week in OJT. The above results imply that vocational training has to be aligned with actual production technologies and work practices to narrow skill gaps and create positive returns.

Approaches to risk ratio estimation in a regression discontinuity design: Application to the prescription of statins for cholesterol reduction in UK primary care.

In recent years regression discontinuity designs have been used increasingly for the estimation of treatment effects in observational medical data where a rule-based decision to apply treatment is taken using a continuous assignment variable. Most regression discontinuity design applications have focused on effect estimation where the outcome of interest is continuous, with scenarios with binary outcomes receiving less attention, despite their ubiquity in medical studies. In this work, we develop an approach to estimation of the risk ratio in a fuzzy regression discontinuity design (where treatment is not always strictly applied according to the decision rule), derived using common regression discontinuity design assumptions. This method compares favourably to other risk ratio estimation approaches: the established Wald estimator and a risk ratio estimate from a multiplicative structural mean model, with promising results from extensive simulation studies. A demonstration and further comparison are made using a real example to evaluate the effect of statins (where a statin prescription is made based on a patient's 10-year cardiovascular disease risk score) on low-density lipoprotein cholesterol reduction in UK Primary Care.

IDENTIFICATION AND INFERENCE IN A QUANTILE REGRESSION DISCONTINUITY DESIGN UNDER RANK SIMILARITY WITH COVARIATES

This study investigates the identification and inference of quantile treatment effects (QTEs) in a fuzzy regression discontinuity (RD) design under rank similarity. Unlike Frandsen et al. (2012, Journal of Econometrics 168, 382–395), who focus on QTEs only for the compliant subpopulation, our approach can identify QTEs and average treatment effect for the whole population at the threshold. We derived a new set of moment restrictions for the RD model by imposing a local rank similarity condition, which restricts the evolution of individual ranks across treatment status in a neighborhood around the threshold. Based on the moment restrictions, we derive closed-form solutions for the estimands of the potential outcome cumulative distribution functions for the whole population. We demonstrate the functional central limit theorems and bootstrap validity results for the QTE estimators by explicitly accounting for observed covariates. In particular, we develop a multiplier bootstrap-based inference method with robustness against large bandwidths that applies to uniform inference by extending the recent work of Chiang et al. (2019, Journal of Econometrics 211, 589–618). We also propose a test for the local rank similarity assumption. To illustrate the estimation approach and its properties, we provide a simulation study and estimate the impacts of India’s 40-billion-dollar national rural road construction program on the reallocation of labor out of agriculture.

Does Schooling Have Lasting Effects on Cognitive Function? Evidence From Compulsory Schooling Laws.

This study assesses whether an additional year of secondary schooling has lasting causal effects on cognitive function. I use data from Understanding Society, the largest longitudinal household study in the United Kingdom, and exploit quasi-experimental variation in schooling from the 1972 raising of the school-leaving age in England and Wales. This reform increased the minimum secondary school‒leaving age from 15 to 16 years. Cognitive function outcomes were measured when participants were aged 48 to 60. Using a fuzzy regression discontinuity design, I show that remaining in school until age 16 improved working memory by one third to one half of a standard deviation. I find limited evidence for causal effects on verbal fluency and measures of numeric ability. Analyses of potential mechanisms showed statistically significant effects of remaining in school until age 16 on the type of occupation entered immediately after leaving school and at older ages. These patterns are consistent with basic education improving cognitive outcomes through occupation choice. The findings are robust to sensitivity analyses.

Does higher education affect pro-environmental behavior? Evidence from household waste recycling in Greece

We examine the effect of higher education attainment on pro-environmental behavior focusing on household waste recycling. To address the endogeneity of higher education attainment, we exploit a set of reforms that increased opportunities for university studies in Greece, affecting cohorts graduating from high school in year 2000. We leverage the exogenous variation introduced by the school enrollment age cutoff and estimate the local average treatment effect of higher education employing a fuzzy regression discontinuity design, comparing educational attainment and recycling behavior between households that were just, and just not exposed to the reforms. We find little evidence that higher education increases the probability of recycling, and no evidence that the share of household waste recycled responds to higher educational attainment. Our results suggest that human capital accumulation alone may not deliver green behavior returns.

Can I live with you after I retire? Retirement, old age support and internal migration in a developing country

AbstractThis study provides causal evidence on the impact of retirement on internal migration in a developing country. Using a fuzzy regression discontinuity design, combined with a nationally representative sample of 228,855 older Chinese adults, we find that retirement leads to an increase in the probability of being a migrant by 12.9 percentage points (an 80% increase in migration). Approximately 38% of the total migration effects can be attributed to intertemporal shifts. The impact is more pronounced for the lower‐educated, those who have restricted access to health insurance and pension and those who come from origins with high accommodation costs. Relying on old age support from adult children in migration is a likely mechanism. Lifting migration restrictions and improving government‐provided benefits for retirees could be helpful to cope with population aging in a developing country.

School starting age policy and students' risky health behaviors.

A school starting age policy may generate disparities in the age at which students enter school, and in the relative age for grade among children born in similar periods. I analyze the impacts of being young-for-grade on students' risky health behaviors. Using a fuzzy regression discontinuity design which exploits the school entry system in South Korea, I find that being younger in the class leads students to begin drinking alcohol earlier. In addition, it increases the likelihood of drinking over the past 30days. Being young-for-grade also impacts the probability of having sexual intercourse during one's high school years. My main findings are driven by both girls and boys. The robustness of my results is supported by several alternative specifications.

The thin blue line in schools: New evidence on school‐based policing across the U.S.

AbstractU.S. public school students increasingly attend schools with sworn law enforcement officers present. Yet little is known about how these school resource officers (SROs) affect school environments or student outcomes. Our study uses a fuzzy regression discontinuity (RD) design with national school‐level data from 2014 to 2018 to estimate the impacts of SRO placement. We construct this discontinuity based on the application scores for federal school‐based policing grants of linked police agencies. We find that SROs effectively reduce some forms of violence in schools, but do not prevent gun‐related incidents. We also find that SROs intensify the use of suspension, expulsion, police referral, and arrest of students. These increases in disciplinary and police actions are consistently largest for Black students, male students, and students with disabilities.

Unplanned Costs and Benefits: Gender and Spousal Spillover Effects of Retirement on Health.

Our study assesses how women and men's health indicators are shaped by their spouse's retirement. The retirement process can reshape the health of a retiree, but these effects can also extend onto the health of spouses. Although past research has largely focused on how men's retirement might negatively shape their wife's health outcomes, it is possible that wives' retirement has detrimental effects on their husband's health as well. Using data from the Health and Retirement Study (HRS), we employ a fuzzy regression discontinuity design to identify the causal effects of spousal retirement on indicators of physical and mental health in married older adults. Our results suggested that men, not women, experience the most negative spousal spillover effects of retirement on their health outcomes. We found the most support for spillover effects on spouses' physical health outcomes. Additionally, men who are not working when their spouse retires experienced the most negative health effects. Women and men's health is differentially affected by spousal retirement, where men might be the most negatively affected by their spouses' transition in the U.S. context. These results contradict conventional wisdom that undergirds numerous untested assumptions underlying prior research on this significant life transition.

Safe at Last? Late Effects of a Mass Immunization Campaign on Households' Economic Insecurity

We study the effects of receiving immunization from COVID‐19 on households' economic insecurity. To provide causal estimates we use a fuzzy regression discontinuity design which takes advantage of the UK's immunization plan. The plan was primarily based on age, granting differential eligibility to proximate cohorts. Our estimated local average treatment effect indicates that the share of households who declared being economically insecure dropped by 41 percentage points among those who received the vaccine due to the eligibility criteria. Using a difference‐in‐discontinuity design we next document that immunization was more salient for women as well as for large households and those with children. Our results suggest that the mass immunization campaign against COVID‐19 had relevant short‐run economic effects, well beyond its expected impact on people's health.

Early dropout, earnings and skills in later life: evidence from Spain

ABSTRACT Most education systems have set a minimum age until which students must stay at school. In the case of Spain, students can drop out the same day they reach that age, even without finishing that academic year. In the present research work, we intend to analyse the influence of early dropout on later life outcomes for the Spanish population, i.e. literacy and numerical skills and earnings. In order to do this, we employ PIAAC 2012 data and a fuzzy regression discontinuity design, focusing on a transition period between education laws: a first period without minimum dropout age and a second period (law approved in 1970) which obligated students to be at least 14 years-old to dropout. Our results show, for this particular cohort of the Spanish population, that dropping out is substantially negatively associated with literacy and numerical skills in around 1.5 standard deviations, while it did not influence future earnings.

The impact of retirement on body mass index in China: An empirical study based on regression discontinuity design

An aging population is an important trend of social development, and it will be China's basic national condition for a long time. However, the pressure on domestic pension payments and economic operations will increase daily. The delayed retirement policy is gradually implemented as a critical initiative to improve capital and labor force allocation. The impact of retirement on residents' Body Mass Index (BMI) and weight has become a focus issue. This paper investigates the mechanism of the impact of retirement on residents' BMI using microdata from the China Family Panel Studies (CFPS) 2018, combined with a fuzzy regression discontinuity design to measure the potential health impact of China's current retirement policy on residents. The study finds that: (1) Retirement has a significant negative effect on BMI for women, with retirement leading to a significantly increased risk of deviation from normal BMI levels and significantly increasing the weight of retired women. However, retirement does not have a significant effect on men. (2) Retirement policies affect residents' BMI to different degrees depending on their family size, with the negative effect on women being more pronounced in smaller family sizes. (3) Female residents who retire to help their children with intergenerational care are more likely to maintain normal BMI levels, significantly positively affecting their potential health. (4) Retirement negatively affects BMI through channels such as significantly reducing exercise frequency among female residents. The study demonstrates that retirement policy impacts the BMI and weight of female residents, so the formulation and implementation of delayed retirement policy should be flexible, and family factors such as family sizes and intergenerational care should be considered appropriately.

Complier and monotonicity for Fuzzy Multi-score Regression Discontinuity with partial effects

In fuzzy regression discontinuity with multiple scores, it is not clear what the monotonicity condition should be and how to define compliers. This paper introduces ‘(semi-) strong monotonicity’, so that the identified effect becomes the effect on the just-treated compliers.

Heterogeneous Effects of Women's Schooling on Fertility, Literacy and Work: Evidence from Burundi's Free Primary Education Policy

Abstract This article investigates the effect of women's schooling on fertility as well as on associated mechanisms by leveraging Burundi's free primary education policy (FPE) of 2005 as a natural experiment. Exogenous variation in schooling is identified through a fuzzy regression discontinuity design. Our results show that educational attainment was positively influenced by Burundi's FPE for women situated at all wealth levels. However, the relevant downstream effects of schooling—measured by fertility, literacy and work outcomes—reveal heterogenous treatment effects which are moderated by women's household wealth. While poor women profit in terms of increases in literacy (6.7 percentage-point increase for each year of policy-induced schooling), remunerated employment opportunities (5.7 percentage-point increase), as well as a reduction in desired and actual fertility outcomes (6.9 percentage-point reduction in teenage childbirth), none of these effects of additional education are observed for women from the wealthier households of our sample. The evidence of such a marked heterogeneity contributes to the growing literature examining the nexus between education and fertility in developing countries and helps to evaluate under which conditions the literature's findings may generalize.