Regression Discontinuity Research Articles

Returns to women’s education within marriage: evidence from a regression discontinuity design study in China

Returns to women’s education within marriage: evidence from a regression discontinuity design study in China

Prevailing Party laws and general election outcomes

AbstractObjectiveThis article examines the incidence and effects of the most common ballot ordering procedure used in U.S. general elections, Prevailing Party laws, which give the most advantageous ballot position to the currently prevailing political party.MethodsPanel regression and regression discontinuity analyses are applied to almost 50 years of county‐level election data from Wyoming.ResultsPrevailing Party laws generally increase the favored candidate's vote share by two percentage points or more, enough to flip the result of roughly 1 percent of major elections nationwide.ConclusionsThe effect of Prevailing Party laws is substantially larger than that of more innocuous ballot ordering schemes, due to “endorsement effects” these other schemes lack. The existing literature, which exclusively analyzes these other schemes, substantially understates the degree to which ballot order can be used to maintain political power.

How Education Shapes Trust: Empirical Evidence from China

The general assumption in the mainstream social capital theory states a likely causal relationship between education and trust in people, while such a conclusion might be contextual. Using the CGSS data and employing a fuzzy regression-discontinuity design where the compulsory education law (1986) in China is taken as an exogenous instrument, I find a positive correlation between education and the level of generalized trust. Further, with respect to another measure of trustthe radius of trust, I find no evidence supportive to significant correlations between education and trust in different distant social groups. However, the second-stage results do not suggest any likely causal effect from education on generalized trust or trust in different relationships, which indicates the OLS results might suffer from omitted variable bias.

Causal Inference in Transport Policy Evaluation: Principle, Method and Application of Regression Discontinuity Design

The methods of causal inference have swept the social sciences in recent years, and regression discontinuity design is one of the typical methods in causal inference. In the field of estimation and impact of traffic policy in developing countries, the research on cause-and-effect relationships is extremely limited. This paper introduces the principles, methods of regression discontinuity design, research application in the field of transport policy, and compares regression discontinuity design with methods of causal inference on the advantages and disadvantages in the field of transport policy. The purpose of this paper is to introduce the regression discontinuity design method into the field of transport policy evaluation, to look forward to its application prospects and to provide basic support for the application of regression discontinuity design in the field of developing countries' transport policy.

The Impact of Criminal Financial Sanctions: A Multistate Analysis of Survey and Administrative Data

We estimate the impact of financial sanctions in the US criminal justice system, leveraging nine natural experiments in a regression discontinuity design framework across a diverse range of enforcement levels ($17–$6,000) and institutional environments. We leverage survey and administrative data to consider a variety of short- and long-term outcomes, including employment, recidivism, household expenditures, and other self-reported measures of well-being. We find robust evidence of precise null effects, including ruling out long-run impacts larger than −$391–$142 in annual earnings and −0.001–0.01 in annual convictions, with no corresponding payment increases despite salient and heterogeneous enforcement mechanisms. (JEL G51, H76, J31, K14, K41, K42)

The intergenerational effects of requiring unemployment benefit recipients to engage in non-search activities

We use a quasi-experimental design and national administrative data to analyze the intergenerational effects of introducing non-search activity requirements for long-term unemployment benefit recipients aged 18–34. The young-adult children of these recipients were in early adolescence when the requirements were introduced. Using a regression discontinuity approach, we find that young adults, particularly men, whose fathers were subject to the requirements have a lower incidence of unemployment benefit receipt compared to those whose fathers were not. More detailed investigation suggests completion of the mandated activities, role modeling, changes in attitudes, improved health, and greater support and stability as potential channels.

Impact of bidding zone re-configurations on electricity prices: Evidence from Sweden

In the European Union's electricity wholesale markets, energy is traded across large bidding zones, which, in most cases coincide with national borders: the so-called zonal markets. Within each market zone, energy flows are supposed to be free of transmission constraints. However, in some cases transmission constraints exist, implying an inefficient mismatch between demand and supply in different areas where the zonal market holds. Does transition to a more homogeneous multi-zonal market change equilibrium prices by facilitating a better balance between demand and supply when transmission constraints hold? We answer this question, by testing if the market re-configuration from one bidding zone to four zones, together with the commitment to not limit export with the neighboring countries, which took place in Sweden in November 2011, implied a change in prices. We perform a Regression Discontinuity in Time (RDiT) to compare change in prices in the four zones of Sweden. We find that, after November 2011, electricity prices increase across all zones with a stronger effect in SE3 and SE4 (southern Sweden).

Proliferation of religious associations and local incumbency advantage: evidence from municipal elections in Turkey

ABSTRACT For two decades, the Justice and Development Party (AKP) in Turkey sustained its electoral performance at the local level. We offer and test a new mechanism that contributed to the party's success in municipal elections. We argue that patronage links with religious civil society served as an important instrument of building and sustaining local incumbency advantage. Employing a close-election regression discontinuity design, we address two questions: To what extent did religious associations proliferate in AKP-controlled municipalities? Was this civil society penetration instrumental for the party’s electoral success? Employing a close-election regression discontinuity design (RDD), we show that in districts with a small-margin AKP victory in the 2014 elections, the number of religious associations in 2016 were significantly higher than in districts where AKP lost narrowly. We also find that the AKP incumbency in 2014 in closely contested districts increased the party’s vote share in the next local elections. Our results empirically demonstrate a new channel of local incumbency advantage. Moreover, we highlight the role of civil society as an agent of clientelistic exchange in competitive authoritarian regimes.

School Starting Age, Fertility, and Family Formation: Evidence From the School Entry Cutoff Using Exact Date of Birth.

In the past 50 years, the age at first birth in Norway and other European countries has shifted, leading to concerns that individuals begin childbearing too late to reach their intended family size. This article analyzes the effect of school starting age on fertility and family formation by utilizing Norway's age-based school entry policy. Using individual-level register data and a regression discontinuity design, we find that being born after the age cutoff for school start results in an increased age at first birth of 2.9 months for women and 4.0 months for men, whereas completed cohort fertility was unchanged. Similarly, being born after the cutoff increased the age at first marriage by 4.7 months for women and 2.4 months for men, with no effect on the overall probability of having a partner. Results show that age at completed education and earnings development are important mechanisms in this fertility postponement. Additionally, we analyze detailed age- and parity-specific effects, providing important insights into how age at starting school affects fertility timing but not overall fertility.

Does household electrification alleviate energy poverty? Empirical evidence from China

In the context of carbon neutrality goals, mitigating energy poverty presents a certain level of complexity. This study employs Prospect Theory to analyze the impact of household electrification on energy poverty. It tests the non-linear relationship between household electrification and energy poverty using Oprobit models, Robin Hood algorithm and Regression Discontinuity Design. Besides, the mediating effect of energy efficiency and employment rate, and the moderating effect of family cohabitation are verified. The results show that there is an inverted U-shaped relationship between household electrification and energy poverty. Furthermore, urban households and those using old furniture experience a stronger increase in energy poverty in the early stages of electrification, while households in areas with long heating degree days experience an increase in energy poverty in the later stages of electrification. Household electrification impacts on energy poverty through energy efficiency and alleviates energy poverty by promote employment rate. Additionally, the higher the degree of family cohabitation, the more it can alleviate the exacerbating effect of household electrification on energy poverty.

The Effect of Blood Lipid Profiles on Chronic Kidney Disease in a Prospective Cohort: Based on a Regression Discontinuity Design

ObjectivePrevious studies on the association between lipid profiles and chronic kidney disease (CKD) have yielded inconsistent results and no defined thresholds for blood lipids. MethodsA prospective cohort study including 32,351 subjects who completed baseline and follow-up surveys over 5 years was conducted. Restricted cubic splines and Cox models were used to examine the association between the lipid profiles and CKD. A regression discontinuity design was used to determine the cutoff value of lipid profiles that was significantly associated with increased the risk of CKD. ResultsOver a median follow-up time of 2.2 (0.5, 4.2) years, 648 (2.00%) subjects developed CKD. The lipid profiles that were significantly and linearly related to CKD included total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), TC/HDL-C, and TG/HDL-C, whereas lowdensity lipoprotein cholesterol (LDL-C) and LDL-C/HDL-C were nonlinearly correlated with CKD. TC, TG, TC/HDL-C, and TG/HDL-C showed an upward jump at the cutoff value, increasing the risk of CKD by 0.90%, 1.50%, 2.30%, and 1.60%, respectively, whereas HDL-C showed a downward jump at the cutoff value, reducing this risk by 1.0%. Female and participants with dyslipidemia had a higher risk of CKD, while the cutoff values for the different characteristics of the population were different. ConclusionThere was a significant association between lipid profiles and CKD in a prospective cohort from Northwest China, while TG, TC/HDL-C, and TG/HDL-C showed a stronger risk association. The specific cutoff values of lipid profiles may provide a clinical reference for screening or diagnosing CKD risk.

Place-Based Policies and Migration: Evidence from India

Abstract Place-based policies may increase migration into eligible areas and, hence, affect their intended aim of reducing economic disparities. Using a regression discontinuity design that exploits an Indian place-based tax-exemption policy, this paper finds that there was an increase in migration for job-related reasons, mainly driven by job transfers. While firm entry and hiring increased in the treated districts, overall employment probability and wages did not change for the average initial resident, although there is some evidence of sectoral shifts in employment. Furthermore, the cost of housing increased in the treated districts. These results provide empirical evidence for the implications of labor mobility for place-based policies.

Can a Driving Restriction Policy Improve Air Quality? Empirical Evidence from Chengdu

Automotive exhaust emissions contribute significantly to air pollution in developing countries. However, the effectiveness of driving restriction policies (DRPs) is unclear, and most research on China emphasizes Beijing. This study used Chengdu, a typical large city in China, to examine the impact of a DRP on air quality. To alleviate potential endogeneity threats, we employed a regression discontinuity design to verify the policy’s effect. The results show that the DRP significantly reduced air pollution levels, effectively improving air quality in restricted areas. The heterogeneity analysis found that (1) the DRP effectively reduced pollution in newly added and original areas, while the air quality in adjacent areas deteriorated; and (2) the DRP significantly improved air quality during peak travel periods but had no significant impact in other periods. Our results indicate that the DRP is an effective tool for urban environmental governance but presents potential negative aspects. Therefore, restricted areas and periods should be carefully considered when designing similar policies. This study provides significant insights into the governance of automotive exhaust emissions pollution for large cities in developing countries.

Transient institutional ownership, costly external finance and corporate cash holdings

AbstractWe investigate and robustly show that transient institutional ownership (IO) has a positive effect on the level and value of corporate cash holdings. Further, using a regression discontinuity design exploiting the Russell 1000/2000 index reconstitution as an exogenous shock to transient IO, we show that the effects of transient IO on cash holdings are causal. Additionally, our analysis shows that transient institutions exacerbate debtholder–shareholder conflicts, thereby increasing the cost of debt. Overall, our results suggest that transient institutions make cash holdings more valuable because financing by debt becomes more costly.

Is Mandatory Community Service a Catalyst for Civic Growth or Fruitless Servitude?: Evidence from a Natural Experiment in South Korea

AbstractDiminishing civic vitality has been reported in numerous societies, irrespective of democratic maturity. Mandatory community service initiatives in schools have garnered attention as a strategy for fostering long-term civic engagement. However, methodological challenges such as selection bias and observation timing have led to inconsistent empirical results. This study leverages a unique natural experimental context where mandatory community service requirements were nationally implemented in middle and high schools in South Korea. Employing a non-parametric regression-discontinuity design, we provide robust evidence that such programs increase the likelihood of volunteering in adulthood among lower socioeconomic groups. We found no indications that these mandates increase or decrease the propensity for civic engagement among higher socioeconomic groups.

How COVID-19 has Impacted American Attitudes Toward China: A Study on Twitter

ABSTRACT Past research has studied social determinants of attitudes toward foreign countries. Confounded by potential endogeneity biases due to unobserved factors or reverse causality, the causal impact of these factors on public opinion is usually difficult to establish. Using social media data, we leverage the suddenness of the COVID-19 pandemic to examine whether a major global event has causally changed Americans’ views of another country. We collate a database of 297 million posts on the social media platform Twitter about China or COVID-19 up to June 2020, and we take tweeting about COVID-19 as a proxy for an individual’s awareness of COVID-19. Using regression discontinuity and difference-in-difference estimation, we reveal that awareness of COVID-19 causes a sharp rise in anti-China attitudes. Our work has implications for understanding how self-interest affects policy preference and how Americans view foreign countries.

Governmental role conflicts and corporate overinvestment: Evidence from a regression discontinuity design

Governmental role conflicts and corporate overinvestment: Evidence from a regression discontinuity design

Infrastructures of race? Colonial indigenous segregation and contemporary land values

We investigate the persistent impact of a colonial segregation policy on land values in modern Mexico City. During colonial times, Indigenous communities were confined, with varying degrees of success, to settlements known as pueblos de indios. Using historical records, we exploit quasi-random variation due to the pueblos’ catchment areas and use a Regression Discontinuity Design to estimate the causal effects of pueblos on land prices. We find a 5% land value penalty for areas affected by the colonial policy. The penalty is exacerbated for the pueblos formerly inhabited exclusively by Indigenous populations. Historical evidence and novel digitized maps reveal that these land value penalties have been driven over the past two centuries by low public goods provision, negative economic expectations, and the historical sorting of working-class individuals who built small housing structures that are second-nature factors. Moreover, in contemporary data, we observe discontinuities in housing overcrowding and public goods quality within the pueblos’ catchment areas. Our results underscore the repercussions of colonial policies on contemporary spatial equilibria, clarifying the mechanisms driving historical persistence and offering implications for urban policies.

COVID-19 Incidence and Age Eligibility for Elementary School

Understanding the role of school attendance on transmission of SARS-CoV-2 among children is of importance for responding to future epidemics. Estimating discontinuities in outcomes by age of eligibility for school attendance has been used to examine associations between school attendance and a variety of outcomes, but has yet to be applied to describe associations between school attendance and communicable disease transmission. To estimate the association between eligibility for elementary school and COVID-19 incidence. This case series used data on all pediatric COVID-19 cases reported to California's disease surveillance system between May 16, 2020, and December 15, 2022, among children within 24 months of the age threshold for school eligibility. Birthdate before or after the age threshold for elementary school eligibility during periods when school was remote vs in person. COVID-19 cases and hospitalizations. Between May 16, 2020, and December 15, 2022, there were 688 278 cases of COVID-19 (348 957 cases [50.7%] among boys) and 1423 hospitalizations among children who turned 5 years within 24 months of September 1 of the school year when their infection occurred. The mean (SD) age of the study sample was 5.0 (1.3) years. After adjusting for higher rates of testing in schooled populations, the estimated pooled incidence rate ratio among kindergarten-eligible individuals (eg, those born just before the age threshold for school eligibility) compared with those born just after the eligibility threshold for in-person fall 2021 semester was 1.52 (95% CI, 1.36-1.68), for in-person spring 2022 semester was 1.26 (95% CI, 1.15-1.39), and for in-person fall 2022 semester was 1.19 (95% CI, 1.03-1.38). Reported incidence rates among school-eligible children remained higher during the month-long winter 2021-2022 school break but were lower during the longer summer break that followed. The findings were unable to establish whether associations between school eligibility and COVID-19 incidence were based on in-school vs out-of-school routes (eg, classrooms vs school buses). The study lacked power to detect associations between school attendance and hospitalization. Results were robust to functional form. A simulation study was conducted to demonstrate bias associated with nonadjustment for differential case acquisition by exposure status. In this case series of children in California, the magnitude of the association between school eligibility and COVID-19 incidence decreased over time and was generally lower than other published associations between out-of-school child social interactions and COVID-19 incidence. This regression discontinuity design approach could be adapted to other geographies and/or disease systems to assess associations between schooling and disease transmission.

The price of attention: an analysis of the intersection of media coverage and public sentiments about eggs and egg prices.

Retail prices for eggs surged during the period from early 2022 to mid-2023 in the U.S. We investigate the impact of egg price fluctuations on online and social media discourse, analyzing the relationship between egg prices and public sentiment. Utilizing online and social media listening data from September 2019 to August 2023, we explore how the total number of statements (i.e., mentions) and sentiment respond to changes in egg prices in the U.S. We find a significant association between increases in egg prices and both the volume of mentions and the sentiment of online discussions. Notably, mentions increased and sentiment became more negative as egg prices rose, highlighting a clear public response to price changes. However, the relationship between egg prices and online and social media attention is complex, which becomes apparent when studying the timing of increased concern with the timing of online news media coverage. Importantly, using regression discontinuity in time, we show that online and social media conversations about eggs and egg prices tend to increase after the rapid rise in online news coverage. Similarly, online and social media conversations about eggs and egg prices tend to decrease after the rapid rise in online news coverage.