Parallel Trends Assumption Research Articles

Promoting corporate financing efficiency through digital finance and sustainable practices

ABSTRACT This paper explores the impact of digital finance (DFI) and environmental performance (EP) on corporate financing efficiency (FE) in China's property sector. The research addresses two main questions: How do digital finance and environmental performance influence corporate financing efficiency? What are the policy implications of these influences for enhancing green transformation in the sector? Employing robust techniques, including fixed-effect regression and a difference-in-differences (DID) approach. The study reveals that DFI and EP have a significant positive effect on FE. The analysis demonstrates that higher levels of DFI and improved EP lead to more effective allocation and utilization of financial resources. The DID results further indicate that an increase in DFI contributes to a significant enhancement in FE with a one-year lag, without violating the parallel trends assumption. The study shows that integrating digital finance and environmental sustainability improves financial efficiency. It urges policymakers to promote investments in digital financial technologies.

Women improving nutrition through self-help groups in India: Does nutrition information help?

Women’s self-help groups (SHGs) are an important platform for reaching poor women in India. Despite SHGs' women-focused programming, evidence of the impact of SHG-based interventions on nutrition outcomes is limited, and most evaluations of nutrition interventions have not examined intermediate outcomes along the impact pathways or outcomes for women themselves. This paper evaluates the effectiveness of an integrated agriculture-nutrition intervention delivered through women’s SHGs in five states in central and eastern India. The interventions involved the delivery of nutrition behavior change communication to groups through participatory approaches, community engagement around key issues, and the strengthening of collective organizations. Our analysis is based on three rounds of rich panel data on close to 2700 rural women and their households from eight districts in these five states and qualitative work from an accompanying process evaluation. Using difference-in-difference models with nearest neighbor matching methods, we present results on women’s anthropometry and diet-related outcomes.We do not observe any improvements in women’s BMI or overall dietary diversity. Although more women in the nutrition intensification arm consumed animal source foods, nuts and seeds, and fruits, this was not enough to increase overall dietary diversity scores or the proportion of women achieving minimum dietary diversity. We measure intermediate outcomes along the program’s impact pathways and find improvements in household incomes, cultivation of home gardens, and utilization of government schemes but not in women’s empowerment. The lack of improvement in anthropometry and diets despite changes in some intermediate outcomes can be attributed to several factors such as low implementation intensity, poor facilitator capacity and incentives, the lack of relevance of the BCC topics to the average SHG member, and resource and agency constraints to adoption of recommended practices. Although we do not have data to test the parallel trends assumption and so do not interpret our results as causal, these findings do suggest that optimism about using group-based platforms needs to be tempered in resource-poor contexts.

공무원 채용시험 과목의 변화가 공공봉사동기에 미치는 영향: 이중차분법을 중심으로

In 2017, the subject ‘Constitution' was added to the Open Competitive Examination for Grade 5 public officials to strengthen the assessment of public service values among applicants. This study examines the impact of this new subject on Public Service Motivation (PSM) among newly recruited Grade 5 public officials. Survey data spanning from 2013 to 2023 from the Korea Institute of Public Administration was utilized to measure PSM. The study employed the Difference-in-Differences (DID) method, a quasi-experimental design technique, to enhance causal inference. The treatment group consisted of new Grade 5 public officials, while the control group comprised new Grade 7 public officials. Visual and statistical tests confirmed the parallel trend assumption. Despite a decade-long decline in PSM scores for both groups, the introduction of the 'Constitution' subject resulted in a 0.127-point increase in PSM scores among new Grade 5 public officials. This suggests that the changes in examination subjects effectively achieved their intended objective by selecting individuals with high public service motivation. Unlike other countries with intense competition for public official positions, Korea did not experience adverse selection issues, where individuals with low PSM were recruited.

Market uncertainties and too-big-to-fail perception: Evidence from Chinese P2P registration requirements

The enforcement of peer-to-peer (P2P) registration requirements in mid-2018 triggered a P2P market meltdown, highlighting the inherent challenge faced by Chinese market participants in distinguishing between genuine and fraudulent fintech firms. The difference-in-difference results suggest that the too-big-to-fail (TBTF) perception can effectively halve investor outflows and borrower outflows during periods of uncertainty. Dynamic analysis further validates the parallel-trend assumption and underscores the persistent influence of TBTF perception. Moreover, the empirical findings suggest that, in the face of a market downturn, fintech market participants become unresponsive to all other certification mechanisms, including venture capital participation, custodian banks, and third-party guarantees.

The impact of patient registration on utilisation and quality of care: a propensity score matching and staggered difference-in-differences analysis of a cohort of 16,775 people with type 2 diabetes

BackgroundIn 2012, Luxembourg introduced a Referring Doctor (RD) policy, whereby patients voluntarily register with a primary care practitioner, who coordinates patients’ health care and ensures optimal follow-up. We contribute to the limited evidence base on patient registration by evaluating the effects of the RD policy.MethodsWe used data on 16,775 people with type 2 diabetes on oral medication (PWT2D), enrolled with the Luxembourg National Fund from 2010 to 2018. We examined the utilisation of primary and specialist outpatient care, quality of care process indicators, and reimbursed prescribed medicines over the short- (until 2015) and medium-term (until 2018). We used propensity score matching to identify comparable groups of patients with and without an RD. We applied difference-in-differences methods that accounted for patients’ registration with an RD in different years.ResultsThere was low enrolment of PWT2D in the RD programme. The differences-in-differences parallel trends assumption was not met for: general practitioner (GP) consultations, GP home visits (medium-term), HbA1c test (short-term), complete cholesterol test (short-term), kidney function (urine) test (short-term), and the number of repeat prescribed cardiovascular system medicines (short-term). There was a statistically significant increase in the number of: HbA1c tests (medium-term: 0.09 (95% CI: 0.01 to 0.18)); kidney function (blood) tests in the short- (0.10 (95% CI: 0.01 to 0.19)) and medium-term (0.11 (95% CI: 0.03 to 0.20)); kidney function (urine) tests (medium-term: 0.06 (95% CI: 0.02 to 0.10)); repeat prescribed medicines in the short- (0.19 (95% CI: 0.03 to 0.36)) and medium-term (0.18 (95% CI: 0.02 to 0.34)); and repeat prescribed cardiovascular system medicines (medium-term: 0.08 (95% CI: 0.01 to 0.15)). Sensitivity analyses also revealed increases in kidney function (urine) tests (short-term: 0.07 (95% CI: 0.03 to 0.11)) and dental consultations (short-term: 0.06, 95% CI: 0.00 to 0.11), and decreases in specialist consultations (short-term: -0.28, 95% CI: -0.51 to -0.04; medium-term: -0.26, 95% CI: -0.49 to -0.03).ConclusionsThe RD programme had a limited effect on care quality indicators and reimbursed prescribed medicines for PWT2D. Future research should extend the analysis beyond this cohort and explore data linkage to include clinical outcomes and socio-economic characteristics.

Advances in Difference-in-differences Methods for Policy Evaluation Research.

Difference-in-differences (DiD) is a powerful, quasi-experimental research design widely used in longitudinal policy evaluations with health outcomes. However, DiD designs face several challenges to ensuring reliable causal inference, such as when policy settings are more complex. Recent economics literature has revealed that DiD estimators may exhibit bias when heterogeneous treatment effects, a common consequence of staggered policy implementation, are present. To deepen our understanding of these advancements in epidemiology, in this methodologic primer, we start by presenting an overview of DiD methods. We then summarize fundamental problems associated with DiD designs with heterogeneous treatment effects and provide guidance on recently proposed heterogeneity-robust DiD estimators, which are increasingly being implemented by epidemiologists. We also extend the discussion to violations of the parallel trends assumption, which has received less attention. Last, we present results from a simulation study that compares the performance of several DiD estimators under different scenarios to enhance understanding and application of these methods.

State mandatory paid medical leave policies and cancer stage at diagnosis in US adults.

11008 Background: Since 2012, several states implemented mandatory paid medical leave policies. Mandatory paid medical leave may alleviate time and financial burdens, enabling patients to undergo screening and/or timely diagnostic workup for new cancer-related symptoms. These policies have been associated with increased cancer screening. We investigated whether state mandatory paid medical leave policies were associated with stage at cancer diagnosis. Methods: Adults ages 18-64 years diagnosed with cancer from 2010-2019 were identified from the Surveillance, Epidemiology, and End Results program. Differences-in-differences (DID) analyses using the updated method of accounting for heterogeneity in policy enactment compared proportions of stage I and IV cancer diagnoses from pre- to post-policy implementation in states with vs. without paid medical leave policies. Analyses adjusted for age, sex, race, ethnicity, rurality, cancer site, county-level income, and time-varying state Medicaid eligibility levels. Since some cities enacted city-wide policies prior to state-wide implementation, cases from affected states between the time of city and state policy enactment were excluded. Sensitivity analyses were conducted excluding counties with cities that enacted paid leave policies. The plausibility of the parallel trends assumption was assessed by comparing stage at diagnosis between state groups in the pre-policy period and was satisfied for all reported results. Results: A total of 1,052,307 adults with cancer were identified. 39.8% and 19.7% of patients were diagnosed with stage I and IV cancers, respectively. In adjusted DID analyses, there was a 1.34 percentage point (PP) increase in stage I diagnoses and a 2.05 PP decrease in stage IV diagnoses associated with mandatory paid medical leave policy implementation (Table). Results were similar in sensitivity analyses. In subgroup analyses, policy-associated stage shifts were seen in males, Hodgkin lymphoma, and liver and lung cancers, with less clear patterns for cervix and pancreas cancers. Conclusions: State mandatory paid medical leave was associated with increased early-stage cancer diagnoses and decreased late-stage diagnoses. Paid leave policies may promote earlier detection of non-screenable cancers by enabling employees to promptly seek care for new cancer-related symptoms. Policies that improve opportunities to seek medical care may facilitate earlier cancer detection and ultimately contribute to better patient outcomes. [Table: see text]

Has the Establishment of National Key Ecological Function Zones Improved Eco-Environmental Quality?—Evidence from a Quasi-Natural Experiment in 130 Counties in Sichuan Province, China

China’s National Key Ecological Function Zones (NKEFZs) currently represent the largest and most extensive ecological conservation policy in China, with one of the core objectives of this policy being to improve eco-environmental quality (EEQ). This study regards the establishment of NKEFZs as a quasi-natural experiment. Based on panel data from 130 counties in Sichuan Province from 2001 to 2021, a multi-period difference-in-differences (DID) model was employed to evaluate the impact of NKEFZ establishment on EEQ. The findings indicate the following: ① The establishment of NKEFZs can significantly enhance the EEQ of the covered areas, albeit as a gradual long-term process. This conclusion not only meets the parallel-trends assumption but also holds true in a series of robustness tests such as placebo tests. ② Mechanism analysis reveals that NKEFZs can enhance EEQ through the effects of optimizing land spatial allocation and upgrading industrial structure. ③ Heterogeneity analysis demonstrates that the beneficial effect of NKEFZs on EEQ varies across different functional zone types, geographic spaces and ethnic regions. Our study not only contributes to the accumulation of empirical evidence and institutional refinement in the sustainable implementation of ecological policies in China but also offers valuable insights and references for other countries in formulating policies for eco-environmental protection.

Do studies evaluating early-life policy interventions fully adhere to the critical conditions of difference-in-differences? A systematic review

ObjectivesTo assess the reporting and methodological quality of early-life policy intervention papers that applied difference-in-differences (DiD) analysis.Study designSystematic review.Data sourcesPapers applying DiD of early-life policy interventions in high-income countries as...

Difference-in-Differences Estimators of Intertemporal Treatment Effects

Abstract We study treatment-effect estimation using panel data. The treatment may be non-binary, non-absorbing, and the outcome may be affected by treatment lags. We make a parallel-trends assumption, and propose event-study estimators of the effect of being exposed to a weakly higher treatment dose for ℓ. periods. We also propose normalized estimators, that estimate a weighted average of the effects of the current treatment and its lags. We also analyze commonly-used two-way-fixed-effects regressions. Unlike our estimators, they can be biased in the presence of heterogeneous treatment effects. A local-projection version of those regressions is biased even with homogeneous effects.

Testing for Equivalence of Pre-Trends in Difference-in-Differences Estimation

The plausibility of the “parallel trends assumption” in Difference-in-Differences estimation is usually assessed by a test of the null hypothesis that the difference between the average outcomes of both groups is constant over time before the treatment. However, failure to reject the null hypothesis does not imply the absence of differences in time trends between both groups. We provide equivalence tests that allow researchers to find evidence in favor of the parallel trends assumption and thus increase the credibility of their treatment effect estimates. While we motivate our tests in the standard two-way fixed effects model, we discuss simple extensions to settings in which treatment adoption is staggered over time.

Простой и сложный метод разности разностей

The paper presents extensions of the popular difference‐in‐differences approach (DD) from 2×2 design on multiple time‐period, multiple groups, fuzzy DD, non‐staggered treatment and approaches to measure distributional treatment effect. The paper describes assumptions for consistent estimation of the treatment effect by two‐way fixed effects model (TWFE) and presents the problem leading to inconsistent estimates justifying the application of alternative estimators. The paper briefly introduces methods developing DD for multiple‐period multiple‐group cases based on TWFE and alternative approaches. The proposed techniques allow treatment evaluation in the frame of DD when canonical TWFE leads to inconsistent estimates. Some approaches allow replacement of the well‐known parallel trend assumption (PTA) for a conditional PTA or time randomisation. The paper refers to implementations of these methods in Stata and R. Simulation modelling demonstrates that the stated properties of the alternative estimators are not always reliable.

Employee stock ownership plans and within-firm income inequality: evidence from China

Purpose Drawing on bargaining power and the inequality aversion perspective, this study aims to probe employees’ influence on addressing income inequality between top executives and nonexecutive employees. Meanwhile, it examines the moderating role of employee-related factors and plan attributes. Design/methodology/approach This study uses a staggered difference-in-differences design with a propensity scoring match approach and verification of the parallel trend assumption to test the hypotheses. Findings The results support the hypothesis that employee stock ownership plans (ESOPs) significantly reduce within-firm income inequality. The negative effect is amplified by both the presence of trade unions and the unemployment rate at the regional level, as well as the duration of the lock-in period and the scale of participants within the stock ownership plan. Practical implications This study has implications for income inequality research and ESOP design and provides theoretical support for policymakers and corporate governance. Originality/value This study contributes to the literature on income inequality by examining the implementation of ESOPs from the employee perspective. Furthermore, it extends the current literature by investigating the strengthening effects of regional factors and ESOP attributes on the relationship between ESOPs and income inequality. The conclusions provide new empirical evidence to promote the effective implementation of ESOPs by combining internal and external factors.

Can environmental information disclosure spur corporate green innovation?

How can the disclosure of environmental information (EID) stimulate corporate green innovation (CGI)? This research challenges the prevailing assumption that environmental regulations impact CGI by influencing corporate compliance costs. Instead, it offers a fresh theoretical framework to explain how EID affects CGI. This study combines signal theory and resource dependence theory to develop a moderated mediation model, illustrating how EID reduces information asymmetry and alleviates corporate financial constraints (CFC). To test these hypotheses, this study utilized data from A-share listed companies spanning the period 2004 to 2017. This study considered the year 2009 as a crucial point of analysis, marking the period before and after the implementation of China's first EID policy in 2008. This study employed a Difference-in-Differences (DID) model. The results reveal that EID has a positive impact on CGI by mitigating CFC, with non-state-owned enterprises (non-SOEs) exhibiting a more pronounced mediating effect. These findings remain robust even when the parallel trend assumption was tested to eliminate interference from other factors. This study unveils the mechanism through which voluntary environmental regulation, represented by EID, influences CGI by mitigating information asymmetry and alleviating CFC. These results deviate from the predictions of compliance cost theory and Porter's hypothesis regarding the impact of traditional environmental regulations on CGI, providing a fresh perspective on the role of voluntary environmental regulation in driving CGI.

Testing Parallel Trends in Differences-in-Differences and Event Study Designs: A Research Approach Based on Pre-Treatment Period Significance

Traditional tests for parallel trends in the context of differences-in-differences are based on the observation of the mean values of the dependent variable in the treatment and control groups over time. However, given the new discussions brought by the development of the event study designs, controlling for observable factors may intervene in the fulfilment of the parallel trend assumption. This article presents a simple test based on the statistical significance of pre-treatment periods which can be extended from the classic Differences-in-Differences up to event study designs in universal absorbing treatments. The test requires at least two pre-treatment periods and can done by constructing appropriate dummy variables.

Universal Difference-in-Differences for Causal Inference in Epidemiology.

Difference-in-differences is undoubtedly one of the most widely used methods for evaluating the causal effect of an intervention in observational (i.e., nonrandomized) settings. The approach is typically used when pre- and postexposure outcome measurements are available, and one can reasonably assume that the association of the unobserved confounder with the outcome has the same absolute magnitude in the two exposure arms and is constant over time; a so-called parallel trends assumption. The parallel trends assumption may not be credible in many practical settings, for example, if the outcome is binary, a count, or polytomous, as well as when an uncontrolled confounder exhibits nonadditive effects on the distribution of the outcome, even if such effects are constant over time. We introduce an alternative approach that replaces the parallel trends assumption with an odds ratio equi-confounding assumption under which an association between treatment and the potential outcome under no treatment is identified with a well-specified generalized linear model relating the pre-exposure outcome and the exposure. Because the proposed method identifies any causal effect that is conceivably identified in the absence of confounding bias, including nonlinear effects such as quantile treatment effects, the approach is aptly called universal difference-in-differences. We describe and illustrate both fully parametric and more robust semiparametric universal difference-in-differences estimators in a real-world application concerning the causal effects of a Zika virus outbreak on birth rate in Brazil. A supplementary digital video is available at: http://links.lww.com/EDE/C90.

How did unmet care needs during the pandemic affect health outcomes of older European individuals?

The first wave of the COVID-19 pandemic left many people with unmet health care needs, which could have detrimental effects on their health. This paper examines the effects of these unmet needs during the first wave of the pandemic on health outcomes one year later. We combine two waves of the SHARE survey collected during the COVID-19 pandemic (in June/July 2020 and 2021), as well as four waves collected before the pandemic. Our health outcomes are four dummy variables: fatigue, falling, fear of falling and dizziness/faints/blackouts issues. Finally, we use OLS regression with individual and time fixed effects for our difference-in-difference analysis, as well as a doubly robust estimator to condition the parallel trend assumption on pre-pandemic covariates. We find substantial effects of having had unmet healthcare needs during 2020 on the probability of having trouble with fatigue and fear of falling one year later. We particularly find strong effects for general practitioner (GP) and specialist care, and in lower extent of physiotherapist, psychotherapist, and rehabilitation care.

Association of state policies modifying Supplemental Nutrition Assistance Program (SNAP) eligibility and cancer stage at diagnosis.

214 Background: Food insecurity is associated with access to care and is particularly prevalent among adolescents and young adults (AYA) with cancer. The Supplemental Nutrition Assistance Program (SNAP) was designed to minimize food insecurity for low-income households. Under the broad-based categorial eligibility (BBCE) policy, states were given the ability to relax requirements for qualifying for SNAP through increasing the income limit or eliminating the asset test. While most SNAP policy changes under the BBCE policy occurred from 2009-2011, it is unknown whether these policies affect access to care among AYA with cancer. We sought to determine whether state SNAP eligibility policies were associated with stage at diagnosis among AYA patients. Methods: We utilized the Surveillance, Epidemiology, and End Results database to obtain data from 2006-2019 for newly diagnosed patients with cancer ages 18-39 years from 11 states. All cancer types were included with the exception of liquid tumors and other unstaged cancers. Difference-in-differences analyses were utilized to evaluate changes in the percentage of localized and distant stage diagnoses from pre- to post-policy enactment between states that enacted (1) asset test elimination or (2) increased income limit policies. Linear probability models were adjusted for state and year fixed effects, race, ethnicity, sex, metropolitan residence, marital status, county-level income, cancer type, and Medicaid expansion effects. We also accounted for state-level clustering via clustered standard errors. Event study analyses were utilized to assess the plausibility of the parallel trends assumption. Results: A total of 157,380 AYAs with cancer were identified. In adjusted difference-in-differences analyses, there was a 0.67 percentage point (PP) (95% CI = 0.15 to 1.19, P=.012) increase in localized stage cancer diagnoses and a -0.81 PP (95% CI = -1.41 to -0.21, P=.008) decrease in distant stage cancer diagnoses in states with increased income limits for SNAP eligibility relative to other states. There were no statistically significant changes in the percentage of localized (0.87 PP, 95% CI = -1.27 to 3.00, P=.43) or distant (0.05 PP, 95% CI = -1.44 to 1.55, P=.94) stage cancers with policies eliminating the asset test. Among patients with cervical cancer, where screening is recommended in this age group, there were extremely similar patterns. Conclusions: State policies increasing income eligibility limits for the Supplemental Nutrition Assistance Program were associated with a higher percentage of localized stage and a lower percentage of distant stage cancers among adolescents and young adults. These data add to a growing body of literature suggesting that policies that address social determinants of health may improve access to care.

Exploring the effect of emission trading system on marginal abatement cost-based on the frontier synthetic difference-in-differences model

Evaluating the cost-effectiveness of emission trading system (ETS) will explore how to control the total emission amount while minimizing the economic cost. Most related studies employ the difference-in-differences (DID) and multi-period DID models. However, this research highlights the superiority of construction technique and estimation results in the synthetic DID model, and applies it to a new scenario for evaluating the cost-effectiveness of China's pilot ETS. From the construction technique, the synthetic DID model synthesizes more desirable control groups and allows more flexibility in matching treatment units with extreme values by setting weight combinations and intercept terms. From the estimation results, the treatment effects are considered more reliable because ideally satisfy the parallel trend assumption. Specifically, the synthetic DID model indicated that the pilot ETS effectively reduced the marginal abatement cost (MAC) by 24.39%, whereas the traditional DID and multi-period DID models may have overestimated the policy effects by 98.48% and 71.09%. The main reason was the MAC has shown a downward trend before the policy implementation. Moreover, the estimation results of heterogeneity analysis and mechanism test in different models were also different. The synthetic DID model integrates the strengths of the DID model and synthetic control method (SCM), making a methodologically grounded contribution to the existing policy evaluation system.

Mechanisms of Forestry Carbon Sink Policies on Land Use Efficiency: A Perspective from the Drivers of Policy Implementation

Rapid urbanization has brought economic dividends to China, but it has been accompanied by inefficient land use. Meanwhile, the mechanism of forestry carbon sinks (FCSs) on land use efficiency (LUE) has not been sufficiently discussed in the context of the pursuit of “carbon neutrality” around the world. Based on the idea of the benefit–cost theory, this study investigated the impact of FCSs on LUE in 30 provincial-level regions (2006–2019) in China using the difference-in-difference model. The results showed that, first, via the mechanisms of public opinion (PO) and rewards and penalties (RP), FCSs could significantly improve the LUE in the regions, and that the former had a greater effect than the latter; second, the tests of the assumption of parallel trends showed that FCSs had a slower effect on the LUE under the PO mechanism than under the RP mechanism; third, the analysis of the LUE showed that the improvement in LUE mainly occurred in the eastern, central, and southwestern regions of China. The conclusions were as follows: (1) FCS is able to promote LUE via both the PO and RP mechanisms; (2) there is a lag in the promotion of LUE by FCS, and the lag is larger with the PO mechanism; and (3) there is spatial clustering in the promotion of LUE by FCS. In line with these conclusions, we propose policy recommendations to better exploit the policy effects of FCSs in three aspects, namely promoting the development of forestry carbon sink projects, improving the relevant mechanisms of FCSs, and improving the mechanisms of PO and RP.