Instrumental Variable Regression Research Articles

The Impact of Social Investment and Green Finance on Sustainable Development: Evidence From Emerging Market Economies

ABSTRACTThis study examines how social investment and green finance affect sustainable development (SDE) in 16 emerging market economies from 2014 to 2022. Truncated regression, Prais‐Winsten regression, and single‐equation instrumental variables (IV) regression models were applied for the empirical analysis. The findings reveal that social investment and green finance have positive effects, whereas poverty and ecological footprint negatively affect SDE. Social investment reduces the negative impact of poverty on SDE by enhancing human development and reducing economic vulnerability, confirming the moderation impact. The Juodis‐Karavias‐Sarafidis (JKS) panel's non‐causality analysis confirms the causal relationship between variables. This study recommends that emerging markets allocate more resources for social and green initiatives and limit their consumption of natural resources to avoid the negative consequences of a high ecological footprint. This study implies that doing and going green is essential for meeting sustainability challenges in the post‐modern era in emerging market economies.

Multivariate proteome-wide association study to identify causal proteins for Alzheimer disease.

Alzheimer disease (AD) is a complex and progressive neurodegenerative disorder that accounts for the majority of individuals with dementia. Here, we aim to identify causal plasma proteins for AD, shedding light on the etiology of AD. We utilized the latest large-scale plasma proteomic data from the UK Biobank Pharma Proteomics Project (UKB-PPP) and AD genome-wide association study (GWAS) summary data from the International Genomics of Alzheimer's Project (IGAP). Via a robust univariate instrumental variable (IV) regression method, we identified causal proteins through cis-protein quantitative trait loci (pQTLs) and (both cis- and trans-)pQTLs. To further reduce potential false positives due to high linkage disequilibrium (LD) of some pQTLs and high correlations among some proteins, we developed a robust multivariate IV regression method, called two-stage constrained maximum likelihood (MV-2ScML), to distinguish direct and confounding/mediating effects of proteins; some key features of the method include its robustness to invalid IVs and applicability to GWAS summary data. Our work highlights some differences between using cis-pQTLs and trans-pQTLs and critical values of multivariate analysis for fine-mapping causal proteins, providing insights into plasma protein pathways to AD.

The Impact of ICT on Regional Economy in Indonesia Through MSEs as Mediators: Application of Causal Mediation Analysis in Instrumental-variable Regressions

Introduction/Main Objectives: The development of information and communication technology (ICT) and small and micro enterprises (SMEs) can encourage regional economic growth. Background Problems: Studies on the impact of ICT on the rural economy at the village level are very limited. Furthermore, the Indonesian study neglects to tackle the endogeneity issue associated with this variable and the indirect effects of ICT on the regional economy. Novelty: Using SMEs as a mediator, this study examines the impact of ICT (internet signal strength) on the village's local economy (nighttime light), both directly and indirectly. ICT is considered to be endogenous. Research Methods: This study employs causal mediation analysis in instrumental-variable (IV) regressions at the village level in 2018 and 2021, using lightning strike intensity as IV. Finding/Results: Internet signal strength can increase the number of SMEs, and this increase can positively and significantly impact the local economy. In addition, the direct impact of internet signal strength on the local economy is significantly negative. However, the total impact of internet signal strength is significantly positive.

Tourism, industrialization, and sustainable economic growth

<p>The coordination between tourism and other industries contributes to the tourism sustainability as well as the sustainable economic growth. This study comprehensively examines the effect of the relationship between tourism development and industrialization on sustainable economic growth based on provincial data of China. By using the coupling coordination degree model and instrumental variable regression method, we found the following: (1) Tourism-industrialization coordination exerts positive and significant effects on sustainable economic growth. Additional robustness checks show that the results are reliable. (2) Economic development, tourism development, and manufacturing innovation moderate the effect of tourism-industrialization coordination on sustainable economic growth. (3) The sustainable economic growth effect of tourism is nonlinear and affected by tourism-industrialization coordination, and industrialization is important for tourism-led sustainable growth. The findings broaden the understanding of the relationship between tourism and economic sustainability from the perspective of industrial coordination. It may contribute to building a sustainable economic development path in developing countries.<strong></strong></p>

The impact of export diversification on business startups in Africa

This paper aims to examine the impact of export diversification on business formation in Africa. The study used a panel dataset comprising 10 African countries (Algeria, Gabon, Lesotho, Morocco, Nigeria, Sierra Leone, Togo, Tunisia, Uganda and Zambia) observed annually from 2008 to 2018. The model was estimated using an instrumental variable regression approach within a count data framework and the difference-in-difference method. Main findings of the study demonstrate that (i) export diversification facilitates the formation of new businesses in the sampled countries, (ii) the odds of a business start-up increase by 1.7 – 3.5 times in response to a percentage point increase in export diversification, and finally (iii) business start-ups are at least 95 percent higher in countries with high export diversification compared to countries with narrow export baskets. These findings emphasize the need for policies aimed at moving away from concentrated export baskets towards more diversified ones to leverage entrepreneurial effort in the selected African countries.

Estimated Reduction in Health Care Spending Associated With Weight Loss in Adults

Recent data from federal surveys show that more than 70% of adults have either overweight or obesity. Overweight and obesity are associated with several expensive chronic conditions, such as type 2 diabetes, heart disease, hypertension, and hyperlipidemia. The prevalence of chronic disease is a key driver of rising health care spending among employers and the Medicare program. To estimate health care spending among adults with overweight or obesity and have employer-sponsored insurance or Medicare and examine the association of higher and lower body mass index (BMI). This cross-sectional study included privately insured adults and adult Medicare beneficiaries with a BMI of 25 and higher. Data from the Medical Expenditure Panel Survey-Household Component were used to tabulate total annual health care spending. Data were analyzed from April 1 to June 20, 2024. The primary intended measures were total annual health care spending, controlling for patient demographics, income, education, and BMI. The baseline health care spending as a percentage reduction in BMI. An instrumental variable regression model with estimated total spending at various levels of BMI was used. This study included 3774 adults who were insured with Medicare (mean [SD] age, 63.1 [11.1] years; mean [SD] percentage female, 50.4% [49.7%]; mean [SD] percentage non-Hispanic Black adults, 17.4% [37.7%]; mean [SD] percentage non-Hispanic White adults, 61.3% [48.4%]). The study also included 13 435 adults who had employer-sponsored insurance (mean [SD] age, 46.3 [6.9] years; mean [SD] percentage female, 47.6% [48.9%]; mean [SD] percentage non-Hispanic Black adults, 11.1% [30.7%]; mean [SD] percentage non-Hispanic White adults, 73.1% [43.4%]). Overall, adults with employer-sponsored insurance with a weight loss of 5% were estimated to spend a mean of $670 (95% CI, $654-$686) less on health care (8% less), and those with a weight loss of 25% spent an estimated mean of $2849 (95% CI, $2783-$2916) less on health care (34% less). Among adults with Medicare who had 1 or more comorbid conditions, a 5% weight loss was estimated to reduce spending by $1262 (95% CI, $1217-$1306) (7% less) and a 25% weight loss was estimated to reduce health care spending by a mean of $5442 (95% CI, $5254-$5629) (31% less). In this cross-sectional study, projected annual savings from weight loss among US adults with obesity were substantial for both Medicare and employer-based insurance.

Robust Permutation Tests in Linear Instrumental Variables Regression

This article develops permutation versions of identification-robust tests in linear instrumental variables regression. Unlike the existing randomization and rank-based tests in which independence between the instruments and the error terms is assumed, the permutation Anderson-Rubin (AR), Lagrange Multiplier (LM) and Conditional Likelihood Ratio (CLR) tests are asymptotically similar and robust to conditional heteroscedasticity under standard exclusion restriction, that is, the orthogonality between the instruments and the error terms. Moreover, when the instruments are independent of the structural error term, the permutation AR tests are exact, hence, robust to heavy tails. As such, these tests share the strengths of the rank-based tests and the wild bootstrap AR tests. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.

Evaluating the Policy Implications of Renewable Portfolio Standards on Grid Reliability in the United States: An Instrumental Variable Quantile Analysis

This paper investigates the influence of Renewable Portfolio Standards (RPSs) on the reliability of grids while considering key indicators, including the System Average Interruption Duration Index (SAIDI), System Average Interruption Frequency Index (SAIFI), and Customer Average Interruption Duration Index (CAIDI), both without and with the inclusion of Major Event Days (MEDs). The study adopted the instrumental variable quantile regression (IVQR) method for estimation purposes. The first-stage 2SLS results confirm the robustness and validity of the instruments used, including the Democrats’ Participation Rate, Real Average Household Income, and Renewable Energy Capacity, which show strong relevance and satisfy the conditions of homogeneity. Our findings demonstrated that the adoption of RPS policies has significantly decreased the interruption of power in all of the considered quantiles. The impact is more potent in states facing more severe reliability challenges. The results proved the important role of the policies of RPSs in improving the stability of grids in the context of states having relatively poor reliability. In summary, the obtained findings demonstrate significant insights related to the policies of renewable energy in the resilience of infrastructure. Finally, the findings suggest that the government authorities should promote the RPS as a tool for enhancing the performance of grids in different states.

Spatial lag quantile regression for compositional data

While research on spatial lag quantile regression models exists, the extension to incorporating compositional data within this framework remains unexplored. The unique characteristics of compositional data present significant issues in constructing such a model. In this paper, we investigate the estimation problems for the spatial lag quantile regression model in compositional data. We propose the constrained two-stage quantile regression (CTS-QR) and constrained instrumental variable quantile regression (CIV-QR) methods based on linear programming. Numerical simulations show that our proposed methods are more accurate than traditional unconstrained estimation methods. A real data application of our methods is also provided.

The welfare effects of digital infrastructure: Micro-level evidence from China

Over the past 40 years of reform and opening-up, China has achieved remarkable progress in rural infrastructure development. Although some empirical studies have demonstrated a positive relationship between the provision of digital infrastructure and increases in agricultural output and farmer income in various developing countries, research exploring the impact of digital infrastructure on rural households’ welfare is rare. This gap represents a significant issue for policymakers and academic researchers in developing countries today. By matching micro-survey data with county-level digital infrastructure investment data, this paper empirically examines the effects and mechanisms of digital infrastructure on the welfare of rural households using ordered probit and mediation models. We found that digital infrastructure has a significant impact on rural households’ overall welfare levels. Mechanism testing revealed that digital infrastructure operates through at least three distinct pathways, positively affecting both the economic and non-economic welfare of rural households. Additionally, the level of human capital and regional economic development exhibit notable heterogeneous effects on the impact of digital infrastructure. The validity of the results was demonstrated through instrumental variable regression and a series of robustness tests. The reason behind this result lies in the fact that the improvement of digital infrastructure in rural areas enhances households’ ability to access resources, develop skills, and engage in decision-making processes. This objectively improves their living standards and meets their needs. The empirical findings of this study have important policy implications for strengthening rural digital infrastructure, further enhancing rural households’ digital skills, and ultimately improving their overall welfare.

How does developmental advising impact college students? Findings from medical students in China

ABSTRACT This study contributes evidence on the impact of developmental advising using a national sample of students in the undergraduate clinical medicine program in China (n = 122,932). Using a combination of instrumental variable regression and structural equation modeling, the study explored beyond low-touch informative intervention and fully presented how developmental advising makes effects. A significantly positive advising effect on both academic and non-academic indicators was found, and it varies by the students’ demographic features, family backgrounds, and pre-college experiences, suggesting its potential to address achievement gaps across students from different backgrounds. In addition, the mediating roles of self-efficacy and learning engagement that enable advising to impact academic improvement were verified, shedding light on how developmental advising impacts students’ development.

Does Retirement Increase Fraud Exposure and Fraud Victimization?—Evidence From China

This study investigates the impact of retirement on fraud exposure and victimization among the older adults in China. Using data from the China Health and Retirement Longitudinal Survey, it employs instrumental variable regression, exploiting the statutory retirement age in China as a source of exogenous variation. The findings indicate that retirement does not significant affect fraud exposure but significantly increases the risk of fraud victimization and associated financial losses for older adults. The primary channels through which retirement influences fraud victimization include changes in post-retirement time allocation and increased psychological vulnerability.

Uneven Burdens: The Intersection of Brownfields, Pollution, and Socioeconomic Disparities in New Jersey, USA

Brownfields, abandoned or underutilized lands having the potential for contamination due to industrial activities, pose health risks to nearby residents and raise environmental justice (EJ) concerns. This study examines the relationship between brownfield distribution and socioeconomic factors in New Jersey, a state with over 700 brownfield sites, using correlation, spatial autoregression, and instrumental variable regression methods. Results show that brownfields are associated with higher combined EJ stressors, indicating compounding environmental pollution and health risks in affected areas. Brownfields are disproportionately correlated with lower household income, higher rates of uninsured individuals, lower educational attainment, and a greater percentage of people of color. When accounting for spatial autocorrelation, relationships between income, education, and brownfield distributions are reduced, reflecting New Jersey’s historical, environmental, and social contexts. Overall, the presence of people of color maintains the strongest association with brownfield distribution. This study highlights ongoing EJ challenges faced by disadvantaged groups and emphasizes the need for targeted mitigation measures to enhance sustainability and equity. New Jersey serves as a critical case study given its ethnic diversity, growing urban population, and status as the state with the highest population density in the United States, illustrating broader trends of disproportionate exposure to environmental pollution among marginalized communities.

Digital financial inclusion and SME market dynamics: an empirical analysis based on big data of business registrations in China

ABSTRACT With the deepening of internet development, digital financial inclusion has gradually become a crucial force in promoting the market entry of small and micro enterprises (SMEs). This study analyzes the impact of digital financial inclusion on SME market dynamics, using big data of business registrations and digital financial inclusion data of 268 cities in China from 2011 to 2017. Employing two-way fixed effects and instrumental variable regression methods, we find that digital financial inclusion significantly facilitates SME market entrance and reduces exiting risk. Moreover, the level of internet development moderates the impact of digital financial inclusion, with regions having high internet coverage showing a stronger effect in promoting entry; while regions with low internet coverage exhibit a more pronounced effect in reducing exiting. Further heterogeneity analysis suggests that the impact of digital financial inclusion primarily reveals with the breadth of coverage and depth of use. The effects are more significant in regions with high levels of innovation and foreign investment. This paper provides new empirical evidence on the role of digital financial inclusion in SME development and proposes relevant policy recommendations to promote sustainable development for SMEs.

Month of Birth, Early Academic Achievement, and Parental Expectations of University Completion: A New Test on Sticky Expectations

Previous studies have shown that educational expectations of individuals with high socioeconomic status (SES) are relatively unaffected by low academic performance, a phenomenon called “sticky expectations.” However, this result might be biased by endogeneity and reverse causality between academic achievement and educational expectations. Using data from the Trends in International Mathematics and Science Study from 11 countries with a strict school-entry rule and building on the well-established finding that children born in the months before the school-entry cutoff underperform at school, we use birth month as an instrument to identify the causal effect of early academic achievement on parents’ expectations of university completion by parental education. Our findings based on the instrumental variable (IV) regression show that the moderation of social origin in the relationship between children’s performance and parental expectations is moderately overestimated in cross-sectional data. Nonetheless, the stickiness of high-SES parental expectations is confirmed in the IV model, proving that parental expectations are less affected by children’s early achievement when the parents are highly educated.

Community Social Capital and Accounting Conservatism

ABSTRACT The combination of peer pressure and who you know in our society can be powerfully influential. Likewise, firms headquartered in high social capital locations can exhibit associated economic outcomes and financial qualities. Our study examines the impact of a critical firm-level qualitative characteristic, community social capital, on firm accounting conservatism. The results of our research show firms headquartered in areas of high community social capital tend to exhibit less accounting conservatism. These findings suggest high community social capital mitigates issues related to contracting efficiency, social disciplinary instruments, and information asymmetry, reducing the corporate impetus for using accounting conservatism. We apply alternative proxies for community social capital, and further robustness analyses include entropy balance matching and instrumental variable regressions. Our study extends and complements literature investigating the effects of various firm-specific traits affecting accounting conservatism and contributes to research examining the effect of community social capital on corporate policies. Data Availability: Data are available from the public sources cited in the text. JEL Classifications: A13; D91; M14; M41; Z13.

Social psychology research in the commercial art market: challenges and opportunities

This study investigates the social psychological dynamics of the commercialized art market, employing a multifaceted approach to uncover the complex interplay of factors influencing artistic valuation and market behavior. Utilizing data from 500 top-selling artists and 1000 major collectors over a decade, we conduct quantitative analyses, including time series analysis, principal component analysis, and network structure examination. Our findings reveal the primacy of artist reputation and market demand in price formation, accounting for 45% of the variance in artwork prices. We identify cyclical market patterns aligned with major auction events and economic trends. Network analysis exposes the critical role of intermediaries and community structures in market interactions. Addressing endogeneity concerns through instrumental variable regression, we establish a robust causal link between artist reputation and market outcomes. This research provides valuable insights for artists, galleries, collectors, and investors, offering a comprehensive framework for understanding the contemporary art market's complexities. Our study contributes to the growing body of literature on art market dynamics, highlighting the intersection of social psychology, economics, and cultural trends in shaping artistic value.

Information and communication technology and female employment in India

We revisit the issue of declining female labour force participation in India by investigating the role of information and communication technology (ICT) adoption. Given that ICT has been rapidly penetrating in India in the past few decades, we explore the impact of ICT ownership and use on female labour market behaviour. We use a nationally representative survey, the India Human Development Survey and estimate instrumental variable regression models and find that ICT adoption increases the likelihood of female employment. Our results show that the positive impact of ICT adoption is limited to individuals who are salaried or are self-employed. Further, we find that ICT adoption increases (decreases) the likelihood of entry into (exit from) the labour market. We contrast our analysis by comparing it with male employment and find evidence of differential impact of ICT adoption. Finally, we find that the effect is limited to women with relatively higher levels of education and those belonging to relatively higher income households. We investigate the potential channels that could drive our results and find that women who adopt ICT have more flexible work options where they spend lesser time to travel to work, have a higher autonomy in work related decisions and a higher willingness to enter the workforce. Thus, ICT adoption by reducing information asymmetries in the labour market could increase female labour market participation.

Examining the joint effect of clinical quality, meaningful use of HIT and patient-caregiver interaction on mortality rates in US acute care hospitals.

Healthcare quality has long been defined and assessed using different theories that outline care delivery as a product of clinical and non-clinical aspects. The clinical aspect is medicinal in nature, while the non-clinical symbolizes social and technological aspects of care. To the best of our knowledge, the joint effect of the clinical, social, and technological aspects of care on outcome quality of care has not been investigated in the literature. The current study empirically investigates, using the joint effect of the clinical, social, and technological care quality dimensions on mortality rates through analyzing longitudinal data from 3081 US hospitals. Six-year data from 3081 acute care hospitals are collected and analyzed using econometric analysis with two-stage least square instrumental variable regression models. Hospitals that jointly focus on clinical, social, and technological care dimensions realize lower mortality rates. Combining clinical quality (CM) with either patient-caregiver interaction (PCI) or meaningful use of health infomration technology (MUHIT) reduces mortality rates. The lowest mortality rate is realized when hospitals combine CM, PCI, and MUHIT. Our study provides empirical evidence on the importance of combining clinical and non-clinical care measures to reduce mortality rates in hospitals. Our results indicate that hospitals that combine dual quality dimensions, clinical quality with either PCI or MUHIT, can also realize improvement in mortality rates. However, the best outcome can be realized by focusing on the triple quality dimensions (CM, PCI, and MUHIT). The study provides pointers to healthcare professionals and policymakers on the impact of non-clinical care on the clinical-mortality link in hospitals.

Unraveling the Effects of Economic Complexity on Governance in Developing Countries. New Evidence from Instrumental Variable Quantile Regression Estimates

This study examines the effect of economic complexity on governance indicators. We use instrumental variable quantile regression to estimate a panel data model for 78 developing countries over the period 2000–2019. The results show that there is a positive relationship between economic complexity and overall governance; however, the effect varies according to the distribution of the different governance indicators. More specifically, our results show that economic complexity improves political governance only for the higher quantiles. The effect varies positively along the distribution of economic governance. For institutional governance, the effect of economic complexity is statistically significant for the median and upper quantiles. Our results also allow us to show that the effects of economic complexity on governance differ significantly between quantiles of the distribution in low- and middle-income countries. Finally, human capital, FDI inflows and reductions in income inequality are the channels through which economic complexity affects governance.