Search

Efficient computational algorithms for bootstrapping linear regression models with clustered data are discussed. For ordinary least squares (OLS) regression, a new algorithm is provided for the pairs cluster bootstrap, along with two algorithms for the wild cluster bootstrap. One of these is a new way to express an existing method. For instrumental variables (IV) regression, an efficient algorithm is provided for the wild restricted efficient cluster (WREC) bootstrap. All computations are based on matrices and vectors that contain sums of squares and cross-products for the observations within each cluster, which have to be computed just once before the bootstrap loop begins. Monte Carlo experiments are used to study the finite-sample properties of bootstrap Wald tests for OLS regression and of WREC bootstrap tests for IV regression.

This paper re-examines the problem of estimating risk premia in unconditional linear factor pricing models. Typically, the data used in the empirical literature are characterized by weakness of some pricing factors, strong cross-sectional dependence in the errors, and (moderately) high cross-sectional dimensionality. Using an asymptotic framework where the number of assets/portfolios grows with the time span of the data while the risk exposures of weak factors are local-to-zero, we show that the conventional two-pass estimation procedure delivers inconsistent estimates of the risk premia. We propose a new estimation procedure based on sample-splitting instrumental variables regression. The proposed estimator of risk premia is robust to weak included factors and to the presence of strong unaccounted cross-sectional error dependence. We prove the consistency of the new estimator, establish asymptotically valid inferences using Wald statistics, verify performance of the new procedure in simulations, and revisit some empirical studies.

This paper analyzes whether money influences election outcomes. Using a new and more comprehensive dataset built from government sources, the paper shows that the relations between money and votes cast for major parties in elections for the U.S. Senate and House of Representatives from 1980 to 2018 are well approximated by straight lines. It then considers possible challenges to this “linear model” of money and elections on statistical grounds, resting on possible endogeneity arising from reciprocal causation between, for example, popularity and votes. The paper develops a spatial Bayesian latent instrumental variable model to tackle this much discussed problem. It checks its results by studying relations between changes in gambling odds and contributions in key elections. Both approaches suggest that reciprocal causation may happen to some degree, but that money’s independent influence on elections remains powerful.

We examine how online customer reviews for one generation of a product affect sales of another generation in the same product series. The main intriguing result is that previous generation valence has a positive impact on current generation sales; however, current generation valence has a negative impact on previous generation sales. The positive impact of previous generation valence becomes even stronger (1) as the uncertainty (standard deviation) in reviews for the current generation increases and (2) when the current generation valence is high. In contrast, it becomes weaker (1) as the uncertainty in reviews for the previous generation increases and (2) when the current generation has been on the market for a longer period of time. Other results are discussed. Our data consist of intergenerational pairs of point-and-shoot cameras on the largest online seller of such devices, Amazon.com. We estimate the current and previous generation models jointly, allowing for errors to be clustered at the daily and product levels. In addition, we address endogeneity concerns over the online word of mouth measures by using instrumental variables. This paper was accepted by Juanjuan Zhang, marketing.

Pharmaceutical firms typically enjoy market exclusivity for new drugs from concurrent protection of the underlying invention (through patents) and the clinical trials data submitted for market approval (through data exclusivity). Patent invalidation during drug development renders data exclusivity the sole source of protection and shifts the period of market exclusivity at the project level. Patent invalidation therefore constitutes a natural experiment that allows us to causally identify how the duration of market exclusivity affects firms' incentives to innovate. In instrumental variables regressions we quantify the effect of a one-year reduction in expected market exclusivity on the likelihood of drug commercialization. We provide first estimates of the responsiveness of R&D investments to market exclusivity expectations and hereby inform the ongoing policy debate on the regulation of intellectual property rights in the pharmaceutical industry.

This paper provides sufficient conditions for identification of a nonparametric regression model with an unobserved continuous regressor subject to nonclassical measurement error. The measurement error may be directly correlated with the latent regressor in the model. Our identification strategy does not require the availability of additional data information, such as a secondary measurement, an instrumental variable, or an auxiliary sample. Our main assumptions for nonparametric identification include monotonicity of the regression function, independence of the regression error, and completeness of the measurement error distribution. We also propose a sieve maximum likelihood estimator and investigate its finite sample property through Monte Carlo simulations.

The increasing rate of obesity in the U.S., particularly among low-income households, necessitates a thorough understanding of the relationship between obesity and in-kind federal benefits such as the Supplemental Nutrition Assistance Program (SNAP), formerly the Food Stamp Program. However, when examining this relationship, the existing literature often ignores evidence that respondents frequently misreport their participation in SNAP. This paper studies the impact of such misreporting on the estimated average treatment effect (ATE) of SNAP participation on adult obesity. Our analysis also synthesizes the current empirical techniques available for estimating ATEs in light of misreported treatment participation, adopting a range of parametric analyses as well as nonparametric bounds. The results highlight the inherent bias of common point estimates when ignoring misreporting, with treatment effects from instrumental variable methods exceeding the nonparametric bounds by over 200% in some cases. Accounting for misreporting, the estimated effects of SNAP participation on obesity are largely inconclusive. We find a slight negative effect of SNAP participation on the probability of being overweight, but the results specific to gender remain inconclusive due to the high rates of misreporting, particularly among men.

Retail distribution is one of the major challenges in emerging economies. These economies are volatile and filled with inefficiencies, and the representativeness of unstructured retail increases the complexity of distribution systems for consumer packaged-goods companies. We analyze 644 brands to extend the existing literature by modeling the retail distribution and market share in an emerging market according to the type of retail channel (full- and self-service channels), moderated by economic fluctuations and the market position of a brand (high- and low-share brands). Our model controls for endogeneity using instrumental variables (IVs) and accommodates heterogeneity across brands and categories by means of a fixed-effects robust regression. Our study highlights that the relationship between distribution and market share exhibits greater convexity in the self-service channel than in the full-service channel. Further, we contribute to the existing research in distribution effectiveness in emerging markets by showing the convex effect of distribution on market share could vary when the economy changes. Distribution gains are more effective in the self-service channel than in the full-service channel in times of economic decline. Also, the results indicate the higher degree of convexity in the relationship between distribution and market share for the self-service channel compared with the full-service channel is increased further for high-share brands than for low-share brands.

This paper examines whether unionization influences firms’ investments in environmental protection, and finds a positive association between unionization and investment in environmental protection. Specifically, the positive association strengthens when firms (i) show higher levels of charitable donations; (ii) are located in regions receiving greater amounts of foreign direct investment; (iii) are located in regions in which political reliance is high; (iv) are located in regions with low antipollution investment. We alleviate endogeneity concerns by controlling for firm-fixed effects, using the propensity-score matching (PSM) method and the instrumental variable approach, and conducting a battery of robustness checks. The results of our additional analysis validate the finding that firms with unions tend to engage in environmental investment. In conclusion, we provide new evidence for the impact of unions on a given firm’s investment in environmental protection.

How do the economic characteristics of regions affect social trust? This article tries to answer this question using a dynamic game model with empirical testing. People in regions with a larger scale of economic activity are usually made up of groups with different historical or cultural backgrounds. The dynamic game with incomplete information will lead to misperceptions among groups and will reduce the mutual trust level. The results of OLS regression reveal that city's economic scale and residents' average social trust are negatively correlated based on the Chinese General Social Survey (CGSS) with a sample size of 40,360 and Map Apps data. This study overcomes the endogeneity problem by using the distance from a city to Beijing and Shanghai as the instrumental variable. Results from the 2SLS analysis confirm that the prior negative correlation remains. A 1% increase in economic scale drops the trust level by 0.05%. In addition, women in large cities have lower trust levels than their male counterparts. Policymakers should aim to reduce the probability of misunderstanding among the groups to increase the level of social trust.

This paper studies Structural Vector Autoregressions in which a structural shock of interest (e.g., an oil supply shock) is identified using an external instrument. The external instrument is taken to be correlated with the target shock (the instrument is relevant) and to be uncorrelated with other shocks of the model (the instrument is exogenous). The potential weak correlation between the external instrument and the target structural shock compromises the large-sample validity of standard inference. We suggest a confidence set for impulse response coefficients that is not affected by the instrument strength (i.e., is weak-instrument robust) and asymptotically coincides with the standard confidence set when the instrument is strong.

We study identification in nonparametric regression models with a misclassified and endogenous binary regressor when an instrument is correlated with misclassification error. We show that the regression function is nonparametrically identified if one binary instrument variable and one binary covariate satisfy the following conditions. The instrumental variable corrects endogeneity; the instrumental variable must be correlated with the unobserved true underlying binary variable, must be uncorrelated with the error term in the outcome equation, but is allowed to be correlated with the misclassification error. The covariate corrects misclassification; this variable can be one of the regressors in the outcome equation, must be correlated with the unobserved true underlying binary variable, and must be uncorrelated with the misclassification error. We also propose a mixture-based framework for modeling unobserved heterogeneous treatment effects with a misclassified and endogenous binary regressor and show that treatment effects can be identified if the true treatment effect is related to an observed regressor and another observable variable., Comment: 25 pages

Given overall lack of gender diversity in the venture capital and entrepreneurship industry shown in Calder-Wang and Gompers (2017) we ask: What promotes greater gender diversity in hiring? Does diversity lead to better firm performance and higher investment returns? In this paper, using a unique dataset of the gender of venture capital partners’ children, we find strong evidence that when partners have more daughters, the propensity to hire female partners increases. Moreover, our instrumental variable results suggest that increased gender diversity improves deal and fund performance. Lastly, the effects are primarily driven by the gender of senior partners’ children.

We exploit an allocation rule set by the ECB for the first series of TLTROs to study the effects of targeted monetary policy on banks’ credit supply to firms. Combining transaction-level data from the Italian credit register and an instrumental variable identification strategy, we find that targeted longer-term central bank liquidity decreased rates and increased loan amounts, also avoiding some unintended consequences of untargeted measures, such as carry-trade strategies and risk shifting. We show that the outward shift in banks’ credit supply was heterogeneous, taking place only in more competitive banking markets, with stronger effects for smaller and safer firms.

As promises are non-binding, they can be either fulfilled or broken. Previous lab and field research mostly examines the factors that explain fulfillment, as promising itself is difficult to randomly assign. We design a novel natural field experiment where subtle changes of words in an invitation email shifts the level of promise making among our subjects to test the causal effect of promises. Promises are positively related to experimental attendance, and using the invitation treatment as an instrumental variable, we recover a positive but imprecisely measured causal impact of promise making on behavior.

The current study investigates whether board age diversity affects company financial performance and risk, and whether diverse work-related values held by directors of different ages are the underlying cause of any effects via a sample of listed companies from Turkey, which has a collectivistic and paternalistic culture. In order to investigate these effects, various 2 Stage Least Squares Instrumental Variables models are employed. In addition, to investigate through which channels age diversity may affect company performance, directors' personal values are also considered via propensity score matching. Even though directors’ values are not observable, it is assumed that their ages reflect their values. For this purpose, the current study utilizes the World Value Survey data. The findings suggest that board age diversity has a positive effect on both company performance and risk, but do not suggest that intra-group conflicts in terms of work-related values are the underlying causes of this positive effect.

We examine the effect of the marginal child and the total number of children on self-reported well-being as a proxy for happiness. Prior literature has not controlled for endogeneity. We propose an instrumental variable approach which remediates the existing endogeneity problem, and generates a non-linear marginal effect and a bimodal distribution which explains why prior research arrived at competing positive, negative, or inconclusive outcomes. We provide evidence in support of a "parenting happiness gap" where happiness declines after the birth of the first child; however, beyond the first child additional children have a significant, positive and increasing effect on well-being. Confirmed by evidence from sociology and psychology literature, we consider possible channels which explain these results and discuss their policy implications in the context of fertility, labor market and health outcomes.

This paper develops an empirical balancing approach for the estimation of treatment effects under two-sided noncompliance using a binary conditionally independent instrumental variable. The method weighs both treatment and outcome information with inverse probabilities to produce exact finite sample balance across instrument level groups. It is free of functional form assumptions on the outcome or the treatment selection step. By tailoring the loss function for the instrument propensity scores, the resulting treatment effect estimates exhibit both low bias and a reduced variance in finite samples compared to conventional inverse probability weighting methods. The estimator is automatically weight normalized and has similar bias properties compared to conventional two-stage least squares estimation under constant causal effects for the compliers. We provide conditions for asymptotic normality and semiparametric efficiency and demonstrate how to utilize additional information about the treatment selection step for bias reduction in finite samples. The method can be easily combined with regularization or other statistical learning approaches to deal with a high-dimensional number of observed confounding variables. Monte Carlo simulations suggest that the theoretical advantages translate well to finite samples. The method is illustrated in an empirical example.

We utilize the implementation of the 1986 Compulsory Education Law in China as a natural experiment to examine the relationship between educational attainment and migrant entrepreneurship. Using data from the 2017 China Migrants Dynamics Survey, results from our preferred two-stage least square model, which uses instrumental variable and difference-in-differences estimators to correct for endogeneity, suggest that having one additional year of education generates a 3.5 percentage points increase in the probability of being an employer entrepreneur vis-a-vis an employee, and a 4.7 percentage points increase in the probability of being an employer entrepreneur vis-a-vis a solo entrepreneur. Having better education, however, does not affect the propensity to become an entrepreneur in general vis-a-vis an employee or a solo entrepreneur vis-a-vis an employee. These results are robust to various checks including alternative estimation methods and ways of defining entrepreneurship. We also explore the channels through which educational attainment influences whether one chooses to become an employer entrepreneur. Our findings suggest that assortative mating, difficulties experienced in the host location, permanent settlement intention and social capital mediate the relationship between better education and the choice of both becoming an employer entrepreneur vis-a-vis a solo entrepreneur and an employer entrepreneur vis-a-vis an employee, while risk preferences are a mechanism through which education affects the probability of becoming an employer entrepreneur as opposed to a solo entrepreneur.

We study whether, how, and why the investment of a firm depends on the investment of other firms in the same product market. Using an instrumental variable based on the presence of local knowledge externalities, we find a sizeable complementarity of investment among product market peers, holding across a large majority of sectors. Peer effects are stronger in concentrated markets, featuring more heterogeneous firms, and for smaller firms with less precise information. Our findings are consistent with a model in which managers are imperfectly informed about fundamentals and use peers’ investments as a source of information. Product market peer effects in investment could amplify shocks in production networks. This paper was accepted by Gustavo Manso, finance.

We analyze the funding of start-up companies across financing rounds, focusing on the dynamic interactions between angel investors and venture capitalists. Using unique data from British Columbia, Canada, we show that angels and venture capitalists are dynamic substitutes. This substitutes pattern applies across the performance range of companies. It is less pronounced for serial angels. An instrumental variable analysis, based on available investor tax credits, suggests that the substitutes pattern is driven by company characteristics. Overall, the evidence points to the existence of parallel streams of angel and venture capital funding, with fewer transitions between streams than is traditionally assumed.

Purpose The purpose of this paper is twofold, namely, to investigate if living and working abroad influences the (subjective) health of return migrants and to understand if there are any spillovers of return-migrant members onto health conditions of the family members left behind. Design/methodology/approach To that end, this paper uses the DoTM (Development on the Move) Migration Survey 2009, as well a propensity score matching to address selectivity on observables and IV (instrumental variables) for the selectivity on unobservables. Findings Results suggest that when equalized on observables, return migrants have better health than non-migrants. However, the reverse causality channel (less healthy individuals are more inclined to return) works to attenuate the true effect of return migration on health. Results further suggest a positive spillover effect of return migration on the health of the family members left behind, being mainly driven by the work of remittances sent while abroad, and not by the returned wealth or by the health knowledge transfer. Originality/value This paper offers at least two novelty lines to contribute to the current sparse of knowledge. First, it is among the scarce papers, and probably the only quantitative one, to investigate the nexus between return migration and health outcomes. Second, it heavily dwells on the role of selectivity (both on observables and unobservables) in determining the true (causal) effect of return migration on health.

We examine how banking supervisors affect credit at the local level by charging fines to individual banks. Using a macro approach to capture the direct effect on the fined bank and the indirect effect on the other banks operating in the local credit market, we estimate reputational, reallocation and balance sheet effects on Italian provinces over the period 2005–2016 by a fixed effects model and instrumental variables. Provincial gross bank loans expand after a fine independently of its size. The impact of fine frequency depends on the size of the provincial banking sector, but neither on bank governance/ownership nor crises. No statistically significant evidence supports reputational or balance sheet effects. Instead, our results suggest that it would behoove bank supervisors to favor frequency over size of bank fines. Bank fines seem to work more like a good housekeeping seal of approval, enhancing transparency and effective banking practices.

We investigate the effects of opioid abuse on municipal finance. We employ instrumental variables, border discontinuity difference-in-differences regressions, and coarsened exact matching to identify consistent causal effects, while controlling for variation in economic conditions and demographics. Opioid abuse lowers credit ratings, increases new offer yields, and reduces bond issuance. Reversal of these effects following effective antiopioid legislation further supports causality. Differential effects due to investor heterogeneity suggest that opioid abuse affects municipal finance through a capital supply channel. Overall, we conclude that local opioid abuse impedes municipalities’ access to capital and thus hurts their ability to provide public services and infrastructure.

We use the presence of a Wikipedia article for initial public offering (IPO) firms to test theories of information asymmetry and investor awareness. While we find limited support for the former, our results provide strong support for theories of investor awareness. Specifically, IPO firms with a Wikipedia article exhibit significantly higher underpricing and offer price revisions than do IPO firms without a Wikipedia article. Investor awareness has positive long-term effects, including greater analyst following and institutional ownership for up to three years after the offering. The effect is robust to firm-specific Google search volume, news coverage, retail trading intensity, social media activity, propensity score matching, and an instrumental variable approach.

Using data from the China Household Finance Survey, we investigate the relationship between family involvement and the subjective well-being (SWB) of the self-employed. Instrumental variables are u...

Many studies use shift-share (or “Bartik”) instruments, which average a set of shocks with exposure share weights. We provide a new econometric framework for shift-share instrumental variable (SSIV) regressions in which identification follows from the quasi-random assignment of shocks, while exposure shares are allowed to be endogenous. The framework is motivated by an equivalence result: the orthogonality between a shift-share instrument and an unobserved residual can be represented as the orthogonality between the underlying shocks and a shock-level unobservable. SSIV regression coefficients can similarly be obtained from an equivalent shock-level regression, motivating shock-level conditions for their consistency. We discuss and illustrate several practical insights of this framework in the setting of Autor et al. (2013), estimating the effect of Chinese import competition on manufacturing employment across U.S. commuting zones.

PurposeThe authors analyse how the receipt of a wealth transfer (inheritance or gift) affects labour force participation in 14 EU countries. They compare the effect of receiving an inheritance or a gift and investigate different behaviours at the gender level and educational level and for elderly individuals.Design/methodology/approachThe authors use data from the Household Finance and Consumption Survey for 14 European countries and adopt an instrumental variable approach. They use information on the type of donor (family and nonfamily) to infer the degree of anticipation of a wealth transfer.FindingsThe authors find that unexpected wealth transfers have a negative impact on labour force participation, with a stronger impact for gifts than for inheritances. For gender, they find larger negative impacts for females than for males, which is in line with a weaker attachment to the labour market. Receiving an unexpected wealth transfer may also result in early retirement.Originality/valueThe paper offers a novel comparison of the effect of receiving an inheritance or a gift on labour force participation using a unique European dataset. The authors investigate whether males and females react differently to the receipt of a wealth transfer and the existence of different responses at the educational level and for elderly individuals.

This paper presents a simple method for carrying out inference in a wide variety of possibly nonlinear IV models under weak assumptions. The method provides a finite-sample bound on the difference between the true and nominal probabilities of rejecting a correct null hypothesis. The method is a non-Studentized version of the S test of Stock and Wright (2000) but is implemented and analyzed differently. It does not require restrictive distributional assumptions, linearity of the estimated model, simultaneous equations, or information about whether the instruments are strong or weak. It can be applied to quantile IV models that may be nonlinear and can be used to test a parametric IV model against a nonparametric alternative. It provides information about the relation between the “degree of weakness” of the instruments and the power of the test. The bound presented here holds in finite samples, regardless of the strength of the instruments.

Can central banks defuse rising stability risks in financial booms by leaning against the wind with higher interest rates? This paper studies the state-dependent effects of monetary policy on financial crisis risk. Based on the near-universe of advanced economy financial cycles since the nineteenth century, we show that discretionary leaning against the wind policies during credit and asset price booms are more likely to trigger crises than prevent them. (JEL E43, E44, E52, E58, F33)

Previous studies show that children in lower socioeconomic status families reveal higher rates of mortality. We complement the income-mortality literature by establishing a causal link between income and child mortality. Our instrument for income is based on time-series global shocks to oil prices combined with the cross-sectional share of employment in manufacturing across US states as their exposure to oil price changes. Using the universe of death records between the years 1975-2004, we find the OLS results of income-child-mortality relationships are under-biased. The 2SLS-IV results suggest that a $1,000 increase in income per capita at the state level reduces child mortality and infant mortality by 0.87 and 0.53 fewer incidences per 1,000 population of age-specific children.

We develop a unifying econometric framework for the analysis of heterogeneous panel data models that can account for both spatial dependence and common factors. To tackle the challenging issues of endogeneity due to the spatial lagged term and the correlation between the regressors and factors, we propose the CCEX-IV estimation procedure that approximates factors by the cross-section averages of regressors and deals with the spatial endogeneity using the internal instrumental variables. We develop the individual and Mean Group estimators, and establish their consistency and asymptotic normality. By contrast, the Pooled estimator is shown to be inconsistent in the presence of parameter heterogeneity. Monte Carlo simulations confirm that the finite sample performance of the proposed estimators is quite satisfactory. We demonstrate the usefulness of our approach with an application to the house price growth for Local Authority Districts in the UK over 1997Q1–2016Q4.

This research examines the relationship between urban polycentric spatial structure and driving. We identified 46 employment sub-centers in the Los Angeles Combined Statistical Area and calculated access to jobs that are within and beyond these sub-centers. To address potential endogeneity problems, we use access to historically important places and transportation infrastructure in the early 20th century as instrumental variables for job accessibility indices. Our Two-stage Tobit models show that access to jobs is negatively associated with household vehicle miles traveled in this region. Among various accessibility measures, access to jobs outside sub-centers has the largest elasticity (-0.155). We examine the location of places in the top quintile of access to non-centered jobs and find that those locations are often inner ring suburban developments, near the core of the urban area and not far from sub-centers, suggesting that strategies of infill development that fill in the gaps between sub-centers, rather than focusing on already accessible downtowns and large sub-centers, may be the best land use approach to reduce VMT.

Do firms in China share rents with their workers? We address this question by examining firm-level panel data covering virtually all manufacturing firms over the period 2000-2007, representing an average of 200,000 firms and 54 million workers per year. We find robust evidence of rent sharing (RS): workers that would move from low- to high-profit firms would see their wages increase by about 45%. The results are based on multiple instrumental variables, including firm-specific international trade shocks. We also present a number of complementary findings that allow us to understand better the nature of RS in the country: RS is weaker in firms with more women and less educated workers; RS involves an element of risk sharing, as wages also decrease when profits fall; RS is lower in regions with more latent competition from rural workers; higher minimum wages tend to reduce RS; and, while employer labour market power reduces wages, it increases RS. Overall, despite its importance, RS in China is smaller and more symmetric than in developed economies, which reflects the weaker bargaining power of its workers and the different scope of its labour market institutions.

Purpose This study aims to investigate the impact of energy access on foreign direct investment (FDI) in an emerging market. Design/methodology/approach The study uses the two-stage least square instrumental variables estimation approach to compute the parameters of the model to account for any potential endogeneity and time persistence in energy access. Findings The results show that energy access significantly influences FDI inflows in Ghana. The results of the study also revealed that natural resources and macroeconomic variables such as real interest rate, gross domestic product growth rate are significant determinants of FDI inflows in Ghana. Practical implications The practical implication of this study is that there is a need for energy sector policy reforms in Ghana that would guarantee a secured and continued supply of energy to enhance energy access to boost FDI. Ghana should aim for a cost-effective, stable and environmentally friendly source of energy as an alternative to hydro energy as the main source of its power generation to promote FDI. Also, Ghana should initiate and implement policies aimed at creating an enabling and stable macroeconomic environment, as macroeconomic factors in this study are found to be drivers of FDI. Originality/value This study provides firsthand information on energy access and FDI from the Ghanaian perspective.

This paper investigates the relationship between economic policy uncertainty (EPU) and corporate intellectual capital (IC) investment. The empirical results, which are based on a wide range of econ...

Despite the increasing interest in analysing determinants of health status and subjective well-being, existing research has been mainly confined to developed countries’ adult population. This paper...

This paper estimates peer effects on children’s school entry age, using a dataset on 4,165 children from rural northwestern China (Gansu province). Instrumental-variable estimation, exploiting vari...

Several studies have estimated effects of body mass index (BMI) on labour market outcomes, and these studies have mixed findings. A significant challenge has been to adequately control for omitted variables, selection, reverse causality, and measurement error. We examine the impact of BMI on income using genetic variants as instrumental variables for BMI. Individual-level pre-tax income from tax records was merged with health survey data containing measured height and weight, and data on genetic variants. The analyses were stratified by sex and a variety of methods were used to explore the sensitivity and validity of the instrumental variable (IV) strategy. For females we found that BMI had a negative effect on the logarithm of income. The effect estimated from the IV models (-0.02) was larger than the effect estimated from naïve ordinary least squares regressions (-0.01). For males, the coefficients for the effect of BMI on income were imprecise, and both positive and negative coefficients were estimated depending on the estimation method. Our results suggest that females are susceptible to reduced income levels following increased BMI.

This study provides new evidence on the long-term impact of education on welfare participation in the US. I exploit historical changes in child labor laws as an instrumental variable for education ...

Using a novel dataset of independent directors’ voting activities on items proposed by managers of Korean firms, we investigate whether independent directors’ dissension in board meetings plays an effective role in enhancing firm value through improved corporate governance. Our results indicate that dissension improves firm value. This finding is robust to different measures of firm value and alternative model specifications including subsample, propensity score matching, and instrumental variable analyses. Overall, we contribute to the understanding of the relation between corporate governance and firm value. Specifically, we provide new evidence that the monitoring by independent directors enhances firm value.

We study how the enforcement of value-added tax (VAT) affects Chinese firms’ evasion of payroll tax, which is collected by weakly empowered agencies. Using the central government’s 2005 repeal of the agricultural tax to measure fiscal squeeze and subsequent enforcement effort of VAT collection, our instrumental variable estimation finds that the VAT enforcement leads to a significant increase in the payroll tax evasion. Examining firm heterogeneity and real responses suggests that increased payroll tax evasion mainly stems from cost optimization by small and cash-constrained private firms. Our paper echoes a growing literature to highlight the importance of understanding compliance spillover across taxes.

This paper exploits a panel of 28 European Union (EU) countries between 1995 and 2016 to analyze whether higher debt resulted in lower private investment - the so called debt overhang effect. We deal with the potential endogeneity between private investment and other macroeconomic determinants by applying an instrumental variable approach (GMM). Our results support the debt overhang hypothesis and indicate that this relationship only works through the public debt channel. In our baseline regression, a 10 percentage point increase in public debt reduced private investment by euro18.32 billion, given the levels of private investment prevalent in 2016. By contrast, private debt does not appear to be a significant determinant of private investment. These results hold after controlling for a number of factors that might have caused public debt to increase and private investment to decrease. While our analysis focuses on the financial sector channel, we find no evidence that public debt tightens the credit constraints for private firms or worsens the public debt overhang. We also show that government bailouts of the financial sector, which could alleviate financial distress and boost credit provision, do not appear to be effective in mitigating the public debt overhang effect. Finally, we find evidence that the financial openness of a country does alleviate the negative impact of public debt on private investment. This might suggest that attracting foreign capital compensates for a contraction in the domestic pool of financial resources due to higher public debt levels.

In structural dynamic discrete choice models, unobserved or mis-measured state variables may lead to biased parameter estimates and misleading inference. In this paper, we show that instrumental variables can address such measurement problems when they relate to state variables that evolve exogenously from the perspective of individual agents (i.e., market-level states). We define a class of linear instrumental variables estimators that rely on Euler equations expressed in terms of conditional choice probabilities (ECCP estimators). These estimators do not require observing or modeling the agent’s entire information set, nor solving or simulating a dynamic program. As such, they are simple to implement and computationally light. We provide constructive arguments for the identification of model primitives, and establish the estimator’s consistency and asymptotic normality. Four applied examples serve to illustrate the ECCP approach’s implementation, advantages, and limitations: dynamic demand for durable goods, agricultural land use change, technology adoption, and dynamic labor supply. We illustrate the estimator’s good finite-sample performance in a Monte Carlo study, and we estimate a labor supply model empirically for taxi drivers in New York City.

Do workers' first jobs affect their careers? Do such first-job effects (FJEs) vary across worker types? If so, can policy improve upon a “free” labor market by altering initial worker-employer matches? We study these questions using Norway's pre-2013 system of assigning doctors to their first job–residencies–through a random serial dictatorship. This generated individual-level variation in workers' choice sets over employers, which we use as instrumental variables to estimate FJEs. We then decompose workers' preferences over first employers into FJEs-on-earnings and employer “amenity value” components, showing how matches and worker welfare changed in the post-2013 decentralized labor market.

This study examines the effect of reliability of electricity on gender differences in socio-economic status using a comprehensive set of labor and non-labor market outcomes in India. Using the temporal variation in household electricity hours from a large gender-disaggregated data set, we examine the effects of electricity reliability with individual fixed effects and instrumental variable regressions. Our analysis reveals contrasting trends with significant progress at the extensive margin of electricity access, but little progress at the intensive margin of reliability, hours of electricity. We find that reliable electrification improves the status of women relative to men through increased employment opportunities and reduced time allocation to home production. For instance, 10 more hours of electricity increases the likelihood of employment in the ‘usual status’ by 2.8 percentage points (pp) for men, and 4.2 pp for women. The analysis is robust to the use of piece-wise linear regression approaches, as well as alternate specifications of the outcome variables. The study recommends considering electricity as a right, and as part of the broader strategy for reducing gender disparities in India.

For a panel of 1880 firms across 21 countries from 2004 to 2008, the impact of firm-level and country-level governance on payout policy is consistent with the La Porta et al. (J Finance 60(1):1–33, 2000) outcome model. In weak legal regimes, dividends increase in firm-level governance. In strong legal regimes, dividends decrease in firm-level governance while share repurchases increase, substituting tax-efficient, flexible share repurchases for rigid dividends. Consistent with the outcome model, payout decreases in growth opportunities when firm-level and country-level governance are high. These results are robust to an instrumental variable approach and alternative firm-level and country-level governance measures.

The present paper assesses the causal effects of education on smoking and self-rated health in Canada. Education is instrumented using the changes in compulsory schooling laws. The sample is restri...

ObjectiveAs elderly parents today will share a longer life with their children than ever before, the spillover effects of children’s human capital on parents’ well-being become increasingly important. This study investigated whether children’s schooling leads parents to give up smoking and whether the effects were moderated by their education or child–parent contact frequency.MethodsUsing data from the China Health and Retirement Longitudinal Study, we conducted an instrumental variable (IV) analysis using China’s compulsory schooling reform in the 1980s as a natural experiment.ResultsThe IV estimates suggest that elderly parents of more highly educated children are more likely to quit smoking. Moreover, the effects are more significant among parents who had not finished primary school and also slightly stronger among parents who live close to their children or meet their children frequently.DiscussionOur findings add to current evidence regarding spillover effects of education on smoking cessation. A child’s education may exert an impact through the spillover of health knowledge.

This paper analyzes the effect of a discrete treatment Z on a duration T. The treatment is not randomly assigned. The confounding issue is treated using a discrete instrumental variable explaining the treatment and independent of the error term of the model. Our framework is nonparametric and allows for random right censoring. This specification generates a nonlinear inverse problem and the average treatment effect is derived from its solution. We provide local and global identification properties that rely on a nonlinear system of equations. We propose an estimation procedure to solve this system and derive rates of convergence and conditions under which the estimator is asymptotically normal. When censoring makes identification fail, we develop partial identification results. Our estimators exhibit good finite sample properties in simulations. We also apply our methodology to the Illinois Reemployment Bonus Experiment.