Use Of Instrumental Variables Research Articles

The income equality dividend in ‘green’ infrastructure: Evidence from the transport and waste sectors in the OECD

As it is exemplified by the Sustainable Development Goals, countries are expected to advance towards both decreasing environmental harmfulness and reducing income inequality. However, although debates can be found on how to succeed in each of these two challenges, less is known about how income distribution is affecting environmental achievements. In this article, the authors propose the concept of an ‘income equality dividend’ regarding adoption of sustainable modes within infrastructure networks in high income countries. Evidence is provided to show that more egalitarian countries do better adopting mature environmental technologies within infrastructure networks, such as municipal recycling and railway passenger transport. Panel data from the OECD, covering between 1996 and 2013, shows that the adoption rates of these environmental technologies are higher in countries with more egalitarian income distributions. The use of instrumental variables allows to prove that the link is exogenous. In addition, adoption of mature environmental technologies seems to respond more strongly to changes in the income shared by different groups, depending on the sector. While railway passenger transport is more strongly linked to changes in shares of the bottom 40 per cent and the top 10 per cent, municipal recycling is linked to income shares by the upper half of the bottom 40 per cent, and not affected at all by the share of the rich. The income equality dividend can be explained both due to changes in consumer behaviour on the demand side, and to institutional factors reflected by income distribution, on the supply side.

Income segregation's impact on local public expenditures: Evidence from municipalities and school districts, 1980–2010

Residential income segregation within local political jurisdictions has risen considerably since 1980 in the U.S. Despite this trend, and a growing body of research on wealth inequality and its impact on public sector size, income segregation's own influence over local public choice has not been thoroughly investigated. Using block-group income distributions for the years 1980, 1990, 2000, and 2010 income segregation is measured for individual U.S. municipalities and school districts, where the spatial distribution of income may carry the most political relevance. Estimates indicate that rising income segregation reduced per-capita spending growth considerably for both municipalities and school districts. These findings are robust to various model specifications and to the use of instrumental variables that adjust for the potential endogeneity of local income segregation. This evidence is consistent with the view that intra-jurisdictional income segregation undermines trust between community members and complicates collective action. However, a-spatial income inequality continues to be positively correlated with local public sector size, as suggested by recent empirical research.

Recent Advances in Empirical Land-Use Modeling

Data sets providing repeated observations of land use at fine spatial scales have enabled a new generation of land-use studies. In the past decade, these analyses have put increasing emphasis on empirical research designs that provide more convincing causal estimates. I review the use of instrumental variables, matching, difference-in-differences, regression discontinuity design, and randomized controlled trials in the recent land-use economics literature, exploring how new data have made possible the use of these research designs. I show that these estimators have produced different results than were obtained with traditional approaches and have provided new insights into important land-use policy issues such as additionality and spillover effects.

Identifying treatment effects in the presence of confounded types

In this paper, I consider identification of treatment effects when the treatment is endogenous. The use of instrumental variables is a popular solution to deal with endogeneity, but this may give misleading answers when the instrument is invalid. I show that when an (unobserved) instrument is invalid due to correlation with the first stage unobserved heterogeneity, a proxy for the instrument helps partially identify not only the local average treatment effect, but also the entire potential outcomes distributions for compliers. I exploit the fact that the distribution of the observed outcome in each group defined by the treatment and the instrument is a mixture of the distributions of interest. I write the identified set in the form of conditional moment inequalities, and provide an easily implementable inference procedure. Under some tail restrictions, the potential outcomes distributions are point-identified for compliers. Finally, I illustrate my methodology on data from the National Longitudinal Survey of Young Men to estimate returns to college using college proximity as a proxy for the instrument low college cost. I find that a college degree increases the average hourly wage of the compliers by 15%–30%.

Alcances de la aleatorización mendeliana para el control de confusores no observables en epidemiología

The Mendelian randomization is an epidemiologic method proposed to control for spurious associations in observational studies. These associations are commonly caused by confusion derived from social, environmental, and behavioral factors, which can be difficult to measure. Mendelian randomization is based on the selection of genetic variants that are used as instrumental variables that influence exposure patterns or are associated with an intermediate phenotype of the disease. The present work aims to discuss how to select the appropriate genetic variants as instrumental variables and to present methodological tools to deal with the limitations of this epidemiological method. The use of instrumental variables for modifiable exposures has the potential to mitigate the effects of common limitations, such as confusion, when robust genetic variants are chosen as instrumental variables.

The impact of genetic diversity of executive board directors on corporate misconduct: Evidence from US banks

We examine the link between the genetic diversities of executive board members and bank financial misconduct. The premise is that genetic diversity results in different perspectives, skills, and abilities that impact on the effectiveness of executive board guidance and monitoring, including with respect to misconduct. Employing a panel of US banks over 1998-2019 we find that adding directors from countries with different levels of genetic diversity is negatively associated with financial misconduct as measured by enforcements and class action litigation against banks by the main regulatory agencies. In addition, the relation between genetic diversity and misconduct is hump-shaped, suggesting that there a trade-off between the beneficial and detrimental effects of genetic diversity on board monitoring and guidance. These results are robust to controlling for bank specific variables, including other board characteristics, and to the use of instrumental variables.

U.S. POLITICAL CORRUPTION AND LOAN PRICING

AbstractUsing U.S. Department of Justice data on state‐level political corruption, we find that banks charge higher loan spreads (all‐in‐drawn spreads) to firms in states with higher corruption and that these effects are more pronounced for firms facing financial constraints but less pronounced for firms experiencing greater external monitoring. These results are robust to additional controls, alternative corruption measures, a measure of the lack of oversight of lobbyist activities, and the use of instrumental variables. Overall, our findings are consistent with the harmful corruption environment hypothesis, which states that banks charge higher loan spreads to firms in states with greater political corruption environments as these firms are susceptible to making suboptimal financial decisions to fend off rent‐seeking behavior.

Political Corruption and Cost of Equity

Using U.S. Department of Justice data on state-level political corruption, we find that, consistent with the Harmful Corruption Environment Hypothesis (HCEH), firms situated in states with higher levels of corruption incur higher costs of equity (CoEs). These results are robust for additional controls, propensity score matching, use of instrumental variables, exogenous shocks, and alternate measures for main dependent and primary independent research variables. Our study extends the stream of literature that investigates the influence of local ethical or trust factors on CoE and complements works by El Ghoul and colleagues and Gupta and colleagues.

Capital adjustment cost and inconsistency in income-based dynamic panel models with fixed effects

Abstract After the seminal work of Nickell (1981), a vast literature demonstrates the inconsistency of ‘conditional convergence’ estimator in income-based dynamic panel models with fixed effects when the time horizon (T) is short but the sample of countries (N) is large. Less attention is given to the economic root of inconsistency of the fixed effects estimator when T is also large. Using a variant of the Ramsey growth model with long-run adjustment cost of capital, we demonstrate that the fixed effects estimator of such models could be inconsistent when T is large. This inconsistency arises because of the long-run adjustment cost of capital which gives rise to a negative moving average coefficient in the error term. Income convergence will be thus overestimated. We theoretically characterize the order of this inconsistency. Our Monte Carlo simulation demonstrates that the size of the bias is substantial and it is greater in economies with higher capital adjustment costs. We show that the use of instrumental variables that take into account the presence of the negative moving average term in the error will overcome this bias.

Family Discourse Right Conflict and Asset Allocation

ABSTRACT This paper uses the data of the China Household Finance Survey(CHFS) 2013 to research the impact of family discourse right conflict on household asset allocation. The results show that discourse right conflict has a significant impact on family investment. In addition, the conclusion that households with a lower degree of discourse right conflict will hold less risky assets remains robust after replacement of the indicators of the conflict variable, changing of the model settings and the use of instrumental variables. We think there may exist an impact mechanism between discourse right conflict and asset allocation. With the increase in the level of discourse right conflict, family members may change their risk preferences because they want to gain higher family status by earning more.

Analysts’ Cash Flow Forecast Accuracy and Recommendation Profitability

We examine whether analysts who supplement their earnings forecasts with more accurate cash flow forecasts generate more profitable recommendations. Previous research using consensus cash flow forecasts, or the issuance of cash flow forecasts fails to document a significant relation between analysts’ cash flow forecasts and their stock recommendation performance. We argue that analysts’ cash flow forecasts differ in quality and hypothesize that the relative accuracy of an individual analyst’s cash flow forecasts is positively associated with the profitability of that analyst’s stock recommendations. We find that when analysts issue both earnings and cash flow forecasts for a firm, cash flow forecast accuracy predicts stock recommendation profitability even after controlling for earnings forecast accuracy. Our findings are robust to the use of instrumental variables and the use of recommendation change in the analysis. Whether analysts also issue sales forecasts and/or long-term growth forecasts does not affect the results.

Experimentation and Startup Performance: Evidence from A/B Testing

Recent work argues that experimentation is the appropriate framework for entrepreneurial strategy. We investigate this proposition by exploiting the time-varying adoption of A/B testing technology, which has drastically reduced the cost of experimentally testing business ideas. This paper provides the first evidence on how digital experimentation affects a large sample of high-technology startups using data that tracks their growth, technology use, and product launches. Despite its prominence in the business press, relatively few firms have adopted A/B testing tools. However, among those that do, we find increased performance on several critical dimensions, including page views, code changes, and new product features. These results are robust to our use of instrumental variables and synthetic control models, where we leverage Google’s launch of a new A/B testing tool. Our results also show that the value of A/B testing increases over time and is positively related to tail outcomes, with adopt- ing firms both scaling and failing faster. Our results inform the emerging literature on entrepreneurial strategy and illustrate the growing importance of digitization and data-driven decision-making.

Closed-Form Algebraic Solutions for Angle-of-Arrival Source Localization With Bayesian Priors

This paper considers the problem of angle-of-arrival (AOA) source localization with Bayesian priors. Extending the so-called pseudolinear estimator (PLE) for AOA localization to the Bayesian framework results in the Bayesian PLE (BPLE). Similar to the PLE, the BPLE suffers from severe bias problems despite enjoying the inherent stability and computational simplicity of a closed-form least-squares estimator. Bayesian priors make bias compensation for the BPLE much more challenging than the PLE. We present a bias analysis for the BPLE from which two direct bias compensation methods are developed. As an alternative to direct bias compensation, a new approach with superior estimation performance and low computational complexity is then proposed by exploiting jointly the use of instrumental variables and 2-D unscented (or cubature) transformation. The resulting estimator, referred to as the bias-compensated Bayesian weighted instrumental variable estimator, enjoys the desirable properties of negligible bias and mean-squared error on par with the computationally demanding an iterative maximum a posteriori estimator. The comparative simulation studies are presented to corroborate the performance advantages of the proposed estimators.

E-commerce as a Potential New Engine for Growth in Asia

The use of e-commerce around the world has accelerated in recent years, with Asia, led by China, spearheading the rise. Using cross-country enterprise survey data, this paper shows that firms engaged in e-commerce have higher productivity and generate a larger share of their revenues from exports than other firms. This is particularly true in Asia, where firms have 30 percent higher productivity and generate about 50 percent more of their revenues from exports. The results presented in this paper are robust to the use of instrumental variables, which highlight possible larger effects of e-commerce on Asian productivity and exports when essential elements are in place for its effective use, such as reliable electricity, telecommunication, and transport infrastructure. Despite the rapid growth of e-commerce in recent years, gaps persist in digital infrastructure and legislation, preventing many Asian countries from fully reaping the potential benefits of e-commerce.

The relationship between the Chinese ‘going out’ strategy and international trade

Abstract This study is the first to estimate a system of simultaneous gravity equations for Chinese exports, imports and foreign direct investment (FDI) using a sample of 167 countries over the period 2003–2012. The main results indicate that trade and outward FDI are complementary. In particular, the authors show that outward Chinese FDI is related to higher exports and imports and that China trades more with countries hosting Chinese FDI. Results are also robust to the use of instrumental variables. Therefore, Chinese investment seems to foster trade.

Growth takeoffs and trade margins: a quantile regression approach

Recent studies find evidence that economic growth takeoffs are strongly positively correlated with large increases in total aggregate exports. These studies often analyze the average relationship between these economic growth takeoffs and total trade flows. Therefore, these studies are somewhat limited because the average relationship masks the potential heterogeneous impact of takeoffs on different levels of the trade distribution. Moreover, endogeneity issues due to a possible bidirectional relationship between trade openness and economic growth also need to be considered. In this paper, we use two quantile regression models—both models allow for panel data and fixed effects, while one also makes use of instrumental variables to address endogeneity concerns. As such, we investigate whether takeoffs affect countries equally when they trade at the lower end of the distribution (countries that trade less with each other) compared to those at the higher end of the distribution (countries that trade more with each other). Our results suggest that economic growth takeoffs affect countries at the lower end of the export distribution (10th percentile) significantly more compared to higher end of the export distribution (90th percentile). Moreover, these results are driven entirely by the extensive margin. We also find evidence that takeoffs affect countries with a lower level of income much more compared to those with a high level of income, with this result also extending to the extensive margin.

Nonparametric Estimates of Demand in the California Health Insurance Exchange

We estimate the demand for health insurance in the California Affordable Care Act marketplace (Covered California) without using parametric assumptions about the unobserved components of utility. To do this, we develop a computational method for constructing sharp identified sets in a nonparametric discrete choice model. The model allows for endogeneity in prices (premiums) and for the use of instrumental variables to address this endogeneity. We use the method to estimate bounds on the effects of changing premium subsidies on coverage choices, consumer surplus, and government spending. We find that a $10 decrease in monthly premium subsidies would cause between a 1.6% and 7.0% decline in the proportion of low-income adults with coverage. The reduction in total annual consumer surplus would be between $63 and $78 million, while the savings in yearly subsidy outlays would be between $238 and $604 million. Comparable logit models yield price sensitivity estimates towards the lower end of the bounds.

Historical Grievances and War Dynamics: Old Land Conflicts As a Cause of Current Forced Displacements in Colombia

We explore the role of unfairly resolved land conflicts on the long-run dynamics of Colombia’s war. We claimed that inadequately or unfairly resolved land conflicts can explain not only the emergence of rebel groups but also their repertoires and targets of violence—notably forced displacements. The identification strategy used combines the use of instrumental variables—distance of the municipality to the railroad in 1900- with a methodology of neighbor-pair fixed effects to test the causal relation between old land conflicts and recent forced displacements. Moreover, we use mediation analysis to test the mechanisms that we hypothesize as links between the independent and dependent variables—presence of armed groups and unfair land reform allocations. Our results show that, indeed, early 20th century land conflicts and recent forced displacements are causally linked, and their connection is mainly produced through the mediation of unfair land reform allocations and the presence of violent armed groups. We argue that unfair land reforms harbor the formation of peasant-based rebel groups and such groups promote displacements in the very same places that experienced old land conflict.

Identifying Treatment Effects in the Presence of Confounded Types

In this paper, I consider identification of treatment effects when the treatment is endogenous. The use of instrumental variables is a popular solution to deal with endogeneity, but this may give misleading answers when the instrument is invalid. I show that when an (unobserved) instrument is invalid due to correlation with the first stage unobserved heterogeneity, a proxy for the instrument helps partially identify not only the local average treatment effect, but also the entire potential outcomes distributions for compliers. I exploit the fact that the distribution of the observed outcome in each group defined by the treatment and the instrument is a mixture of the distributions of interest. I write the identified set in the form of conditional moment inequalities, and provide an easily implementable inference procedure. Under some tail restrictions, the potential outcomes distributions are point-identified for compliers. Finally, I illustrate my methodology on data from the National Longitudinal Survey of Young Men to estimate returns to college using college proximity as a proxy for the instrument low college cost. I find that a college degree increases the average hourly wage of the compliers by 15-30%.

Peer and parental influence in academic performance and alcohol use

This paper analyzes the interactions between peers and parents in the development of adolescent's cognitive skills, proxied by the academic performance of adolescents, and the predisposition to risky behavior, proxied by the frecuency of alcohol consumption. We estimate a technology of skill formation that identifies peer effects on the basis of quasi-random assignment of students across classes, time varying data, and the use of instrumental variables. We find that both peer and parental socialization efforts have a positive influence over adolescents’ academic skills, and that these effects are complementary: as parents invest more, peers amplify their effects. Our results suggest that peer effects are heterogenous (boys, popular students and high achievers are more likely to be influenced by peers academic abilities). On the other hand, we do not find robuts evidence of peer effects on the frecuency of alcohol consumption. We do find, however, that parental investment decreases alcohol consumption.