Construct Instrumental Variables Research Articles

Causality between depression and ankylosing spondylitis in a European population: Results from a Mendelian randomization analysis.

The aim of this study was to explore the application of Mendelian randomization (MR) Egger and inverse variance weighted (IVW) in a causal effect on depression and ankylosing spondylitis (AS). Instrumental variables (IVs) were determined using genome-wide association studies. The 2-sample MR analysis was conducted by MR Egger to test the causal effect between depression and AS. The pleiotropy of potential instrumental variables was evaluated. The results of MR Egger and IVW were further compared. A total of 3 single nucleotide polymorphisms as the construct IVs were included. IVW results showed a significant causal effect between depression and AS (P < .001). Depression could promote the risk of AS (odds ratio = 1.060, 95% confidence interval: 1.026-1.094). However, the MR Egger showed no causal effect (P = .311). Heterogeneity statistics suggested that no heterogeneity was existed (P > .05). It was also suggested that there was no horizontal pleiotropy in IVs (MR Egger intercept: -0.0004, P = .471). Reverse MR analysis suggested that there was no causal effect between AS and depression (P > .05). Gene expression quantitative trait locus (QTLs) suggested that rs2517601 and RNF39 were positively correlated (beta = 1.066, P < .001). Depression may be one of the causes of AS by MR analysis in a European population. We can estimate the causal effect based on IVW when horizontal pleiotropy is very tiny.

Image features and demand in the sharing economy: A study of Airbnb

Peer-to-peer marketplaces and the sharing economy are reshaping many markets. Airbnb is attractive to many customers because of its non-standardized, diverse selections and superior value over hotels, but Airbnb properties also come with more uncertainty than hotels. Images of the property can help resolve uncertainty. This study focuses on the background image (the image displayed in search results) and examines how the background image’s content features (living room, bedroom, and interior design) and aesthetic features (clarity, brightness, and contrast) affect the booking rate during a 16-night end-of-year holiday period. The authors develop a model that includes the amount of visual information, property characteristics, and host characteristics, and correct for endogeneity by using a pair of markets (New York City and San Francisco) to calculate propensity scores and construct instrumental variables. The results show that having a background image that features the living room and shows more interior design elements increases the booking rate, while featuring a bedroom decreases it. The effects of the content features are larger than the effects of the aesthetic features, and the effects are economically significant: for example, the effect of featuring the living room in the background image translates into a 35% increase in the booking rate, which amounts to $728 more revenue during the holiday period.

Hazed and Confused: The Effect of Air Pollution on Dementia

Abstract We study whether long-term cumulative exposure to airborne small particulate matter (PM2.5) affects the probability that an individual receives a new diagnosis of Alzheimer's disease or related dementias. We track the health, residential location, and PM2.5 exposures of Americans aged sixty-five and above from 2001 through 2013. The expansion of Clean Air Act regulations led to quasi-random variation in individuals’ subsequent exposures to PM2.5. We leverage these regulations to construct instrumental variables for individual-level decadal PM2.5 that we use within flexible probit models that also account for any potential sample selection based on survival. We find that a 1 µg/m3 increase in decadal PM2.5 increases the probability of a new dementia diagnosis by an average of 2.15 percentage points (pp). All else equal, we find larger effects for women, older people, and people with more clinical risk factors for dementia. These effects persist below current regulatory thresholds.

The impacts of highways on firm size distribution: Evidence from China

AbstractThis paper examines the impact of highway construction on the size distribution of manufacturing firms using Chinese firm and highway network data. I construct instrumental variables using historical and counterfactual road approaches to identify the causal effects of highway construction. I find that highway construction can increase firm size dispersion and market concentration. Areas near highways have a higher number of firms, more outsourcing activities, and more large firms; while areas further from highways have a higher share of new entry firms.

Instrumental‐variable estimation of exponential‐regression models with two‐way fixed effects with an application to gravity equations

SummaryThis paper introduces instrumental‐variable estimators for exponential‐regression models that feature two‐way fixed effects. These techniques allow us to develop a theory‐consistent approach to the estimation of cross‐sectional gravity equations that can accommodate the endogeneity of policy variables. We apply this approach to a dataset in which the policy decision of interest is the engagement in a free‐trade agreement. We explore ways to exploit the transitivity observed in the formation of trade agreements to construct instrumental variables with considerable predictive ability. Within a bilateral model, the use of these instruments has strong theoretical foundations. We obtain point estimates of the partial effect of a preferential‐trade agreement on trade volume that range between 20% and 30% and find no statistical evidence of endogeneity.

Blockchain and Firm Total Factor Productivity: Evidence from China

This paper creatively constructs blockchain development indicators using geographical characteristics to investigate the influence of blockchain development on the total factor productivity of listed companies. Our findings reveal that local blockchain development can significantly promote the improvement of the firms’ total factor productivity. To alleviate endogeneity, this paper combines exogenous policy and geographic distance to construct instrumental variables. Moreover, the positive influence is more pronounced in non-SOEs, non-excess capacity industries, and samples with high initial productivity. After the robustness test, the results are still valid. The aforementioned results provide practical implications for Chinese listed companies to lay out digital business.

The Impact of City Cluster Development on the Inter-City Disparity: Evidence From China

City clusters are important carriers for China’s future urbanization and vital areas for promoting coordinated regional development and common prosperity. The authors first divide the city clusters into Types I, II, and III according to the characteristics of their development stages, and measure the degree of development of city clusters by using city-level night time light (NTL) data and inter-city commuting time, and empirically identify the types of China’s city clusters. On that basis, the authors examine the influence of city cluster development on the inter-city disparity from a perspective of urban agglomeration. The results show that the development of Type I city clusters can lead to the widest inter-city disparity, and that of Type II city clusters ranks the second place in this regard, while that of Type III city clusters can narrow the inter-city disparity. Furthermore, using the 268-year historical geographic data of the Qing Dynasty, the authors construct instrumental variables to address the endogeneity issues, and find that the above conclusions are robust. This study also finds that the factor allocation for the development of Types I and II city clusters is oriented towards labor, while that for Type III city clusters is oriented towards capital. Besides, different types of city clusters vary in the extent of correcting factor misallocation for their development. The differences in the orientation of factor allocation and the extent of factor misallocation correction constitute the main mechanism for the heterogeneous influence of city cluster development on the inter-city disparity. China’s city cluster development strategy should adopt a category-based approach to avoid homogenization, conform to the objective laws, and make greater use of central cities’ spatial spillover effects.

Association of psoriasis with risk of COVID-19: A 2-sample Mendelian randomization study

With the growing pandemic of COVID-19, psoriasis has been reported to be linked with COVID-19 from genetic and epidemiological perspectives, especially in patients receiving systemic treatments.1-3 However, traditional epidemiology is inevitably affected by confounding bias. Mendelian randomization (MR) is an approach based on genome-wide association studies to construct instrumental variables (IVs) and can effectively control the confounding bias of observational studies. IVs refer to variables that only affect the outcome through risk factors, and MR uses single-nucleotide polymorphisms as IVs to identify risk factors.

The Health Effect of the Number of Children on Chinese Elders: An Analysis Based on Hukou Category.

Based on the 2018 China Health and Retirement Longitudinal Study (CHARLS 2018), from the perspective of urban-rural disparity, this paper investigates how fertility affects Chinese elders' health. We exploit the enactment of the one-child policy in 1979 to construct instrumental variables capturing the health effect of having only one child rather than multiple children. The empirical results show that the health condition of rural elders having only one child is worse than elders having multiple children, while the negative health effect of lower fertility becomes statistically insignificant for urban elderly parents. After considering the selection on both levels and gains, the results are still robust in marginal treatment effect (MTE) estimation. We investigate the potential mechanism in four ways, the results suggest that having only one child instead of multiple children depresses the upstream intergenerational transfer payments more for rural parents; ameliorates offspring's educational attainment more for urban parents; improves housing conditions more for urban elders; and decreases the visit frequency of children to both urban and rural parents. Our findings have important implications, in the context of increasing population aging, the urban-rural inequality caused by the hukou system has been magnified by the declining fertility rate. The Chinese government should pay more attention to rural elders with only one child, and more public-funded socioeconomic resources are needed for one-child parents in rural areas to improve their health. Moreover, the empirical results also imply that urbanization in China may be able to soften the health deterrent effect of lower fertility.

Network externalities and consolidation in the dating website industry

Network externalities—i.e., effects on consumers' alternative-specific pay-offs of other consumers' decisions—arise in many economic settings. This paper considers the identification of discrete-choice models featuring network externalities and then uses such a model to analyze how network externalities in the market for dating websites in the United States contribute to the effects of consolidation in that industry. My identification analysis shows that discrete-choice models with network externalities are generally not identified with market-level data alone, but that microdata allows the researcher to construct instrumental variables that identify the model. Using rich online browsing data, I estimate a model of consumer choice of dating website and find evidence of considerable network externalities. In my preferred specification, an inframarginal user of a site values a 10% increase in the site's usership at $11.11/month, which is about one third of the most popular site's price. I use my estimates to assess whether the observed extent of network externalities imply that increased market concentration would benefit consumers. I find that welfare losses from decreased variety and increased prices outweigh the gains from network externalities associated with a move from the observed market structure to monopoly. Platform differentiation may explain why platform integration has not followed consolidation in the dating websites industry.

Remittances channels and the physical growth of Honduran children

Ensuring the availability of food and other resources for young children is important for sustaining physical growth. We examine the role of remittances and its associated implications in determining heights and weights of 4459 children aged 0–5 years in Honduras in 2004. To address the endogeneity problem with household remittance receipt, we take advantage of the timing of Hurricane Mitch in 1998 to construct instrumental variables that are exogenously related to migration decisions made before children included in the later survey were conceived. We find that children are significantly taller and heavier for their age and gender in households receiving remittances. Further investigation of household spending indicates significant changes in food purchases and dietary diversity. Households receiving remittances are more likely to include fish, fruits, and meats in their diets. Additional findings also indicate that households receiving remittances spend absolutely more on food, health care, education, and durable goods. Overall, the findings provide strong evidence that remittances change household consumption and increase children’s body sizes.

Learning to Trust Flu Shots: Quasi-Experimental Evidence from the 2009 Swine Flu Pandemic.

This paper studies consumer learning in influenza vaccination decisions. We examine consumer learning in influenza vaccine demand within a reduced form instrumental variable framework that exploits differences in risk characteristics of different influenza viruses as a natural experiment to distinguish the effects of learning based on previous influenza vaccination experiences from unobserved heterogeneity. The emergence of a new virus strain (influenza A H1N1/09) during the 2009 'Swine flu' pandemic resulted in two different vaccines being recommended for distinct population subgroups with some people, who were not usually targeted by seasonal vaccination programs, being specifically recommended for the new Swine flu vaccine. We use these differences in vaccination targeting to construct instrumental variables for estimating the effect of past influenza vaccination experiences on the demand for pandemic vaccine. We find large causal effects of previous seasonal vaccination on pandemic vaccination. Causal effects of past influenza vaccination experiences on perceived vaccination safety are likely to be an important pathway linking past vaccination experiences with future vaccine uptake. Our results suggest a significant role of learning in vaccination decisions. Current efforts to expand seasonal vaccination may thus have potentially important long-term effects on future influenza vaccination levels and pandemic preparedness. Copyright © 2016 John Wiley & Sons, Ltd.

Effects of Agricultural Productivity Shocks on Female Labor Supply: Evidence from the Boll Weevil Plague in the US South

In the beginning of the 1890s, counties located in the Cotton Belt of the American South were hit by an agricultural plague, the Boll Weevil, that adversely affected cotton production and hence the demand for labor. We use variation in the incidence of the Boll Weevil multiplied with counties’ initial cotton share to construct instrumental variables estimates of the labor supply curve. Controlling for county and state-by-time fixed effects, we find a significant positive response of labor supply to changes in labor income. The effect is particularly large for females, consistent with evidence that females had a comparative advantage in picking cotton.

The Impact of Taxes on Charitable Giving: Empirical Evidence from the Korea Labor and Income Panel Study

Households’ or individuals’ decision regarding charitable giving may differby type of recipient of the gift. In light of the relative paucity of empirical research on the impact of tax incentives on charitable giving outside Western countries, empirical research on this topic in South Korea is valuable in order to compare effects across difference tax regimes and in different institutional environments. We use the Korean Labor and Income Panel Study (KLIPS), whose panel structure helps alleviate the omitted variable bias that has often appeared in previous literature using cross-sectional data. This study aims to perform a robust estimation of tax price and income elasticities for charitable contributions in South Korea. First, we use exogenous changes in tax rates resulting from Korean Tax Reform to construct instrumental variables (IVs) for the change in the price of giving. Two tests are undertaken to determine whether the IVs are weak or not: a size-corrected test of a weak IV robust inference for the linear instrumental variable model with autocorrelation and heteroscedasticity recently devised by Finlay and Magnusson; and the LIML CUE-GMM estimation. We find that our instruments are not weak. Following Smith and Blundell, and Rivers and Vuong , we then estimate the random effect (RE) Tobit Model using acontrol function based on the IVs. Using the procedure developed by Mundlak, we estimated a fixed effect model from the RE Tobit model. The tax price and income elasticities from the pseudo-fixed effect Tobit model are found to be significant and the magnitudes are similar to those from the GMM fixed effect and CUE-GMM models. To investigate additional features of the conditional distribution of charitable giving in South Korea, we use the Censored Quantile regression with instrumental variables (CQIV) recently proposed by Chernozhukov, Fernandex-Val, and Kowalski. These estimate indicate that the price elasticity of charitable giving is very heterogenous among donors, while income has a quite uniform and positive effect over the whole range of the giving distribution significantly.

The Causal Effects of an Industrial Policy

Business support policies designed to raise productivity and employment are common worldwide, but rigorous micro-econometric evaluation of their causal effects is rare. We exploit multiple changes in the area-specific eligibility criteria for a major program to support manufacturing jobs (Regional Selective Assistance). Area eligibility is governed by pan-European state aid rules which change every seven years and we use these rule changes to construct instrumental variables for program participation. We match two decades of UK panel data on the population of firms to all program participants. IV estimates find positive program treatment effect on employment, investment and net entry but not on TFP. OLS underestimates program effects because the policy targets underperforming plants and areas. The treatment effect is confined to smaller firms with no effect for larger firms (e.g. over 150 employees). We also find the policy raises area level manufacturing employment mainly through significantly reducing unemployment. The positive program effect is not due to substitution between plants in the same area or between eligible and ineligible areas nearby. We estimate that per job of the program was only $6,300 suggesting that in some respects investment subsidies can be cost effective.

Economic growth, size of the agricultural sector, and urbanization in Africa

This paper uses variations in international commodity prices and rainfall to construct instrumental variables estimates of the within-country effect that changes in the size of the agricultural sector and GDP per capita growth have on the urbanization rate. For a panel of 41 African countries during the period 1960–2007, the paper’s three main findings are that: (i) decreases in the share of agricultural value added lead to a significant increase in the urbanization rate; (ii) conditional on changes in the share of agricultural value added GDP per capita growth does not significantly affect the urbanization rate; (iii) increases in the urbanization rate had a significant negative average effect on GDP per capita growth.

An instrumental variables approach to estimating tax revenue elasticities: Evidence from Sub-Saharan Africa

This paper exploits the significant response of real GDP growth of Sub-Saharan African countries to exogenous international commodity price and rainfall shocks to construct instrumental variables estimates of the tax revenue elasticity IV estimates yield that a 1% increase in GDP increases tax revenues by up to 2.5%.

Long-term Consequences of Vietnam-Era Conscription: New Estimates Using Social Security Data

We use the draft lottery to construct instrumental variables (IV) estimates of the impact of Vietnam-era military service on veterans' Social Security (SSA) earnings through 2007. We also use SSA data to construct IV estimates for employment (as measured by an indicator for positive earnings) and disability status (as measured by an indicator for social security disability program application). New findings for recent years show surprisingly rapid convergence in veteran and nonveteran earnings: by the early 1990s, there was no longer a substantial Vietnam-era conscription penalty. The IV estimates also show no effect on employment or disability rates.

Constituents’ Responses to Congressional Roll‐Call Voting

Do citizens hold their representatives accountable for policy decisions, as commonly assumed in theories of legislative politics? Previous research has failed to yield clear evidence on this question for two reasons: measurement error arising from noncomparable indicators of legislators’ and constituents’ preferences and potential simultaneity between constituents’ beliefs about and approval of their representatives. Two new national surveys address the measurement problem directly by asking respondents how they would vote and how they think their representatives voted on key roll‐call votes. Using the actual votes, we can, in turn, construct instrumental variables that correct for simultaneity. We find that the American electorate responds strongly to substantive representation. (1) Nearly all respondents have preferences over important bills before Congress. (2) Most constituents hold beliefs about their legislators’ roll‐call votes that reflect both the legislators’ actual behavior and the parties’ policy reputations. (3) Constituents use those beliefs to hold their legislators accountable.

Wage shocks and consumption variability in Mexico during the 1990s

This paper presents evidence on the relationship between shocks to relative male wages and changes in household consumption in Mexico during the 1990s decade, which is a period characterized by high volatility. Apart from performing analysis of this type for Mexico for the first time, the paper has mainly two contributions. The first is the use of alternative data sources to construct instrumental variables for wages. The second is to examine differences across four consumption categories: nondurable goods, durable goods, education, and health. Our results for nondurable goods consumption reject the hypothesis that Mexican households are able to ensure idiosyncratic risk. For the comparisons across consumption categories, the conclusion is that households in Mexico tend to react to temporary shocks by contracting the consumption of goods that represents longer run investment in human capital, which makes them more vulnerable in the future.