Quantitative Economics May 2019 is now online
TABLE OF CONTENTS, May 2019, Volume 10, Issue 2
Full Issue
Articles
Abstracts follow the listing of articles.
Jump factor models in large cross‐sections
Jia Li, Viktor Todorov, George Tauchen
On optimal inference in the linear IV model
Donald W. K. Andrews, Vadim Marmer, Zhengfei Yu
A more powerful subvector Anderson Rubin test in linear instrumental variables regression
Patrik Guggenberger, Frank Kleibergen, Sophocles Mavroeidis
Identification of a nonseparable model under endogeneity using binary proxies for unobserved heterogeneity
Benjamin Williams
The long run health consequences of rural‐urban migration
Janna E. Johnson, Evan J. Taylor
Uncertainty about future income: Initial beliefs and resolution during college
Yifan Gong, Todd Stinebrickner, Ralph Stinebrickner
The right stuff? Personality and entrepreneurship
Barton H. Hamilton, Nicholas W. Papageorge, Nidhi Pande
Optimal unemployment insurance with monitoring
Ofer Setty
Financial frictions, trends, and the great recession
Pablo A. Guerron‐Quintana, Ryo Jinnai
Communication and behavior in organizations: An experiment
Piotr Evdokimov, Umberto Garfagnini
------------------
Jump factor models in large cross‐sections
Jia Li, Viktor Todorov, George Tauchen
Abstract
We develop tests for deciding whether a large cross‐section of asset prices obey an exact factor structure at the times of factor jumps. Such jump dependence is implied by standard linear factor models. Our inference is based on a panel of asset returns with asymptotically increasing cross‐sectional dimension and sampling frequency, and essentially no restriction on the relative magnitude of these two dimensions of the panel. The test is formed from the high‐frequency returns at the times when the risk factors are detected to have a jump. The test statistic is a cross‐sectional average of a measure of discrepancy in the estimated jump factor loadings of the assets at consecutive jump times. Under the null hypothesis, the discrepancy in the factor loadings is due to a measurement error, which shrinks with the increase of the sampling frequency, while under an alternative of a noisy jump factor model this discrepancy contains also nonvanishing firm‐specific shocks. The limit behavior of the test under the null hypothesis is nonstandard and reflects the strong‐dependence in the cross‐section of returns as well as their heteroskedasticity which is left unspecified. We further develop estimators for assessing the magnitude of firm‐specific risk in asset prices at the factor jump events. Empirical application to S&P 100 stocks provides evidence for exact one‐factor structure at times of big market‐wide jump events. Factor model panel high‐frequency data jumps semimartingale specification test stochastic volatility C51 C52 G12
---
On optimal inference in the linear IV model
Donald W. K. Andrews, Vadim Marmer, Zhengfei Yu
Abstract
This paper considers tests and confidence sets (CSs) concerning the coefficient on the endogenous variable in the linear IV regression model with homoskedastic normal errors and one right‐hand side endogenous variable. The paper derives a finite‐sample lower bound function for the probability that a CS constructed using a two‐sided invariant similar test has infinite length and shows numerically that the conditional likelihood ratio (CLR) CS of Moreira (2003) is not always “very close,” say 0.005 or less, to this lower bound function. This implies that the CLR test is not always very close to the two‐sided asymptotically‐efficient (AE) power envelope for invariant similar tests of Andrews, Moreira, and Stock (2006) (AMS). On the other hand, the paper establishes the finite‐sample optimality of the CLR test when the correlation between the structural and reduced‐form errors, or between the two reduced‐form errors, goes to 1 or −1 and other parameters are held constant, where optimality means achievement of the two‐sided AE power envelope of AMS. These results cover the full range of (nonzero) IV strength. The paper investigates in detail scenarios in which the CLR test is not on the two‐sided AE power envelope of AMS. Also, theory and numerical results indicate that the CLR test is close to having the greatest average power, where the average is over a specified grid of concentration parameter values and over a pair of alternative hypothesis values of the parameter of interest, uniformly over all such pairs of alternative hypothesis values and uniformly over the correlation between the structural and reduced‐form errors. Here, “close” means 0.015 or less for k ≤ 20, where k denotes the number of IVs, and 0.025 or less for 0 < k ≤ 40. The paper concludes that, although the CLR test is not always very close to the two‐sided AE power envelope of AMS, CLR tests and CSs have very good overall properties. Conditional likelihood ratio test confidence interval infinite length linear instrumental variables optimal test weighted average power similar test C12 C36
---
A more powerful subvector Anderson Rubin test in linear instrumental variables regression
Patrik Guggenberger, Frank Kleibergen, Sophocles Mavroeidis
Abstract
We study subvector inference in the linear instrumental variables model assuming homoskedasticity but allowing for weak instruments. The subvector Anderson and Rubin (1949) test that uses chi square critical values with degrees of freedom reduced by the number of parameters not under test, proposed by Guggenberger, Kleibergen, Mavroeidis, and Chen (2012), controls size but is generally conservative. We propose a conditional subvector Anderson and Rubin test that uses data‐dependent critical values that adapt to the strength of identification of the parameters not under test. This test has correct size and strictly higher power than the subvector Anderson and Rubin test by Guggenberger et al. (2012). We provide tables with conditional critical values so that the new test is quick and easy to use. Application of our method to a model of risk preferences in development economics shows that it can strengthen empirical conclusions in practice. Asymptotic size linear IV regression subvector inference weak instruments C12 C26
---
Identification of a nonseparable model under endogeneity using binary proxies for unobserved heterogeneity
Benjamin Williams
Abstract
In this paper, I study identification of a nonseparable model with endogeneity arising due to unobserved heterogeneity. Identification relies on the availability of binary proxies that can be used to control for the unobserved heterogeneity. I show that the model is identified in the limit as the number of proxies increases. The argument does not require an instrumental variable that is excluded from the outcome equation nor does it require the support of the unobserved heterogeneity to be finite. I then propose a nonparametric estimator that is consistent as the number of proxies increases with the sample size. I also show that, for a fixed number of proxies, nontrivial bounds on objects of interest can be obtained. Finally, I study two real data applications that illustrate computation of the bounds and estimation with a large number of items. Nonseparable model unobserved heterogeneity latent variable binary measurement error C14 C35 C38
---
The long run health consequences of rural‐urban migration
Janna E. Johnson, Evan J. Taylor
Abstract
Rural‐urban migration is an integral part of the structural transformation as societies move from a traditional agricultural economy to a modern economy. This process has many potential consequences for migrants. Our study focuses on the lifetime health effects of the large mid‐20th century migration out of rural U.S. Northern Great Plains states, primarily to urban locations in the West and Midwest. An analysis of marginal treatment effects (MTEs) shows that (a) migrants are positively selected, and (b) the causal impact of migration is decreased longevity. Our evidence suggests that elevated mortality among migrants is linked to increased smoking and alcohol consumption. Rural‐urban migration mortality marginal treatment effects C31 I12 R23
---
Uncertainty about future income: Initial beliefs and resolution during college
Yifan Gong, Todd Stinebrickner, Ralph Stinebrickner
Abstract
We use unique data from the Berea Panel Study to characterize how much earnings uncertainty is present for students at college entrance and how quickly this uncertainty is resolved. We characterize uncertainty using survey questions that elicit the entire distribution describing one's beliefs about future earnings. Taking advantage of the longitudinal nature of the expectations data, we find that roughly two‐thirds of the income uncertainty present at the time of entrance remains at the end of college. Taking advantage of a variety of additional survey questions, we provide evidence about how the resolution of income uncertainty is influenced by factors such as college GPA and college major, and also examine why much income uncertainty remains unresolved at the end of college. This paper also contributes to a literature interested in understanding the relative importance of uncertainty and heterogeneity in determining observed earnings distributions. Income uncertainty uncertainty resolution heterogeneity expectations data J30
---
The right stuff? Personality and entrepreneurship
Barton H. Hamilton, Nicholas W. Papageorge, Nidhi Pande
Abstract
We construct a structural model of entry into self‐employment to evaluate the impact of policies supporting entrepreneurship. Previous work has recognized that workers may opt for self‐employment due to the nonpecuniary benefits of running a business and not necessarily because they are good at it. Other literature has examined how socio‐emotional skills, such as personality traits, affect selection into self‐employment. We link these two lines of inquiry. The model we estimate captures three factors that affect selection into self‐employment: credit constraints, relative earnings, and preferences. We incorporate personality traits by allowing them to affect sector‐specific earnings as well as preferences. The estimated model reveals that the personality traits that make entrepreneurship profitable are not always the same traits driving people to open a business. This has important consequences for entrepreneurship policies. For example, subsidies for small businesses do not attract talented‐but‐reluctant entrepreneurs, but instead attract individuals with personality traits associated with strong preferences for running a business and low‐quality business ideas. Entrepreneurship personality socioemotional skills latent factors J23 J24 J31 J32
---
Optimal unemployment insurance with monitoring
Ofer Setty
Abstract
I model job‐search monitoring in the optimal unemployment insurance framework, in which job‐search effort is the worker's private information. In the model, monitoring provides costly information upon which the government conditions unemployment benefits. Using a simple one‐period model with two effort levels, I show analytically that the monitoring precision increases and the utility spread decreases if and only if the inverse of the worker's utility in consumption has a convex derivative. The quantitative analysis that follows extends the model by allowing a continuous effort and separations from employment. That analysis highlights two conflicting economic forces affecting the optimal precision of monitoring with respect to the generosity of the welfare system: higher promised utility is associated not only with a higher cost of moral hazard, but also with lower effort and lower value of employment. The result is an inverse U‐shaped precision profile with respect to promised utility. Unemployment insurance optimal contracts moral hazard job‐search monitoring D82 E24 J64 J65
---
Financial frictions, trends, and the great recession
Pablo A. Guerron‐Quintana, Ryo Jinnai
Abstract
We study the causes behind the shift in the level of U.S. GDP following the Great Recession. To this end, we propose a model featuring endogenous productivity à la Romer and a financial friction à la Kiyotaki–Moore. Adverse financial disturbances during the recession and the lack of strong tailwinds post‐crisis resulted in a severe contraction and the downward shift in the economy's trend. Had financial conditions remained stable during the crisis, the economy would have grown at its average growth rate. From a historical perspective, the Great Recession was unique because of the size and persistence of adverse shocks, and the lackluster performance of favorable shocks since 2010. Endogenous productivity financial friction great recession liquidity shocks trend shift E22 E32 E37 G01 04
---
Communication and behavior in organizations: An experiment
Piotr Evdokimov, Umberto Garfagnini
Abstract
We design a laboratory experiment to study behavior in a multidivisional organization. The organization faces a trade‐off between coordinating its decisions across the divisions and meeting division‐specific needs that are known only to the division managers, who can communicate their private information through cheap talk. While the results show close to optimal communication, we also find systematic deviations from optimal behavior in how the communicated information is used. Specifically, subjects' decisions show worse than predicted adaptation to the needs of the divisions in decentralized organizations and worse than predicted coordination in centralized organizations. We show that the observed deviations disappear when uncertainty about the divisions' local needs is removed and discuss the possible underlying mechanisms. Communication coordination decentralization experiment C70 C92 D03
