论文检索

We propose a practical and robust method for making inference on average treatment effects estimated by synthetic controls. We develop a K-fold cross-fitting procedure for bias correction. To avoid the difficult estimation of the long-run variance, inference is based on a self-normalized t-statistic, which has an asymptotically pivotal t-distribution. Our t-test is easy to implement, provably robust against misspecification, and valid with stationary and nonstationary data.…

We develop methods for inference following sequential experiments by studying the asymptotic properties of tests. We find that the large-sample power of any test can be matched by a test in a suitable limit experiment involving Gaussian diffusions. This establishes that a fixed set of statistics are asymptotically sufficient for testing; these are the number of times each treatment has been sampled and the final value of the score/efficient influence function process for…

We develop an asset demand system to analyze the equilibrium relation between international portfolio holdings and flows, exchange rates, and asset prices across all countries. We introduce a nested logit model of asset demand, for which we develop a new identification strategy by instrumental variables. Averaged across years and issuer countries, the demand elasticities are 27.9 for short-term debt, 3.2 for long-term debt, and 1.2 for equity. These demand elasticities are…

We consider estimation and inference for a regression coefficient in panels with interactive fixed effects (i.e., a factor structure). We show that previously developed estimators and confidence intervals (CIs) are heavily biased and size distorted when some of the factors are weak. Combining the theory of minimax linear estimation with a nuclear norm bound on the error of an initial estimate of the interactive effects, we propose estimators with improved rates of…

The current polarization of elites in the U.S., particularly in Congress, is frequently ascribed to the emergence of cohorts of ideologically extreme legislators replacing moderate ones.Politicians, however, do not operate as isolated agents, driven solely by their preferences.They act within organized parties, whose leaders exert control over the rank-and-file, directing support for and against policies.This paper shows that the omission of party discipline as a driver of…

We study the impact of financial aid on upward mobility. We estimate the causal impacts of a Colombian policy targeting high-achieving, low-socioeconomic-status students on later-life educational and labor market outcomes. Using regression discontinuity design, difference-in-differences, and population-wide administrative microdata, we find that the policy boosted attendance and completion at high-value-added colleges, particularly in STEM fields. Improved college quality…

We state conditions under which choice data suffice to identify preferences when consumers may not be fully informed about attributes of goods. Our approach can be used to test for full information, forecast how consumers will respond to information, and conduct welfare analysis when consumers are imperfectly informed. In a lab experiment, we successfully forecast the average response to new information when consumers engage in costly search. In data from Expedia, our method…

How efficiently do markets reallocate capital in booms and busts? Using a novel dataset of offshore drilling contracts, I examine the role of matching in shaping industry reallocation. Oil companies search and match with capital (rigs) in a decentralized market. I find that oil and gas booms increase the option value of searching, which leads agents to avoid bad matches, reducing mismatch through a sorting effect. I provide an identification strategy to disentangle…

We address the estimation of endogenous treatment models with social interactions in both the treatment and outcome equations. We model the interactions between individuals via a game theoretic approach based on discrete Bayesian games. We employ a nested pseudo joint likelihood algorithm when obtaining the model’s parameters and provide relevant treatment effects and procedures for their estimation. Our empirical application examines the impact of an individual’s exercise…

Using a model of maternal labor supply and investment in children, this paper synthesizes the findings from three separate welfare reform experiments across six sites. The proposed model maps variation in experimental design to parameters that define labor supply behavior, childcare use, and the importance of money and childcare arrangements in the development of child skills. The estimation procedure—which aggregates available evidence to identify the model’s key causal…

We develop a distribution regression model under endogenous sample selection. This model is a semiparametric generalization of the Heckman selection model, which accommodates much richer patterns of heterogeneity in the effect of covariates and selection process and allows for departures from Gaussian errors while maintaining the same level of tractability. The model applies to continuous, discrete, and mixed outcomes. We provide identification, estimation, and inference…

We develop a new method for selecting the penalty parameter for ℓ1-penalized M-estimators in high dimensions, which we refer to as bootstrapping after cross validation. We derive rates of convergence for the corresponding ℓ1-penalized M-estimator and also for the post-ℓ1-penalized M-estimator, which refits the nonzero entries of the former estimator without penalty in the criterion function. We demonstrate via simulations that our methods are not dominated by cross…

We develop a generalized theory of rationed fertility to analyze treatment effect heterogeneity in the child quantity-quality trade-off. An exogenous increase in fertility can be either desired or undesired. We derive a positive rationing income effect on child quality for desired fertility increases but a negative rationing income effect for undesired fertility increases. We propose an econometric framework to identify treatment effects of desired and undesired fertility…

Governments use policy experiments to facilitate learning, but the nature and effects of these experiments remain unclear. We analyze China's policy experimentation since 1980—among the most systematic in history—and document three facts. First, most experiments exhibit positive sample selection. Second, local politicians exert excessive efforts during experiments that are not replicable during policies’ national rollout. Third, the central government is not fully…

We empirically investigate how the performance of small and medium-sized enterprises (SMEs) changes when gaining access to market information. To do so, we evaluate the impact of an information program diffused by a bank among its SME customers. Adopting firms gained access to reports with rich information about their own clientele and that of nearby establishments. While we find that adoption is associated with a 4.5% revenue increase, our instrumental variable results…

We develop a new method to construct instruments in a broad class of economic environments. In the economies we study, a few large firms, industries, or countries account for an important share of economic activity. Idiosyncratic shocks to these large players significantly affect aggregate outcomes and are valid instruments. We provide a methodology to extract these idiosyncratic shocks to create granular instrumental variables (GIVs), which are size-weighted sums of…

We develop and estimate a model of student study time choices on a social network. The model is designed to exploit unique data in the Berea Panel Study. Study time data allow us to quantify an intuitive mechanism for academic social interactions: own study time may depend on friend study time in a heterogeneous manner. Social network data allow us to embed study time and resulting academic achievement in an estimable equilibrium framework. We develop a specification test…

This paper examines a commonly used measure of persuasion whose precise interpretation has been obscure in the literature. By using the potential outcome framework, we define the causal persuasion rate by a proper conditional probability of taking the action of interest with a persuasive message conditional on not taking the action without the message. We then formally study identification under empirically relevant data scenarios and show that the commonly adopted measure…

Karl Marx’s high academic stature outside of economics diverges sharply from his peripheral influence within the discipline, particularly after nineteenth-century developments rendered the labor theory of value obsolete. We hypothesize that the 1917 Russian Revolution is responsible for elevating Marx into the academic mainstream. Using the synthetic control method, we construct a counterfactual for Marx’s citation patterns in Google Ngram data. This allows us to predict how…

We study partial identification of the preference parameters in the one-to-one matching model with perfectly transferable utilities. We do so without imposing parametric distributional assumptions on the unobserved heterogeneity and with data on one large market. We provide a tractable characterization of the identified set under various classes of nonparametric distributional assumptions on the unobserved heterogeneity. We reexamine some of the relevant questions in the…

We present a mechanism based on managerial incentives through which common ownership affects product market outcomes. Firm-level variation in common ownership causes variation in managerial incentives and productivity across firms, which leads to intraindustry and intrafirm cross-market variation in prices, output, markups, and market shares that is consistent with empirical evidence. The organizational structure of multiproduct firms and the passivity of common owners…

The economic consequences of large-scale government investments in education depend on general equilibrium effects in both the labor market and the education sector. I develop a general equilibrium model capturing the consequences of massive countrywide schooling initiatives. I provide unbiased estimates of the model’s elasticities, using a regression discontinuity derived from Indian government policy. The earnings returns to a year of education are 13.4%, and the general…

Macroeconomists increasingly use external sources of exogenous variation for causal inference. However, unless such external instruments (proxies) capture the underlying shock without measurement error, existing methods are silent on the importance of that shock for macroeconomic fluctuations. We show that, in a general moving-average model with external instruments, variance decompositions for the instrumented shock are interval-identified, with informative bounds. Various…

We study how different forms of communication influence inflation expectations in a randomized controlled trial using nearly 20,000 US individuals. We elicit individuals’ inflation expectations and then provide eight different forms of information regarding inflation. Reading the actual Federal Open Market Committee (FOMC) statement has about the same average effect on expectations as simply being told about the Federal Reserve’s inflation target. Reading news articles about…