Research
Description of Research Interests
My research revolves around methodological problems directly inspired by economic applications, often with panel data, and spanning topics such as education policy, agricultural development, the allocation of resources without money, or the management of electrical grids.
Published Research (with short descriptions)
Braun M. and V. Verdier, Estimation of spillover effects with matched data or longitudinal network data, Journal of Econometrics, 2023.
How variation over time in relationships can identify peer effects, even in the presence of selection on unobservables (homophily) and when peer effects occur through unobservable characteristics.
Verdier, V. and C. Reeling, Welfare effects of dynamic matching: An empirical analysis, Review of Economic Studies, 2022.
In simplified settings, allocation without money may be efficient if access can be rationed over time (Jackson and Sonnenschein (2007)). Using an empirical model that can match observed behaviors and outcomes, we find that a real-world dynamic mechanism used to allocate hunting licenses is more efficient than static alternatives but not fully efficient.
Paper, Appendix, Replication archive.
Verdier, V., Average Treatment Effects for Stayers with Correlated Random Coefficient Models of Panel Data, Journal of Applied Econometrics, 2020.
How to extrapolate from "movers" to "stayers" with correlated random coefficient models or in difference-in-differences frameworks.
Paper, Appendix, Replication archive.
Reeling, C., V. Verdier., and F. Lupi, Valuing Goods Allocated via Dynamic Lottery, Journal of the Association of Environmental and Resource Economists, 2020.
We compare the non-market valuation of resources allocated by dynamic allocation mechanisms using a dynamic discrete choice model --- similar to the model developed in Verdier and Reeling (2020) --- with existing approaches to value these resources.
Verdier, V., Estimation and Inference for Linear Models with Two-Way Fixed Effects and Sparsely Matched Data, Review of Economics and Statistics, 2020 (lead article).
New standard errors for two-way fixed effects regressions that are robust to heteroscedasticity and dependence in the errors and are appropriate for sparsely matched data (e.g., relatively few workers per firm or relatively few students per teacher).
New estimator for dynamic models with two-way fixed effects and sparsely matched data.
Paper, Appendix, Replication archive.
Verdier, V., Local Semi-Parametric Efficiency of the Poisson Fixed Effects Estimator, Journal of Econometric Methods, 2018.
The Poisson fixed-effects estimator is a maximum-likelihood estimator based on distributional assumptions. It was previously known to be consistent even if these distributional assumptions do not hold. This paper shows that it is also efficient even if these distributional assumptions do not hold.
Verdier, V., Estimation of Dynamic Panel Data Models with Cross-Sectional Dependence: Using Cluster Dependence for Efficiency, Journal of Applied Econometrics, 2016.
How to leverage cross-sectional dependence (e.g., spatial or cluster dependence) to estimate dynamic models of panel data more efficiently.
Paper, Appendix, Replication archive.
Working Papers
Holland, S., E. Mansur, V. Verdier, and A. Yates. 2023. A New Estimator for Marginal Emissions.
Marginal emissions are the environmental effect of electricity consumption. We propose a new estimator for marginal emissions that relies on two economic constraints (supply = demand and an assumption of monotonicity) to regularize a high-dimensional aggregation problem that would occur with existing methods.
Paper (most up to date). NBER working paper version.
Work in Progress
Peter, K. and V. Verdier. 2023. Local Average Treatment Effects without Instrumental Variables: Using Repeated Cross-Sections.
Nhu, C., C. Reeling, and V. Verdier. 2023. Additionality of Conservation Practices.
Sensitivity of Average Treatment Effects Estimators to Double Model Misspecification.
A Simple Regression-Based Approach to Difference-in-Differences with Covariates, Heterogenous Effects, and Possibly Multi-Valued Treatment.
Inactive Manuscripts
A Flexible Correlated Random Effects Approach to Identification and Estimation of Partial Effects with the Logit Fixed Effects Model
Abstract: The Logit fixed effects model for binary outcomes with panel data relies on a linear model for a latent variable which includes an additive unobserved heterogeneity term and additive transitory shocks that are assumed to be serially independent and to follow the logistic distribution. This model has proved to be very popular in empirical work, mainly due to the fact that a conditional maximum-likelihood estimator, the Logit fixed effects estimator, has been shown to be consistent for the coefficients on observed covariates in the model for the latent variable, even in the absence of restrictions on the distribution of unobserved heterogeneity conditional on the covariates. While these coefficients determine the sign of the effects of the covariates on the outcome variable, they are not sufficient to calculate the magnitude of these effects. Here I show that the distribution of partial effects conditional on covariates is identified as long as the distribution of unobserved heterogeneity conditional on covariates is restricted to belong to a fairly general class of distributions and restrictions on the support of the observed covariates hold. I also show that results coming from the proof of identification can be used to motivate a relatively simple estimator of average partial effects. The results are shown to extend to the ordered Logit fixed effects model. I conclude with a simple empirical example that studies the effect of number of children and husband's income on women's labor force participation.
Additional Moment Conditions for Non-Linear Models of Panel Data with Sequential Exogeneity
Abstract: I consider instrumental variable estimation of non-linear panel data models with multiplicative unobserved effects where the instrumental variables are sequentially exogenous. Existing estimators for these models suffer from a weak instrumental variable problem that can cause them to be too inaccurate to be reliable. In this paper, I present additional sets of restrictions that can be used for more precise estimation. Monte Carlo simulations show that using these additional moment conditions improves the precision of the estimators significantly and should facilitate the use of these models.