Xavier D'HAULTFOEUILLE

Grade : Administrateur hors classe

Mail : xavier.dhaultfoeuille[arrowbase]ensae.fr

ResearchEducationJobsPublicationsTeachingWork in progressOther

Research Interests

[haut]

CV

Econometric theory (in particular identification issues)

Empirical industrial organization


Biography

Education

[haut]

2014: HDR (habilitation to supervise PhD students)

2009: PhD Thesis "Essay on some Identification Issues in Economics", supervised by  Jean-Marc Robin.


Jobs

[haut]

20011-now: professor in economics at CREST-ENSAE.

2008-2011: researcher on firms strategies issues at INSEE.

2005-2008: teaching assistant at ENSAE.

2000 - 2003: researcher on data collection issues at the methodological unit of INSEE.



Publications

[haut]

- The Provision of Wage Incentives: A Structural Estimation Using Contracts Variation, with Philippe Février, forthcoming in Quantitative Economics.

To what extent do people react to incentives? Are observed contracts (nearly) optimal? We answer to these questions using a nonparametric principal agent model and an exogenous variation in contracts between the French national institute of statistics and its interviewers.

- Automobile Prices in Market Equilibrium with Unobserved Price Discrimination, with Isis Durrmeyer and Philippe Février. forthcoming in Review of Economic Studies. Supplementary material.

We consider inference on a demand and supply model for differentiated products with price discrimination that is unobserved by the econometrician. We show how to extend BLP's GMM estimation to this setting, using restrictions on marginal costs. We apply our framework to the French automobile market.

- Fuzzy differences-in-differences with Stata (2019), with Clément de Chaisemartin and Yannick Guyonvarch. Stata Journal (19). Fuzzydid Stata package available from the SSC repository. You can fin the files to replicate the application on Clément's webpage.

This paper presents the Stata command fuzzydid, which computes various estimators of the LATE and LQTE for fuzzy DID designs, following our paper "Fuzzy DID". It can handle non-binary treatments, multiple periods and groups, covariates and partial identification. 

- A cautionary tale on instrument vector calibration for the treatment of unit nonresponse in surveys (2019), with David Haziza and Eric Lesage. Journal of the American Statistical Association (114).

We show that the calibration method based on instruments, proposed by Deville (2002), leads to a large variance when the instrumental variable are poorly related to the calibrating variables. If the exclusion restriction is violated, the bias is also large under the same condition.

Fuzzy Differences-in-Differences (2018), with Clément de Chaisemartin. Review of Economic Studies (85). Supplementary Material. See above for the Stata package and the corresponding paper.

In many applications of the DID method, the treatment rate only increases more in the treatment group. In such fuzzy designs, we show that the popular "Wald-DID" (the DID of the outcome divided by the DID of the treatment) identifies a LATE only if two homogeneous treatment effect assumptions hold. We then propose two alternative estimands that do not rely on such assumptions.

Extremal Quantile Regressions for Selection Models and the Black-White Wage Gap (2018), with Arnaud Maurel and Yichong Zhang. Journal of Econometrics (203). Supplementary MaterialStata code and documentation.

We consider models with endogenous selection and no instrument nor large support regressors. Identification relies on the independence between the covariates and selection, when the outcome tends to infinity. We propose a simple estimator based on extremal quantile regressions and apply it to the  evolution of the black-white wage gap in the US.

Identification of Additive and Polynomial Models of Mismeasured Regressors Without Instruments (2017), with Dan Ben-Moshe and Arthur Lewbel. Journal of Econometrics (200).

Suppose that Y = g(X*) + h(Z) + U, E(U|X*,Z)=0 but X* is measured with error. We show that g and h can be identified nonparametrically without side information provided that, basically, Z affects X*. A similar result holds when Y=P(X*,Z) + U, with P polynomial.

Measuring Segregation on Small Units: A Partial Identification Analysis (2017), with Roland Rathelot. Quantitative Economics (8) (The supplement and code can be found following the link).

Suppose that an individual in a small unit j (a classroom, a small firm...) belongs to a minority with a probability pj. To measure segregation of this minority, one would ideally use an inequality index on the pj, but they are unobserved. Using the observed proportion instead leads to an overestimation. The segregation indices are actually partially identified. We provide tractable bounds and develop inference.

Disentangling Sources of Vehicle Emissions Reduction in France: 2003-2008 (2016), with Isis Durrmeyer and Philippe Février. International Journal of Industrial Organization (47).

We study the factors of the decrease in average CO2 emissions of new cars in France between 2003 and 2008. We show that the evolution of consumers' preferences account for 43% of this decrease, and that these changes follow two environmental policies put in place during this period.

A Convenient Method for the Estimation of the Multinomial Logit Model with Fixed Effects (2016), with Alessandro Iaria. Economics Letters (141). Supplement. Matlab code.

We propose a computationally convenient alternative to the conditional MLE for fixed effect multinomial logit models.

Identification of Mixture Models Using Support Variation (2015), with Philippe Février. Journal of Econometrics (189).

Suppose that observed variables (X1,...,XK) are independent conditional on a continuous and unobserved variable X*. We show that the distributions of Xi conditional on X* are identified if the bounds of the conditional support of Xi are strictly increasing with X*. We also develop a test of this condition.

Identification of Nonseparable Triangular Models with Discrete Instruments (2015), with Philippe Février. Econometrica (83). Supplementary materials. Old, longer version.

Consider a model Y = g(X,U) with X endogenous, and suppose that we have instruments Z such that X = h(Z,V). If Z is independent of (U,V) and  both g(X,.) and h(Z,.) are strictly monotonic, then g can be partially or pointly identified if Z is binary. It is fully identified in general if Z takes three values or more.

- The Environmental Effect of Green Taxation: the Case of the French “Bonus/Malus” (2014), with Pauline Givord and Xavier Boutin. Economic Journal (Features, 124).

In 2008 was introduced in France a feebate system for new automobiles. We investigate the effect of this policy on CO2 emissions. We find that the policy actually led to an increase in these emissions, mostly because of a substantial increase in the sales of new automobiles.

Inference on an Extended Roy Model, with an Application to Schooling Decisions in France (2013), with Arnaud Maurel. Journal of Econometrics (174). Web appendix.

Consider an extended Roy model where a binary decision depends on expected gains and an unobserved cost. The model is identified without instruments if, basically, the unobserved cost only depends on covariates. Applying our results to French data, we show that nonpecuniary components are a key factor for going to college.

- Another Look at Identification at Infinity of Sample Selection Models, (2013), with Arnaud Maurel, Econometric Theory (29).

The sample selection model can be identified without instrument if basically, the probability of selection, conditional on the potential outcome and covariates, does not depend on covariates as the potential outcome tends to infinity.

- On the Completeness Condition for Nonparametric Instrumental Problems (2011), Econometric Theory (27). 

Sufficient conditions for the completeness condition (E(g(X)|Z) = 0 => g(X)=0) used in nonparametric IV problems are given. It holds in particular under a large support condition on n(Z) and technical restrictions on V in the generalized additive model X = m(n(Z) + V). 

- A New Instrumental Method for Dealing with Endogenous Selection (2010), Journal of Econometrics (154).  

Consider the sample selection model under the nonstandard IV restriction that D is independent of Z conditional on Y. Nonparametric identification is achieved under a completeness condition between Y and Z. Partial identification can also be obtained if one replaces independence by monotonicitiy restrictions.

- Identification of Peer Effects Using Group Size Variation (2009), with Laurent Davezies and Denis Fougère, Econometrics Journal (12). 

A linear-in-means model close to the one of Manski (1993) is identified provided that we observe groups with three distinct sizes. This applies even if one does not observe all members of the group, and can also be extended to binary outcomes.

- Measuring the Evolution of Complex Indicators: Theory and Application to the Poverty Rate in France (2008), with Fabien Dell, Annales d’Economie et de Statistique (90). SAS macros to compute inequality indicators and their variances (details on the macros are available at the end of this document (in French), written with Fabien Dell, Philippe Février and Emmanuel Massé).

 

In French:

La régression quantile en pratique (2014), Economie et Statistique (471) with Pauline Givord.

A survey on quantile regression, with in particular some discussons on interpretation and solutions agains endogeneity.

Le coût du bonus/malus écologique : que pouvait-on prédire ? (2011), Revue économique (62),  with Isis Durrmeyer and Philippe Février.

In 2008 was introduced in France a feebate system for new automobiles. While the policy was expected to be neutral, it turned out to cost more than 200 million euros. It was actually difficult to predict ex ante this cost using a standard discrete choice model of automobile choice.

 

 


Teaching

[haut]

Semi and nonparametric econometrics at ENSAE (3rd year) and Paris Sacler Master (M2): quantile regression, IV in nonlinear / nonparametric models,  partial identification

Econometrics 2 at ENSAE (2nd year).

Econometrics 1 at the Master in Economics (1st year).


Work in progress

[haut]

Papers under revision:

Two-way fixed effects estimators with heterogeneous treatment effects, with Clément de Chaisemartin. R & R to American Economic Review.twowayfeweights and did_multipleGT Stata packages available from the SSC repository

We show that if treatment effects are not constant, regressions with groups and time fixed effects identify weighted averages of treatment effects across groups and time periods, with potentially (many) negative weights. We suggest sensitivity checks and better estimands under testable restrictions on the design.

Nonlinear Difference-in-Differences in Repeated Cross Sections with Continuous Treatments, with Stefan Hoderlein and Yuya Sasaki. Revision requested by Journal of Econometrics.

We study the identification of nonseparable models with continuous, endogenous regressors using repeated cross sections. Several treatment effect parameters are identified under, basically, a weak stationarity condition on the unobservables and time variation in the distribution of endogenous regressors.  

Working papers

Empirical Process Results for Exchangeable Arrays, with Laurent Davezies and Yannick Guyonvarch (this paper supersedes Section 3 of "Asymptotic results under multiway clustering").

We show the weak convergence of empirical processes and some bootstrap processes with multiadic (e.g., dyadic) data or multiway clustering. These results imply asymptotic normality and the validity of the bootstrap for a large class of nonlinear estimators. We illustrate our results with trade data.

- Asymptotic results under multiway clustering, with Laurent Davezies and Yannick Guyonvarch.

Given the paper above, this paper is now mostly obsolote, except for the new variance estimator, which is positive by construction. Simulations suggest that it works well even with a fairly small number of clusters.

Rationalizing Rational Expectations? Tests and Deviations, with Christophe Gaillac and Arnaud Maurel. Corresponding R package.

We construct the best possible test of rational expectations (RE) when we observe expectations on future variables on a certain sample, and realizations of that variable on another sample of different units. We also construct minimal deviations from RE to assess the sensitivity of structural models to this assumption.

In French :

Faut-il pondérer ? Ou l'éternelle questin de l'économètre confronté à des données de sondage, with Laurent Davezies.

A note with on whether we should use  survey weights in econometrics. The note is complete but most plausibly will never be published.


Other

[haut]

Grants and awards

2017- : Principal investigator of the ANR project "Otelo" (On Treatment Effects estimation using LOngitudinal data).

2015: Dennis J. Aigner Award (with Arnaud Maurel) for the best applied paper published in Journal of Econometrics between 2013 and 2014.

 

Editorial activities

1. Associate editor for Review of Economic Studies (2016-), Econometric Theory (2017-), The Econometrics Journal (2016-), Annals of Economics and Statistics (2011-2012, 2016-).

2. Former co-editor of Annals of Economics and Statistics (2013-2015).

3. Referee for American Economic Review, Annals of Statistics, Bernoulli, Econometrica, Economic Journal, Journal of Applied Econometrics, Journal of Business and Economic Statistics, Journal of Econometrics, Journal of the Royal Statistical Society (series B), Macroeconomic Dynamics, Oxford Bulletin of Economics and Statistics, Quantitative Economics, RevStat, The Rand Journal of Economics, The Review of Economics and Statistics.

 

Visits (one week or more)

Yale University (Cowles Foundation), April 2018.

Toulouse School of Economics, April-May 2015.

Boston College, January – June 2013.

 

PhD students supervision

Lucas Girard, “Data quality and identification” September 2017-now.

Ao Wang, “High-dimensional statistics in empirical industrial organization”, September 2016-now.

Jérémy L’Hour, “Treatment Effects Evaluation with High-dimensional Data”, September 2015-now. 

Benjamin Walter, “Two essays on the market for Bitcoin mining and one essay on the fixed effects logit model with panel data”, September 2015-August 2018.


"Le centre de la Recherche en Économie et Statistique ne peut être tenu responsable pénalement des infractions aux lois que pourrait contenir cette page personnelle qui est sous la responsabilité de son auteur."