Abstract: Does working while in school smooth students' transition into the labor market? We provide evidence on this question by leveraging a one-year work-study program that randomized job offers among over 90,000 student applicants in Uruguay. Program rules forbade employers from employing participants in the same job after program completion, and less than 5 percent of participants ever worked in the same firm again. Two years after the program, participants had 8 percent higher earnings. Our results suggest that the program's focus on work-related skills was a key mechanism for earnings impacts.

Media: vox, j-pal.

Abstract: We quantify the implications of voter bias and electoral competition for politicians' gender composition. Unfavorable voters' attitudes towards women and local gender earnings gap correlate negatively with the share of female candidates in Parliamentary elections. Using within-candidate variation across the different polling stations of an electoral district in a given election year, we find that female candidates obtain fewer votes in municipalities with higher gender earnings gaps. We show theoretically that when voters are biased against women, parties facing gender quotas select male candidates in the most contestable districts. We find empirical support for such a strategic party response to voter gender bias. Simulating our calibrated model confirms that competition significantly hinders the effectiveness of gender quotas.

Media: vox, Bocconi knowledge, IPP (french), IPP (english).

Abstract: This paper measures the job-search responses to the COVID-19 pandemic using real-time data on vacancy postings and job ad views on Sweden’s largest online job board. First, new vacancy postings drop by 40%, similar to the US. Second, job seekers respond by searching less intensively, to the extent that effective labour market tightness increases during the first three months after the COVID outbreak. Third, they redirect their search towards less severely hit occupations, beyond what changes in vacancies would predict. Overall, these job-search responses have the potential to amplify the labour demand shock. 

Abstract: We relate gender differences in willingness to commute to the gender wage gap. Using French administrative data on job search criteria, we first document that unemployed women have a lower reservation wage and a shorter maximum acceptable commute than their male counterparts. We identify indifference curves between wage and commute using the joint distributions of reservation job attributes and accepted job bundles. Indifference curves are steeper for women, who value commute around 20% more than men. Controlling in particular for the previous job, newly hired women are paid after unemployment 4% less per hour and have a 12% shorter commute than men. Through the lens of a job search model where commuting matters, we estimate that gender differences in commute valuation can account for a 0.5 log point hourly wage deficit for women, that is, 14% of the residualized gender wage gap. Finally, we use job application data to test the robustness of our results and to show that female workers do not receive less demand from far-away employers, confirming that most of the gender gap in commute is supply-side driven.. 

Media: Atlantico, Alternatives économiques

Abstract: This article analyses the effectiveness of hiring credits. Using comprehensive administrative data, we show that the French hiring credit, implemented during the Great Recession, had significant positive employment effects and no effects on wages. Relying on the quasi-experimental variation in labour cost triggered by the hiring credit, we estimate a structural search and matching model. Simulations of counterfactual policies show that the effectiveness of the hiring credit relied to a large extent on three features: it was non-anticipated, temporary and targeted at jobs with rigid wages. We estimate that the cost per job created by permanent hiring credits, either countercyclical or time-invariant, in an environment with flexible wages would have been much higher.

Media: Les échos, Le Monde

Abstract: Although the reservation wage plays a central role in job search models, empirical evidence on the determinants of reservation wages, including key policy variables such as unemployment insurance (UI), is scarce. In France, unemployed people must declare their reservation wage to the Public Employment Service when they register to claim UI benefits. We take advantage of these rich French administrative data and of a reform of UI rules to estimate the effect of the Potential Benefit Duration (PBD) on reservation wages and on other dimensions of job selectivity, using a difference-in-difference strategy. We cannot reject that the elasticity of the reservation wage with respect to PBD is zero. Our results are precise and we can rule out elasticities larger than 0.006. Furthermore, we do not find any significant effects of PBD on the desired number of hours, duration of labor contract and commuting time/distance. The estimated elasticity of actual benefit duration with respect to PBD of 0.3 is in line with the consensus in the literature. Exploiting a Regression Discontinuity Design as an alternative identification strategy, we find similar results.

Media: vox

Abstract: Since the late 90s, Regression Discontinuity (RD) designs have been widely used to estimate Local Average Treatment Effects (LATE). When the running variable is observed with continuous measurement error, identification fails. Assuming non-differential measurement error, we propose a consistent nonparametric estimator of the LATE when the discrepancy between the true running variable and its noisy measure is observed in an auxiliary sample of treated individuals, and when there are treated individuals at any value of the true running variable — two-sided fuzzy designs. We apply our method to estimate the effect of receiving unemployment benefits.

[download Matlab programs]

Abstract: Recent empirical literature finds very limited average effects of generous unemployment benefits on match quality. This study examines those effects in a setting where they could be large. We focus on workers with low employability and evaluate the impact of a large increase in potential benefit duration from 7 to 15 months. Our regression discontinuity design does not elicit significant short-term or medium-term effects on either employment duration or wages, whereas we find large positive effects on unemployment and non-employment duration.

Media: France inter

Abstract: We propose a novel selectivity correction procedure to deal with survey attrition in treatment effect models, at the crossroads of the Heckit model and the bounding approach of Lee (2009). As a substitute for the instrument needed in sample selectivity correction models, we use information on the number of prior calls made to each individual before obtaining a response to the survey. We obtain sharp bounds to the average treatment effect on the common support of responding individuals. Because the number of prior calls brings information, we can obtain tighter bounds than in other nonparametric methods.

[download Stata programs] Media: world bank blog

Abstract: We evaluate an experimental program in which the French public employment service anonymized résumés for firms that were hiring. Firms were free to participate or not; participating firms were then randomly assigned to receive either anonymous résumés or name-bearing ones. We find that participating firms become less likely to interview and hire minority candidates when receiving anonymous résumés. We show how these unexpected results can be explained by the self-selection of firms into the program and by the fact that anonymization prevents the attenuation of negative signals when the candidate belongs to a minority. 

Media: Libération, France Inter

Abstract: In this paper, we shed light on the relative contribution of the separation and job finding rates to French unemployment at business cycle frequencies by using administrative data on registered unemployment and labor force surveys. We first investigate the fluctuations in steady state unemployment, and then in current unemployment in order to take into account the unemployment deviations from equilibrium. Both data sets lead to quite similar results. Both rates contributed to unemployment fluctuations during the nineties (50:50 split), while in the last decade the job finding rate was more significant and explained around 65% of the French unemployment fluctuations. In particular, the last business cycle episode, including the last recession, exacerbated the role of the job finding rate. We then show that the predominant role of the job finding rate in the last decade holds when the economy is hit by aggregate business cycle shocks, moving unemployment and vacancies along the Beveridge curve.

Media: Libération, France Inter

Abstract: France and Spain have similar labour market institutions and their unemployment rates were both around 8% just before the Great Recession but subsequently that rate has increased to 10% in France and to 23% in Spain. In this article, we assess the part of this differential that is due to the larger gap between the firing costs of permanent and temporary contracts, and the laxer rules on the use of the latter in Spain. A calibrated search and matching model indicates that Spain could have avoided about 45% of its unemployment surge had it adopted the French employment protection legislation.

Media: vox

Abstract: We analyze the consequences of counseling provided to job seekers in a standard job search and matching model. It turns out that neglecting equilibrium effects induced by counseling can lead to wrong conclusions. In particular, counseling can increase steady state unemployment although counseled job seekers exit unemployment at a higher rate than the non-counseled. Dynamic analysis shows that permanent and transitory policies can have effects of opposite sign on unemployment.

Media: vox

Abstract: How distortive are labor taxes that entitle workers to social benefits? I provide new empirical evidence of bunching in U.S. earnings distribution that show how workers draw a link between taxes and benefits. I find that partially unemployed workers value future preserved benefits when they bunch at the kink of the unemployment insurance benefit-withdrawal schedule. I then extend the bunching formula of Saez (2010) to a dynamic setting that accounts for the value of future benefits tied to taxation. This yields new tests of tax-benefit linkage based on bunching heterogeneity. I verify in quasi-experiments that UI extension programs that decrease the value of future benefits lead to more bunching and to lower labor supply. Last, a quantification exercise of the dynamic bunching formula provides extra support for a strong tax-benefit linkage.

Previously circulated under the title: Partial Unemployment Insurance

Abstract: This paper studies the effect of hiring difficulties on firms' outcomes. We use a shift-share identification strategy combining occupation-specific changes in the difficulty of filling job vacancies within a local labor market (the shifts) and variation across firms in their pre-sampled occupation mix (the shares). We show that hiring difficulties have negative effects on firms' employment, capital, sales, and profits. Firms partially adjust to hiring difficulties by increasing wages and the retention rate of incumbent workers, and by lowering their hiring standards. We then document larger effects of hiring difficulties for labor-intensive firms, firms in expanding sectors, and for non-routine cognitive, high-skill, high-wage, and specialized occupations. Taken together, our findings indicate that hiring difficulties are an important determinant of the growth and profitability of firms across time and space.

Abstract: In this paper, we provide a new set of stylised facts on firm provision of on-the-job training and local labor market competition by exploiting the language used in job vacancies. We take a supervised machine learning approach to identify training offers in more than 12 million US job vacancies. We show our measure correlates well with established on-the-job training measures at the occupation, industry, and regional level. We find that around 20% of job posts offer on-the-job training, with an upward trend over the last decade. Training offers are positively correlated with local labor market concentration, a finding that is robust to an instrumental variables strategy based on the local differential exposure to national firm-level trends. Moving from the first to the third quartile of labor concentration increases training by almost 5%. We interpret our results through the lens of a directed search model where training acts to reduce the queue to fill a vacancy and training has a greater expected benefit to the employer in less competitive labor markets given the lower separation rates.

Abstract: We explore how Artificial Intelligence can be leveraged to help frictional markets to clear. We design a collaborative-filtering machine-learning job recommender system that uses job seekers' click history to generate relevant personalized job recommendations. We deploy it at scale on the largest online job board in Sweden, and design a clustered two-sided randomized experiment to evaluate its impact on job search and labor-market outcomes. Combining platform data with unemployment and employment registers, we find that treated job seekers are more likely to click and apply to recommended jobs, and have 0.6\% higher employment within the 6 months following first exposure to recommendations. At the job-worker pair level, we document that recommending a vacancy to a job seeker increases the probability to work at this workplace by 5\%. Leveraging the two-sided vacancy-worker randomization or the market-level randomization, we find limited congestion effects. We find that employment effects are larger for workers that are less-educated, unemployed, and have initially a large geographic scope of search, for jobs that are attached to several jobs, and are relatively older. Results also suggest that recommendations expanding the occupational scope yield higher effects.