Ongoing research
Long, specific, or work-based: Heterogeneous employment effects of adult training by design
Profiling jobseekers with survey data: the predictive ability of personality and expectancies
Ongoing research
Long, specific, or work-based: Heterogeneous employment effects of adult training by design
Profiling jobseekers with survey data: the predictive ability of personality and expectancies
[01/2025] Which companies use employment subsidies for young workers?
(Submitted)
Active labour market policies like employment subsidies require the involvement of employers to be implemented. However, little is known about the types of firms that participate in these programs. This article merges administrative data of subsidy recipients of a program targeted to young workers from Andalusia (Spain) with balance sheet data of Spanish non-financial companies obtaining a sample of 115,867 units. In a first stage, we avoid modelling assumptions by estimating summary statistics on the socioeconomic features of the companies that used the subsidy. In a second stage, to measure the associations with flexible models, we estimate generalized additive models with restricted cubic splines. The model estimates indicate that there are two variables with outstanding prediction ability of participation in employment subsidies: the temporary rate of the workforce and the gross value added of the firm. Thanks to our flexible model, we detect that the relationship between using a subsidy and the temporary rate is non-linear, specifically hump-shaped. Considering interquartile variations in the covariates, a rise in the temporary rate or the gross value added is associated with a substantial increase in the odds of getting an employment subsidy for young people. Our results show the usefulness of generalized additive models to shed light on the actual recipients of the subsidies. This may be of interest when policymakers expect employers to informally train the workers or if they want to foster innovation or certain economies of scale.
[06/2024] From rules to forests: rule-based versus statistical models for jobseeker profiling (with Christoph Kern)
(Accepted by JLMR) [Working paper and code]
Public employment services (PES) commonly apply profiling models to target labour market programs to jobseekers at risk of becoming long-term unemployed. Such allocation systems often codify institutional experiences in a set of profiling rules, whose predictive ability, however, is seldomly tested. We systematically evaluate the predictive performance of a rule-based profiling procedure currently implemented by the PES of Catalonia, Spain, in comparison to the performance of statistical models in predicting future long-term unemployment (LTU) episodes. Using comprehensive administrative data, we develop logit and machine learning models and evaluate their performance with respect to both discrimination and calibration. Compared to the current rule-based procedure of Catalonia, our machine learning models achieve greater discrimination ability and remarkable improvements in calibration. Particularly, our random forest model is able to accurately forecast LTU episodes and outperforms the rule-based model by offering robust predictions that perform well under stress tests. This paper presents the first performance comparison between a complex, currently implemented, rule-based approach and complex statistical profiling models. Our work illustrates the importance of assessing the calibration of profiling models and the potential of statistical tools to assist public employment offices in Spain.
[01/2024] More money, more effect? Employment effects of job search services of different intensity
(Submitted). Version 2 - April 2025. [Working paper and code]
Despite being widely regarded as effective labor market policies, the impact of job search services on employment remains contested. Recent research challenges the assumption that the intensity of such programs is directly related with its effectiveness. We evaluate the effects of two treatments of an Italian active labour market program called Assegno per il Lavoro. Each intervention is made up of a voucher to fund job search assistance and a performance-based payment related to job search intermediation. Participants are assigned to a certain group with a certain treatment endowment that is increasing as a function of a scoring variable. Leveraging this design, we applied a regression discontinuity analysis to estimate effects on both employment duration and employment quality. First, being assigned to an intermediate-dose treatment slightly increased the probability of finding an open-ended employment, but this did not generate an increment in the average of days worked. Second, being assigned to a high-dose treatment only augmented the mean and the median of the distribution of days worked during the second semester and it did with a greater prevalence of short temporary contracts. Third, different treatment assignments barely changed the reception of job search assistance, so the impacts stem from job search intermediation. The main policy implication is a shift from a focus on intensity to a focus on other design features.
[2025] Outsourcing and time-related dimensions of job quality: Evidence from the Facility Management business (with Alejandro Godino)
Published in F. P. García Márquez, B. Lev, & H. Liao (Eds.), Outsourcing. Cases and Studies in Using Operations Research and Management Science Methods (pp. 21-41). Springer. [link, preprint, code]
[2024] Las políticas activas de empleo en el marco de las políticas de empleo
Published in Anuario IET de Trabajo y Relaciones Laborales and available here
[2024] Efectos de la asistencia en la búsqueda de empleo: una revisión sistemática para España
Published in Gestión y Análisis de Políticas Públicas and available here
[2024] El diseño de los subsidios al empleo del Servei Públic d'Ocupació de Catalunya (2016-2018)
[2025] Què funciona en polítiques actives d'ocupació per a persones amb trastorns de salut mental? Una revisió sistemàtica de la literatura (with David Gallardo and David Saeteros) [Available soon]
[2023] PerfilaSP - Anàlisi de variables sociològiques i psicològiques com a predictores d’ocupabilitat en el perfilatge dels Serveis d’Ocupació de Catalunya (with Oscar Molina) [link]
[2022] Informe d'ampliació de l'avaluació de Reempresa-SOC (with Joan Miquel Verd). [link]
[2021] Avaluació del disseny, la implementació i l'impacte del programa Reempresa-SOC (with Joan Miquel Verd and Caterina Molero). [link]
[2020] Where is fiscal stimulus more effective against COVID-19 crisis? An answer in terms of employment. [link]