1-2 of 2 Results

  • Keywords: program evaluation x
Clear all

Article

Machine Learning in Policy Evaluation: New Tools for Causal Inference  

Noémi Kreif and Karla DiazOrdaz

While machine learning (ML) methods have received a lot of attention in recent years, these methods are primarily for prediction. Empirical researchers conducting policy evaluations are, on the other hand, preoccupied with causal problems, trying to answer counterfactual questions: what would have happened in the absence of a policy? Because these counterfactuals can never be directly observed (described as the “fundamental problem of causal inference”) prediction tools from the ML literature cannot be readily used for causal inference. In the last decade, major innovations have taken place incorporating supervised ML tools into estimators for causal parameters such as the average treatment effect (ATE). This holds the promise of attenuating model misspecification issues, and increasing of transparency in model selection. One particularly mature strand of the literature include approaches that incorporate supervised ML approaches in the estimation of the ATE of a binary treatment, under the unconfoundedness and positivity assumptions (also known as exchangeability and overlap assumptions). This article begins by reviewing popular supervised machine learning algorithms, including trees-based methods and the lasso, as well as ensembles, with a focus on the Super Learner. Then, some specific uses of machine learning for treatment effect estimation are introduced and illustrated, namely (1) to create balance among treated and control groups, (2) to estimate so-called nuisance models (e.g., the propensity score, or conditional expectations of the outcome) in semi-parametric estimators that target causal parameters (e.g., targeted maximum likelihood estimation or the double ML estimator), and (3) the use of machine learning for variable selection in situations with a high number of covariates. Since there is no universal best estimator, whether parametric or data-adaptive, it is best practice to incorporate a semi-automated approach than can select the models best supported by the observed data, thus attenuating the reliance on subjective choices.

Article

The Economic Implications of Training for Firm Performance  

Pedro S. Martins

A small literature on the relationship between employee training and firm performance is currently emerging. This line of research is particularly promising given the underexplored potential of training to drive productivity, wages, and employment. Until recently, training was regarded as a costly and risky investment because workers may leave their firm after being trained. However, studies on labor and education economics have found that training results in high returns for firms and that the costs of training can be recouped in a relatively short time. These results follow from different econometric identification approaches, including a small but growing number of randomized controlled trials. Moreover, most training is of a general nature and therefore applicable in other firms, which is at odds with the original theory of training but consistent with novel models that emphasize labor market power. There are a number of possibilities for future research, including a better understanding of the heterogeneity and patterns of training contents and formats across firms and workers, the differentiation of the effects of training along such dimensions, the role of labor market competition in driving training, the extent to which the productivity effects of training are shared with employees, the role of labor market institutions (including minimum wage, collective bargaining, and occupational licensing) in the dimensions above, and the firm performance effects of training provided to unemployed job seekers (as opposed to employees). Evaluation of the public training programs developed during the Covid-19 pandemic crisis and new forms of training in the context of the growth of remote work also merit further investigation.