Modelling the Learning Impacts of Educational Disruptions in the Short and Long Run

In this paper, I propose a new framework for analysing the short and long-run effects of temporary educational disruptions on the learning progression of children. The framework integrates into a coherent model recent advances in the literature on learning acquisition (Kaffenberger, 2021; Kaffenberger and Pritchett, 2020b, 2021) and the literature on estimating the immediate costs of instructional disruptions (Neidhöfer et al., 2021). The integrated framework includes explicit modelling of continuous parental investments, filling a gap in the literature related to the Potential Pedagogical Function and other explicit models of learning progression and acquisition. In the same way, the model considers the role of economic resources as part of the resources employed by parents to mitigate the effects of a temporary shock in instruction., expanding the notion of attenuation capacity developed by Neidhöfer et al. (2021). Finally, I take this framework to the data to estimate the potential effects of the instructional disruption caused by the COVID-19 pandemic in Mexico. The estimates suggest that, for the Mexican cohort affected by the instructional disruption, the potential persistent loss in learning with respect to the counterfactual lies on average between 20% and 90% of the learning acquired during a usual school year, depending on the effectiveness of the remote learning policies implemented during 2020 and 2021. These results already consider the mitigating role of parental investments. Furthermore, my results suggest substantial variation between inhabitants from different regions of the country and inside inhabitants of the same region, being the South of the country the region where the losses are the largest. (Stone Center on Socio-Economic Inequality Working Paper)