Too little but not too late: nowcasting poverty and cash transfers' incidence during COVID-19's crisis.

Abstract

The economic crisis triggered by COVID-19 has caused a world-wide economic downturn, and the deepest GDP contraction in Latin America since the beginning of the XXth century. One of the most dramatic outcomes of the crisis is the increase in poverty, but its extent will remain unknown until household income data is collected and analyzed. We propose a simple approach to provide early estimates, micro-simulating the short-run effect of the crisis on the poverty rate. It combines household level micro-data, estimates on the feasibility of working from home, information on key public policies (e.g., cash-transfers, unemployment insurance), and forecasts of GDP contraction. This approach, which can be easily adapted and applied to different countries, allows to nowcast the current poverty level and the poverty-reducing effect of public policies, while providing full micro-macro consistency between heterogeneous impacts on households and the shock to aggregate GDP. Moreover, it enables to estimate the effect on informal and self-employed workers, of utmost importance in developing countries. We illustrate the methodology with an application for Uruguay, finding that during the first full trimester of the crisis, the poverty rate grew by more than 38%, reaching 11.8% up from 8.5%. Moreover, cash transfers implemented by the government in the period had a positive but very limited effect in mitigating this poverty spike, which could be neutralized with additional transfers worth under 0.5% of Uruguay’s annual GDP.