The effects of COVID-19 on employment, labor markets, and gender equality in Central America

Abstract

Abstract This study considers the economic impact of coronavirus disease 2019 (COVID-19) on commercial enterprises in four Central American countries – El Salvador, Guatemala, Honduras, and Nicaragua. At the time of analysis, neither the pandemic nor its economic consequences had fully run their course. It is not, therefore, a definitive analysis, but it is important to try to draw important lessons as soon as possible. The main focus of the study was the initial impact on labor markets. The analysis was based on World Bank Enterprise Surveys undertaken before the outbreak of COVID-19 and follow-up surveys on the effects of the pandemic, also undertaken by the World Bank (Source: Enterprise Surveys, The World Bank, http://www.enterprisesurveys.org). These were combined with data on both government containment measures and rates of morbidity and mortality. The use of enterprise data to analyze labor market issues has some limitations but also many strengths. The data are useful for analyzing the consequences for gender equality in employment. Since the demand for labor is a derived demand, firm-level data provide a clear link to labor market effects. The pandemic has caused a significant loss in sales for many firms. This has created a loss of liquidity, which, in turn, has caused some firms to reduce employment, working hours, and wages. Government containment measures necessary to save lives, such as temporary workplace closures, have added to the burden for both firms and employees. The study starts by using the surveys to identify the important stylized facts. Although some issues are already well documented anecdotally through media reports, this method provides a more evidence-based approach. It also helps identify several issues, such as the impact on gender equality, which has received less journalistic attention. The study is further supported by a regression analysis (ordinary least squares and seemingly unrelated regression equations models) of several key outcomes (changes in sales, employment, the share of females in employment, and expectations of firm survival). A limitation of such an analysis at any enterprise level is heterogeneity and, consequently, a risk of sample selection bias. To provide robustness checks, we use a matching approach. The results suggest that a significant proportion of surviving firms are vulnerable to permanent closure. The ability of firms to retain labor depends on sales, which are affected by both the pandemic itself and the government containment measures. Only a small proportion of firms have received government support, and there is evidence that it could help both firm survival and the retention of labor. There is some doubt whether the four countries have the institutional capacity to provide effective support. If such doubts prove well founded, then support may need to be externally driven.