Income Loss and COVID-19: Evidence from Tunisia

Abstract

The purpose of this study is to analyze and to predict the individuals’ income loss associated with education, job sector, income brackets during the COVID-19 pandemic in the Tunisian context. A direct survey was conducted in 1842 Tunisian active worker and self-employment aged over 20 years (mean=35.61% female) between December 20, 2020, and February 8, 2021 in Grand Tunis region. Multinomial logistic regression had been used to assess the COVID-19 impact on an individual’s income change in the Tunisian context. The education attainment, job sector and income level had been mobilized to explain the income loss. We find that the education attainment, job sector and income predictor variables are statistically significant. In particular (1) the log-odds will decrease of being in ‘Partial Income Loss’ versus ‘No Income Loss’ class if the responder has a university degree. (2) The log-odds of being in ‘Partial Income Loss’ class relative to ‘No Income Loss’ class will increase if the individual is not in the Job state sector (3) The log-odds will decrease of being in ‘No Income’ versus ‘No Income Loss’ class if the responder has a secondary, or a university degree. Our predictions point out that four fifth of the responders losing their income temporarily or permanently during the year 2020. Tunisian government has to assist the vulnerable social classes and reduce the inequality between social classes.