Abstract
This paper examines income dynamics among Vietnamese workers during the COVID-19 pandemic and the subsequent recovery period. Using the Labor Force Survey (LFS) conducted by the General Statistics Office of Vietnam from 2020 to 2023, the study analyzes more than 1.5 million individual observations. A multinomial logit model is employed to predict the probability of income gains and losses based on demographic characteristics, human capital, employment attributes, and institutional factors. The results show that income differentiation mirrors labor market segmentation. First, female workers are more likely to experience income increases, while male workers have a higher probability of income losses, with odds ratios ranging from 1.04 to 1.07 and 0.87 to 0.99, respectively. This suggests that women may have stronger long-term income resilience. Second, individuals with higher educational attainment are less likely to suffer income declines and are also less likely to recover income losses in the post-pandemic period, highlighting the protective role of education against economic shocks. By focusing on income movements rather than income levels, the study offers new evidence that women do not necessarily face higher income risk; instead, they demonstrate relatively strong income adjustment capacity. These findings contribute to understanding how Vietnamese workers adapt to income uncertainty in the digital era.