We characterize the salient features of the distribution of (log) earnings of formal workers in Mexico using social security records for the period 2005–2019. The analysis is based on a nonparametric approach and is focused primarily on the properties of the distribution of earnings changes. We find strong evidence of deviations from normality of this distribution in terms of negative skewness and high kurtosis, with these deviations varying with income and along the worker's life cycle. A comparison between results obtained with administrative data and household survey data suggests that this latter source of information is inadequate to fully capture the evolution of inequality and the properties of earnings changes as nonresponse is nonrandom and concentrated among formal and highly educated workers—likely the highest earners. Due to the relative size of the informal sector in the Mexican economy, which results in a large number of workers maintaining a weak attachment to formal employment, we also study the impact of transitions out of and back into formal employment on wages earned in the formal sector. We document that workers who exit formal employment experience a significant wage penalty upon reentry taking, on average, 3 or more years to achieve comparable preexit wage levels.