How does higher education shape social mobility in countries where elite colleges are public and tuition-free? Using linked microdata spanning several decades, we follow Brazilian high school graduates through college and into the labor market to study income segregation, mobility, and the distributional incidence of public spending. We show that elite public colleges have relatively low mobility rates, despite high returns, because they enroll few disadvantaged students. Using college-level financial data, we document that per-student government transfers are much higher at elite publics, yielding a highly regressive pattern of public expenditure in higher education: the top 10% receive 6.75 times as much as the bottom 10%. Finally, exploiting differential exposure to a nationwide affirmative-action reform, we quantify causal effects on income composition in each college tier. We find that, in the absence of the policy, the top 20% would have increased its share of total government spending by 14%

Job displacement generates large and persistent income losses. This paper examines how different public policy designs mitigate the consequences of job loss in Uruguay. Using matched employer-employee records, we combine difference in differences and RDD designs to estimate both the effects of job displacement and the insurance value of two policies: (i) a traditional contributory unemployment insurance program and (ii) a non-contributory conditional cash transfer program that is widely available to poor households. First, we document sharp earnings declines following displacement and incomplete recovery even five years later. We then compare the insurance value of alternative policies. We show that access to the non contributory cash transfer program provides an effective alternative to UI for vulnerable and middle income workers facing layoffs. In particular, access to the cash transfer is associated with faster earnings recovery and improved medium run labor market outcomes.

We study how firms adjust to more generous paid sick leave. Exploiting a 2011 Uruguayan reform that gradually raised the benefit cap, we implement a difference-in-

differences design with employer–employee data, comparing firms with pre-reform exposure measured by the share of workers eligible for higher benefits. More-exposed firms experience a large increase in sick-leave incidence among eligible workers, with little change in spell duration. Earnings growth is 1–6% lower, while employment rises by 3% relative to less-exposed firms.

We study how an expansion to paid maternity leave affects women’s labor market trajectories in an economy with substantial informality. Our setting is Chile’s 2011 reform that extended maternity leave from 12 to 24 weeks based on a child’s date of birth. The reform creates quasi-experimental differences in incentives to participate in formal employment around childbirth. We combine administrative birth records with longitudinal social security data and government social registry, where we observe

workers’ formal and self-reported employment and earnings, to track women’s employment before and after the birth event. Using an event-study framework, we analyze dynamic adjustments in both labor force participation and the allocation between formal and informal work, allowing for anticipatory responses when eligibility depends on prior contributions. The analysis provides new evidence on how contributory maternity benefits shape employment trajectories around childbirth in high-informality environments.