Labor force participation and retirement behavior in Brazil

This dissertation uses Censuses and Household surveys to analyze the evolution and the determinants of retirement of Brazilian workers from 1960 to 2000. The first analytical chapter uses historical data on labor force participation to estimate trends in retirement for these workers. I find a steady and fast decline in the labor force participation of older workers and that these workers leave the labor force at ages in which pension benefits are first available. I also find that the expected length of retirement has more than doubled from 1960 to 2000, and most of the change is explained by the decline in labor force participation rates. This finding, in conjunction with international comparison, imply an impressive transformation in the Brazilian labor market. These trend together with population aging and longer life expectancy reduce the ratio between workers and retirees affecting the basic premise of the PAYGO system.;The second analytical chapter estimates social security financial incentives to early retirement using contemporary techniques developed by Gruber and Wise (1999). I find that for the median worker the implicit tax on continued work amounts to over one-fifth of his potential earnings by age sixty-five. The replacement rate, implicit tax on work, and pension accrual rates in Brazil compare to the levels observed in the US and Canada. These findings, in conjunction with the evidence from the previous chapter, suggests that the pension system in Brazil creates incentives to early retirement and it is as generous as systems from countries with greater levels of income and economic development.;The third chapter uses matched and unmatched household survey data to study the determinants of male retirement over the past two decades. I find educational, occupational and regional gradients of retirement behavior in Brazil. I observe that more educated workers, and those in the formal sector have higher retirement probabilities than less educated ones and those in the informal labor market. The fourth chapter presents a new angle to the analysis of retirement. In this chapter, I use Brazilian data to examine how social security financial incentives and personal characteristics affect one's own and spouse's retirement decisions. I find that couples synchronize retirement, and that they respond similarly to their own characteristics. Moreover, I find that wives are more responsive to husbands' incentives than vice-versa. Policy simulations show that equalizing male and female normal retirement ages would have a large impact on male's retirement but a small direct impact on female retirement.