The United States has experienced over the past forty years an apparent correspondence between the pattern of retirement among men aged 55-69, and the proportion of workers aged 25-34 working part-year and/or part-time. The latter was an effect of overcrowding among the baby boomers as they moved through the labor market. The former is hypothesized here to be a function of the increasing difficulty older men experienced in obtaining "bridge jobs" – part-year and/or part-time – between career and retirement. It has been demonstrated in a series of studies that a large proportion (as many as two-thirds) of older men – especially those in lower-wage jobs – seek such bridge jobs before retirement. And in many cases these bridge jobs are not in the same industry or even occupation as the career job, leading one to suspect that in many cases there might be little transfer of skill or human capital. If this is the case, then the older workers would at least to some extent be in direct competition with younger workers for these jobs. Given difficulty in finding bridge jobs, a higher proportion of older workers might choose to enter retirement directly from career jobs, skipping the bridge jobs. A relative cohort size measure – the number of 25-34 year olds working part-year and/or part-time, relative to the number of older men, at the state level – has been shown here to be highly significant – both statistically and substantively – in explaining changes in older men's annual hours worked, labor force participation, and propensity to retire, and propensity to claim Social Security benefits. In general terms, relative cohort size can be said to have generated between 25-40% of the observed changes in these variables, with the strongest effects being on the propensity to claim Social Security benefits. Somewhat weaker effects were found for older women, in a companion to this study.