I hypothesize that in times of economic slow-down, welfare regimes have contrasted answers in terms of insiders/outsiders balances and conseequences for different age groups. Conservative Mediterranean welfare regimes are characterized by skyrocketing youth unemployment, promote youth dependence on family and develop generous schemes of early retirement. Other welfare regimes tend to respect a better between-cohort balance, with more within-cohort inequality in Liberal regimes and less in the Nordic regimes. To test these hypotheses, we make use of new age-period-cohort models to show that some birth cohorts are significantly advantaged over others in terms of their living standard (real incomes after transfers and taxes, adjusted for household size). We look into the cases of the United States, France and Germany. After controlling for education, gender, and immigration status, different cohorts diverge from the general trend of economic growth in all three countries. However, despite such fluctuations, virtually every successive birth cohort in Germany and the US experienced increasing mean disposable incomes relative to the preceding birth cohort. For France however, we find that, while disposable income increased throughout the 20th century, cohorts born before 1950 profited from this, while cohorts born after 1950 experience no gains in real living standards over previous cohorts. Thus, while economic growth seems to benefit all birth cohorts in the US and Germany, pre-1950 birth cohorts in France seem to have monopolized lucrative positions and social transfers to the detriment of post-1950 birth cohorts. As we focus on post-tax, post-transfer income, which can be influenced by social policy, we show how different countries gave diverging answers to the general economic slowdown of the post 1970 era. While the US and Germany did not obviously burden some birth cohorts more than others, when economic problems started to arise in the 1970s, the burden of adjustment to more difficult conditions fell disproportionately on the shoulders of post-1950 birth cohorts in France, leading to strong inequalities between cohorts.
