This is a study of aging, inequality, and welfare state regimes. The project will use micro-level data with detailed income measures from the Luxembourg Income Study (LIS). The project selects five high income nations (Canada, Germany, Sweden, the United Kingdom, and the United States) with data over a 15-20 year period from the mid-1970s to about 1990. Analyzing income changes during old age for multiple nations then involves the following steps: 1) describing within-cohort changes in income levels and inequality for pre-retirement age men and women compared to post-retirement age men and women; 2) estimating the changes in the effects of education and gender -- variables that generally remain stable during the age span studied -- on income outcomes for the same age groups; 3) estimating the effects of the stable background variables with controls for additional variables that change from middle-age to old age -- labor force status, hours worked, family status, and disability -- to help identify the potential sources of changes in the effects of the stable background variables; 4) replicating the models separately for private and public income sources; and 5) comparing the differences for the five nations. Because cohorts come to depend increasingly on public transfers during old age, the analysis of changes in income levels and inequality as cohorts age can address three theoretical issues. First, it can address arguments concerning the way one component of social inequality, old age, reshapes or fails to reshape the influence on economic outcomes of other ascribed and achieved characteristics such as gender and class. Disadvantages of old age may level, heighten, or maintain the gender and class advantages or disadvantages brought into old age. Second, by considering changes in the stratification processes as a shift occurs in the primary sources of income, it can addresses debates over the progressivity, regressivity, or continuity of public transfers relative to market income. Third, with attention to multiple nations with distinct types of public programs and old age policies, it can address debates over the extent of national differences in the ability of welfare states to reduce, heighten, or maintain the gender and class-based market inequalities brought into old age. The analysis will compare the determinants of the distribution of household and personal income in both absolute and relative terms. It will use regression techniques to analyze income variance, and it will use information-based inequality techniques to analyze relative income shares. The analysis can use the same techniques to further examine the contribution of different income sources to the changes in income levels and distribution over the later stages of the life-cycle. The extent to which the changes show declining inequality, increasing inequality, or stable inequality will provide support for debates over the impact of aging and public income support on the structure of stratification. The project will thus contribute in several ways to the understanding of social stratification. First, it relates economic well-being in old age to stratification in middle age. With age viewed dynamically, sources of inequality structure the life course in ways that make the isolated study of cross-sectional age groups problematic. Second, the project recognizes that structural sources of gender inequality make for different economic experiences -- and require separate study -- of men and women during the aging process. The well-established vulnerability of older women to financial problems reflects more than individual risks; it also reflects the different and unequal legal, governmental, and institutional structures men and women face. Third, the project examines the structural sources of age and gender stratification with data on nations that have substantially different institutions of economic support for men and women in old age. In so doing, it supplements a huge comparative research tradition that often focuses on class to the exclusion of gender and age.
