Marital History and Retirement Security: An Empirical Analysis of the Work, Family, and Gender Relationship

This dissertation investigates the relationship between marital history and individuals' retirement resources, namely Social Security, employer-sponsored pensions, and non-housing wealth. Prior research provides a foundation for understanding marriage's positive relationship to retirement security, and suggests that marriage is financially beneficial and can even lessen some external factors that would otherwise damage a family's financial situation. Yet changing demographics, with fewer people in first marriages and rising numbers of individuals experiencing divorce and choosing to remain unmarried, suggest our understanding of this relationship for today's retirees may be limited. The purpose of this research is to identify which aspects of complex marital histories are associated with individuals' retirement security, paying particular attention to gender differences. Using data from nine waves of the Health and Retirement Study (1992-2008), four facets of marital history are examined: marriage type, frequency, timing, and duration. Currently married and currently unmarried respondents are separated during the analyses in order to adequately capture the association between previous marital events and retirement resources. The results indicate that marital history is associated with Social Security, employer-sponsored pensions, and non-housing wealth differently, and that these relationships vary by gender and current marital status. The findings provide support for the argument that marital history, and in particular marital duration, has a strong relationship to retirement resources. Contrary to expectations, currently married women with longer marriages have less Social Security and pension income than married women who experienced shorter marriages. Marital history has no relationship to the retirement security of married men. For the unmarried groups, never married men have the lowest odds of receiving an employer-sponsored pension and have less non-housing wealth than both divorce and widowed men. Unmarried women's retirement security is associated with the type of disruption experienced; women with multiple past marriages have more resources if they are currently widowed but less if they are currently divorced. Further study is needed to understand how and why complex marital history factors have a relationship to retirement finances, and to expand our knowledge about certain understudied populations such as remarried women and never married men.

retirement security