Retirement

Old age security programs define age-based windows for retirement, but they usually allow some choice in the exact timing of the retirement transition (although sometimes at the cost of reduced benefits). Consequently, factors other than age and employment history can and often do affect retirement decisions. Studies addressing motives for retirement indicate that marital and family reasons play some role in retirement decisions, especially among women (Clemens 1997; Disney et al. 1997; Ruhm 1996).

The specific marital characteristics entering retirement decisions include marital status, spouse's employment/retirement status, spouse's health, spouse's economic situation, and the quality of the marital relationship. Marriage and marital history are pertinent for retirement benefits (spouses can rely at least partially on current or former partners' pensions), are linked to employment history, and influence individuals' attitude toward retirement. Most studies suggest that married persons are more prone to retire (Flippen and Tienda 2000; Miniaci 1998; Szydlik and Ernst 1996) although contrary evidence exists as well (Lindeboom 1998; Ruhm 1996) for married men.

Perhaps the most widely studied phenomenon is spouses' tendency toward joint retirement (Allmendinger 1990; Blau 1998; Disney et al. 1997; Henkens 1999; Miniaci 1998; Pepermans 1992; Zweimüller et al. 1996). Explanations for this trend refer to joint leisure preferences of spouses, shared economic restrictions, and similarity in spouses' background characteristics such as age and education, as well as traditional gender roles that preclude continued employment of typically younger wives after their husbands' retirement (Gustman and Steinmeier 1994; Henkens et al. 1993). Indeed, several studies show that already retired men pressure their employed wives to leave the labor force (Skirboll and Silverman 1992; Szinovacz 1989). However, there is also evidence for wives' influence on their husbands' retirement decisions (Henkens and Siegers 1994; Smith and Moen 1998). Whether spouses are in fact able to implement joint retirement depends on a variety of circumstances, including the age difference between spouses, their economic situation, their health, and family obligations (Allmendinger 1990; Arber and Ginn 1995; O'Rand et al. 1992).

Research further suggests cross-influences of spouses' economic situation (wages, Social Security and pensions, Medicare eligibility) on each other's retirement timing, although these effects are often weak (Allmendinger 1990; Burkhauser et al. 1996; Madrian and Beaulieu 1998; Zweimüller et al. 1996). In addition, the spouse's health plays a role in retirement timing. A spouse's illness can either delay or hasten the partner's retirement depending on the balance between caregiver burden and financial needs associated with the spouse's disability (Hayward et al. 1998; Honig 1996; Szinovacz and DeViney 2000).

The attractiveness of retiring is also influenced by the quality of the marital relationship. Spouses who enjoy a close relationship, have joint hobbies, or desire more time with their partner are more inclined to retire, whereas couples in conflict-laden relationships may dread spending more time together and hence delay retirement (Henkens 1999; Honig 1998; Naegele and Voges 1989, as cited in Kohli et al.1989; Szinovacz and DeViney 2000). Some husbands also fear that retirement could undermine their power position in the marriage and postpone retirement for that reason (Szinovacz and DeViney 2000).

Not only marital but also family circumstances can impinge on retirement decisions. Financial obligations, especially for dependent children, may preclude early retirement (Miniaci and Stancanelli 1998; Pienta et al. 1994; Szinovacz, DeViney, and Davey 2001; Talaga and Beehr 1995), whereas the burden of care for frail relatives sometimes entices married women to retire (Miniaci and Stancanelli 1998; Zimmerman et al. 2000). Whether closeness of ties to relatives including adult children affects retirement transitions remains virtually unexplored. Preliminary evidence suggests that individuals who lack extended family ties (for instance, the childless) may delay retirement (Szinovacz, DeViney, and Davey 2001).