Political Behavior

Older people vote at higher rates than people in younger age groups. Studies of voting participation over several decades have shown that voter turnout is lowest among young adults, increases rapidly up to ages thirty-five to forty-five, and then continues to increase (more slowly), declining only slightly after the age of seventy or eighty in the United States (Miller and Shanks), and at somewhat younger ages in other industrial nations (e.g., see Myers and Agree). Consequently, the percentage of the total vote cast by older people in elections is greater than their proportion of the voting-age population. In the 1996 U.S. presidential election, for example, people age sixty-five and older made up 16.5 percent of the voting age population, but cast 20.3 percent of the vote; the turnout rates were 32 percent among those age eighteen to twenty-four, 49 percent among those age twenty-five and forty-four, 64 percent among those age forty-five to sixty-four, and 68 percent among those sixty-five years and older (Binstock, 2000).

Why do older people turn out to vote at higher rates than middle-aged and younger people? Although the connection between age and voting participation has been investigated a great deal, overall the reasons for this relationship remain a source of controversy. Alternative explanations for age group differences in turnout have helped to define the issues, but they have not resolved them.

Some scholars (e.g., Miller and Shanks) hypothesize that the relatively high voting rate of older Americans during the past several decades can be attributed to the movement of successive birth cohorts through the life course. They focus on the contrasting participation rates of the cohort that was first socialized to U.S. politics during the New Deal, and subsequent cohorts whose political attitudes and behavior have been shaped by the effects of historical periods and specific political events that they have lived through (e.g., Vietnam and Watergate) at different ages. Supporting this view is the fact that during these decades the rates of age-group participation have been dynamic, not static, as the various cohorts have entered different stages of the life course. From the 1972 election through the 1996 election, the participation rate of persons age sixty-five and older increased by 6.5 percent, while the rates for all other age groups declined—by 9 percent for those forty-five to sixty-four, 21.5 percent for those twenty-five to forty-four, and 34.7 percent for those eighteen to twenty-four years old (Binstock, 2000). But other analysts (e.g., Rosenstone and Hansen; Teixera) suggest that the contribution of cohort replacement to voting turnout rates may be overestimated. Moreover, a study of voting turnout in Sweden and Germany over four decades found age-group differences to be similar to those in the United States, despite the fact that cohorts in these three nations experienced different political events distinctive to their respective countries (Myers and Agree).

The alternative explanation is that age group differences in voting participation are attributable to life-course effects that correspond to changing characteristics, needs, and incentives of people as they grow older. One contributing factor is that interest in politics (as well as knowledge about it) increases with age and declines only slightly at advanced old ages (Strate, Parrish, Elder, and Ford; also see MacManus). Another contributing factor to the higher voting rate of older persons—related to interest in and knowledge about politics—is age-group differences in voting registration, an essential precursor to voting. Persons who are comparatively well informed about politics and public affairs are more likely to register and vote (Flanigan and Zingale). A study of voter registration and turnout in U.S. national elections (Timpone) found that increased age (from age eighteen to eighty-eight) is monotonically related to being registered and that another aging-related factor, length of residence in one's own home, also has a substantial influence. An additional explanation for the relatively high voting rates of older people is that as they reach old age they have more of a personal stake in government programs, such as Social Security, that provide old-age benefits. Still another contributing factor is the relatively strong partisan attachments of older people (discussed above), because there is a well-established connection between the strength of political party identification and higher rates of voting (Flanigan and Zingale; Rosenstone and Hansen).

Although older people vote at a high rate, they are as diverse in their voting decisions as any other age group, despite the frequent efforts of political candidates to sway them with campaign issues focused on government old-age programs. The votes of older people generally divide along the same partisan, economic, social, gender, and other lines as those of the electorate at large. Accordingly, the various cohorts of older Americans during the past fifty years, for example, have tended to distribute their votes among presidential candidates in roughly the same proportions as other age groups do; exit polls show sharp divisions within each age group, and very small differences between age groups (see Campbell and Strate; Binstock, 1997a). A great deal of empirical evidence indicates that the situation is similar throughout Europe (Walker and Naegele). One exception to this general pattern in the United States is that older partisans are less likely than younger ones to abandon their political party to vote for an independent candidate (Flanigan and Zingale) because of their comparatively stronger attachment to their parties than younger age groups. This tendency was clear in the 1980, 1992, 1996, and 2000 U.S. presidential elections, in which an independent received more than a negligible percentage of votes. The older the age group, the less heavily it voted for the independent candidate (Connelly).