For the first time since the 1930s, participation in Republican primaries exceeds participation in Democratic primaries, according to a report by the Center for the Study of the American Electorate at American University.
The study, which looked at elections held through Sept. 1 of this year, found that more than four million more voters cast ballots in Republican primaries than in Democratic primaries. While it is unclear whether higher levels of Republican primary participation spell doom for the Democrats in November, a closer look at the data shows reasons for leaders of both parties to be concerned — the number of nonvoters continues to outpace voters. In a primary season where the narrative tends to be about partisanship and anger, the statistics through the end of the summer suggest that voter participation remained relatively consistent with the last couple off-year election cycles.
According to the study, voter turnout rebounded slightly from the 2006 primary season (which, with slightly less than 17 percent of the voting age population voting, holds the record for the lowest turnout for a midterm election on record). It is consistent with turnout from the 1998 and 2002 primaries, in which slightly less than 19 percent of the population voted. Compared to 1994, participation in the 2010 primary runs about two percentage points lower.
Since 1966, when more than 20 percent of the voting-age population voted in Democratic primaries, participation among Democratic primary voters has steadily declined. With the exception of a slight upward blip in 2002 (only one-third of a percentage point), a smaller and smaller percentage of the voting age population has voted in a Democratic primary each year. On the Republican side, the trend appears to be substantially flatter, especially across the last three decades. Still, participation in Republican primaries is down by one-third between 1966 and 2006 (and down by one-sixth between 1966 and 2010).
Why is participation in partisan primaries on the decline? Are voters increasingly apathetic, or channeling their anger elsewhere? The most likely answer seems to be that fewer voters are eligible to participate in primary elections. Evidence on changing partisan identification, coupled with state-by-state eligibility requirements for participation in primary elections, suggests that declining turnout in partisan primaries reflects voters’ retreat from partisan identifications.
A minority of states conduct open primaries, in which registered voters can vote in a partisan primary regardless of whether they’re registered with that party. In Georgia, for instance, registered voters can select whether they want to vote in the Democratic or Republican primary, regardless of their party affiliation.
The majority of states, however, hold closed primaries. In closed primaries, like those held in Florida. registered Republicans vote in Republican primaries, registered Democrats vote in Democratic ones, and those not affiliated with a political party can’t vote in primary elections.
With closed primaries dominating the political landscape, long-term trends in self-identified partisan affiliation suggest that fewer people are eligible to vote in partisan primaries. According to the General Social Survey, more than 47 percent of Americans self-identified as Democrats in 1972 (the inaugural year of the General Social Survey). That number declined to about 30 percent of Americans in 2006, before bouncing back up six percentage points with the Obama election in 2008. Republican self-identification, on the other hand, rose through the 1970s and the 1980s, peaking at about 33 percent in the early 1990s before experiencing a parallel decline to Democratic self-identification in the last two decades (with a small resurgence during the Bush presidency).
Although the General Social Survey data for 2010 is not available yet, recent Gallup polls on Partisan affiliation suggest that the percentage of self-identified Democrats has returned to about 30 percent, and about 28 percent of Americans self-identify as Republicans.
The tricky part for primary elections, though, involves how organizations measure voter turnout. Voter participation rates, like those used in the study by the Center for Study of American Elections, use the voting age population, rather than the number of voters eligible to vote in the primary, as the denominator.
Assuming that the decline in partisan identification is accompanied by a decline in partisan registration, declining levels of participation in partisan primaries reflects a declining pool of eligible voters. (Of course, declines in partisan registration likely lag somewhat behind Democratic self-identification.) The trends, though, are pretty clear in the graph below. Over the last three decades, the percentage of self-identified Democrats has steadily declined alongside the percentage of the voting-age population that participates in Democratic primaries. On the Republican side, partisan self-identification peaked in the early 1990s – as did the percent of the electorate voting in Republican primaries – before declining.
Although these national trends help explain declining participation rates in partisan primaries nationwide, they mask substantial variation in state-level trends. Participation rates fluctuate substantially within states, reflecting, in large part, the electoral competitiveness of individual races. In the state-level data from the 2010 primary season, one state stands out for its remarkable increase in voter turnout: Colorado. Participation in the 2010 primary elections in Colorado increased by 114 percent over 2006. Surely, competitive races on both sides of the aisle, as well as for both statewide offices, drove turnout. But, for the first time, Coloradans had the option of voting by mail, rather than voting in person. In fact, in almost three-quarters of Colorado counties, all-mail primaries required voters to vote by mail. Like most states, Colorado has a closed primary that restricts participation to registered partisans. Yet, the state’s effort to make voting easier seems to have generated real returns for participation rates. More than twice as many voters voted in Colorado’s 2010 primary than voted in the 2006 primary election.
Absent a reversal in trends in partisan identification, efforts to make voting easier could be the antidote to primary seasons continually marred by declining levels of participation.