It is common to hear pundits, politicians, and experts decry low voter turnout in the United States relative to other democracies. There are many reasons to desire higher voter turnout in all elections, but primary election turnout in particular is more in need of attention than general election turnout. It is far too low considering the importance of primaries in choosing representatives at all levels of government.
The U.S. election process is typically comprised of two components: a nominating contest, in which parties select their standard-bearers, and a general election, in which those party standard-bearers compete for elected office. In the United States, general elections are usually conducted in November of even-numbered years, though some states hold their statewide contests on odd-numbered years.
This paper focuses on the nominating contests held during midterm election cycles. Focusing specifically on midterm election cycles, which tend to see lower voter turnout than during presidential election years, gives researchers the purest view of participation in elections for Congress.
There are limitations to studying midterm nominating contests. Unlike presidential cycles, which can be compared every four years, midterm cycles are not on the same schedule. Each state has two senators, so over a 12-year span, there will be two midterm cycles with a statewide Senate election and one without a Senate contest. Similarly, about one-third of states do not have statewide gubernatorial elections that coincide with midterm federal contests. The lack of statewide races is known to depress turnout and should be considered when taking a national view of turnout during the nomination process.
The vast majority of nominating contests used for congressional elections are primaries. While some states allow parties to choose other means of selecting nominees, all states in 2018 used primaries. During this election cycle, there were states in which parties used conventions to whittle down the number of candidates eligible to appear on a primary ballot. Additional research is needed to determine how this may affect turnout during the primary election contest that is open to the public.
The majority of this analysis covers turnout during primary nominating contests, which the Bipartisan Policy Center’s Commission on Political Reform identified as far too low. As the commission stated in its report Governing in a Polarized America, “Increasing participation in party primaries is good for the parties as well as the country, and setting higher turnout goals for primaries should be a national priority.”
The commission also found that low-turnout midterm primaries erode the credibility of U.S. democracy and may allow more extreme candidates to reach general elections and attain office. Higher participation means that the primary electorate would more likely match that of the general electorate and the population at large.
This paper examines turnout during the 2018 primary elections, conducted in 49 states and the District of Columbia, compared with turnout during the 2014 and 2010 midterm election cycles. (Louisiana holds its primary on Election Day.) The paper will then analyze how some recommendations made by BPC’s Commission on Political Reform show promise as ways to increase voter participation. There will also be a brief summary of some other factors that are correlated with higher turnout but for which further research is needed to justify policy change.
Summary of Findings
Turnout of all eligible voters in 2018 was 19.9 percent. That compares with 14.3 percent in 2014 and 18.3 percent in 2010.
During the 2018 primary election contests, 46,287,000 ballots were cast. Of these, 23,001,000 ballots were cast for Democratic candidates and 20,462,000 were cast for Republican candidates. Therefore, in 2018, 9.9 percent of eligible voters cast a vote for a Democratic candidate, 8.8 percent for a Republican candidate, and 80 percent cast no vote at all.
The states that already exceed the BPC Commission on Political Reform’s target of 30 percent primary turnout by 2020 are Missouri, Montana, Oklahoma, Washington, and Wyoming.
The states with the least turnout in 2018 are Iowa, New Jersey, New York, North Carolina, and Virginia. Each had 12 percent participation or less.
Initial evidence confirms that some of the commission’s recommendations to increase primary turnout work. Those recommendations include adopting open primaries and consolidating primary election dates.
Other policy reforms that were not considered by the commission in 2014 but that are found to correlate with higher primary participation include holding primaries for state offices at the same time as federal offices, holding primaries in summer, allowing voters to cast ballots in uncontested races, and reconsidering nominating conventions.
Factors that impact turnout but remain outside the control of policymakers include the presence of statewide contests on the ballot in a given year, the opportunity for every eligible voter to participate in a primary, and higher levels of partisanship.