It Takes (At Least) Two to Do Democracy
by Barry C. Burden, Julia Nestel, and Rochelle Snyder
Every definition of democracy has at its core the requirement that voters be offered choices at the ballot box. Without a choice between at least two candidates or parties, a voter cannot hold incumbent office holders accountable for their performance or change the direction of government.
It follows that a key responsibility of a political party is to nominate candidates for elected offices. Setting aside concerns about the two-party system itself, the parties can be judged by whether they actually field candidates so that all voters have at least one alternative to the person representing them. Using this yardstick, it is quickly apparent that democracy is flourishing in some parts of the country but floundering in others.
Democracy at Two Levels
Building on prior research, we have collected data on whether congressional and state legislative elections were uncontested from 1968 to 2020. The uncontested rate is computed as the percentage of seats with only one candidate listed on the ballot. (In multimember districts, the rule is generalized to when the number of candidates is less than the numbers of seats up.)
Usually this means one Democrat and one Republican are running. In a small number of cases in states using the “top two” primary, it means two candidates of the same party. There are also many minor party and independent candidates contesting seats.
The figure below shows that the contestation facet of democracy is functioning rather differently at the federal and state levels.
In elections for the U.S. House of Representatives, the share of seats that is uncontested is low and has been steadily falling for decades. In 2020, only 13 out of the 435 districts failed to list at least two candidates on the ballot.
But state legislative elections tell a different story. Examining all races for the lower chambers in the states, the share of uncontested seats is not only high but has been moving in the opposite direction from Congress. The share of state legislative seats without a contest rose from one fifth to more than one third. The vigorous 2018 and 2020 elections reversed the trend somewhat, but it is not yet clear if these two most recent election cycles are a new normal or merely aberrations.
In cases where general elections are uncontested, one might hope that there are at least some choices for voters in the primary. Unfortunately, research shows that contestation in primaries is not sufficient to compensate for its absence in November. There is no real substitute for the lack of an opponent on election day.
Both major parties are at fault for the lack of opposition in state legislative elections. Of the 4,710 partisan seats up in 2020 (i.e., not Nebraska), Democrats had 3,966 candidates on the ballot while Republicans had 3,901. A total of 785 non-major party candidates also appeared on ballots.
The level of contestation is not due so much to the overall political competitiveness of a state but whether the party balance in the legislature makes it appear “flippable,” especially to the minority.
This may seems like a lot of candidates, yet millions of American nonetheless had no choices in the general election about who would represent them in the state capitol. This is because parties have all but given up in some states. For example, of the 75 state House districts with elections in 2020 Rhode Island, Democrats fielded candidates in 71 of them but Republicans only ran in 26.
Not All States
The levels of contestation are thus highly variable from one state to the next. As the figure below shows for the 2020 elections, the best performing states demonstrate that a lack of competition is not a necessary aspect of state legislative elections. All of the (nonpartisan) seats in the Nebraska unicameral were contested, as were all 60 seats in the Oregon House of Representatives, and all but one of the districts in the Michigan House. On the other end of the spectrum, in nine states more than half of districts had only one candidate on the ballot. At the most extreme, voters were given choices in 22 out of 60 state house districts in Wyoming and 41 out of 160 districts in Massachusetts.
Rates of contestation tend to be quite durable from one election cycle to the next, suggesting that there are enduring features of states that affect how many candidates run. Our prior research points to the size of the majority in the state legislature as a key factor behind the amount of contestation. Seats are less likely to be challenged when it seems that one party has a lock on the legislature. The level of contestation is not due so much to the overall political competitiveness of a state but whether the party balance in the legislature makes it appear “flippable,” especially to the minority.
In a few cases state parties seem to have made intentional decisions about recruiting candidates. For example, between 2018 and 2020, the number of Democrats running in Tennessee fell by a third. But these kinds of leaps are the exception rather than the norm.
Surprisingly, we find that reforms such as term limits and public campaign funding (at least as it exists) do little to influence contestation. Reformers are likely to have more success tackling two aspects of the system that affect the supply of candidates. First, the method of redistricting itself is a natural target. The uncontested seat rate drops in the first election after a decennial redistricting, suggesting that altering the way that state legislative district lines are drawn could affect whether the parties field candidates. Second, there may be other innovations such as ranked choice voting that could invite more -- and more kinds of -- candidates to run for office. Without a shakeup in the system, control of Congress may remain competitive but too many voters will continue to lack the fundamental requirement of choice on the ballot.
Julia Nestel is an undergraduate student majoring in political science and Spanish at the University of Wisconsin-Madison.
Rochelle Snyder is a doctoral student in political science at the University of Wisconsin-Madison completing a dissertation titled “Constitutive Voices: Constituent Service and Representation in the United States Congress.”