The First Shall Be Last: The Dangerous Decline in Primary Turnout
Voter turnout statewide primaries in 1966: 33.5%
Voter turnout in 38 states with such primaries in 2006: 15.4%
(the lowest in history)
Ratio of voter turnout for those over 70 vs. those under 45 in Oregon"s 2006 primaries: 3.5 to 1
As the presidential nomination season is heating up, have you ever thought about the word "primary"? Going to the dictionary, one finds that the first definition is, "first or highest in rank, quality, or importance."
Well, in the case of voter turnout, the first shall be last.
Primary elections indeed determine who wins the great majority of our nation's elections for Congress and state legislature - most districts are safe for one party, and winning the primary is tantamount to winning the general election. But in contrast to turnout in November elections, turnout in primaries is sinking like a stone.
Plunging Turnout in Primaries: In 2006, for example, the Committee for the Study of the American Electorate reports that turnout in statewide primaries around the nation plunged by one-sixth to an all-time low of 15.4%, down from the previous low of 18.6% in 2002 - and less than half of the 33.5% turnout in 1966.
Oregon had the nation's sixth highest primary turnout in 2006 at 23.5%. Of those Oregon voters, more than two-thirds were older than 57 years old, with more total voters older than 70 than under 45 -- and those older voters turned out at a rate of three and a half times the rate of the younger set. Political consultant Rick Ridder estimates that the median age of voters in primaries is more than 60 - some 10 years older than the median age of voters in general elections.
These kinds of trends are similar by race and income. As one example, FairVote analyzed relative turnout in Oakland (CA) in the June 2004 primary and the November 2004 general election. Voter turnout in June was at least 47% lower than turnout in the November general election in every census tract predominantly made up of racial minorities. In contrast, turnout was 32% lower in predominantly white census tracts.
The Mounting Cost of Primaries: Primaries aren't cheap for taxpayers, either. In Washington State, for example, the cost of the 2004 congressional primaries was estimated at more than $10 million, approaching $2 per resident.
This all begs the question, just why is the government involved in primaries anyway? Primaries are a process by which private associations - e.g., political parties - select their nominees. Given how general elections often are lopsided, one might think it makes sense to force parties to be as democratic as possible in choosing their nominees. But when 85% are skipping the show, something is clearly not working.
Maine Secretary of State Matt Dunlap is one person who is asking the question. The top elections official in a state consistently among those with the highest turnout, Dunlap last month proposed Maine"s government get out of the business of regulating and paying for party primaries altogether.
The Pierce County Solution: Across the nation in Washington, Pierce County (home of Tacoma and nearly 800,000 people) last year voted to showcase a new approach in 2008: giving voters in county elections a full choice ballot in a "blanket general" election in November, with instant runoff voting in place to determine a majority winner. Voters will have full voter choice in a high-turnout general election.
Recommended by Robert"s Rules of Order and the League of Women Voters in several states, instant runoff voting (IRV) allows voters to rank their candidates in order of preference. If the voter"s first choice candidate doesn"t make the cut, their second choice is counted instead, and so on until one candidate earns an absolute majority of votes and is declared the winner of the race.
If the decisions are made by those who show up, as the old saying goes, it makes sense in a democracy to have a voting system that helps to ensure that more people participate. IRV has the potential to do just that.
Read more about instant runoff voting at instantrunoff.com or fairvote.org/irv. Previous editions of Innovative Analysis can be found here.
Comment on The First Shall Be Last: The Dangerous Decline in Primary Turnout
Current Discussion
-
Posted by AllAboutVoting, 09/08/2007 (3 years ago)
-
Alan who comments at F&V would be able to summarize the history of STV in Oz for everyone. Ireland elects its president under IRV, and that country has multi-party elections. IRV does not prop up a two-party system. Single-member districts are the key disincentive for multi-party elections.
Posted by Jack, 09/08/2007 (3 years ago) -
@Dan: IRV is used in Australia and they have an even stronger 2 party system. Stronger than ...? Stronger than the U.S.? I don't think so. Strong than they had before IRV? That would go back to approximately 1918. I don't know the answer, but I suspect that IRV in Australia did not destroy a pre-existing multiparty system. Does anybody know the history in Australia?
Posted by Bob Richard, 09/07/2007 (3 years ago) -
Dan, Yours is a common misconception. Voters actually _cannot_ fearlessly vote their sincere preferences with IRV. They are strategically forced to top-rank their favorite of the two "electable" (major party) candidates. This page explains why: http://rangevoting.org/TarrIrv.html Clay Shentrup San Francisco, CA clay@electopia.org 415.240.1973
Posted by Clay Shentrup, 09/07/2007 (3 years ago) -
Vague hypotheticals aside, the point is not to enhance or hinder the two parties, but to offer the chance for people to vote their conscience without feeling compelled to do otherwise because of a fear of vote splitting. IRV allows more voices into a debate without the danger of spoilers.
Posted by Paul, 09/07/2007 (3 years ago) -
Instant Runoff still has problems. For example lets say that you have 3 canidates, A B and C. Everyone votes for A or C as their first choice and B as their second choice. B is eliminated in the first round because B got no first choice votes. The rest of the country is divided between A and C and one side wins with a 51% vote, while 49% of the people have their worst pick. What you really want to optimize is voter satisfaction. Range voting, where each individual rates each candidate on a scale from strongly against to strongly support and the candidate with the highest average vote would work. There would be no NEED for primaries because you would have no 3rd party spoiler effect. IRV is used in Australia and they have an even stronger 2 party system.
Posted by Dan, 09/07/2007 (3 years ago)
>[Paul] IRV allows more voices into a debate without >the danger of spoilers. This is only true while third parties are small. If a third party gets larger nasty spoiler effects are likely to occur in IRV. From Minguo.info : >IRV can prevent the spoiler effect, in which a minor >party splits the vote with a major party and tilts >the election toward the other major party, hurting >its own cause. IRV therefore seems to allow >supporters of minor parties to cast protest votes >without "wasting" their votes. This advantage is >illusory, however, because it applies only as long >as those minor parties are sure to lose. As soon as >one of those minor parties threatens to become a >major party and actually win, its supporters vote >for them at the risk of hurting their own cause, >just as in the current plurality system. >Under IRV, votes for minor parties are therefore >symbolic at best, or dangerous at worst. An example of this follows at the link above. More to the point, the idea of getting rid of primaries and instead use a different voting system for the general election (as apparently is happening in Pierce County) seems very questionable to me. I say this as a Californian who saw the disfunctional CA governor recall race with 100+ candidates in it. There is a value to political parties to concentrate their support in one candidate.