Pundits are busy parsing the results from three recent statewide primaries in Ohio, North Carolina and Indiana held last Tuesday. Much of the focus has been on how well candidates backed by the Tea Party did. Of more interest to me, however, is the relative turnout among Democrats and Republicans. At first glance, the results don’t seem to bode well for the Democrats. Thus, Reid Wilson, writing for Hotline On Call, in his article titled “Dem Turnout Falls Off a Cliff”, notes that “Turnout among Dem voters dropped precipitously in 3 statewide primaries on Tuesday, giving the party more evidence that their voters lack enthusiasm ahead of midterm elections.”
Well, maybe. I’ve written before about the dangers of generalizing from a limited base of comparison, particularly given the idiosyncratic nature of midterm primary elections. With this in mind, is it the case that headlines trumpeting a collapse in Democrat turnout on Tuesday are justified? Note that Wilson draws his conclusion in part by comparing turnout in Tuesday’s Senate primaries with turnout in previous statewide elections for different offices, such as Governor, in these states. This comparison is not ideal – far better, I think, to try to compare turnout to previous midterm Senate primary races. That is precisely what I’ve tried to do below, although I caution that I had to piece together data from different sources and so can’t be completely certain of the accuracy of all these figures, although most are drawn directly from state records.
To start, I thought it might be interesting to see how the overall statewide turnout in these three states compares to previous midterm primaries, including the last “wave” midterm election in 1994. In the following table, I’ve pasted the overall turnout (not just turnout for Senate races) of registered votes in the statewide primary elections in all three states for elections in 1994, 2002, 2006 and 2010 – all midterm elections.
|State||1994 – Primary Turnout Statewide||2002 – Statewide Primary||2006 – Statewide Primary||2010 – Statewide Primary|
|Indiana||31%||22%||19%||Not Released Yet|
It does not appear, then, that turnout in these three set of statewide elections is abnormally low, although neither does it approach the levels achieved in the “wave” election of 1994, at least in Indiana and Ohio. (Interestingly, North Carolina did not have much of a turnout boost in 1994.) But this does not tell us whether Democrat turnout has dropped off while Republican turnout is up. Unfortunately, none of these states provides a partisan breakdown of turnout. To get some leverage on the question, I went back to look at actual turnout in previous Senate primaries in these states. All data is drawn from state election records, but in most cases I had to add up totals by hand, so reader beware. Note that all I can provide for now is absolute vote totals – not percentage turnout by party.
Here’s what I found, broken down by state:
OHIO – DEMOCRAT SENATE PRIMARY COMPARISON 2010, 2006, 1994.
In 2006, 747,404 votes were cast in the Ohio Democratic primary race for the Senate, won by Sherrod Brown, although it was not a tightly contested race. In the more hotly contested 1994 Democrat Senate primary, however, in which Joel Hyatt narrowly beat Mary Boyle, there were 934,891 votes cast. This past Tuesday, however, despite a closer primary battle between Lee Fisher and Jennifer Bruner, turnout in the Democrat primary only numbered 673, 597. So it appears there was a drop in Democratic turnout this past Tuesday from previous years.
INDIANA – REPUBLICAN SENATE PRIMARY 2010, 2006 and 1998 (partial)
In what was essentially a three-candidate contest on Tuesday for the Republican Senate nomination, 527,036 voters cast ballots in an election won by former Senator Dan Coats. In 2006, incumbent Senator Richard Lugar ran essentially unopposed in the Republican primary and drew 393,960 votes. In 1998, when Coats decided to step down as Senator, Republicans waged a tightly contested primary fight eventually won by Paul Helmke, who was defeated by Evan Bayh in the general election. Helmke received 129,309 votes in the Republican primary compared to his main rival’s 124,005, but I have not been able to get the overall totals for that primary as yet. With this caveat, it looks like turnout was up in Indiana among Republicans on Tuesday.
NORTH CAROLINA – DEMOCRAT SENATE PRIMARY 2010, 2002, 1998
Despite a closely contested Democratic primary between Elaine Marshal and Cal Cunningham this past Tuesday, only 423,453 voters cast ballots, with no candidate reaching the 40% threshold for avoiding a runoff election. In 2002, in contrast, 645,070 turned out in the Democrat Senate primary won by Erskine Bowles, who then went on to lose to Republican Elizabeth Dole in the general election. In 1998, John Edwards (remember him?) won the North Carolina Democrat Senate primary in which 540,031 voters cast ballots. Again, it appears that Democrat turnout is down last Tuesday.
It does appear, then, that historically speaking Democrat voter intensity, at least in these two Senate primaries in Ohio and North Carolina, was down this past Tuesday, while Republican turnout was up in Indiana. But I caution you not to read too much into this quite yet. I am not that familiar with all the candidates involved in these races, nor can I say much about contextual factors in each state that might have influenced overall turnout during past years. These are three races with their own idiosyncrasies, so I hesitate to project voter enthusiasm by party for the 2010 midterms.
Nonetheless, if I were a Republican I would be pleased. If a Democrat – maybe not. As a political scientist, I want to see more data!
UPDATE: the link to the Wilson article I cite above was not working – it should be fixed now.