Why Turnout Should Be Typical for the RI Primary

I’m seeing some chatter on Twitter suggesting incredulity at the idea of the total turnout for this primary breaking 20% or reaching the Secretary of State’s estimate of roughly 140,000 voters. Actually, that’s about right on the money.

Something that helps generate turnout is competition. A competitive election drives coverage, increases the amount of voter contact from campaigns (ensuring more voters show up at the polls), and generally gets people more excited to go vote. But that doesn’t tell the whole story.

For example, people have been pointing to elections data supplied by The Providence Journal‘s Katherine Gregg about turnout percentages:

The problem with it is (as Lake Research’s Matt McDermott points out) that it’s comparing September primaries. While this seems logical, it’s actually a bad way to estimate turnout for this year. You’ll notice that turnout crashes every presidential election year, even 2008, when Clinton and Obama energized massive turnout in one of the hardest fought primary elections in U.S. history. That’s because the presidential primary took place on March 4; turnout was a very high 32.5%. (You can pull all this data from the bi-yearly Countbooks issued by the Board of Elections.) But that’s not typical. Primaries tend to be more or less over by the time Rhode Island votes (as an example in 2004, the turnout was a scant 6.05% and about half that in 2012).

So obviously you can’t use 2012 presidential primary data. But why not the September data from the presidential years? Well, because it doesn’t cover the whole state. In 2012, the “competitive” election was Cicilline versus Gemma. Cicilline absolutely crushed Gemma, despite the sheer amount of ink spilled predicting his political demise (I was among the ink-spillers). The only statewide election was Sheldon Whitehouse’s easy nomination. 2008 was even simpler. Jack Reed crushed the living embodiment of the case against democracy Chris Young. 2004 was even worse, etc., etc.

So that leaves us with only three primaries to compare: 2002, 2006, and 2010. 2010 has the lowest turnout, and it’s important to interrogate why that’s so.

In 2010, a “competitive” elections was in Congressional District 1 for Patrick Kennedy’s seat. David Cicilline still won by 14 points. Jim Langevin had a spirited challenger whom he still triumphed over by 23 points. In both cases, the big names triumphed. Statewide candidates also tended to have big double-digit triumphs; only the Attorney General had a seriously competitive race (between Peter Kilmartin, Stephen Archambault and the late Joseph Fernandez), where over 95,000 votes were cast. That’s 20,000 more votes than Frank Caprio received in his uncontested waltz to being the nominee for governor. The other well-covered and “competitive” race was the nomination for Mayor of Providence. Angel Taveras won that handily. But it’s important, because it forced a lot of turnout in a Democratic stronghold. In 2002, there were competitive nominations for governor and mayor, and real turnout in Providence was even higher than in 2010.

Compare this to 2006, when the big driver of turnout was the Republican nomination for U.S. Senator between Chafee and Cranston’s Stephen Laffey. While turnout was similar to 2010 and 2002, the Republican share of that turnout was around 40%, over twice the share they usually receive. I’ve previously pointed out that Providence Republicans make up a small share of Republican primary voters statewide, and 2006 was no exception to this rule. It turns out that the Democratic-leaning voters of Providence aren’t going to go vote in a Republican primary that’s largely a clash between the Republican strongholds of Warwick and Cranston. However, a Republican senator who bucked the Iraq War and made a name as a maverick politician in year in which national Republicans were disastrously unpopular was worth supporting to many voters in the state during the primary.

But those statewide races drive votes, and 2014 has a large share of competitive statewide races, the likes of which we haven’t seen for a while. All but one of our statewide executive offices have competitive Democratic primaries, along with the chief executive of Providence. The chief executive of Warwick also faces a spirited primary, which is good news for the Republican primary (though not necessarily for any specific Republican candidate on the ballot there).

So, where does all this leave us?

1. We should expect 2014’s turnout in Providence to more closely mirror 2002 than 2010 or 2006.
2. We should expect Warwick to matter more in the Republican primary.
3. If we average the turnout from the last three gubernatorial-year primaries (working out to ~20.5%) and do some math against the number of eligible voters, we should expect over 150,000 voters in this election.
4. We should also expect a more typical share of those votes to go to the Democratic primary; something around 83%. Republicans should have a more typical 17% share of the vote. The competitiveness of Republican gubernatorial primaries does not appear to have an impact on Republican shares of primary votes.