Regular readers of this blog were likely not surprised by Tim Pawlenty’s decision to drop out of the race for the presidency after his disappointing third-place finish in Saturday’s straw poll. In an earlier post I had noted, based on the debate and polling data, that his candidacy was in trouble. Meanwhile, Michele Bachmann tried to capitalize on the shower of media coverage, including this harsh grilling by David Gregory on Sunday’s Meet the Press show, which accompanied her victory.
In their rush to make sense of Saturday’s results, however, the media largely ignored one of the more interesting facets of the Iowa straw poll: that voters had to pay to play. Because the straw poll doubles as a Republican fundraiser, participants were required to pay a $30 fee to vote. No fee, no vote. There’s a pretty good chance, of course, that candidates took on the costs of not only busing their supporters to the polls, but also paying their entrance fee. Which raises the obvious (to me, if not to the media) question: is there any relationship between the results of Saturday’s straw poll and the amount of money each candidate had on hand to spend? In the table below, I ranked the candidates by the votes they received compared to the amount of discretionary cash they had on hand as of June 30, which is the last quarterly fundraising reporting period. I only include the five candidates who actively competed in the events leading up to the straw poll (so, no Mitt Romney). Note that I don’t have fundraising data on Herman Cain or Jon Hunstman.
|Name||Votes||Percentage||Cash On Hand|
What do you know? There is an almost one-for-one correspondence between the amount of discretionary cash a candidate has and their performance in the straw poll – the more cash on hand, the better the candidate did! Most importantly, Bachmann led the pack in discretionary cash and in votes received. So, did Bachmann win because her campaign personally footed the roughly $150,000 it cost people to vote for her? Probably not. Instead, the link between the two measures – votes received and cash on hand – is likely that they both reflect the candidate’s ability to attract support from the issue and partisan activists who dominate these types of events. Keep in mind that Saturday’s event drew less than 17,000 people. Those who do show up are committed partisans – not Joe and Jane Six-pack. They are the same people who are likely to contribute money to these candidates early in the nominating process.
My point is that Bachmann, and Paul, performed well in Ames not necessarily because of any broad-based support among Iowan voters, but because they do disproportionately well among the core group of committed activists who both attend straw poll events and give money to candidates. It bears repeating that, media reports to the contrary notwithstanding, all those “small donors” the Obama campaign so loudly proclaimed were an indication of their grass roots support among new voters in 2007-08 in fact indicated nothing of the sort – these people making small donations are typically the hard-core ideologues that provide much of the electoral impetus for the polarized nature of Washington politics today.
Does the relationship between discretionary cash and votes received hold for previous straw polls? It’s not quite as strong in 2007 and 1999, the two previous straw polls for which I have data, but it’s not non-existent either. Note that in 2007 Romney was the clear discretionary cash leader and he won the Ames straw poll rather convincingly. To be sure, there’s not a one-to-one correspondence for candidates below Romney between vote share and cash on hand (Ron Paul in particular throws the figures off). Note that I don’t include John McCain, Rudy Giuliani or Fred Thompson since they didn’t compete in the event.
|Name||Votes||Percentage||Cash on Hand (millions)|
We see similar results in 1999; if we exclude Steve Forbes, who was largely a self-funded candidate, we see that Bush was the clear discretionary cash leader, and he easily won the straw poll. Again, the correspondence becomes a bit murkier for candidates below the winner.
|Name||Votes||Percentage||Cash On Hand|
|George W. Bush||7,418||31.3||6,355,370|
If we pool the results from the three straw polls and run a simple regression, cash-on-hand is a statistically significant predictor of vote percentage (adjusted R-squared = 0.47), based on 19 data points. Anna Esten charts the pooled relationship below.
Of course, if I’m right and the relationship is not simply a function of candidates buying votes, but instead a reflection of support among party and issue activists more generally, a better predictor of vote share might be the share of money each candidate raised in small donations. If I get a chance, I’ll run those figures as well.
Pending those results, however, the bottom line is this: in explaining the results of the Ames straw poll, cash-on-hand isn’t everything. But it does seem to correlate at least in part with the candidates’ straw votes totals. And that means, consistent with my earlier post on this topic, I would be skeptical of overstating the significance of Saturday’s outcome. Rather than a measure of general support in Iowa, the results instead are a better gauge of the intensity of support among core activists for particular candidates. This is not insignificant, of course, since these intensely committed activists form the shock troops that can be counted on to support a candidate throughout the campaign. But to win the party’s nomination, candidates need to broaden their support beyond this core group.
Addendum: According to this CNN article, Bachmann purchased 4,000 tickets to dole out to potential supporters, and Paul paid for a couple of thousand. See: http://politicalticker.blogs.cnn.com/2011/08/13/bachmann-paul-camps-getting-supporters-to-the-straw-poll/