By Mark Mellman - 05/27/14 08:05 PM EDT
For the last several weeks I’ve been trying to correct a meme that has been abroad in the land among journalists, academics, donors and even campaign strategists. In only slightly oversimplified form, progenitors argue that, in this hyperpartisan world, political persuasion doesn’t happen and that the only way for Democrats to win is to change the electorate through get-out-the-vote (GOTV) efforts.
Part of the misunderstanding arises from an uncomfortable fact that few in our business are willing to confront: Much of what we do is not ultimately dispositive. Both GOTV and persuasion strategies have limited impact. Circumstances beyond anyone’s direct control — incumbency, partisanship, the economy and candidate quality, among others — usually do far more to determine electoral outcomes than anything campaigns and their consultants do.
Winning requires understanding what’s possible in each circumstance. Persuasion can happen, but fewer voters are persuadable. I mentioned in last week’s column our client Sen. Heidi Heitkamp (D-N.C.) as evidence persuasion can happen. Apparently at least 12 percent of North Dakota voters were persuaded to cast ballots for both the senator and GOP presidential nominee Mitt Romney in 2012.
But Heitkamp is an outlier. Very few candidates can persuade that many to split their tickets. Recall that in Florida’s 13th District, David Jolly persuaded 1.4 percent of those who had voted for Alex Sink in her gubernatorial bid to desert her as she ran for Congress. It was enough to determine the outcome, but it is a small number.
Professor Seth Hill of Yale estimated that 4.4 percent of California’s electorate switched their votes from John Kerry in 2004 to Arnold Schwarzenegger in 2006, and that 9.2 percent switched from a non-Democrat in 2006 to Barack Obama in 2008. Meanwhile 4 percent of the Florida electorate switched from the Democratic gubernatorial candidate in 2006 to Sen. John McCain (R-Ariz.) in his 2008 presidential bid. Recall that academic studies on persuasion yielded average effects of just under 4 percentage points, with some producing no change at all.
In the turnout world, studies point to smaller effects, with a good program increasing turnout by 2.8 percentage points. But those numbers are not added directly to the bottom line. Logic and evidence suggests that kind of turnout increases yields by about a point (or less).
Of course, both increasing turnout and persuading voters cost money. And what they cost varies. Estimates suggest it costs between $47 and $87 to turn out a voter who would otherwise not come to the polls. For the sake of argument, let’s take the midpoint, $67 — at that price, turning out 200,000 costs about $13.4 million.
How much does it cost to persuade 200,000? Alas, no one knows for sure, so comparisons become difficult.
We do know that allowing an opponent to outspend you on TV can be politically deadly, though estimates of the impacts vary.
Across several studies, being out-communicated by 500 gross rating points can reduce a candidate’s vote total by 5 to 14 percentage points. In a race where 2 million votes are cast, if one takes the midpoint of the effect, that’s about the same impact as turning out 200,000 voters at a cost of $13.4 million. How much achieving that kind of TV superiority costs varies dramatically from market to market.
In the end, this column won’t provide a cookbook — and that’s the point. The debate between GOTV and persuasion is sterile and unproductive. Both can be effective, and both have limits. Both surely cost money. Good strategists figure out how best to allocate resources for the particular race in which they are involved.
But anyone who says one strategy works and the other doesn’t is not knowledgeable enough to earn a candidate’s trust.
Mellman is president of The Mellman Group and has worked for Democratic candidates and causes since 1982. Current clients include the majority leader of the Senate and the Democratic whip in the House.