House Elections Aren't Rigged for Incumbents

Voters tend to respond to things they know about candidates.
Being the incumbent didn't help House Majority Leader Eric Cantor fend off a primary challenge.

Ezra Klein has a nice item about whether elections are rigged for incumbents, following up on a poll result showing many voters believe incumbents have an unfair advantage. It's true that House incumbents win at very high rates, more than 90 percent of the time. Does that mean elections are "rigged"?

Not really.

The key here is that we're talking about districted elections in the first place. We know one thing about all House incumbents: they won the district the last time. That means either they had some sort of advantage the time before (and if they've been in for multiple terms, going back to their initial election), or they are excellent politicians, or both. Either way, it's hardly a surprise, or a problem, that the next election tracks the results of the previous one.

And voters don't reward incumbency per se. Instead, as Gary Jacobson shows, voters tend to respond favorably to positive things they know about candidates, and unfavorably to negative things. For House elections, which the news media largely ignore, the practical upshot of this often comes down to voters knowing one or two positive things about the incumbent (she got a grant to add a new lane to the freeway, his office helped my neighbor deal with Social Security, she showed up a local event and was really nice to Aunt Bertha), and nothing about the challenger.

What could change that? Klein suggests ending gerrymandering. Although that might mean fewer lopsided districts in terms of partisanship, it wouldn't affect the main dynamic very much.

The way to get rid of any incumbency advantage would be to make voters choose among party lists, rather than individual candidates. But the cost would be the loss of particularized, individual representation.

Barring that, the way to give challengers a real chance is to get money to them so they can run real campaigns. These days, both parties (sensibly, from their point of view) target only a handful of competitive seats, and put all available resources into them. That's why in most House races, no one knows anything about the challenger. My own preference would be partial public financing of House elections: Give every major party nominee $250,000 or even $500,000, and then let them raise money above that (I'd remove all other limits except disclosure requirements). That's not going to result in a lot of incumbents losing, because most members of the House do a good job of representing their districts - again, for the same reasons they won in the first place. But at least it would make it more likely that a fully engaged campaign could break out, and that would give challengers a chance.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.