With resignations in Congress piling up and rumors that two or three dozen members could face charges of sexual misbehavior in the near future, it’s not too early to start thinking through the political impact. Could such a wave of resignations affect swing the House?
Nate Silver thinks it “probably helps Democrats’ chances” to do so, further explaining:
1 — GOP has more exposure to losses because they hold most of the swing seats in the House
2 — Voters will “punish” members accused of harassment by enough to mitigate their incumbency advantage
3 — Failing to oust creeps from the party (while D’s do) would hurt GOP brand
Now, we have no idea if that pattern will hold, and in fact later Thursday Arizona Republican Trent Franks resigned for sexual harassment. But it’s still worth working through how this would play out because it’s a good way of discussing how House elections work.
First, Silver’s initial assumption is a mistake. If 40 members of the House are engulfed in scandals, they are unlikely to be split evenly between the parties. Right now, there are 240 Republicans and 193 Democrats in the House (with each party defending one open safe seat). But if all 40 of the hypothetical scandals are about men (not an unreasonable assumption) then the relevant numbers are the 218 Republican men and 133 Democratic men. So we could expect, all else being equal, about 25 Republicans to face trouble for every 15 Democrats.
We could then add his first condition: Republicans overall are defending more difficult seats, as any majority party will. Beyond that, more of the Republican seats are likely to be competitive than the Democratic seats given this is a midterm election with a Republican in the White House, especially a very unpopular Republican.
So what happens from there? Fifteen new Democratic open seats in November 2018 would probably mean relatively little risk for Democrats. Of 14 seats Democrats are currently vacating, Cook Political Report considers 9 safe, 1 likely Democratic, 1 lean Democratic, and 3 toss-ups (by contrast, Cook has the open Republican seats as 13 safe, 2 likely Republican, 2 lean Republican, 3 toss-up, and one 1 lean Democratic). So another set of open seats would probably put few if any in danger. It’s true that resignations this far from the November elections would still require a round of special elections to select someone to complete the unfinished terms, so Republicans would have two chances to win. But then again if a Republican won a special election in a normally Democratic or swing seat in February or March, Democrats would have a chance to reclaim it in November. Overall, it just wouldn’t cost the Democrats many, if any, seats.
As far as the Republican-held seats are concerned, Silver got some pushback on Twitter about the idea that incumbent Republicans would be hurt by scandal given that Donald Trump won and Roy Moore is ahead in the polls right now. What we’re interested in, however, isn’t who wins, but how scandal affected the margin of victory. In the presidential election, both candidates were scandal-plagued, and Trump had so many scandals that it’s hard to isolate the effects of any of them. Moore, too, has a history of being a weak statewide Republican in Alabama. As it was, Trump did somewhat worse than fundamentals-based projection systems based on the economy and other non-candidate factors predicted, and Moore is likely to do much worse than typical Alabama Republicans. So neither of these contests is evidence that the normal rules don’t apply.
Gary Jacobson has shown that the incumbent advantage may be primarily based on voters typically learning more positive things than negative things about their members of the House, while they are unlikely to know much about the challenger. Now, candidate factors are not as important as party loyalty in House elections, but they do have an effect, and Silver is correct that scandal can wipe out the incumbency advantage, or even worse.
According to Cook, Republicans are defending 54 seats that are classified as toss ups, lean Republican or likely Republican (compared to 14 on the Democrats’ side). Making any of those a little more vulnerable can have an outsize impact on the fight to control the House.
As far as Silver’s last point: Right now, I’m extremely skeptical that to GOP brand will sustain any damage at all. Moore and Franken will have faded from the headlines months before voters start paying attention to the 2018 elections, and the impact (outside of Trump) is likely to be muddled anyway.
Overall? A sudden scandal hitting 40 incumbent male members of the House would play out very well for the Democrats — unless they were very unlucky in which seats were affected.
Jonathan Bernstein is a Bloomberg View columnist. He taught political science at the University of Texas at San Antonio and DePauw University and wrote A Plain Blog About Politics.