The Race for the House, Part One

Ticket-splitting dwindling but still vital; GOP got a little more of it in key 2016/2020 races

Facebooktwitterredditpinterestmail

KEY POINTS FROM THIS ARTICLE

— Any analysis of House elections needs to start with the presidential race, as voting for the former has become so intertwined with the latter.

— That said, some voters do still split their tickets, and they may decide the House majority.

— Despite worries that Donald Trump would tank Republican House chances in both 2016 and 2020, that did not come to pass, and Republican House candidates were likelier to run ahead of Trump in key races than Democratic candidates were to run ahead of either Hillary Clinton or Joe Biden.

Ticket-splitting in a straight-ticket era

With Donald Trump appearing well on his way to a third straight Republican presidential nomination, his lone remaining major rival, Nikki Haley, is arguing that he would not just lose the presidential election, but also oversee a Republican defeat down the ballot, including in the race for control of the U.S. House of Representatives.

This question—whether Trump would cost Republicans in down-ballot races—is one that was on the minds of many in both 2016 and 2020, when Trump, respectively, won and then lost competitive presidential races. And it’s on our minds now as we survey the House battlefield.

Why start a House analysis with a focus on the presidential race? Because, as we frequently note in the Crystal Ball, we are in a highly nationalized era of elections, with the presidential results having a large influence on what happens in down-ballot races. Respected congressional scholar Gary Jacobson observed after the 2020 election that the correlation between the district-level presidential and House of Representatives results was an “astonishing” 0.987 (on a scale that ranges from -1 to 1). Just 16 districts produced split results that year: 9 Republicans won districts carried by Joe Biden for president, and 7 Democrats won districts carried by Donald Trump. That was the lowest such total of the postwar era, surpassing the previous low of 26 in the 2012 election.

And yet, just because there is not nearly as much ticket-splitting as there once was doesn’t mean that no one splits their ticket. These voters still exist, and their decisions are crucial at a time when the House is so evenly divided and winnable for either side—even if that side’s presidential nominee is losing.

This brings us back to Trump, and the question of the down-ballot impacts of his general election candidacy. For all of the GOP concerns about Trump, Republicans have done fairly well in the House both times he was on the ballot—and also generated a bit more ticket-splitting in their favor than Democrats did in theirs in the key races.

At the topline level, Republican House candidates performed better in the aggregate House vote in each election compared to Trump’s presidential popular vote performance.

In 2016, Trump lost the national popular vote by 2 points to Hillary Clinton. At the same time, Republicans won the aggregate national House vote by a point. In 2020, Trump lost the popular vote to Joe Biden by about 4.5 points, while Republicans lost the aggregate House vote by 3 points. This gap in favor of the GOP also exists when one corrects for uncontested seats in each election and considers only the two-party vote. According to calculations by the late political scientist Theodore S. Arrington, Republicans won this estimated iteration of the two-party House vote by about 1.5 points in 2016 and lost it by a little more than 2 points in 2020. The two-party presidential margin was basically the same as the all-party margin reported above in both 2016 and 2020, so the basic story remains—House Republicans ran a bit ahead of Trump in aggregate in both years, although by a little more in 2016 than 2020.

In terms of net change each year, Republicans lost 6 seats in 2016, but that still left them with 241 seats and the majority (and some of those losses were directly caused by mid-decade redistricting in a couple of states). Meanwhile, in 2020, Republicans netted 13 seats and cut the Democratic majority to just 222-213. One could reasonably argue that Trump was a bigger problem for Republicans when he was not on the ballot than when he was—Democrats won the House in a wave in 2018 at the midpoint of his presidency, for instance, and the Trumpiest candidates also cost the Republicans some key 2022 House and (especially) Senate races.

It’s possible that a different Republican presidential nominee would have done better than Trump in either or both of 2016 and 2020, and therefore provided more of a lift to Republican candidates. But it’s also possible such a replacement candidate would have done worse, too.

In both elections, Republicans were likelier to run ahead of Trump than Democrats were to run ahead of either Clinton (in 2016) or Biden (in 2020) in the most closely-watched districts.

We went back and looked at our final House ratings in both cycles and compared the House margins to the presidential margins in the districts we rated as most competitive (leaning one way or the other in our final ratings—many of these districts would have been Toss-ups throughout the cycle).

In 2016, 40 races were in the leans category in the final ratings. The GOP House candidate’s margin was better than Trump’s in 27 of those 40 races, while the Democratic House candidate’s margin was better than Clinton’s in 13 of the 40. Four years later, in 2020, there were 51 races in the leans category: The Republican margin was better than Trump’s in 32, compared to 19 Democrats who ran ahead of Biden. On average in the 40 2016 districts, Republicans ran 5.5 points ahead of Trump; in the 51 2020 districts, the average Republican overperformance was a more modest 2.1 points.

This is important to note because Republicans may very well need to once again run ahead of Trump in more places than Democrats run ahead of Biden, as they are more overextended into Biden-won territory than Democrats are overextended into Trump-won territory.

Based on the 2022 results, Republicans won 18 Biden-won districts and Democrats won 5 Trump-won districts. One of these Biden/Republican districts, NY-3, is currently vacant and will be filled in what both sides regard as a highly competitive special election next Tuesday. The sheer fact that Republicans won 13 more of these crossover seats in 2022 than Democrats did, combined with the reality that Democrats need to only win 5 more seats than they won in 2022 to flip the House, gives Democrats an obvious opening.

Of course, we haven’t even mentioned what the polls are saying right now—namely, Trump is polling better in the presidential race now than he did in either 2016 or 2020. If this persists, the operative House question may not be whether Republicans can run ahead of Trump, but rather whether Democrats can run ahead of Biden. It also may be that some of these Biden-won 2020 districts end up voting for Trump, and if Biden is the one who loses, it’ll be the Democrats who may need to produce more ticket-splitting than Republicans to flip the House. And perhaps that ends up being the dynamic, particularly if Trump asserts himself as a favorite in the popular imagination, and some small but important group of voters compensate for that belief by voting Democratic down the ballot in a bid to block a President Trump from having full control of Washington (we’ve written about this dynamic in previous cycles, based on a paper by political scientist Robert Erikson about ticket-splitting and voter expectations of the presidential race).

Haley and her campaign may be right—Trump could eventually lose the presidential race and drag down-ballot Republicans down with him. But that is far from a certainty, and Trump may win the presidency and restore unified GOP control of Washington—or lose the presidency in such a way that House Republicans can once again score a greater level of ticket-splitting in the key districts than Democrats, perhaps even allowing them to hold the House majority amidst a presidential defeat.

This is the first part of a two-part Crystal Ball analysis of the race for the House. Tomorrow we’ll take a look at some key districts, including the looming NY-3 special.