The most under-covered story of this endless election is the battle for the House of Representatives. While it’s true that based on what we know right now Democrats are unlikely to take back the House, that statistical likelihood is being covered as a near certainty. Yet a shift of a few points in the national mood in the Democrats’ direction could completely reshape the post-election political scene. Considering that control of Congress is the difference between whether a President Clinton could enact her far reaching agenda or become further entangled in the legislative morass of the last six years, the possibility of such a shift is of major importance. If Democratic turnout is higher and Republican enthusiasm is tepid—and there is real reason to think it will be—the greatest drama as election night turns into the small hours of the early morning might be over control of the House.
The national polling data and the current state of play in races around the country suggest that Democrats will gain seats in the lower chamber, but fall short of the 30 pickups needed to regain control. Yet, the amount of coverage national media outlets and respected nonpartisan congressional analysts have given to the simple possibility that Democrats could take the House is far too low. Even though the national polling averages have remained steady in this election, the amount of uncertainty in the polls is far higher than four or eight years ago. A landslide victory at the presidential level would likely push Democratic House gains into the 25-35 seat range. That scenario is very much in play.
First, the cold, hard math. Republicans hold 246 House seats, with a vacancy in another heavily GOP district. Democrats have 186 seats, with vacancies in two strongly Democratic districts. With 218 seats needed for a majority, Democrats need to pick up victories in 30 of the approximately 47 GOP districts with competitive races. And that assumes that Democrats win back the 13 seats they currently hold but are in danger of losing. These 60 competitive races are listed in the chart below.
At first blush, this task might not be so daunting. Wave elections in 2006, 2008, 2010, and 2014 have brought significant changes. The Democrats won 30 seats in 2006 and another 21 in 2008. Republicans bounced back with an historic gain of 63 in 2010, before adding another 13 in 2014. All of that, however, obscures two unique challenges facing Democrats this decade: further partisan geographic sorting and widespread Republican gerrymandering. That is, due to shifts in both the types of people who identify as Democrats and in the geographic movements of those people, Democrats are increasingly located in smaller and smaller areas of the country. The consequence of this phenomenon is that no matter how one draws districts, Democrats are disproportionately likely to be found in heavily blue areas, while Republicans have shallower but broader support.
This trend was then exacerbated by the 2010 census and subsequent redistricting process. The 2010 elections gave Republicans governorships and state legislatures across the country right before the federal congressional maps were to be redrawn. Unsurprisingly, Republicans (as Democrats have done in decades past) used the redistricting process to partisan advantage. The interplay of self-sorting and partisan redistricting means both darker blue districts and more uniformly and solidly red ones. According to the Cook Political Report’s Partisan Voting Index (PVI)—which measures how congressional districts voted in the last two presidential races compared to the national average—38 districts in the country vote at least 25 points more Democratic than the national average, while only 16 seats have a Republican PVI of +25 or greater. At the same time, though, there are 149 districts with a Democratic leaning PVI of +6 and a whopping 183 GOP districts with a similar PVI in their favor.
In practice, this means that congressional Democrats nationwide must significantly outpoll their GOP brethren to gain seats. In 2012, Democratic House candidates won 1.37 million more combined votes in the nation’s 435 districts than Republicans did. Even so, it was Republicans who won a healthy majority, winning 234 races to just 201 for the Democrats. A Democratic majority of the two-party vote (and an overall 1.2% advantage) yielded only 46% of the available seats. Simply put, Democrats are at a significant disadvantage when it comes to control of the House of Representatives.
The polling data as it stands right now suggests that Democrats are a little short. Right now, Democrats enjoy a 4.3% advantage in the generic congressional ballot per the Huffington Post’s pollster aggregation model (the generic congressional ballot asks voters which party they are likely to vote for in their local congressional race). This metric is historically very accurate in predicting the overall structure of the post-election House, even though it says little about individual races where candidate strength, fundraising, and local factors can have outsized influences. Simulations by the Princeton Election Consortium suggest that Democrats are likely to wrest control of the House from the GOP if they win the congressional popular vote by at least 8%. A popular vote margin of between 6% and 8%, however, would make control of the House too close to call. So, if the generic ballot holds at current projections, Democrats will likely make gains but fall well short of control. Larry Sabato’s website for the University of Virginia Center for Politics places the likely Democratic gains at between 10 and 15 seats.
So, why, then, is the House in play? To be overly simplistic, 4.3% plus uncertainty plus Trump might equal 6-8%. That is, Democrats already enjoy a solid advantage, polling has been fairly erratic this cycle (i.e., a higher standard deviation than normal), and the unique candidacy of Donald Trump might very well depress Republican turnout. Hillary Clinton right now leads Trump by 5.5% according to FiveThirtyEight and by 6.5% per the Huffington Post in the national four-way race. That lead is even higher when limited to a simple Clinton-Trump head-to-head contest, likely because Gary Johnson and Jill Stein voters dislike Clinton but prefer her to Trump. If there is a modest polling error underestimating her vote, or if more Johnson and Stein voters elect to cast a meaningful vote (already Johnson and Stein are dropping precipitously in the polls as Duverger’s Law foretold they would), or turnout adds to her margins, a 10% Clinton win is within the realm of possibility. Indeed, early vote data and public polls suggest that Trump is winning a far smaller share of the Republican electorate than past GOP nominees and his erratic behavior and recent scandals may suppress turnout as well.
The assumptions that the Democrats are too far away to have any chance at the House ignore this uncertainty and instead examine only the dynamics of individual races. But House-race polling is often sparse and unreliable. In 2006 and 2010, complacent incumbents thought to be untouchable were caught flat footed against the national environment. How are the tens of GOP representatives in PVI R+3-6 seats going to survive if Clinton is winning nationally by 8-11%? Ticket splitting is becoming increasingly rare and it’s far from clear that voters are willing to separate the politics of Trump from down-ballot Republicans.
Take New Mexico’s second congressional district as an example of how this might work. NM-2 is an R+5 district, meaning that Mitt Romney won it 52-45 while losing the country 51-47. During the 2012 election, incumbent representative Stevan Pearce won the district 59%-41%, running seven points better than Romney. Though his seat is now considered safe, Pearce is a hard-right candidate who drew a capable challenger. And if Clinton shifts the national vote to D+10 and Pearce drops a few percentage points of the 7% ticket splitters from 2012, then he’s suddenly looking at a likely 51-53% vote share. Sure, that puts him on track for reelection, but don’t forget that in addition to the 47 GOP-held seats identified by prognosticators as currently competitive, there are another 50 or so Republican seats in R+3-7 districts where GOP incumbents usually win 58-61% of the vote. Democrats don’t need to beat Pearce to win back the House; they just need to beat 15% of representatives like Pearce.
Again, a Democratic House majority is neither a given nor even more likely than not. But a landslide presidential victory would most certainly put the House in play. With Hillary Clinton poised to win the presidency by a comfortable margin and Democrats in great shape to win the Senate, the most dramatic race on election night could be whether Democrats can eke out a narrow majority in the House.