Democrats are wrong about why Republicans control the House
Many Democrats and liberal-progressive pundits are obsessed these days with the idea that they are being robbed of victories by the structure of the American electoral process. One element of this complaint is the charge that the Republican dominance of the House of Representatives is due to partisan gerrymandering. Former attorney general Eric Holder has pledged to dedicate himself to attacking gerrymandering, and former president Barack Obama has likewise indicated that he will make it a major cause of his post-presidency, the most directly partisan focus of any former president in living memory. A November court decision striking down Wisconsin’s state-assembly districts, solely on grounds that they unduly favor Republicans, represents a renewed legal assault on the practice, despite the Supreme Court’s skepticism over the past 30 years that the courts have any principled and workable way to referee purely partisan district-drawing fights.
Gerrymandering is no novelty; it’s as old as American democracy itself. The practice is named for Elbridge Gerry, who signed the Declaration of Independence, was a delegate to the Constitutional Convention, served in the first Congress, helped write the Bill of Rights, and was our fifth vice president (under James Madison). To anyone over the age of 40 (including Obama and Holder), a sudden Democratic enthusiasm for stopping gerrymanders reeks of hypocrisy and partisan opportunism. Democrats dominated the House of Representatives for four uninterrupted decades from 1954 to 1994, and they did so with the considerable help of partisan gerrymanders, abetted by entrenched Democratic control of many state legislatures. A number of states in the South, in particular, had continuously Democratic-run state legislatures from the end of Reconstruction into the 21st century. One of the first big gerrymandering controversies of the current era was a 2003 effort by Texas Republicans, after capturing control of the state legislature for the first time in 130 years, to replace a lopsided 1991 Democratic gerrymander. Democratic legislators fled the state in an effort to prevent its legislature from revising the 1991 map.
Moreover, Obama and Holder are both ardent fans of interpreting the Voting Rights Act to require race-conscious gerrymandering to create “majority minority” districts — i.e., districts in which a majority of voters are members of the same racial-minority group. In the redistricting of Illinois’s state legislature after the 2000 census, Obama insisted on protecting his own racial group, telling the Chicago Defender that, “while everyone agrees that the Hispanic population has grown, they cannot expand by taking African-American seats.” The VRA’s requirement that legislatures engage in racial gerrymandering — sometimes compelling the construction of districts shaped even more oddly than Gerry’s original salamander — makes it illegal in many states to draw House-district lines without regard to outcomes.
Of course, Republicans — like Democrats, then and now — want to draw district lines to help them win more seats in Congress. But the evidence shows that the advantage conferred on today’s House Republicans by partisan gerrymanders is neither large nor historically unusual, and is outweighed by the natural advantages of geography.
First, a bit of history. Elections analyst Sean Trende has noted that, from 1942 to 1992, “the Democrats had a huge advantage in seats won vs. their popular-vote share, averaging 5 percent,” and that “the discrepancy was less than 3 percent on just five occasions.” Trende reached this conclusion by comparing the Democrats’ share of House seats to the Democratic share of the two-party popular vote. Using Trende’s measurement in today’s House, a five-percentage-point advantage is worth an extra 22 House seats. Today’s Republican majorities regularly enjoy a smaller edge than the Democrats averaged over that half century of dominance. The Republican advantage was 4.9 points (worth an extra 21 seats) in 2016, when Republicans won 55.4 percent of the seats with 50.6 percent of the two-party vote. It was smaller in the three previous elections: 3.8 points (17 seats) in 2014, 4.4 points (19 seats) in 2012, and 2.1 points (9 seats) when the Republican majority was first elected in 2010, before the latest round of redistricting.
Not so long ago, Democrats were pocketing winnings from their own advantage. Democrats had a 3.8-point advantage in 2008, giving them 15 more seats than they’d have won just from their share of the national two-party popular vote. Those 15 extra seats were more than twice the seven-vote margin by which the House majority elected in 2008 passed Obamacare, yet few liberal critics of gerrymandering seem to think the democratic legitimacy of Obamacare is somehow in question as a result.
More fundamentally, however, partisan gerrymandering is not the sole reason Republicans have 15 to 20 more House seats than you’d project just from a proportional application of the national popular vote. It’s probably not even the most important reason. While Republicans do benefit from a number of partisan gerrymanders, so do the Democrats, and other factors are at work, ranging from natural population distribution to federal restrictions derived from the Voting Rights Act.
Political scientists Jowei Chen of the University of Michigan and Jonathan Rodden of Stanford University’s Hoover Institution conducted an in-depth study in 2013 of legislative-district lines, using computer simulations of precinct-by-precinct voting patterns to map alternative redistricting plans without regard to partisanship or race. What they concluded was that Republicans have a natural advantage conferred by the “human geography” of Democratic voters’ concentrating disproportionately in overwhelmingly liberal urban districts while Republican voters are more evenly distributed in the suburban, exurban, small-town, and rural districts.
Chen and Rodden found that this “unintentional gerrymandering” produced an average Republican bias of five points nationwide, or seven to eight points in such states as Pennsylvania and Georgia.In Florida, their main test case, they used the 50–50 precinct-by-precinct Bush and Gore votes and found, based on random computer simulations of a map of 25 districts (the number of House districts in the 2002 redistricting), that Republicans hadan average of 61 percent of the House seats. As Chen and Rodden explained in early 2014: “In the vast majority of states, our nonpartisan simulations produced Republican seat shares that were not much different from the actual numbers in the last election.” That’s before even considering the impact of the VRA.
Moreover, Chen and Rodden found the trends to be most pronounced in states with rapid urbanization — and the 2016 election featured an unusually stark urban–rural split. House Republicans also benefit from the advantages of incumbency, in many cases dating from before the friendly maps drawn after 2010. Paul Ryan, for example, was originally elected in a Democratic-leaning district in 1998; in 2016, running in a district rated R+3, he won by 35 points.
Gerrymandering has also been falsely accused (by former president Obama, among others) of being the key factor in the increasing ideological and partisan polarization of Washington. But political scientists who have tried to measure this have concluded that polarization has little to do with gerrymandering, noting that the increased polarization of swing-district representatives and senators mirrors that of safe-district House members.
And while the 241–194 House majority elected in 2016 benefited from a 21-seat advantage (even as House Republicans had 50.6 percent of the two-party national vote), a look at the numbers suggests that most of this advantage was owing to factors other than gerrymandering.
First, the winner-take-all nature of House races, along with the limited number of districts, means that a party that wins a majority of the two-party popular vote in a state will almost always take more than its share of the House races in that state. In Nevada, Democrats won 50.5 percent of the two-party vote but took three of the four House seats; in New Hampshire, Republicans got 48.4 percent of the vote and lost both races. The party that won a majority of the two-party vote won more seats than its share of that vote would have predicted in 46 states and as many as the vote share would have predicted in Vermont, where the Democratic candidate ran unopposed to win the state’s only House seat. In Maine, the only state with an evenly divided House delegation, Democrats won 52 percent of the vote and carried one of the state’s two House seats. The only states in which the party that won the two-party vote failed to win a majority of House seats were Wisconsin and Virginia.
Seven states have only a single congressional district. In 2016, Republicans gained one more seat in those states than their share of the two-party popular vote would have suggested: They won five of the seven seats (71.4 percent) compared with their 52.9 percent of the two-party vote — an 18.5-point advantage. That has nothing to do with redistricting. On the whole, opportunities for serious gerrymandering require that a state be divided into more than a handful of districts. Yet, overall, across the 24 states that elect five or fewer House members, Republicans won 40 of 65 House seats (61.5 percent) with 55.3 percent of the two-party vote, a 6.3-point advantage that yielded them an extra four seats. That’s a small-state advantage, not gerrymandering.
Even in larger states, as in the Electoral College, winner-take-all elections tend not to produce precisely proportional results. In California, one of four states where district lines are drawn by a nominally nonpartisan commission, Republicans got 35.2 percent of the two-party vote but won just 26.4 percent of the House seats, a net 8.8-point, five-seat advantage for Democrats. That’s despite the fact that Republicans ran no candidates in nine of the state’s 53 districts, which inflated the Democrats’ popular-vote advantage (Republicans got 40.4 percent of the vote in the other 44 districts but still won just 31.8 percent of those 44 seats). Republicans won over a third of the vote across Connecticut, Maryland, and Rhode Island, and won just one out of 15 House races in those states.
Across the 17 states in which they hold a majority of the House seats, Democrats won 61.7 percent of the two-party vote but 72.9 percent of the House seats, yielding an extra 19 seats. They won more than their proportional share in every one of these states but Vermont. Democratic gerrymanders in such states as Illinois and Maryland were one reason for this. Another was simply the diminishing returns in winner-take-all races when a party slips below 45 percent of the two-party vote statewide. That happened to Republicans in twelve states, to Democrats in 24 states. No amount of creative mapmaking can make many district majorities out of 35 percent of the statewide vote.
What about the two states in which the results bucked the statewide popular vote? Republicans in Wisconsin won five of the state’s eight House seats (62.5 percent) with just 47.9 percent of the statewide vote. How? They let House Democrats run unopposed in two of those eight districts, one of which (Ron Kind’s D+5 third district) Trump actually won by five points. In the other six, Republicans won 58.5 percent of the two-party vote and carried five districts. Had they run candidates in those two districts, a reasonable estimate would have them winning more than 52 percent of the two-party vote statewide, comparable to Republican majorities of the two-party vote in the Senate and presidential contests, which were 51.7 percent and 50.4 percent, respectively. Even if House seats were awarded proportionally in Wisconsin, Democrats would have won only one additional seat. And proportional districts would be hard to draw: Hillary Clinton carried just twelve of Wisconsin’s 72 counties, mostly in the Madison, Milwaukee, and Eau Claire areas; there were only two places in the state where she carried adjacent counties.
In Virginia, Republicans won seven of eleven seats (63.6 percent) with 49.8 percent of the statewide two-party House vote. The map was likely a larger factor in Virginia than in Wisconsin, but once again Republicans would probably have won a majority of the two-party vote just by running a candidate in every district. Democrat Gerry Connolly ran unopposed in a district where Trump got 29 percent of the two-party vote. Republicans won seven of ten contested seats with 53.4 percent of the two-party vote, and would have cleared 51 percent statewide if they’d won just the Trump voters in Connolly’s district. Democrats in Virginia are also highly concentrated in a few urban areas; despite winning the state, Hillary Clinton carried just 30 percent of its 133 counties, and no adjacent counties outside the Virginia Beach, Richmond, and D.C.-suburbs areas.
Political professionals put great effort into favorable gerrymanders, just as they do with many aspects of campaigning that matter at the margins. But the margins matter only when the race is very close. If Democrats want to retake the House, their best bet is a combination of the factors that benefited them in 2006: a national Democratic wave and the recruitment of more House candidates appealing to voters outside the big blue cities. Complaining about a practice that dates back to the Founding Fathers won’t make either of those things happen.