US House elections are equally unfair to everybody

The United States Capitol Building in Washington, D.C., on Dec. 25, 2018.
The United States Capitol Building in Washington, D.C., on Dec. 25, 2018.
Evan Golub/Zuma Press/TNS

If seats in the U.S. House of Representatives were allotted by statewide vote totals, Democrats would have 24 more members than the 235 they actually sent to Washington in the 2018 midterm elections. Among their shortfalls were four seats in Texas, three in North Carolina and two each in Indiana and Ohio. Due to shameless partisan mapmaking, Wisconsin Republicans were able to turn 46 percent of the votes into 62 percent of the state's delegation. 

But wait! Despite their losses from contorted districts, Democrats won 53 percent of House votes nationally and ended up with 54 percent of the membership. That seems a pretty equitable result.

Or take another measure. A basic stratagem in drawing district lines for partisan advantage is to force the other party to waste votes. That's what happened in November in notoriously gerrymandered North Carolina, where Democrats expended 544,235 votes on each seat they won and Republicans got theirs with 189,644 apiece. Yet nationally, Democrats averaged 255,197 votes per seat, almost identical to the Republicans' 253,994. I can't think of another year when the two sides were as close by that yardstick.

Here's the reason: Republicans fared almost as badly in heavily Democratic states as Democrats did in Republican-dominated ones in the comparison of seat allocation to statewide vote totals. In bright-blue states, the GOP got 21 fewer places than their votes warranted. Notably, they were two down in New York and Illinois, three down in New Jersey, and a dazzling 10 down in California.

How this occurred differed from the manipulations faced by Democrats. California's congressional districts are crafted by an independent panel. In New Jersey, New York and Illinois, the districts that Democrats won had previously gone to Republicans. That itself should be a sign of nonpartisan mapping.

So what brought about the GOP's underrepresentation? I found two causes, neither of which has received analysis or attention, largely because the publicly available tallies don't include those telling statewide vote totals or in-state party percentages. Only by calculating those figures, as I have now done, can a full picture emerge of what happened on Nov. 6.

First, Republicans chose to sit out 38 races. (In 2016, they passed on 22 to the Democrats' 24.) Thus the GOP not only surrendered seats, but also a chance to lift their total vote. Among the races they skipped were four in Texas, six in New York and eight in California. All these no-shows laid bare a lack of drive to be a national presence, masked by the hubris in the White House. I estimate that these blanks caused a dip of 2,473,998 votes cast for Republican House candidates.

Democrats opted to contest 432 of the 435 House seats. Gary Jacobson, a political scientist at the University of California at San Diego, said this was their most ambitious showing since 1974. It evinced an urge to show the flag, no matter how hostile the terrain. (In a sprawling Texas district that includes Midland, the Democrat settled for 18 percent of the tally.) This choice, plus record turnouts, gave Democrats their midterm electoral margin. It also attested to an energy that House Republicans couldn't match.

The second reason is more recondite. In 2018, Democrats increased their victory margin in some fairly competitive states, notably by doing well in reddish suburbs. (Tipped hats are also due for wins in the GOP strongholds of Kansas, Oklahoma, and South Carolina.) In the dozen states where Democrats had their best tallies, they carried an arresting 125 of the 148 districts. In Connecticut and Massachusetts, they took every one. Yet these states also had sizable pools of Republicans — in fact, 11,801,177 in the states together _ who came away with only 23 House successes.

GOP supporters in these states sunk 629,842 of their votes into each of their seats, while Democrats got theirs for 175,936. Few gerrymander templates attain that ratio. In California alone, Republicans ended up with only seven of the 53 places in their state's delegation. Geared to votes, they would have had 10 more.

Much is said about how Democrats coalesce in urban areas, where their votes are clustered in over-large majorities. It appears that Republicans are also assembling with fellow thinkers, even if the milieus can't be as readily identified as a Phoenix or a Philadelphia. As a result, Democrats have felt less need to manipulate maps. (In my reading, only Maryland is truly culpable on this score.)

As matters stand, there's no partisan consistency to the relationship between votes and seats. It comes close to pure equality in Republican Florida and Democratic Colorado. It's most divergent in Democratic New Jersey and Republican Kentucky.

Of course, nonpartisan maps are possible. Pennsylvania's, which was imposed last summer by its highest court, is an excellent model. But it's unclear how even a fair-minded map would aid California's Republicans, apart from adopting multi-member districts and preferential ballots. At the Supreme Court, justices have already voiced unease about having to umpire cartographic disputes.

So does the current composition of the House mirror the electorate's will? In an odd way, it does. Both parties' deficits in some states are redressed elsewhere. It's almost as if Arkansas Democrats have a voice via Rhode Island. Still, state by state, representation is woefully out of kilter. In 31 states, district maps help one party, albeit not always by willful design. The fact that averages are similar should not deter moves for nonpartisan commissions and elemental fairness.