The lesson of 2018 is that the political class is addicted to drawing lessons. Every two years, after the ballots are counted and the winners declared, our reporters, pundits, officials, activists, and analysts turn immediately to the next election. What do these results portend? Will Trump be reelected? Will the suburbs stay Democratic? This emphasis on the future allows the political class to indulge in its favorite activity: mindless speculation. For once, it might be more useful to look backward rather than forward. History has much to tell us.
What it says is that the midterm was about average. The New York Times projects the Democrats will pick up some 35 seats, giving them at least a 12-seat majority in the 116th Congress. The fundamentals pointed to this result. Only 2 of the last 14 presidents (FDR and GWB) have gained House seats in their first midterm. Republican losses are in line with historical trends for a president with less than 50 percent support. The Democratic gain is a few seats higher than in 2006, while less than Republican gains in both 1994 (54 seats) and 2010 (63 seats).
President Trump's approval rating in the exit poll was 45 percent. This is better than Reagan's approval in 1982 (42 percent) and about the same as Clinton's in 1994 (46 percent) and Obama's in 2010 (45 percent). Trump's approval is less than that recorded for Jimmy Carter in 1978 (49 percent) and George H.W. Bush in 1990 (58 percent). Carter and Bush lost seats in Congress too. Donald Trump may be an extraordinary man, but in political terms he is an ordinary president.
The difference between the House and Senate results is unusual. Not since 1970 has a president's party lost seats in the House while gaining them in the Senate. Nor were Democratic gains in statehouses as large as expected. At this writing, they have won the keys to seven more governor's mansions, but lost important contests in Ohio, Iowa, and New Hampshire. (Republican leads in Florida and Georgia have not been certified.) Democrats also won hundreds of state legislative seats, but nowhere near the amount needed to overcome the losses they experienced during the Obama presidency. The split decision makes a kind of sense: This year's Senate map favored Republicans, even as a shift among suburban voters and independents helped Democrats.
The high number of House Republicans who did not seek reelection, combined with a liberal gusher of money, was a boon for the party of Pelosi. The Democrats out-raised and outspent Republicans in what the Center for Responsive Politics says is the most expensive midterm ever. This advantage was especially pronounced in the House, where Democrats raised $951 million to the Republicans' $637 million. Money isn't dispositive. But it helps.
The 2018 election was not an outlier. The rule of divided government is one norm President Trump hasn't violated. Since 1968, no period of unified government has lasted more than four years. As John F. Harris and Charlie Mahtesian of Politico observed: "In the 38 years since Ronald Reagan's victory, presidents have faced having at least one chamber of Congress controlled by the opposition party in 28 years." I was born midway through Reagan's first year in office and have lived through almost every possible configuration of government. In January we'll be back where I began, with a Republican president, Democratic House, and Republican Senate.
After two years of "fire and fury," "little rocket man," Comey and the Mueller "witch hunt," political violence, "fake news," Gorsuch and Kavanaugh, a tax cut, an economic boom, the caravan, and "horse-face," not much has changed. Trump's 45 percent approval is about the same as his 46 percent of the popular vote in 2016. In the end, there probably will have been a five-point shift toward the Democrats in the national House vote, giving them their best result in 10 years. This is a significant win, but also likely to be temporary. Two years after the Democratic highs of 2008, Republicans came back and won 52 percent of the House vote. As Morris Fiorina of the Hoover Institution reminds us: This is an era of unstable majorities.
The result was not a win for President Trump. The loss of the House means an end to conservative legislation and the beginning of investigations. Still, this year could have been much worse for Republicans, for Trump, and for conservatism. It wasn't. So it is a loss with an asterisk, and within historical expectations.
This hasn't stopped pundits from suggesting that Trump may lose his job in 2020. Who knows? The president's advisers must want him to improve his numbers among the overlapping categories of independents, white voters with college degrees, and suburbanites. Trump narrowly won all three groups in 2016. Republicans lost them bigly this year.
Predictions two years ahead of an election are worthless. President Trump is polarizing and does not command majority support. Neither did Reagan, Clinton, and Obama at this point. All three were reelected. The Republicans fared poorly in the swing states of Wisconsin, Michigan, and Pennsylvania. Democrats lost all three in 2010 before Obama won them two years later.
Nor were Wisconsin and Michigan as bad as Republicans feared. Governor Scott Walker barely lost reelection to a third term. John James appeared out of nowhere to come within six points of a three-term incumbent whose margins in 2006 and 2012 were 16 and 21 points, respectively. Look at these losses in relation to the Senate victories in Missouri, Indiana, and North Dakota—the Florida result is not yet official—and to Mike DeWine and Kim Reynolds winning governorships in Ohio and Iowa. The working-class white voter has not abandoned Donald Trump's GOP.
Midterms are referenda on incumbents. Presidential elections are choices. Such was the case in 2016, and so will it be in 2020. Democratic primary voters may follow their passions and nominate a man or woman whom swing voters find extreme, unlikable, or unpalatable. The Democratic nominee may perform badly. Trump may be effective, as he almost always is, at branding his opponent and framing the debate. The political environment might include domestic growth and no new wars. If these conditions pertain, the president will be reelected. The path is there. It always is.
One other thing history teaches: Pundits will ignore history, over-interpret elections, and argue that the next election will be the one where we finally reinvent the political wheel in our closely divided, incredibly heterogenous, rich, grand, beautiful country.