The Real Reason Trump Can't Break the GOP
“There is a belief in many quarters that the Republican Party is about to disappear.” --Review of Reviews, 1893
Political scientists Jacob Hacker and Paul Pierson’s recent piece on the future of the GOP starts out well enough. The first half points out, correctly, that despite predictions of doom for the GOP in 2009 and 2013, the party is actually in reasonably strong shape today. As David Byler and I put it one year ago, when presenting our partisan index in the wake of the 2014 elections: “Overall, this gives the Republicans an index score of 33.8. This is the Republican Party’s best showing in the index since 1928, and marks only the third time that the party has been above 15 in the index since the end of World War II.” Notably, as Hacker and Pierson observe, this occurred despite the GOP having done absolutely nothing that was recommended in the wake of its losses in 2008 and 2012.
But the piece quickly goes off the rails. They see these successes as a product of three factors: GOP strength in rural areas, which gives it an advantage in the House and Senate; the midterm electorate, which is less demographically diverse than the presidential electorate; and Republican anti-government zeal, which plays well in midterm elections.
The GOP’s strength in rural areas certainly helps it in Congress. But Republicans won the popular vote by six points in 2010 and 2014: two of the four largest popular vote margins for Republicans since 1928. There’s certainly a skew between the Republicans’ popular vote margins and the number of seats that they win in Congress, but that is historically the case (Democrats benefited in the pre-1994 era). Regardless, you can’t explain away six-point national wins on the basis of regional variation in support.
You could explain this away, however, by noting the demographic differences between the presidential and midterm electorates, which is the second argument. But as Hacker and Pierson note, over time the nation’s changing demographics should work against Republicans in the midterm electorate as well. Indeed, the racial composition of the 2014 electorate was nearly identical to that of the 2008 electorate, one in which it was supposed to be nearly impossible for an unreconstructed Republican Party to prevail. It is somewhat anomalous for the GOP to be winning historic margins if long-term demographic trends were driving midterm elections against them at the same rate as presidential elections, with only a modest difference in the baselines. Moreover, as Harry Enten and I have noted, demographic differences between midterm and presidential electorates simply don’t account for nearly as much as the conventional wisdom seems to hold.
That leaves us with the third option: that the GOP’s anti-government attacks render it well-suited for midterm successes. This is possible, even if it is difficult to measure. A better explanation seems to be that presidents’ parties almost always do poorly in midterm elections, whether due to the removal of presidential coattails, a tendency of the American people to hedge their bets, or some combination of the two. This is what political scientists have long surmised, and it can explain elections like 1998 without resorting to special pleading: President Clinton had no coattails to remove in 1998 (indeed the Democratic Party was still below its historic strength in the House in the wake of the 1994 disaster), and the American people were in no mood to hedge their bets or punish the president’s party in the middle of a historic economic expansion.
So why has the GOP had such a hard time winning presidential elections? There’s a simple explanation. Consider Alan Abramowitz’s “Time for Change” model, which looks at the relationship between incumbency, presidential job approval, GDP growth in the second quarter, and election outcomes. I’ve approximated his inputs, as well as the predicted and actual outcomes from 1980 to 2012. I think I’ve specified the GDP variable differently than he did, so this isn’t an exact replication of the model, but it is pretty close (note: The “predicted” and “actual” vote shares refer to the vote shares for the incumbent party, as do the job approval numbers):
Here’s why I think the GOP hasn’t won many presidential elections recently: It wasn’t supposed to. It lost 1992 because the public was tired of GOP control and George H.W. Bush was pretty unpopular. It lost 1996 because the economy was growing strongly and President Clinton, a first-term president, was popular.
It won in 2000 amid a strong economy and with a popular president in the White House, but it wasn’t really supposed to; this works against the idea that there is some sort of inherent GOP disadvantage in presidential elections. In 2004 a modestly popular first-term president won re-election in a growing economy, while in 2008 Republicans lost amid an economic collapse (no surprise there). Despite the idea that President Obama was somehow uniquely weak in 2012, in fact he performed almost exactly how the fundamentals suggested that he should.
But this works both ways. Republicans didn’t win from 1980 to 1988 because of the unique strength of Ronald Reagan; they won because we were in a recession in 1980 and had massive growth in 1984. Likewise, George H.W. Bush benefited from strong growth in 1988 and a popular incumbent.
I’m not a “fundamentals fundamentalist”; campaign effects do affect outcomes, and there are decent reasons to believe that Donald Trump would underperform fundamentals significantly. But the truth is, Republicans losing the popular vote in six of the last seven elections is as meaningless as Democrats losing the popular vote in seven of the 10 elections from 1952 to 1988 (eight if you count Richard Nixon as the winner of the popular vote in 1960). As a matter of fact, even if we decided elections by coin flips, we’d expect to see runs like this with almost the exact same frequency as we actually see them.
So even if Trump is the Republican nominee, and costs the GOP an election (and it is worth remembering that right now the fundamentals merely point to a very close presidential election), the GOP would probably be in a position to bounce back quickly.
This leads to the real reason that Trump is unlikely to hurt the GOP permanently: In the grand scheme of things, he is really not a big deal. For example, after the Watergate debacle and election of Jimmy Carter, Everett Carll Ladd Jr., wrote: “The Republican party cannot find, outside of the performance of its presidential nominee, a single encouraging indicator of a general sort from its 1976 electoral performance. . . . [W]hat we see manifested here is a secular deterioration of the GOP position. The Democrats have emerged almost everywhere outside the presidential arena as the ‘everyone party.’” Yet four years later, the Republicans won the first of three landslide presidential wins, and 18 years later it began dominating congressional races.
From 1928 to 1932, the Republicans presided over the near-obliteration of the American economy. After the 1936 elections, there were almost as many Democratic senators – 76 – as there were Republican representatives -- 88 – prompting Ogden Mills to tell Herbert Hoover that “[h]ow to revitalize the Republican party under such conditions looks almost impossible.” Yet six years later Republicans won the popular vote for the House by a healthy margin, and 10 years later it controlled both chambers of Congress.
Or consider 1912, when a split in the Republican Party between progressives and conservatives led to a massive rupture. Republicans controlled only 31 percent of the House seats after that election (131 in today’s terms). Their presidential nominee finished third, winning a paltry 23 percent of the popular vote and just eight electoral votes (Utah and Vermont held steady for William Howard Taft). Yet just four years after that, Republicans were just 2,000 votes in California from winning the presidency. Republicans won the largest popular vote margin in history in 1920, with a decidedly mediocre nominee.
But probably the greatest example of party durability is seen by the Democrats, who, in 1866, found themselves on the wrong side of the Civil War. That is, as we dryly noted when I practiced law, a bad fact. The party didn’t even field a nominee in 1872, instead endorsing the Liberal Republican candidate Horace Greeley (who died before the Electoral College voted). Yet in 1874, the party picked up an astonishing 94 seats (which would be 140 seats under today’s terms). Two years after that, Democrats won a majority of the popular vote (yet lost the Electoral College in a disputed election). They won the presidency outright in 1884, less than a decade after the end of Reconstruction.
So even assuming Trump loses badly (I think he will, but it is too early to say anything definitively), it is nevertheless unlikely that he destroys the GOP. The American public has historically had a decidedly short-term focus on elections. If the Republicans can come back rather quickly from overseeing the Great Depression, and if Democrats can come back rather quickly from losing the Civil War, we should expect any damage caused by Trump to be fleeting.