Five Observations on the Politics of Impeachment
As the House initiates an official impeachment inquiry and moves toward a likely vote to impeach, here are five observations on the politics of the moment:
Our previous impeachments shed little light on this one. It’s generally unwise to make confident predictions without a relatively large data set to draw upon. Here, we have only had four examples of serious impeachment proceedings.
The first two – John Tyler and Andrew Johnson – do not help us understand our present politics. Both involved presidents who had weak relationships with their respective parties and who had succeeded to the office after presidents of the opposing party had died while serving. Perhaps more importantly, both took place in the mid-1800s, long before the arrival of reliable public opinion polling.
The other two examples are more relevant, but in both cases it is nearly impossible to isolate the specifics of impeachment from the broader political context. Richard Nixon’s case would seem to offer a good example for Democrats, as his job approval suffered an astonishing decline from 67% at his second inauguration to a mere 24% as he left office after being told by GOP leaders that he would not survive a Senate trial. Republicans were crushed in the ensuing midterm elections, and Nixon’s successor, Gerald Ford, lost to an inexperienced Southern Democrat named Jimmy Carter two years later.
It is not, however, clear how much of this is attributable to impeachment itself. Nixon had the advantage of running for reelection amid 5% economic growth, which doubtless juiced his job approval numbers. But the economy began turning against Nixon at the beginning of his second term. The easy money policies pursued by the Federal Reserve began to catch up with the country, as the monthly inflation rate brushed up against 1% in March 1973 and almost hit 2% in August. Overall inflation for the year was almost 9%, and the federal funds rate hit 11%. In 1974, inflation surged to over 12%.
At the same time, in October of 1973, OPEC instituted an embargo of oil to the United States, causing widespread gas shortages. The country entered a serious recession, which lasted until 1975 – the longest recession the country had experienced since the Great Depression.
To be clear, my position is not that impeachment was irrelevant, as Nixon’s job approval declined before the recession took hold (though it’s also true that the president’s job approval fell below 50% before the Watergate committee began televised proceedings in May of 1973). The point is simply that we don’t know whether the president’s job approval would have collapsed, and remained depressed, without rising inflation, a serious recession, and a menu of other scandals that cropped up (such as the bombing of Cambodia).
On the other hand, Republicans would like to point to the impeachment of Bill Clinton as an example that benefited the incumbent party. After all, in the midst of the impeachment imbroglio Democrats picked up House seats, and Bill Clinton’s job approval soared.
This too misses the overall economic context. In a time of prosperity and rapid economic growth, Clinton’s job approval would be expected to soar, and Democrats would be expected to perform well. In fact, in early 1998 I suggested that Democrats might become the first presidential party to pick up seats in a midterm since 1934. This was before the discovery of Monica Lewinsky’s famous blue dress; my focus was on economic performance, a number of poorly situated Republican retirements, and the GOP’s overextension into Democratic territory as a result of the 1994 red wave.
Even the strength of Clinton’s job approval was not entirely unexpected. Research by Brian Newman, first presented in a graduate seminar I took with him and later published in Political Research Quarterly, suggests that Clinton did suffer in the polls. His job approval dipped when impeachment news dominated -- but the economy buoyed it in the interim. We also should not forget that Al Gore lost his bid for Clinton’s third term, in part because he felt compelled to distance himself from Clinton’s scandals. A successful presidential trip to Arkansas (then a swing state) was probably all Gore needed to secure the presidency.
There are two final observations worth considering here. First, none of these impeachment proceedings were initiated while the Senate was controlled by the president’s party. Second, the fact that Democrats have been discussing impeachment since before Trump was inaugurated will possibly affect the way independents and Republicans evaluate the charges.
The polls are difficult to interpret. Polling since the impeachment inquiry was announced has shown an uptick in support for impeachment, in some cases dramatically so. This is not entirely surprising. A hefty amount of political science literature, most notably by John Zaller, has demonstrated the way in which public opinion is formed by elite cues. Most people are disconnected from politics to varying degrees, and are content to follow the lead of like-minded party and opinion leaders (this is, of course, oversimplifying greatly). The theory has its problems, but it certainly explains some shifts in recent years.
It isn’t surprising, then, to see weak Democratic support for impeachment when the party leadership opposed it, followed by strengthening support as the party leadership decided to support it. A recent Quinnipiac poll showed voters split over whether to impeach and remove Trump, a significant shift from findings from just a week ago. But almost all of the movement was among Democrats, who went from supporting impeachment by a 50-point margin to supporting it almost unanimously. There was some movement among independents, but a large number of independents are really “closet partisans.”
Additionally, there may be a polling effect where supporters of the president become less likely to respond to the polls; there was likely some of this in the 2016 election in the wake of the “Access Hollywood” tape and in some of the polling of the debates.
In other words, Democrats have picked up some proverbial “low-hanging fruit.” Whether they are able to expand the support for impeachment, and to what degree, remains to be seen. If the public remains split, it is unclear that Democrats would benefit from impeachment.
The distribution of impeachment votes will affect the way impeachment is evaluated. We remember Clinton’s acquittal by the Senate as a foregone conclusion, but there was some concern about it at the time. One of the concerns was that if any Democratic senator supported his removal, others might follow suit. The key vote was Sen. Robert Byrd, who later claimed that announcing his vote to acquit felt like sandpaper in his throat. In the end, all Democrats in the Senate (and most Democrats in the House) voted against removal, reinforcing perceptions that impeachment was largely a partisan endeavor.
Would a trial in 2020 follow a similar approach? Again, no one knows for sure. If the evidence becomes clear enough, one could see Republicans like Mitt Romney, Lisa Murkowski or Susan Collins break ranks with their party. Such a vote, however, would put these senators’ political future at stake.
More interesting, I think, is what the Democrats do. There are currently around 30 House Democrats representing districts that lean Republican, although some of these districts (including suburban districts in Texas) are less Republican now than the presidential topline might suggest. Only one Republican (John Katko) represents a Democratic-leaning district. There are also some Senate Democrats, like Joe Manchin of West Virginia, Doug Jones of Alabama, and Jon Tester of Montana, who will likely face enormous pressure from voters back home to oppose removing the president.
We could end up with a situation similar to what we encountered in 1999, where the vote is largely party line, with a few significant defections from the opposition party (five Republican senators joined 45 Democrats in voting against the obstruction of justice charge, while 10 Republican senators voted against the perjury charge). Of course, we could end up with a dam breaking, with a lopsided vote in either direction, depending how events unfurl. This, in turn, will affect how the impeachment is viewed by the public, and its effect on 2020.
Keep an eye on the president’s job approval. A theme running through the observations above is that impeachment is, at its core, a political proceeding, rather than a legal one. To this end, we should keep a close watch on the president’s job approval. If it were to collapse, as Nixon’s did, Republicans in otherwise-safe districts may begin to feel that the upsides to voting to convict outweigh the potential primary threat. Because of this, the most significant impeachment survey question from the past week is Quinnipiac’s job approval measurement, which was toward the high end of that pollster’s measurements; the recent Monmouth poll shows no change in the president’s job approval.
It is still early in this process. Above all, it is important to remember that it is still early in the process. Although time seemingly passes quickly in the Trump presidency, it has only been a little more than a week since the public learned of the whistleblower complaint. Whether the whistleblower comes forward to testify, whether others can attest to the allegations in that complaint, and whether other shoes drop (it seems inevitable that they will) are all going to impact how this plays out.
This is a fascinating, and in many ways terrifying, political moment. Events are moving quickly, and likely will continue to do so until Election Day. Analysts on both sides would do well to wait a week to two before even attempting to divine how these events are likely to play out.