The House may well vote to impeach President Trump. What remains uncertain is whether the Senate will remove him. Senate Republicans hold 53 seats, and removal requires two-thirds of senators voting, or 67 votes to convict if all 100 senators vote.
If all 45 Democrats and the two independents vote to convict, 20 Republicans would have to switch sides to vote against the president. Is that possible?
The Senate has conducted only two presidential impeachment trials in its history, giving us not much to go on. But impeachment attempts around the world can teach us a lot about what’s likely to happen here.
Parties ‘fire’ their prime ministers at low cost. That’s not true with presidents.
In parliamentary systems, the legislative majority party selects the government leader. Those parties can “fire” their own prime minister if he or she grows unpopular. That only requires a majority vote of the prime minister’s own members of parliament. The party stays in power and can pick a new prime minister.
Such intraparty disputes pushed out U.K. prime ministers Margaret Thatcher in 1990 and Theresa May earlier this year. About 30 percent of all prime ministers worldwide leave office this way; the rest either leave when their party loses an election or when the opposition party holds — and wins — a no-confidence vote, sometimes with the support of a minority of the prime minister’s own party.
That’s not true for political parties in presidential systems, where the legislature and the president are elected separately. Parties can’t easily remove an unpopular inhabitant of the executive branch. Granted, many presidents have been threatened with impeachment, and several removed. Of 223 presidents elected in 53 democracies between 1946 and 2007, 31 faced formal impeachment proceedings; 11 of those presidents were removed.
But the president was almost always kicked out by the opposition, not his or her own party. Of those 223 elected presidents, only one was removed with the support of a majority of the president’s own party. That came in 1999, when Paraguay’s governing Colorado Party supported efforts to remove President Raúl Cubas Grau for abuse of power and obstruction of justice. In 2016, that happened again — for perhaps only the second time in world history — when more than half of the legislators in South Korean President Park Geun-hye’s own party voted to impeach and later to remove her for corruption.
Why a president’s party isn’t likely to convict
Unlike prime ministers, who serve at their party’s behest, presidents have their own source of legitimacy: popular election. That makes impeachments extraordinary constitutional procedures. What’s more, if a party supports impeaching its president, it is admitting that the party and its voters supported not just an incompetent who should be replaced but someone whose wrongdoing was so serious that it verges on criminal. Doing so hurts the party — both in its opponents’ eyes and, more important, among its base.
Sometimes, politicians believe the president’s wrongdoing to be so egregious that they must vote according to their consciences. But doing so typically splits their party. This undermines the party’s electoral message and ability to mobilize supporters, which harms its electoral prospects. In Paraguay, the president’s party lost, and has never regained, its majority in both of Paraguay’s legislative chambers, which it had previously held for decades. Similar results followed after elements of a president’s party supported impeachment efforts (which did not always pass) in Sri Lanka in 1991; Venezuela in 1993; South Korea in 2004; and Nicaragua in 2006.
This is why a president’s party so infrequently supports impeachment: Doing so tends to hurt the party more than it helps. Only in dire circumstances are a president’s co-partisans likely to jump ship, hoping that blaming the president for the country’s ills will spare them from voters’ wrath. In Venezuela in 1992, the country’s economy was nearing collapse before the president’s party abandoned him. In South Korea in 2016, millions had been protesting for months, and presidential approval had plummeted to 5 percent, before members of the president’s party voted to impeach.
Abandoning Trump would probably not help Republican elected officials stay in office — unless Republican voters’ support for the president evaporates. That hasn’t happened. Roughly 40 percent of voters — and 90 percent of Republican voters — have said they approve of Trump’s performance quite steadily throughout his term.
So what brings down approval ratings?
President Richard M. Nixon won a landslide reelection in 1972. His approval rating hit 67 percent at his second inauguration in January 1973; by October it had declined to just 27 percent, where it remained until his exit. Unlike Trump’s, Nixon’s popularity gradually declined even among Republicans — from almost 90 percent to about 50 percent over the same period. That may be why several Republicans voted to move forward with a Judiciary Committee recommendation to impeach Nixon and did not subsequently suffer at the polls.
Why such a difference between Nixon’s crimes and the allegations against Trump? One key reason is that partisanship is stronger today, with voters less likely to shift opinions about “their” party. That’s reinforced by partisan news feeds, with Republicans and Democrats consuming different sources of information. But there’s another factor. In 1973-1974, the U.S. was in a recession. Nixon’s approval rating went down as worries about the economy went up. Republican voters might have continued to back Nixon had the economy been strong.
Would a recession hurt Trump’s popularity among Republicans? Given today’s partisanship, that’s hard to say. But comparative evidence suggests that unless Trump’s popularity among Republican voters drops, turning on Trump would probably hurt Republican politicians, both individually and as a party, with those voters. Unless that shifts, Republican leaders are likely to stick with Trump, no matter the evidence against him.
David Samuels is Distinguished McKnight Professor of Political Science at the University of Minnesota.