Democrats debating whether to impeach Donald Trump may be misreading the evidence from the last time the House tried to remove a president.
It’s become conventional wisdom—not only among Democrats but also among many political analysts—that House Republicans paid a severe electoral price for moving against Bill Clinton in 1998, at a time when polls showed most of the public opposed that action.
But that straightforward conclusion oversimplifies impeachment’s effects, according to my analysis of the election results and interviews with key strategists who were working in national politics at the time. While Republicans did lose House seats in both 1998 and 2000, Democrats did not gain enough to capture control of the chamber either time. And in 2000, lingering unease about Clinton’s behavior provided a crucial backdrop for George W. Bush’s winning presidential campaign—particularly his defining promise “to restore honor and dignity” to the Oval Office.
Matthew Dowd, a senior strategist for Bush’s 2000 campaign, told me that Democrats today “are learning the wrong lessons” from Clinton’s impeachment by neglecting to consider how it shaped both election cycles, especially the presidential race. In January 2001, almost exactly two years after House Republicans defied public opinion to impeach Clinton, the GOP controlled the White House, the House, and initially the Senate. (Within months, the Republican Jim Jeffords of Vermont would switch parties, shifting control to the Democrats.) “Having gone through all that,” Dowd said, “I think the Democrats are way too skittish on impeachment.”
Tad Devine, a senior strategist for Al Gore, the Democratic nominee in 2000, concurs. Bush’s ability to tap the public’s dismay over Clinton’s personal life “more than anything else got in our way in terms of winning the election,” he told me. Even if the Senate doesn’t convict Trump, Devine believes, impeachment in the House could offer Democrats a similar chance to highlight the aspects of Trump’s volatile behavior that most alienate swing voters.
“If impeachment is done properly, then the Democratic nominee will be talking about it [next year] and not be running away from it in the general election,” he said. “I don’t look at it as something that is going to derail a Democratic nominee. Just like we saw in 2000, an impeachment inquiry could very badly damage somebody who is associated with it.”
The notion that impeaching Clinton hurt the Republican Party isn’t entirely a myth. The House Republican majority voted to formally begin an impeachment inquiry in October 1998, just weeks before the midterm elections. GOP leaders confidently predicted that public revulsion with Clinton would lead to big Republican gains. “The Republicans were all full of themselves going into the election,” says then–Democratic Representative Martin Frost of Texas, who chaired the Democratic Congressional Campaign Committee that year. “They expected to pick up 20 or 30 seats.” Instead, in November, Democrats gained five—the first time a president’s party had won House seats in the sixth year of his tenure since Andrew Jackson in 1834.
But it’s easy to overstate the magnitude of the GOP’s backslide in 1998. In the Senate, Democrats gained no seats that year, leaving the Republican majority intact. Nor did the five-seat House loss cost the GOP its majority in that chamber. Republicans still won more of the total national popular vote in House races than Democrats. Swing voters didn’t stampede away from the GOP; in exit polls, Republicans still narrowly beat Democrats among independent voters. And while impeachment provoked big turnout from African Americans, Clinton’s most passionate supporters, overall, turnout that year was very low.
The midterm election was widely seen as a red light from the public on impeachment. But Republicans barreled ahead and voted in mid-December to remove Clinton anyway. On the day they did so, there was about as much public support for impeaching Clinton as there is today for impeaching Trump. A Gallup poll at the time showed that 35 percent of the public overall backed impeachment, including 40 percent of independents. In a CNN poll this week, 41 percent of the public supported impeachment, including 35 percent of independents. Overall, Clinton’s public support in Gallup polling was much stronger at the time (63 percent job-approval rating) than Trump’s is now (40 percent).
After the Senate refused to remove him in early 1999, Clinton’s job approval remained about a buoyant 60 percent through the rest of his presidency. But that strong number coexisted with substantial public disapproval of the personal behavior that the Starr investigation and the impeachment inquiry had highlighted.
Those personal doubts about the outgoing president cast a huge shadow over the election to succeed him—a dynamic usually omitted in the equation portraying Clinton’s impeachment solely as a self-inflicted wound for Republicans. Those doubts led the Gore campaign to conclude they could neither campaign with Clinton nor deploy him extensively as a surrogate.
“It wasn’t just impeachment, it was the feeling that Clinton wasn’t telling the truth to people,” Devine said. “Gore was hurt by that association. Everybody who said the economy was so good, you should just run on Clinton’s record—they weren’t sitting in focus groups in swing states, listening to these swing voters who were concerned there would be a continuation of that [behavior].”
Both Devine and Dowd noted that Bush’s constant pledge to restore presidential honor and dignity skillfully tapped that unease without fully embracing the divisiveness of impeachment itself. “It became a very valuable tool even though Bush didn’t go around saying ‘impeachment, impeachment,’” Devine said. “He took the bad stuff from impeachment and put it front and center [in the campaign]. What could Gore say? ‘I’ll restore honor too’?”
The exit polls in 2000 showed how much that refrain helped Bush with voters conflicted about Clinton. Gore carried 85 percent of the voters who both approved of Clinton’s job performance and expressed favorable views of him personally. But Bush carried one-third of the voters who liked Clinton’s performance but disliked him personally. That’s a much higher-than-usual level of defection from the president’s party among voters who approve of his performance, and in 2000, those voters represented about one-fifth of the entire electorate.
Bush reaped another benefit from the impeachment, Dowd believes: high turnout among Republicans frustrated that Clinton remained in office. “The lack of success on impeachment created a greater hunger, a greater motivation among the base to exact punishment in some way,” said Dowd, now the chief political analyst for ABC News. “And Al Gore became part of that process.”
Despite widespread satisfaction with the economy, all of these factors helped Bush finish just behind Gore in the popular vote and win the Electoral College (after the disputed recount in Florida). Though Republicans lost another two House seats in 2000, they again maintained control of the chamber’s majority—and again narrowly won the nationwide House popular vote. Only in the Senate did the GOP slip, losing five seats that initially created a 50–50 tie until Jeffords’s switch gave Democrats a majority.
Both of the impeachment-related dynamics that benefited Bush in 2000 could help Democrats in 2020, Dowd believes. An acquittal in the Republican Senate after the Democratic House votes to impeach Trump would likely spur turnout next year among Democrats determined “to finish the job,” he maintained, just as it did among Republicans 20 years ago. And impeachment hearings could focus attention on personal behavior by the president that dismays swing voters satisfied with the economy—just as the Clinton impeachment effort did. “It will remind people of all the things related to him that they don’t like,” Dowd said.
However impeachment affects the 2020 presidential landscape, House Democrats also must worry about its influence on their majority. If any backlash develops, the most exposed would be the 31 House Democrats now holding seats that Trump carried in 2016. Here again the 1998 experience may offer a relevant precedent. It’s true, as I wrote at the time, that four of the five Republican House incumbents who lost that fall represented districts that voted for Clinton in 1996. But they still constituted only a tiny fraction of the 91 House Republicans who held Clinton districts, in an era when split-ticket voting was much more common.
Devine believes that Democrats in Trump districts could defend a decision to impeach if the party can conduct the inquiry in a disciplined manner, downplaying partisanship and stressing how Trump threatened the rule of law by trying to obstruct the Russia investigation. “Those people were elected, in large measure, because voters in those districts wanted a return to normalcy, order, a government of laws,” he said. “If you can fit it into that frame, I think voters will give you a lot of room on that.”
Frost, the former Democratic representative, also thinks the party may have more latitude than it believes to shape public opinion on impeachment. “There is clearly a downside for Trump if … we can ever get the information out on television and establish the obstruction-of-justice acts,” he says. “What’s the point of the Constitution if a guy can obstruct justice and not play a political price for it? I think, from my party’s standpoint, we can’t walk away from this at this point.”
The political world has changed significantly since 1998. Key among those changes is the consolidation of a conservative-media infrastructure that dominates communication to the GOP rank and file. That dynamic means Trump is even less likely than Clinton to suffer major erosion of support from his base, and thus also from his party’s representatives in Congress. And Trump has repeatedly demonstrated, with the help of the conservative-media ecosystem, that he can energize his supporters by portraying attacks on him as efforts from disdainful “elites” to suppress their influence. That could allow him to wave impeachment as a bloody shirt to spur turnout from his base in 2020. Even swing voters uneasy about Trump might also recoil from the sheer level of political conflict that impeachment would inevitably ignite in today’s combustible media environment.
All of that suggests it’s not a guaranteed political winner for House Democrats to impeach Trump when there’s virtually no chance the Senate will vote to remove him. But the full ledger on Clinton’s impeachment invalidates the common assumption that impeachment without removal is a guaranteed political loser. Considering both the 1998 and 2000 elections, there’s considerable evidence that the struggle actually helped the GOP; at worst, its political impact was equivocal. Which means that, on impeachment, House Democrats may have more leeway than they believe to do what they think is legally and morally right.