Impeaching the President: Mapping the Political Landscape in the House of Representatives

Abstract

Our main objective is to explain why and under what conditions legislators (co)sponsor impeachment resolutions against the president of the United States. We examine the impact of political, institutional, and economic variables, suggested by presidency and impeachment scholarship, on the behavior of 2,044 legislators who were members of the House of Representatives between 1973 and 2019. To test these hypotheses, we use two-level logistic regressions and an original dataset of all presidential impeachment resolutions filed in the House during that period. Our primary findings are that impeachment resolutions against the president are driven by individual-level variables such as representatives’ ideological extremity and partisanship. Our analysis also shows that the occurrence of major presidential scandals incentivizes representatives to target the chief executive. Surprisingly, we find no statistical evidence for the effects of divided government, presidential approval, and macro-economic factors. Our results suggest that representatives use impeachment resolutions both for position taking and for rendering the president accountable.

Notes

1 There is no requirement about a minimum percentage of legislators that need to formally back an impeachment resolution.

2 Likewise, in his review of the period 1877–2006, Clark (2009, 974) concluded that “the sheer number of Court-curbing proposals that are introduced in Congress with great regularity, but never earn so much as a committee hearing, suggests that Court curbing is driven at least in part by interested contingents or groups from a member’s constituency.”

3 Grimmer (2013, 625) also finds that “legislators from districts composed of copartisans … will emphasize their more extreme positions to constituents” and are more likely to make them public. A similar explanation was set forth by Caldeira and Zorn (2004).

4 Lewis et al.’s (2019) DW-NOMINATE first dimension scores vary from strong liberal (−1) to strong conservative (+1).

5 Rep. Green sponsored and co-sponsored three and four impeachment requests, respectively, against Trump between 2017 and 2019. Green’s ideology score is −0.438.

6 We looked for the terms “impeach,” “impeachment,” “impeaching,” or “impeached,” filtering them by “Bill Type” (resolution), “Status of Legislation” (introduced), and “Chamber of Origin” (House).

7 Here are examples to illustrate each type of House Resolution (HR): (a) impeachment request: “Impeaching William Jefferson Clinton, President of the United States, for high crimes and misdemeanors,” HR 611 (Nov 16, 1998); (b) impeachment investigation: “Directing the Committee on the Judiciary to undertake an inquiry into whether grounds exist to impeach William Jefferson Clinton, the President of the United States,” HR 304 (Nov 5, 1997).

8 Originally, impeachment actions aimed at POTUS were (co)sponsored by 328 representatives. However, three co-sponsors were dropped from our sample due to non-existent data on their ideology scores.

9 When it comes to chief sponsors of impeachment resolutions against POTUS, only four Republicans have taken on that role between 1973 and 2019: Paul Norton “Pete” McCloskey, Jr. (CA-17) in 1973 against Nixon, Henry J. Hyde (IL-6) twice in 1998 against Clinton, Bob Barr (GA-7) also against Clinton, and Tom Cole (OK-4) in 2019 against Trump. Only one Democrat joined any of those GOP impeachment resolutions (Congressman Gary Eugene “Gene” Taylor of Mississippi who co-sponsored Barr’s resolution against Clinton), and only Barr garnered numerous Republicans as co-sponsors (31) on his anti-Clinton resolution. By contrast, during the same period, 59 impeachment resolutions against POTUS had Democrats as chief sponsors (42 of them targeted Nixon).

10 We address this issue at the end of the analysis and findings section. The supplementary statistical results for presidential impeachment resolutions lend support to our approach (see Table C1).

11 We tried to disaggregate the data between sponsors vs. co-sponsors, but a lack of cases—especially of resolutions without sponsors (n = 39)—prevented us from generating feasible regression models. Still, it is worth considering that members of Congress tend to follow patterns of “mutual co-sponsorship” in their networks, which means that legislators tend to reciprocate when they receive support for their bills (Fowler 2006, 484). We do believe that this is a research avenue worth pursuing via more qualitative analysis.

12 −1 equals “most liberal” and +1 equals “most conservative.”

13 As the literature on presidential instability in Latin America suggests, only remarkably notorious scandals in which the president (or close collaborators) are implicated have the potential to produce legislative backlash or public outrage (Martínez 2020; Pérez-Liñán 2007). Thus, we only considered the most prominent political scandals in recent U.S. history. These are the presidents and years linked to scandals identified by Barberio (2020): Richard Nixon’s Watergate scandal (1972–1974), the Iran-Contra affair during the second Reagan administration (1986–1987), the Clinton-Lewinsky scandal (1998), and the Trump-Ukraine scandal (2019).

14 After all, the Constitution does not specifically describe what constitutes “high crimes and misdemeanors,” and failing to advance an impeachment resolution (and reach a positive vote in the House) hardly represents a significant political defeat for its (co)sponsors.

15 The Democratic Party’s ideology mean was −0.323, and the standard deviation was 0.172. On the other hand, only two moderate conservatives signed resolutions against Nixon: Republican Paul Norton “Pete” McCloskey, Jr. (0.028 ideology score) and Democrat Dale Milford (0.012 ideology score).

16 The Republican Party’s ideology mean and standard deviation are 0.402 and 0.145, respectively. In December 1998, Republican Henry J. Hyde sponsored impeachment request HR 611 (no-cosponsors), which eventually reached a vote in the House. Interestingly, Hyde’s ideology score (0.312) was more moderate than his party’s mean (0.402).

17 The Democratic Party’s means for 2017–2018 and 2019 were −0.391 and −0.370, respectively. The Democratic Party’s standard deviations for 2017–2018 and 2019 were 0.114 and 0.122, respectively.

18 Democrats in the House disagreed on whether to further consider Green’s impeachment article or dismiss it, which led to a split vote of 95 in favor vs. 132 rejecting it (Hirschfeld Davis and Fandos 2019).

19 Brindisi declared that “it’s very frustrating for me—someone coming from a district that was one of the districts that helped get us into the majority—having so much focus on things like impeachment or other issues that are divisive” (Ferris 2019).

20 Democrat Anthony Brindisi lost his reelection bid to Republican Claudia Tenney, the former incumbent, in the 2020 New York’s 22nd congressional district election. By contrast, Al Green was reelected in 2020 for the Texas’s 9th congressional district with 75.5% of the vote.