Skip to main content
Menu
What Makes A GOP Leader Resist Trump?

Republican voters have rallied behind Donald Trump. House Speaker Paul Ryan has endorsed him after a few weeks of dithering, and Sen. Marco Rubio, who once called Trump a “con man,” has offered to speak on his behalf at the Republican National Convention. In all, GOP officials have saluted the Trump candidacy at an amazing clip given Trump’s controversial, often racist rhetoric.

But Illinois Sen. Mark Kirk’s reversal in the wake of Trump’s remarks against Judge Gonzalo Curiel may signal that some of the support Trump has received is still fragile. South Carolina Sen. Lindsey Graham is calling on other Republicans to unendorse Trump. And dozens of other prominent elected Republicans are holding out on their party’s presumptive nominee.

Sixty, to be exact. By my count, that is the number of Republican governors, U.S. senators and U.S. representatives whose public position is that they outright oppose Trump or that they are not yet ready to commit one way or the other. Thirteen are in the #NeverTrump club (four senators, one governor and eight House members), and 47 are in the “maybe later/maybe never” camp, spearheaded by Sen. Ted Cruz and, until recently, Ryan.

So who are these 60 Republicans, and what distinguishes them from their more Trump-friendly counterparts?

Women, minorities, Republicans who represent states or districts with large Latino populations, centrists, trade-friendly Republicans and immigration-friendly Republicans have been more skeptical of Trump, according to my analysis. A larger share of men, non-Hispanic whites, Republicans whose states or districts are redder than average, and Southerners have embraced Trump.

Many of these variables are closely correlated with each other (Southerners are more conservative and represent redder areas, for instance), so it’s difficult to pinpoint which among them best explain the variations in these Republicans’ support for Trump. But taken together, they paint a clear picture of a group of 60 holdouts that is less uniformly white and male, less conservative, more open to immigration and trade, and somewhat more vulnerable to Democratic-leaning voters. An alternative Republican Party, if you will.

Two methodological notes before we delve into the details: (1) I am ignoring differences in the level of enthusiasm within the Trump-supporting group,1 and (2) this analysis covers 319 of the GOP’s 331 governors, senators and representatives. Twelve Republican House members have made no public remark on Trump in recent months, and I do not include them among the holdouts.2

Women and minorities: Among the Republican governors, senators and representatives with some public position on Trump, 83 percent of non-Hispanic white men support him. The same can be said of only 68 percent of women, 60 percent of minorities and 57 percent of Hispanics.

POSITION ON TRUMP
DEMOGRAPHIC OPPOSED UNCOMMITTED SUPPORTING
Men 4% 13% 83%
Non-Hispanic whites 4 14 83
All 4 15 81
Women 3 29 68
All minorities 10 30 60
Hispanics 14 29 57
Demographics of GOP governors, U.S. senators and U.S. representatives by Trump support

As a result, the group of 60 holdouts is more diverse than the group of Trump supporters: Women and minorities make up 23 percent of holdouts and 12 percent of supporters.

Trump’s victory in the primary has raised questions about whether the GOP will take a hit this fall among women and minorities. In that context, the disparities within this most elite group of Republican politicians are striking.

And some Republicans are explaining their reluctance to support Trump by directly pointing to their own ethnicity or to Trump’s language toward minorities and women. As The Washington Post reported in April, New Mexico Gov. Susana Martinez “told the crowd of about 60 wealthy GOP backers that, as a Latina, she was offended by Trump’s language about immigrants.” And Trump has chosen to attack Martinez, putting him in open conflict with the country’s first and only Latina governor.

Hispanic residents: The 60 holdout Republicans represent areas with larger Hispanic populations. The average share of Hispanic residents in the areas they represent is 15.9 percent, versus 10.5 percent for those supporting Trump. Among just the 13 Republicans who outright oppose Trump, the number jumps to 20.2 percent.3

Of the five Republicans who represent areas where more than 60 percent of residents are Hispanic — Reps. Mario Diaz-Balart, Carlos Curbelo, Ileana Ros-Lehtinen (FL-25, FL-26, FL-27), Rep. Will Hurd (TX-23) and David Valadao (CA-21) — all are holding out on Trump except Diaz-Balart, who said in May that his “intention is to vote for the Republican nominee.”

nichanian-nevertrump-1

Diaz-Balart’s congressional district, which is 71 percent Latino and located in southern Florida, is therefore the area in the country with the highest share of Hispanic residents that is held by a Republican who plans to vote for Trump. Next up: New Mexico’s 2nd District (Rep. Steve Pearce) and Texas’s 27th District (Rep. Blake Farenthold). Mitt Romney carried all three districts in 2012. If we restrict ourselves to the most evidently vulnerable Republicans — those who represent areas carried by President Obama in 2012 — the Trump supporter whose constituency has the highest share of Latino residents is Rep. Jeff Denham in California’s 10th District.

Partisan leanings: The 60 holdouts represent areas that tend to vote more Democratic than those represented by Trump supporters. Romney received 53 percent on average in holdout territories and 58.3 percent on average in Trumpland. And in the areas represented by the 13 #NeverTrump Republicans, he received just 48.6 percent on average.4

nichanian-nevertrump-2

The discrepancy is clear when we focus on those Republicans who represent states and districts Obama carried in the 2012 presidential election. Only 58 percent of them are supporting Donald Trump, versus 85 percent of those who represent states or districts carried by Romney.

One of the most Republican regions of the country — the Deep South5 — is the most united around Trump. Of its 44 GOP governors, senators and representatives, the only two holdouts are Graham, who clashed with Trump during the presidential primaries and compared him to Joe McCarthy after Trump’s remarks against Curiel, and former South Carolina Gov. Mark Sanford, who now serves in the House.

The two Republicans with the most-Democratic constituencies — Maryland Gov. Larry Hogan and Massachusetts Gov. Charlie Baker — are not supporting Trump. But three of the next four — Illinois Gov. Bruce Rauner, New Jersey Gov. Chris Christie, Maine Gov. Paul LePage (this included Kirk before he unendorsed) — support him despite representing states in which Romney received less than 41 percent of the vote.

Immediate electoral considerations: Only some of the Republicans running statewide in Democratic-leaning areas are holding out on Trump. Of the six Republicans running for U.S. Senate seats in states where Obama won a larger share of votes than he got nationally in 2012, two are not supporting Trump: Kirk and Pennsylvania Sen. Pat Toomey, who has not ruled out backing Trump but says he is “not there yet.” But New Hampshire Sen. Kelly Ayotte, Iowa Sen. Chuck Grassley, Wisconsin Sen. Ron Johnson and Rep. Joe Heck, who is running for Nevada’s open Senate seat, are supporting him. Other Republican senators who may be vulnerable this fall, such as North Carolina Sen. Richard Burr and Arizona Sen. John McCain, are also supporting the presumptive Republican nominee.

The picture on the House side is closer to what is expected: Barely half of the representatives running for re-election in bluer-than-average districts support Trump.

Eight representatives are supporting Trump while preparing to face voters in districts where Obama got a higher share of the vote than he did nationally in 2012. One is Rep. Rod Blum, who represents Iowa’s 1st Congressional District, the most Democratic district won by a Republican in the 2014 midterms (Romney received just 42.5 percent of the vote there). The others are Reps. Cresent Hardy (NV-04), Frank LoBiondo (NJ-02), Elise Stefanik (NY-21), David Young (IA-03), Tom MacArthur (NJ-03), Pete King (NY-02) and Dan Donovan (NY-11).

This analysis provides some support for the hypothesis that Republicans who are not currently in office or are not looking to get elected have an easier time resisting the pressure to unite around their party’s presumptive nominee. Republicans who are facing voters this year are more likely to be supporting Trump. Within that group, 83 percent have fallen in line behind their party’s presumptive nominee, compared with 76 percent of those who are retiring or who are not up for re-election until 2018 or 2020. In fact, three of the 20 House Republicans who are retiring from Congress are outright opposing Trump (Reps. Reid Ribble, Scott Rigell and Richard Hanna), compared with five among all other 226 House Republicans. And after all, the highest-profile Republicans to oppose Trump for now — George W. Bush, Jeb Bush, Romney — get to watch from the sidelines.

Ideology: Many of the most centrist Republicans are not supporting Trump. So it is no surprise that GOP congresspeople holding out on him are as a group less conservative than those who support him, as measured by an average of their DW-Nominate scores.6

And nearly half of the representatives who are holding out on Trump belong to the Republican Main Street Partnership, a group that houses Congress’s more moderate Republicans.

But if you put aside the most moderate tier of Republicans, the propensity to resist Trump is relatively constant across the rest of the party’s ideological spectrum. Support for Trump does dip in the most conservative tier — those with the highest DW-Nominate scores — but only slightly.

nichanian-nevertrump-3

In fact, one striking finding is that Trump has won over nearly all members of the House Freedom Caucus. Despite months of questions as to how these conservative Cruz allies would react to him, all but six are already supporting him. Some of the remaining holdouts, such as Rep. Tim Huelskamp (KS-01), explain their reluctance to endorse him by citing what they see as Trump’s insufficient conservatism.

That speaks to the difficulty of capturing what motivates opposition to Trump through a conventional centrism/conservatism binary. Among openly anti-Trump Republicans are some with the lowest DW-Nominate scores, such as Rep. Bob Dold of Illinois, and some with the highest, such as Huelskamp and Rep. Justin Amash of Michigan.

Two issues on which Trump has catalyzed GOP divides that do not fit neatly into these classifications are immigration and trade — so let’s turn to those.

Immigration: As a proxy for attitude toward immigration, I used the Senate’s 2013 vote on comprehensive immigration reform and the House’s 2015 vote to end the Deferred Action for Childhood Arrivals program. Of the senators who favored immigration reform and House members who favored deferring deportation, only 49 percent support Trump, compared with 88 percent of those who opposed immigration reform or opposed deportation deferrals.

So by this measure, a majority of immigration-friendly Republicans are holding out on Trump.

This isn’t a simple story about their constituencies’ partisanship: 57 percent of immigration-friendly Republicans in states/districts won by Romney are supporting Trump, while 80 percent of immigration-hostile Republicans in states/districts won by Obama are.

You may have expected Republicans who are open to immigration to be more skeptical of Trump. But the magnitude of the difference confirms that immigration has become one of the party’s defining fault lines, as recent analyses of Republican politics have clearly shown.

POSITION ON TRUMP
OPPOSED UNCOMMITTED SUPPORTING
Anti-immigration reform 2% 11% 88%
Anti-trade 3 9 88
Pro-trade 8 17 75
Pro-immigration reform 18 33 49
Positions of GOP senators and representatives by Trump support

Trade: Trade is the other big issue that gets cited to explain Trump’s anti-establishment appeal. To measure Republican congresspeople’s attitude toward trade, I looked at two issues: how they voted on giving Obama fast-track authority over the Trans-Pacific Partnership negotiations and how they voted on reauthorizing the Export-Import Bank. I am defining “pro-trade” Republicans as those who supported both fast-track authority and the Ex-Im Bank.

Pro-traders are a clear minority among Republican congresspeople supporting Trump (38 percent). But they are a clear majority among all those who are holding out (61 percent). In fact, 10 of the 12 #NeverTrump congresspeople are pro-traders. (Amash and Sen. Ben Sasse of Nebraska are the only “Never Trump” congresspeople who didn’t support both measures, though Sasse voted for fast-track authority.)

A favorable outlook toward trade may well be the most interesting characteristic shared by most Trump holdouts but not most Trump supporters.

Footnotes

  1. Many Republicans insist on such distinctions. “To me, support versus endorse are two totally different things,” Wisconsin Sen. Ron Johnson said recently, a position echoed by Sens. Kelly Ayotte and John Hoeven of New Hampshire and North Dakota, respectively. Many others avoid Trump’s name, declaring support for only “the nominee,” or combine their statement of support with highly critical remarks on Trump’s policies and demeanor. New Jersey Rep. Frank LoBiondo’s statement is a typical example: “I have always supported the Republican nominee and will support the Republican nominee this year. I am, however, concerned with the negative tone and personal attacks that have come to define this presidential race on both sides.” For the purpose of this analysis, a minimal expression of support is sufficient.
  2. This is because for some representatives it is unclear how much their silence has to do with their stance toward Trump and how much it has to do with their flying under the radar and receiving less media coverage than Republicans in different media markets. It’s not clear how many would at least declare tepid support for “the nominee” if they were pressed more directly than they have been.
  3. Source: U.S. census (2010-14 American Community Survey, five-year estimates).
  4. For presidential election results by congressional districts, I am using Daily Kos Elections’ calculations.
  5. Louisiana, Mississippi, Alabama, Georgia and South Carolina.
  6. DW-Nominate’s first dimension, which I am using here, uses roll call votes to rate members of Congress on a scale of -1 to 1, where a higher score indicates a more conservative ideology.

Daniel Nichanian is a postdoctoral fellow and lecturer in political science at the University of Chicago

Comments