Republicans lost control of the House in 2018 and now could lose the Senate this year. Their fortunes seem tied to Trump and his agenda, but new research suggests they would be better off trying to distinguish themselves from him and his policies. Sarah Treul finds that votes to repeal Obamacare cost Republicans seats in Congress in 2018. They did not listen to their constituents at Town Hall meetings and the repeal effort resulted in lower vote share. Andrew Ballard finds that Trump endorsed a lot of Members of Congress in 2018 but it actually hurt those endorsees, stimulating the opposition more than the supporters.

GuestsSarah Treul, University of North Carolina; Andrew Ballard, American University 

Studies: “The Electoral Consequences of Roll Call Voting: Health Care and the 2018 Election” and “Be Careful What You Wish For: The Impacts of President Trump’s Midterm Endorsements.”


Matt Grossmann: This week on The Science of Politics, how Republicans lost 2018 by being too close to Trump. For the Niskanen Center, I’m Matt Grossmann. Republicans lost control of the House in 2018, and now could lose the presidency and the Senate this year. The fortunes of Republicans in Congress seem tied to Trump and his national agenda, but new research suggests that they would be better off trying to distinguish themselves from him and his policies. Today, I talked to Sarah Truel of the University of North Carolina, about her political behavior article with Austin Bussing, Will Patton and Jason Roberts, the electoral consequences of roll call voting healthcare and the 2018 election.

They find that votes to repeal Obamacare cost Republican seats in Congress. They didn’t listen to their constituents at town hall meetings and the repeal effort resulted in lower vote share, even though it didn’t pass. I also talked to Andrew Ballard of American University about his legislative studies quarterly article with Hans Hassell and Michael Heseltine. Be Careful What You Wish For: The Impacts of President Trump’s Midterm Endorsements. They find that Trump endorsed a lot of members of Congress in 2018, but it actually hurt them stimulating the opposition more than the supporters and tying legislators to an unpopular president. Truel found that Republicans thought Obamacare repeal would help them, but it hurt them.

Sarah Treul: The article focuses on the electoral ramifications for voting to repeal Obamacare, the Affordable Care Act. Ever since 2010, when the Affordable Care Act passed and particularly after the Republicans took over the majority in those midterm elections in 2010, there were numerous calls to repeal and replace Obamacare. It was a big part of the platform in 2010 itself, in those elections. This of course was easy for Republicans to say prior to President Trump, assuming the presidency. In fact, when Obama was in the White House, anything that came to his desk would of course be vetoed. But the truth was that even in those repealed votes themselves, a lot of what Obamacare offered the American people was beneficial and was actually popular, even though there was never brought electoral support for the entire policy. The individual components of the law were pretty popular.

So the question here is, many members of Congress thus face a tough choice on these repeal votes. These repeal votes that is particularly after President Trump assumed the presidency, was in the White House. You either vote with your constituents and think about keeping some parts that were popular with regards to the Affordable Care Act, or you could vote with the Republican party. So what we’re going to do in this article is use public opinion data. Measure it using the number of town halls held by these Republican members as a way to assess whether members of Congress who heard from angry constituents were more likely to oppose repeal.

Now we don’t find this, but then, what’s pretty neat is we’re able to use data on House primary elections following those votes. So the House primary elections in 2018 and also general election results to see if the member’s position on repeal affected her electoral success in those elections. And here we do find clear evidence that a member’s position on healthcare repeal had electoral consequences in 2018. Particularly this was true in the general election. So members of Congress who don’t respond to the wishes of their constituents, according to our article here, especially, or at least on landmark or consequential legislation or important legislation. However, you want to define that, do suffer electoral consequences. So I think this is pretty neat with regards to showing that the electoral connection, which we think is pretty important in a democratic system between members and their constituents is alive and well, at least on these types of votes.

Matt Grossmann: And Ballard says, that also happened with Trump endorsements.

Andrew Ballard: In this paper, we wanted to figure out what the effect of an endorsement from President Trump was in the 2018 midterms. And what we found was that an endorsement of a Republican candidate by the president helped those candidates raise funds for their reelection efforts or to get elected to Congress for the first time. But it also gave a boost to their opponents. And that ultimately we predicted that the effects on election day were negative. That an endorsement from the president was actually harmful to the election efforts, to the electoral prospects of Republican candidates that he endorsed.

Matt Grossmann: Although, voters don’t know about most that happens in Congress. Truel, says sometimes they do notice.

Sarah Treul: No doubt members face a complex information problem in seeking to figure out what voters are going to think about any particular vote. If they’re going to care about any particular vote, et cetera. And you’re absolutely right that the conventional wisdom suggests that overly partisan voting records can lead to consequences, negative consequences at the ballot box. And I think this research does a nice job when it comes to thinking about some legislation might be different, right? And I would emphasize here the role of particularly important legislation. That some of the signals that members get can be really muddled on that important legislation, but these conflicting partisan and constituency signals need to be parsed out by these numbers.

So the question then is what then? And I think today’s members and also scholars sometimes forget about this conflict. And we think most districts today are safe for one party or the other. Or we think that elections are highly nationalized, where voters cast ballots just along party lines. But it’s really still not that simple as that. This is especially the case with the increase in attention on primary elections and primary elections have become more of a threat to incumbents.

So in this particular case, I vote to keep Obamacare ran the risk of garnering you a primary challenger from your extreme flank in the Republican primary, but yet a vote to repeal might have hurt you in the general election against Democrats. And this is something I’m happy to talk more about. I do a lot of my research on primary elections, and I think we’re kind of just at the tip of the iceberg with regards to seeing increased numbers of primary challengers. But there’s no doubt in my mind that, you used the term members running scared earlier. And I think members are running scared, voting scared when it comes to the threat of primary challengers today.

Matt Grossmann: Ballard was addressing previous assumptions that endorsements were helpful.

Andrew Ballard: The conventional wisdom is that presidents give a bump to the candidates that they endorse. But I think that particularly again, in the new more nationalized environment, that opens up the endorsement and the candidate who is endorsed to more attacks and to giving the opposition side more traction to gain from an endorsement. And so, one of the things that was quite novel about what we did was also considering the effects on the components of endorsed candidates. And we really think that that does reflect changes since anybody had done this analysis. There was a Hernson piece from the early aughts, which was sort of the last, at all comprehensive work on endorsements that we could find.

Matt Grossmann: But he found Trump endorsements were different and may have had different effects.

Andrew Ballard: The way that President Trump chose to endorse candidates is fundamentally different than any president had ever endorsed candidates before. Twitter came online in 2007, so it existed for the entirety of Barack Obama’s presidency, but he never endorsed a candidate explicitly on Twitter. Whereas Trump did most of his endorsing on Twitter, which is a departure from the traditional norm of the president goes and stands with a candidate at a rally or goes and jointly holds a fundraiser with a candidate in the district. This suggests that we are in a new era of how presidents interact with congressional elections. The scholars have talked for a long time about how congressional elections have become more nationalized. And now at least this president, and it wouldn’t surprise me if future presidents are quick with the trigger on an endorsement in a very nationalized context. And particularly that people responded in a very nationalized way.

We found effects, in terms of campaign finance for both the amounts and raised but those were a little bit less consistent. And where we found really consistent effects were that both candidates who were endorsed and their opponents saw increased numbers of endorsements, and the vast majority of those were individual endorsements rather than pack endorsements. So what this says to us is that, okay, the president uses his worldwide megaphone to endorse a candidate via Twitter, and then people will react to that by going and giving some money to the candidate that they prefer in that election. Whether that’s the candidate that the president endorsed or not. And then the other implication is that unpopular presidents probably need to be very strategic and judicious with their usage of their platform to endorse candidates because in certain situations it can do more harm than good

Matt Grossmann: Truel was addressing a strange timeline where Obamacare hurt the Democrats in 2010. It was easy to run against until Republicans gained power.

Sarah Treul: So what we have is the Affordable Care Act of course, passing back in March of 2010. Democrats went on to take a blood bath in those midterm elections, following that vote. I think the Republican party picked up over 60 seats if I recall, right. And of course gained the majority in the House. And then from 2011, so following that 2010 midterm all the way up through 2016, there were over 60 attempts by the Republicans and the majority to repeal Obamacare. Then of course, the presidential election of 2016 happens. President Trump is in the White House. And now, as I stated earlier, there’s big consequences for these votes. So no longer is a vote to repeal or to replace going to just be dismissed at the president’s desk. Now it really could be signed into law.

And so based on this, we look at two important dates in this history. The first one happens in March. And so the date that we use here is March 24th of 2017. So here again, President Trump is in the White House and on March 24th of 2017, why we focus on that date, was the first attempt by the Republicans to really put forward a new package, right? And offer it as a vote to replace Obamacare with Trump in the White House. And what’s interesting here is that speaker, Paul Ryan actually has to pull that bill off the floor, which of course is nothing that any speaker of the house ever wants to do, sort of to admit failure in that regard. And to admit that you don’t have the votes to follow through with this piece of legislation.

And so on that date, in March 24th, the New York Times had sort of been following a whip count with regards to who was leaning towards supporting it, who was definitely supporting it, who was definitely a no, or who was unsure. And so we’re able to use those data there to get some sense of members, raw preferences.

And then the other date that’s important here is May 4th. So, a little less than two months later, May 4th, 2017. And this is the date that the Republicans did put forward a successful repeal vote and what this did and why it was able to work this time, as opposed to in March, is that they altered the bill to provide subsidies for people with preexisting conditions. Again, going back to one of those parts of Obamacare that was incredibly popular was issues related to preexisting conditions.

Also though important to note is in this version of the bill on May 4th, in addition to offering the subsidies to states, the House Freedom Caucus actually also wrote into the bill that states could opt out of those subsidies, right? To get that part of the Republican Party on board in passing this repeal as well. So we analyze that date too, which is the May 4th date. And that vote, which did pass, was a vote of 217 to 213. So it was incredibly close. There were 20 Republicans who voted against the American Healthcare Act on that date.

The final vote there in May was 217 to 213. Then what’s interesting is that’s all happening in 2017 and more than a year later, we’re facing the next midterm election in 2018. And what’s interesting here is that for the first time, and of course it was in primary elections before actually the general election, people were really still focused on this issue. It was not at all resolved. And in fact, you could even look at the world today and say here in 2020 we still have not resolved the issue of healthcare.

And so when we look at the elections that were happening in 2018, both primary elections and of course the general election in November, healthcare was incredibly salient in the minds of the American voter. In fact, 2018 Gallup poll data found that just before the midterm election, 80% of Americans were saying healthcare was the most important issue to their vote. There were other data out there that suggest that the majority of the ads that were aired in 2018, which my understanding was the largest number of ads we’d ever seen before, a majority of those ads focused on healthcare.

The Wesleyan Media Project found that over 50%, almost 60% of pro Democratic ads that were run in that election cycle focused on healthcare. And that’s a huge number. Again, according to the Wesleyan Media Project, the next category after that, most focus on healthcare, was just at like 15% and that was taxes, right? So the majority of all of the ads the American people were seeing were focused on healthcare. And it really was still the predominant issue in the minds of voters going into the voting booth in November and in the primaries proceeding that.

Matt Grossmann: Let’s dig into the project. They found Republicans were not responsive to town hall meetings, just like Democrats in 2010.

Sarah Treul: This come about in a discovery of data that was being collected by the town hall project. I think this came on the heels of the Tea Party Movement to some extent as town halls kind of took off as a forum for communication with voters. And the town hall project tracked every kind of town hall and other types of correspondence. We limited our data here to just those in-person town halls that were happening with a member back in the district. And what’s interesting here is we wanted to see whether or not, of course, these numbers were responsive to the town halls.

And for most media accounts of what was occurring in these town halls and most congressional accounts of what were occurring in these town halls, they seem to suggest that it was an angry public. It was an angry public at Republicans who were trying to repeal and/or replace Obamacare. And given that that was kind of the context of these town halls, we are sort of making an assumption here that the town halls were filled with constituents that were saying, don’t change healthcare. Like don’t go back, right? Don’t take away now, my preexisting condition changes. Don’t take back, my children can stay on my medical insurance. Don’t change all these things that have now been given to us, right? We want to have a health exchange. We want to have all the things that Obamacare was allowing us.

And so what we wanted to see is what the effect then was on this public sentiment, this public anger, if you will, on the members’ votes. And those two particular dates that I had previously talked about. The poll bill date in March and also the May vote. What we find though is that on the poll bill in March, the more town halls that a member held actually made the member of Congress more likely to be in favor of repeal, which is not at all what we would have expected in this case showing that members of Congress were not being persuaded by their constituents, at least not in this venue of course.

And then for the May vote, we find that the number of town halls held was not statistically distinguishable from zero. And so, the implications of this are, what are members paying attention to? Maybe it’s not these types of town hall events. And in fact in the article, we talk a little bit about members of Congress, particularly Republicans in this case, coming out and saying things like, “Well, I didn’t pay attention to the people in my town hall because I don’t even think they were constituents,” or they weren’t real people who vote for me is the implication there.

Sarah Treul: And so, when you kind of backfill the story in that regard, okay, sure, why would you pay attention to people that are not your voters or not your core base? And so, those are the findings that we find with regards to the town hall and the number of town halls that these Republican members were holding.

Matt Grossmann: But Republicans did respond to their district’s presidential voters.

Sarah Treul: We do find that members were responsive to the presidential vote. And probably again, this is because it’s a known quantifiable measure of party support. And in our world of nationalized elections, this might be the thing that candidates look to when they’re trying to figure out what the chances are back home of being reelected or winning decisively or scaring off potential challengers. So in this case, I think it’s not surprising that presidential vote really seem to matter a lot in predicting support for repealing healthcare. I think this is likely, again, the result of members of Congress being motivated by the next election. And at this point they needed to really think about what’s the best tool, what’s the best way that I can try to assess my chances of viability or my chances of success, and presidential vote is certainly one of them.

Matt Grossmann: Treul says Republicans might have been responding to potential primary threats, although they didn’t materialize yet.

Sarah Treul: I wouldn’t say necessarily that the conventional wisdom about primaries is wrong. I just think, as I stated earlier, that we’re at the tip of the iceberg when it comes to primary challenges. Post 2014 in Eric Cantor’s loss in his primary as majority leader of the house, members, when you talk to them, are increasingly fearful of these challenges. We talked about the kind of known running scared philosophy that exists in the literature of that conventional wisdom. And I think today more and more members are running scared in the primaries.

Even if they’re actually not having quality challengers emerging, they’re afraid of it happening. And I think a lot of them are spending time trying to figure out how I can ward off one of those challengers from even coming to the table, from someone even filing to run against me. And so, I think as the financial backing of primary candidates and also how they can be recruited lags behind the sphere to some extent, but as that catches up, we’re going to see primary elections really take off with regards to being a true threat. And by true threat, I mean not just running scared but a true threat in that these incumbents are actually going to lose their primary elections at increasing rates as we move forward once we figure out how to kind of move beyond party, which should not be hard in today’s open money environment.

Matt Grossmann: In 2018, Democrats ran against Obamacare repeal and they won on it.

Sarah Treul: Where we find no huge effect with regards to primary elections, we certainly find that there was an effect with regards to general election results, depending on your vote on the May 4th repeal bill. And I think this is because this became that highly salient roll call that media interest groups, everyone, is kind of looking to gravitate toward. It became the thing that people ran their campaigns on. It became the thing that even outside money spent its advertising money on since this is what Democrats ran on, right?

This is probably where this looks slightly different than what Republicans ran on in a primary, and also we’re dealing with a lot less money in primary elections compared to the general election. But when we move to that general election, we’re seeing Democrats with over a third of all Democratic ads being focused on healthcare. And more than half of those ads, or half of all the ads, I should say, after labor day were focused on healthcare.

The Republican Party was being sort of picked apart for its stance and voting for what they labeled the American Healthcare Act and being against Obamacare, being against the Affordable Care Act. And according to our findings, this cost the Republicans nearly 5% of the vote in their districts, holding all else equal. And I think this is really interesting. It’s certainly not to the extent that we saw Democrats punished for passing Obamacare back in 2010. And there’s some nice research on that by Brendan Nyhan and others. But at the same time, it’s a real effect and it did cost them seats. And in this world of insecure majorities, in this world where any party can become the majority, these types of salient votes, like at this kinds of attention in an election, are really consequential.

Matt Grossmann: Strangely the public reacted against both Democrats passing Obamacare and Republicans repealing it or trying to due to rising public dissatisfaction.

Sarah Treul: These kinds of findings I shared, Democrats are punished in 2010 for supporting healthcare reform. Republicans are punished in 2018 for supporting new healthcare reform if you will, or overturning bad healthcare reform in their eyes. It is probably unique to healthcare to some extent, but I also think the broader point here is that legislating is really tough, right? There’s the classic saying of course that we all know so well equating writing laws with making sausage, and these salient votes are in the limelight. They are the ones that people are gravitating toward. And that includes, of course, the media. We’re knee deep in sausage, if you will.

And I also think there’s just so much distrust today about Washington, particularly about Congress, and that this distrust leads us to gravitate, us being the American voter, when an issue becomes highly salient and the media and interest groups cling onto it, that this is also something challengers can emphasize and they can really play up the fact that we are so dissatisfied and so distrusting of our institutions, of our individual members, and they can capitalize on that. And they can say, “I’m here as the new candidate. I’m here as the outsider candidate. I’m here as the antiestablishment candidate and I’m ready to shake things up.” And we’re all waiting for that in some way, shape or form based on the data that just suggest that we are so dissatisfied and have so little trust in our American political institutions.

Matt Grossmann: People did vote on this issue because it was very salient in 2018.

Sarah Treul: It was really a hard thing for candidates to move away from simply because Democrats were pounding Republicans on it in their ads, outside groups were advertising on it. The media was focused on it. Media has been focused on healthcare since 2009. And I’m still not certain that it’s dissipated. And so, we did add onto this article an individual level analysis looking at CCES data to ask respondents if they favored… using this question from the CCES if they favored a partial repeal of the Affordable Care Act. It said also some things that it did, right?

Sarah Treul: Repealed both the individual and employer mandates, it cut Medicaid payments, and I think it ended the so called Cadillac plans. We do find that when there’s agreement in those things between the voters’ response on that CCES question and their member, unsurprisingly that those people support their member and that they agree. I think this is just all more evidence and why we did this is just to, again, suggest that this vote was really salient. So even if you’re agreeing at the individual level, it’s not like you could be convinced by this outside message that maybe we do need a new candidate or something. If you actually agree on the policy, and this is where maybe policy is more meaningful than I gave it credit for and…

In the policy, and this is where maybe policy is more meaningful than I gave it credit for in my last answer, that people do care when they know about something, and maybe they’re becoming well-informed simply because these votes are salient.

Matt Grossmann: Ballard looked at another factor tying Republican members of Congress to their party, Trump endorsements. Trump endorsed many 2018 candidates, although he was still trying to be strategic.

Andrew Ballard: President Trump was very active in endorsing candidates. He endorsed 80 candidates overall throughout the election cycle, and 45 of those 80, he only endorsed on Twitter. That doesn’t mean the 35 candidates. Also, in addition to endorsing many of them on Twitter, he did endorse them in person, either at a fundraiser or a rally. He was much more likely to endorse candidates for Senate of the 34 Republican candidates for Senate in a general election. He endorsed 62% of them versus only 15% of House candidates. He was also slightly more likely than average, so if he endorsed 80 candidates, there were 430 total Republican candidates in the general elections. That’s 19%. He only endorsed 50% of House candidates, he endorsed 24% of incumbents, he endorsed 8% of challengers, and he endorsed 27% of candidates who were running at open races. It really did seem like he was trying to both pick winners and to influence elections, right? Open seats tend to be some of the more competitive elections, but incumbents are also much more likely to win.

We also found that he was more likely to endorse candidates in places that had a historic Republican advantage in terms of the two party vote share from the Cook PVI scores. Again, on the side of looking at more competitive races, Senate races tend to all else equal be more competitive and are seen as more important than House races, and he was much more likely to endorse candidates there. So those two pieces that he was endorsing candidates in more competitive elections and trying to choose winners to at least in part bolster the power of his endorsements, but we also think that he was very much trying to endorse candidates who were loyal to him at some point. There are all sorts of anecdotes, both from Trump himself and from various political operatives that the president values loyalty. Extremely highly. Anytime a candidate or a member of Congress turns on him, he is quick with the vitriol. He was probably also endorsing candidates who he thought either were or would be loyal to him.

Matt Grossmann: They looked for several potential electoral effects, but overall they found it hurt Republicans.

Andrew Ballard: We wanted to be really thorough in our analysis of the electoral effect, so we did look at those three types of outcomes, both in terms of turnout, vote share, and win probability. It turned out we actually broke it down even farther to look at the turnout as a proportion of 2016 turnout, so the relative turnout to the most recent election in the district and the raw turnout. So just were there more people who went to vote? In terms of relative turnout, we didn’t really find any effects. In the House there were pretty much no effect whatsoever. The point estimate for the regression coefficient for the average treatment effects among the treated is positive in the Senate and the P-value is relatively low but doesn’t meet conventional estimates of statistical significance.

Since turnout is something that is an incredibly messy thing to measure, we’re not putting as much stock in that particular finding. We did find that raw turnout increased in Senate elections as a function of a presidential endorsement, and then across the board, we found decreases in vote share, so in both the House and the Senate, and then in the House that translated to a decrease in win probability. But not at the Senate because the air estimates were very wide even though that point estimate was negative. What does that mean? To us, when you bring all these things together, it means that endorse candidates got a boost in terms of the attention paid to the race and the number of donations that they brought in, but so did their opponents.

Then that didn’t really translate to huge differences or any differences in most cases in the number of people that went to vote in various places. But it did tend to mean that the endorsed candidate got a lower percentage of the vote than we would have expected absent an endorsement, and at least in the case of the House, that decreased the probability that those candidates would win their election, which is pretty big. The president all else equal might have cost his party some seats.

Matt Grossmann: He doesn’t know for sure if it was the endorsement itself or other ties to Trump.

Andrew Ballard: We don’t explicitly have a test for how Trumpian a candidate is. There are various ways that we toyed at trying to get with that using a measure like percentage of times that a candidate votes with the president’s preferred position when the president makes their position known on a bill. But there tends to be such low variance in that across both parties these days that we didn’t really think that that was going to be able to tell us anything. So this is one of those questions that we don’t know the answer to. I can speculate. I do think that a presidential endorsement from an unpopular president over a national media in a time of more nationalized elections, all else equal, it makes sense to me that that is going to hurt the endorsed candidate.

Matt Grossmann: It might’ve been specific to 2018 where Trump put Republicans in a difficult place.

Andrew Ballard: If we are trying to generalize the way that we should be looking at it is how popular is the president and how nationalized and strong the parties are and how nationalized the electoral context is, right? The number of safe House seats keeps going down, but does that mean that the president has more sway over an election in those close races? It’s hard to say, and that’s a good question to keep going for. The comparison to George Bush in 2002 is an interesting one because I would all else equal if you could make a like-to-like comparison and expect endorsements in that situation to have been more effective, at least effective in the way that they were meant to be.

One of the interesting aspects of this that I think we didn’t get as much of a chance to touch on largely due to data availability is that I don’t think the takeaway here should be that President Trump’s … should necessarily be the endorsements from President Trump were a huge damaging force because … But then it does put a number of Republicans in a tough position, because I think that there are lots of districts in which an endorsement from President Trump is an incredibly helpful tool in the primary election, but that it would be detrimental in the general election because Trump does have extremely high approval rates among Republicans. I think that that threat is of either endorsing an opponent or not giving an endorsement is real in a primary election. It’s just that in 2018, that was harmful in the general elections,

Matt Grossmann: Treul agrees that members need to separate themselves in the general, but might need Trump in the primary.

Sarah Treul: We always have known who compile these overly partisan voting records often do suffer the consequences of that at the ballot box. So there’s certainly some truth and probably what Ballard finds that members do need to separate themselves a little bit more. It’s just a big question of where that line is and how you draw it. But I would also exercise caution in overstating that especially given primary elections today. I think about what a Republican primary looks like right now, and you can make the same argument over on the Democratic side with different issues, whether it’s identity politics or something else, and those primary electorates are very different than in a general election and electorate.

So a Trump endorsement there is going to matter in a very different way than it would even in a safe Republican district at the general election level. I think that’s incredibly important to remember, and I don’t want to overemphasize primaries too much, and again there’s so few findings that say, “Oh yeah, this is really an important thing to look at.” But like I said, I really think we’re going to start seeing primary elections take off with regards to changing what our electoral outcomes look like in this country.

Matt Grossmann: She says it’s hard to be a moderate in today’s world, or just separate yourself from your party.

Sarah Treul: I think there’s definitely some truth in members were not able to separate themselves based on their ideology alone, or be able to claim that they’re moderates after this vote. I think that’s true across the board and not just on this vote, right? In our dichotomous world today where we’re so polarized, it’s very hard to maintain true moderate identity, if you will. I think the final vote on the American Health Care Act, which was the May 4th vote, made this repeal [inaudible 00:33:57] in some ways less ideological, but of course setting the moderate aside but definitely more partisan, right? That was because of the concessions, of course, that were offered to the House Freedom Caucus to bring them on board, and this is where again you can’t really be a moderate because all of a sudden you’re voting for something that also the extreme part of your party is supporting.

So prior to making those kinds of concessions to the HFC, as was the case in March when they had to pull the bill off of the floor, the HFC was united with the Democrats, right? In their opposition to the repeal, but obviously for very different reasons. But all of this becomes very muddled when you start constructing legislation to get the majority and all of a sudden it’s the Republican bill. To your point, I think there’s some truth in the piece from 2010 that talks about ideology and it’s really hard to separate and parse these things out now in this highly partisan world that we live in.

Matt Grossmann: Ballard agrees it’s hard for Republicans, especially to get out from under the thumb of Trump.

Andrew Ballard: One of President Trump’s greatest skills is keeping media attention on himself, and so any political figure is going to be more closely tied … any Republican political figure at least is going to be more closely tied to President Trump than many other Republican than they would have been to many other Republican presidents, I think. My guess is to what’s going to happen in the future is that it could go either way. If Trump wins reelection, I think that the Republican Party will continue to trumpify and that the shadow that he casts on over national politics will continue to get larger. If Trump doesn’t win reelection, then I would expect a lot of Trump Republicans to either try and distance themselves or they’ll get beat.

Matt Grossmann: He gives the example of Senator Thom Tillis, who has to win Trump supporters and swing vote.

Andrew Ballard: Thom Tillis needs a lot of support in North Carolina from Republicans, but he also needs support from a number of independence, because North Carolina is an interesting state. There were tens of thousands of people in 2016 who voted for both Donald Trump and Roy Cooper, the now governor and Democrat in North Carolina. So the independent vote or the swing vote is as alive and well in North Carolina as it is in any other state, and so if I’m Cal Cunningham, I am trying to tie Thom Tillis as much to President Trump as possible. And if I’m Thom Tillis, I’m trying to distance myself from Trump while remaining true to the sort of conservative values that got him elected in the first place.

Matt Grossmann: Ballard says all of the articles point out that tying candidates to unpopular aspects of the president can work.

Andrew Ballard: I think that the Sarah Treul piece is really interesting, because, as you know, the main finding is that votes to repeal Obamacare hurt Republicans in 2018, which is exactly the opposite of what Brendan Nyhan and his coauthors found in their 2012 piece for voting to pass Obamacare for Democrats, which was also found to be detrimental to their electoral prospects. And so I think that all of these pieces are sort of speaking a similar language, saying that if you can tie the candidate, not just necessarily to the president, but to unpopular aspects of the president, that in a lot of cases that’s going to be really helpful.

Now, I think that a Democrat running in rural Tennessee is going to be using these particular parts of campaign strategy differently than a candidate running in downtown LA, but in close races, I think that this can be a really powerful thing. That when the president is seen as weak or potentially detrimental to electoral chances, that you try and tie your opponent to that negative thing as much as you can.

Matt Grossmann: Democrats don’t have the one issue to pin Republicans on in 2020, Treul says, but that might not matter.

Sarah Treul: I don’t think we have that one vote, per se, that we’ve seen in past elections, that is going to be highly salient, and I think it’s really going to be an election, both at the congressional level and the presidential level, that’s going to be run on the state of the economy, and how did you, maybe as a governor of a state, or you in Congress, handle locking down your state and opening your state back up? And are you concerned about health? Are you concerned about other essential things such as the economy or education? And I think we’re not going to see the same types of advertising on a particular issue that we did before.

Now, that leads to your question about, does this hurt Democrats who were able to so nicely capitalize on healthcare, and Republicans trying to change your healthcare back in 2018? I’m not sure it does, right? I’m not sure that they’re going to necessarily need that kind of issue. I’m not sure if they brought that kind of issue back up it would have any traction, because so many Americans’ minds are on, “Will I have a job come winter? Will I have a job this summer? How am I going to put food on the table?” And so I don’t see, again, those single issues mattering all that much, and I don’t necessarily see them hurting one party more than the other at this particular time.

Matt Grossmann: Endorsements might not matter as much in 2020 either, Ballard says, when few candidates will be able to divorce themselves from Trump.

Andrew Ballard: I would think that in a presidential election year, that endorsements of congressional candidates from the president would have less of an effect than in the midterms, because presidential coattails are going to wash out some of those potential effects. I do think that there were instances in 2018 of candidates, either trying to walk back an endorsement from the president, there were some candidates who were surprised to be endorsed by the president, and I wouldn’t be surprised if there are some instances of that in 2020 as well.

The president has already started endorsing candidates. I think that President Trump probably still believes that his endorsements can only do good. I think that seems to be part of the president’s psyche, but the degree, if the folks around him think that the research that we did is plausible, which I think it is, and the electoral situation has some of those similar elements in 2020 than 2018, which I also think it does, with the caveat that I just mentioned, that effects in a presidential election year are probably going to be less than in a midterm election year of these endorsements. Then if those two things are true, you could imagine a situation in which the president’s handlers and advisors are urging him not to just endorse candidates willy nilly. Not that that’s what he was doing. He was actually reasonably systematic about it in 2018, but I wouldn’t be surprised if the president is a bit more judicious with his endorsements in 2020, or not. Honestly, it could go either way. Trying to predict what President Trump’s going to do is an exercise in something.

Matt Grossmann: Ballard’s next step is thinking about how minority parties concentrate on elections.

Andrew Ballard: This work sits a bit with a nascent book project that the Hans Hosel and I are starting on the relationship between party support in elections and loyal behavior from members once in Congress, particularly looking at the link between party directed campaign finance, and various legislative behaviors, with the idea that we’re able to say something broader about the relationships between the majority party and the minority party in Congress. And specifically that, and this is a very simplified version of the overall thrust that, majority parties are relatively more able to focus on giving to candidates and trying to compose a delegation that is useful for passing particular policies, whereas minority parties are, again, relatively, because they obviously care about passing policy and winning reelection, but minority parties are relatively more likely to care about winning real, composing a slate of candidates and a delegation that is going to be better at winning reelection and regain the majority party status.

And that might not sound exactly new. The biggest twist on it is that we’re directly comparing how those sorts of electoral and campaign behaviors then manifest as loyal party behaviors in Congress, and that’s part of my broader research agenda on political parties, where I’ve done a fair amount of work on and continue to sort of think about what the role of the minority party is in Congress, both in terms of influencing policy and in trying to position in elections.

Matt Grossmann: And Treul is thinking ahead to the next presidency, which might have to learn again that the public reacts negatively to dramatic change.

Sarah Treul: There is always more political risk and more political downside, and you have less political capital to spend on major legislation. I think it makes a strong case for doing things in a smaller piecemeal fashion, which is ironic, given that so much that Congress is doing today is done using omnibus legislation due to lack of ability to get appropriations bills through quite frequently. And so I think I would go back to Congress and say, “Let’s just do things small.” Right? “Let’s make small changes at the margins on popular things.” Whether that’s a Democratic Congress or Republican Congress, that’s going to be your best method by which to keep the American public on board and not at least have the American public think that you are changing all of American democracy as we currently know it, which I think is the reaction that we’re seeing in some of these back and forth. Whether it’s punishing Democrats in 2010, or it’s punishing Republicans in 2018, it’s all this result of too much change, too fast. And I think again, slowing it down, which is of course what the founding fathers always intended, might be the most beneficial thing to do.

Matt Grossmann: There’s a lot more to learn. The Science of Politics is available biweekly from the Niskanen Center. I’m your host, Matt Grossmann. Please review our recent episodes at or anywhere you find your podcasts. Thanks to Sarah Treul and Andrew Ballard for joining me. Please check out The Electoral Consequences of Roll Call Voting and Be Careful What You Wish For, and then listen in next time.

Photo Credit: The White House from Washington, DC / Public domain