Trump has promised to remake the federal bureaucracy in his own image and go after his political opponents and the media in his 2nd administration. But there are signals that public protest and civil society mobilization are subdued. How much do we have to fear further democratic backsliding under Trump 2? Brendan Nyhan finds expert consensus on many reasons to be concerned but also evidence that experts were too pessimistic about the likelihood of bad actions the first time.
Guest: Brendan Nyhan, Dartmouth
Study: Bright Line Watch
Transcript
Matt Grossmann: Threats to democracy in the second Trump administration, this week on the Science of Politics. For the Niskanen Center, I’m Matt Grossman. Trump has promised to remake the federal bureaucracy in his own image and to go after his political opponents in the media in his second administration. But there are signals that public protests and expert and civil society mobilization are subdued. How much do we have to fear further democratic backsliding under Trump too. This week I talked to Brendan Nyhan of Dartmouth about his new evidence from Bright Line Watch. He finds expert consensus on many reasons to be concerned, but also evidence that experts were too pessimistic about the likelihood of bad actions the first time, and some Trump actions may have public support.
I last spoke to Brendan right before the Biden took office, and we review how our thinking has evolved. You should enjoy our conversation. So we last reviewed the first Trump administration right before January 6th, and I was perhaps a bit premature in declaring most of our institutions resilient to the round one, but I wanted to get your update on the strength of US democracy after Trump’s first term, and how much our verdict should be colored by January 6th.
Brendan Nyhan: Yeah, well, lots happened in that time. It’s hard to capture everything. January 6th obviously was very disconcerting. Interestingly, the experts we surveyed at Bright Line Watch rated US democracy more positively after the 2020 election and after January 6th, I think because of the fact that Donald Trump was able to be defeated and then that the January 6th attempt to overturn the counting of the election results was also turned back. I myself was not so sanguine about those events because they called into question that the fundamental democratic compact of agreeing to respect the results of the election in a profound way we’d never seen before and mobilize violence against the election process in a way we’d never seen before. The fact that the effort was turned back, I suppose, can be seen as reassuring, but if you had told any of us that that’s where we would be back in 2015, we would’ve been shocked and appalled.
Moving forward since then, of course, Trump isn’t removed from office as a result of the second impeachment, nor is he disqualified from running for office again, which turns out to be an incredibly consequential failure on the part of Senate Republicans, even the ones who didn’t want him to be removed from office for those last few weeks. And that obviously has shaped the last four years profoundly. Biden’s term is more like a normal presidency, and of course, we can talk about that, but I don’t know how much his experience in office really should cause us to update our evaluations of US democracy either way. When we’re thinking about its strength, the question became whether the Republican Party was able to reject a candidate who had tried to overthrow our democracy. They weren’t.
And now, the question we face is whether the institutions are strong enough to resist pressure from an authoritarian leader in a second term where he will face fewer constraints in Congress and a more effective and compliant staff and bureaucracy given the expectation that there won’t be people like Milley and Kelly stopping Trump from doing things that would’ve violated the Constitution or the law. I don’t know how much exactly we’ve learned about that to this point, but I guess I would say I’m concerned. I think worst-case scenarios are easy to overstate, but it’s also important to keep in mind how serious the situation is.
Matt Grossmann: But a lot of the threats that you surveyed about are about the January 6th or the first term. He’s going to suspend the prosecutions that’s essentially already been dropped. He will pardon himself or he will have the case dropped, he’ll pardon the January 6th defendants. So how much are we going to be refighting or extending the fights of the first term versus evaluating new threats in the new term?
Brendan Nyhan: Yeah, it’s a fair point. I guess what I would say is I want to be careful not to move past those events related to January 6th that we expect because they are quite consequential. The DOJ policy of not prosecuting a sitting president, of course, never contemplated a scenario like the one we’re currently in. It therefore spares Trump from having to potentially pardon himself or direct DOJ to drop prosecutions against him. But he’s nonetheless indicated, for instance, that he would either pardon January 6th defendants who’ve already been convicted or drop prosecutions against them if they haven’t. If he does so, that undermines DOJ independence in a profound way. So I’m very concerned about the precedents that could be set here, even if they seem kind of baked in at this point. The president making these kinds of political determinations about who should and should not be prosecuted directly is a violation of one of the most important post-Watergate norms and opens the door to other kinds of interference in domestic law enforcement and prosecution and criminal prosecutions that could be incredibly damaging.
Once that door is open, it’s hard to know how we keep Trump from walking through it again and again. We know even in his first term, he directed a number of investigations to be opened against his political enemies. We know even the threat of those investigations can have profound effects on people’s willingness to stand up to him, their willingness to even participate in the political process at all. And that’s not the only risk that people perceive, of course, there’s also other kinds of ways he can exploit state power against his opponents or wield them on behalf of his allies. But I think the way those kinds of actions can shape the incentives of domestic political actors remains underappreciated. And it’s especially hard to understand or study because so much of it is anticipatory. We all can see the way that the billionaires are kissing his feet right now, and it seems to be because they fear the kinds of deployment of state power against them. That is possible, right? Does it mean that Mark Zuckerberg is going to be frog-marched out of Meta headquarters in handcuffs?
No, but that kind of make an example of just a few people can be quite powerful. So even these starting points could really shape our politics and can contaminate them an important way. We’re really losing something when people have to think, should I participate in politics given the risk that I could be financially ruined by an investigation or even prosecuted or even jailed and convicted? That has not been on the table in the post-Watergate United States, and now it is, and it will be as soon as those first norms start to fall. So I don’t want to minimize that point. I apologize for belaboring it, but I think it’s really critical to think about it. Some of the other kinds of scenarios that we surveyed people about once we adjust for the kind of expert tendency towards pessimism that we can talk about, the adjusted probabilities are under 50%. There’s lots of things Trump will not follow through on just like in his first term. So it will be important to calibrate and to think really carefully about what the risks are.
But I don’t want to lose sight of the fact that even some of these events being more unlikely than likely, it doesn’t make it okay. Right? The analogy I always use is if you were at the airport and someone said, “Well, there’s only a 20% chance the pilot of the airplane will take the controls and crash the plane into the sea,” you would of course never get on the airplane and you would never fly with that airline again, and you would demand that he be dragged off the plane immediately. So even small risks can be profoundly unacceptable. And that’s unfortunately the place where we’re in, where we’re bargaining with ourselves saying, “Well, maybe the risks aren’t as bad as the absolute worst-case scenario.” And it will be very easy to talk ourselves into that kind of rationalization, especially early in his term, when it will seem like the world hasn’t ended a month or two into. And his defenders will say, “Look, it’s normal politics, he hasn’t always gotten his way.” But as we saw during his first term, that’s a really important analytical mistake.
Matt Grossmann: So you’re still concerned, your experts are still highly concerned, but the public signals of resistance or mobilization do seem a lot less subdued than in the first Trump administration. The protesters, the expert in civil society mobilization, the media concern are all way less immediate than for the first Trump administration. So that could suggest that we’ve dulled ourselves. It could suggest that people don’t really believe the threats that they proffered in the campaign, or it could suggest that they just don’t see the constant outrage as a solution to those problems. What do you think?
Brendan Nyhan: I think all of those accounts have some validity. I’ve been struck by the lack of intensity of opposition, and even though I talked about the ways in which certain kinds of political actors are stepping out of the fray possibly for reasons we should be concerned about, I don’t think that explains a lack of intensity among the Democratic base or their elected leaders. There seems to be something about the manner of Trump’s victory that he could win, again, knowing everything we know and win in a more convincing fashion. Obviously, it’s still one of the closer races we’ve had where he wins the popular vote, but narrowly he sweeps the battleground results, the battleground states, but very narrowly, that seems to have created a kind of legitimacy, especially in conjunction with him being reelected after having lost, after having been impeached twice that is demobilizing to his opponents.
They’re not certain of how to rally opposition to him, but the rallying will no doubt come. I’ll be interested to see what form it takes. You can certainly imagine there will be a lot of exhaustion going through this whole process again. So I don’t know, for instance, if we’ll ever see the scale of what we saw in the first Trump term where the Women’s March, for instance, was a non-trivial percentage of people who live in the United States participated in those rallies. Very large numbers of people protested against the so-called Muslim ban as well, the kinds of mobilizations we saw. At this point, it seems those seem quite unlikely in the near term, but we’ve certainly seen in the past that Trump can inspire some very strong feelings. So I’m hesitant to rule anything out.
Matt Grossmann: As you say, the manner of the election may matter here. He won the popular vote, he gained ground with racial minorities, which might undermine some of the stories we initially told about the Trump election. And there’s been a lot of global trends where Trump seems less like an outlier this time. So to what extent has that changed how scholars and the media react to Trump and should it?
Brendan Nyhan: We have to hold two ideas in our head at the same time. The first is why did Trump win analytically? And the second is, should we be concerned about the manner in which he exercised his power even though he won office in a legitimate way? So not only did he win the Electoral College, but he won the popular vote, that takes the popular vote objection off the table from last time. It appears as in many other countries that voters did not place a great deal of weight on threats to democracy, at least outside of the Democratic base. That unfortunately is something that we’ve seen a great deal of scholarship on in the intervening years since Trump was first elected. And it appears to be a common feature of human beings, at least in the 20th and 21st centuries. Democracy has not been a first order concern at the ballot box, and that means we’re vulnerable to candidates like Trump, and of course, people have every right to be concerned about the state of the economy, for instance, and the other kinds of factors that played into Trump’s victory.
So I think it’s absolutely fine to say there are reasons Trump won that have nothing to do with threats to democracy. And of course, it is fundamental to the democratic process to respect the outcome, but this is how candidates like Trump anti-system candidates create these profound dilemmas for democracy, because of course, Trump was legitimately elected and legitimately elected leaders should be treated as such. That’s precisely what Republicans and Trump in particular tried to deny to Joe Biden in his presidency. At the same time, Trump is not just another president, and so we have to think about him in a different way, right? In particular that his electoral victory does not provide legitimacy to the actions he might take that would undermine American democracy. And that’s, I think, the balance we have to strike going forward.
Matt Grossmann: So your latest survey found that public faith in elections is, I don’t know, back to normal, but rebounded significantly basically because Republicans see the 2024 elections as legitimate and are much less concerned about various forms of voter fraud, that might seem like success. On the other hand, it seems to reinforce the idea that Republicans are happy with democracy when they win, and Democrats really didn’t. They did react, but nowhere near the level that Republicans did to losing. So how should we interpret that?
Brendan Nyhan: Yeah, that’s right. There’s good news and bad news. The good news is confidence in the electoral vote count is higher than it’s been since before the 2020 election. If you were concerned that Americans didn’t trust elections, then you should be encouraged by the numbers we see now, the problem is that’s entirely dependent, it seems on the outcome. At least that’s what we would worry based on what we saw after 2020. So Democrats did not replicate what Republicans did after 2020. They did not express a widespread lack of confidence in American elections. They did not embrace claims of widespread fraud, and that’s brought us to this point. The concern is first that Democrats may over time move in the Trumpian direction. We saw people trying to get various kinds of fraud and conspiracy claims off the ground, for instance, on social media. They didn’t get any traction this time, but it’s certainly possible to imagine they will in the future.
And secondly, we worry that Republicans won’t react as they did in this election in a future election that they lose. Now, no one knows how that will play in an election that doesn’t involve Donald Trump. We’ve certainly seen people try and fail to use his playbook successfully. So it’s possible the future Republicans who lose will not be as successful as Trump was at generating widespread dissatisfaction and distrust in the Republican base. But we have to worry, and there’s a kind of dilemma here. If the Trumpian pattern persists where essentially you get quiet and calm, if Republicans win and you get legitimacy threats all the way up to violence and insurrection, if Republicans lose and that kind of heads we win, tails you lose game is not one that democracies can long sustain comfortably. So I do fear that Trump playbook being reused in the future and worry that this is not as stable and quiescent as it might seem in the long term.
Matt Grossmann: So one of the most mentioned threats of the second Trump administration is about mass deportation. You included it as a potential threat on your list, but also the public and experts seem to agree that there was something of a mandate for a change in immigration policy. It’s also an area on which the President has long held more power than on other issue areas. It’s a issue on which Joe Biden made policy changes that resulted in some real world changes in conditions that some people credit as a mattering in the election. So on the one hand, this is viewed as a big use of executive power, something that would really change the nation. On the other hand, it seems like a facilitation of the election results in the area that people most identified with Trump’s victory. What do you think?
Brendan Nyhan: Yeah, this is another Trumpian democratic dilemma. There’s no way around. On the one hand, the president has, of course, as you suggested, a good deal of power when it comes to immigration, immigration enforcement. We saw Barack Obama deport very large numbers of people during his presidency, not in the way that Trump has promised to, but in a volume that provoked that was quite striking and provoked very little concern from Democrats who have expressed a great deal of opposition to deportation when done by Trump and even by Joe Biden. So there’s something about this issue that is challenging. I mean, if you are not legally allowed to be in this country, it is not necessarily a kind of core democratic right for those people to stay. And so we see our experts being more divided on the threat to democracy posed by deportations.
And as you noted in terms of democratic responsiveness, both our public respondents and our expert respondents in our most recent Bright Line Watch survey said that to the extent that Trump has a mandate, deportations on immigration are rated as one of the areas where his claim to such a mandate would be strongest. So with all of that said, though, I think we should worry about the manner in which Trump has promised to do deportations as well as the potential volumes he’s describing, that Obama deportations were centered on people who had criminal charges. We’re talking about non-border enforcement here. In Trump’s first term, there was a child separation policy that was widely seen as inhumane in terms of border enforcement. And I think there’s concern about inhumane illiberal types of deportations that would implicate core liberal democratic ideals.
And that’s, I think, a real fear to worry about, especially because Trump has promised to mobilize federal law enforcement and even military units in the service of this deportation. And so now, we’re talking about an authoritarian leader deploying federal force domestically in an unprecedented way. And we can imagine the kinds of concerns that that would raise. And again, as I said, during Trump’s first term a lot, imagine if you saw this in another country, an authoritarian leader deploying the military domestically and removing people, law-abiding people living their normal lives and potentially moving them in vast numbers into camps because that’s what’s been openly contemplated.
Now, there’s lots of reasons to imagine that may not take place. In part, it’s simply so expensive that it’s not clear where Trump would get the money, but I nonetheless think it’s both normal. It’s normal, and it’s okay to have a debate about immigration as a policy issue. It’s a very challenging policy issue. There have always been deportations. Every president has engaged in deportations. The question is, is the way Trump is promising to policy issue a matter of democratic concern, even if he has some electoral mandate or legitimacy around acting in that domain? And I think the answer is yes. I think it is reasonable to be concerned when someone who talks about immigrants in the way he does promises mass deportations using military and law enforcement power that would remove people and put them into camps in unprecedented numbers, I think, even just saying the sentence, the question answers itself.
Matt Grossmann: A place where there’s both motivation and there appears to have been considerable planning compared to the first administration is in trying to remake the administrative part of government. They’re going to reinstate Schedule F in some form, making more government officials fireable and replaceable. Those in some ways respond to Trump’s very specific concerns about he couldn’t fire people who were going against him in the first administration, but they have found common cause with a broader conservative argument that they sort of always lose the administrative part of the state and have been doing so for some time, and it never seems to react to the conservative electoral victories. So it seems like there’s going to be a big push here, and it could affect all manner of other kinds of threats in the next administration, even if it doesn’t actually result in mass firings, but just kind of a few particular firings or replacement that make everyone else respond differently in their job. So how should we think about that as it’s going about? What would be the real threats from those changes?
Brendan Nyhan: Yeah, again, a complicated issue that raises real dilemmas. I mean, I vividly remember reading a book about someone who worked in the Department of Education in the ’80s describing taking over an office within the education department and finding someone who ran a small business off their desk and couldn’t be fired, basically just sat there and did no work. There’s lots of reasons to think that civil service reforms that of the sort experts might propose could be appropriate. Am I convinced that the current federal government personnel rules are optimal? No, it’s not my area of expertise, but I certainly can imagine reasons to make changes. Trump often complains about the Senate confirmation process and using acting officials. That’s another area that experts have long identified as problematic, and one where we can imagine good faith kinds of changes. The challenge as always with Trump is to what end these changes might be directed, and is this even being approached in good faith in any serious sense?
Will these be serious policy reforms of the sort that experts might think would improve American government or they are mechanisms to generate loyalty in a personal sense in the sense it undermines the rule of law and empowers Trump to act in illiberal ways in establishing a kind of personalist regime? And that’s the kind of fear I think it’s appropriate to have. Now, is he going to get that far in four years? Probably not. We know how resistant to change you and I do as political scientists, the federal government is, there’s lots of constituencies that have stakes in the activities undertaken by every part of the federal bureaucracy, and they will make it very challenging to make the sorts of wholesale changes that people like Elon Musk and his allies are fantasizing about. Nonetheless, I think it’s reasonable to worry about the ways in which this Schedule F action could be exploited.
You can also, of course, come up with perfectly justifiable appropriate versions of it, but the track record of Trump in a governance role suggests there’s no reason to believe that any of these actions will be undertaken in a serious way. There were real accomplishments of the first Trump administration, but they worked around him or operated essentially behind his back, people who were able to do things that stayed off the radar or in areas that he wasn’t concerned about. Operation Warp Speed is an amazing accomplishment, but I don’t know how much we can imagine that as a kind of model for these other areas where we’re more likely to get this sort of chaotic Trump approach to governance, which is far less likely, I think, to lead to anything valuable in a Warp Speed type process.
Matt Grossmann: Those seem like two poles of what might happen. But I guess I’m interested more in the intermediaries. I don’t think anybody’s going to look at these reforms and say they’re the expert consensus on reforming personnel and contracting. On the other hand, I think there’s going to be lots, I predict lots of media stories about particular instances where it’s not so much just he’s created a personalist regime, but he is undermining the bureaucracy and trying to get obedience from below by making a few people fall who are working across purposes from his goals, some of which are maybe personal, but some of which are policy goals. But I’m just interested in how we should interpret these as they happen since I’m expecting quite a few.
Brendan Nyhan: Yeah, no, that’s a great point. So I can imagine both happening, and I guess it depends on your priorities about the relative magnitudes. So I think your model is probably a better one than mine that we shouldn’t think about mass removals. We should think about a few high-profile cases that are used to create fear and compliance, and in some cases, that may induce a kind of democratically appropriate responsiveness to normal conservative policy goals being undertaken by a Republican president where we would say it’s appropriate for the bureaucracy to carry out the legitimate democratic policy goals of a Republican president. That’s fine, that’s normal, and the fact that the president has more leverage over the bureaucracy in that sense seems perfectly defensible on the one hand. The other is the people who might have stood up to Trump when he proposes actions that would violate norms or laws if those people become fearful and compliant, and how we weigh the relative magnitudes of those.
I guess it is going to depend on our priorities at this point. I guess, I’m obviously more concerned about the latter when we’re thinking about kind of given the persistence of the ways institutions work and the experience we’ve seen in other countries, damaging institutions in these ways is hard to come back from, the lesson in other countries is when the rule of law is diminished, it’s not easily restored. So I’m very risk averse on undermining the protections of bureaucratic neutrality. I want more responsiveness on the margin, but I’m not sure I’m willing to trade it off one-to-one given the risks of personalist rule and all the attendant downsides.
Matt Grossmann: Well, one place where there’s clear concern is about using the Justice Department to go after political opponents, and some of these debates might play out there. Republicans, of course, say that this is a turnabout because the previous administrations Justice Department did go after Trump, you and I probably view the prosecutions in a more favorable light, but I’m at least willing to say that at the state level, the New York prosecution which is the only one that actually resulted in conviction would not have been pursued if it was a non-political opponent. So we do have at least an example of that kind of overreach. And more broadly, there’s a lot of talk about what happened in the first Trump administration, the Russia investigation, the Ukraine impeachment, the reaction to January 6th, at least with people saying it didn’t seem to work to regularly go after Trump on these issues. So I guess what should we learn about that if and when Trump does go after some political appointees, some political opponents via the Justice Department, but also how opponents of Trump should react to new scandals that arise while he’s in office?
Brendan Nyhan: Yeah, this is a tough one. I’ll say, you were describing in your case, and I think that’s important. I think legal accountability for January 6th in the classified documents’ case seems very appropriate. Those seem to have been quite strong cases. And the fact that Trump did not face full legal accountability before running again is a kind of profound failure on the part of Merrick Garland in the Justice Department in my view. But in your case was very, I’m not a lawyer, but seems very weak. Our experts have always expressed very mixed views about the merits of that case. And I was expressing concern about New York Attorney General candidates campaigning on promises to investigate Trump several years ago. I think that’s a real violation of democratic norms to use prosecutorial power against your opponents. And Democrats are likely to get sucked into that vortex if they’re not careful in just a way that Trump has. So I think we should be very cautious about that and really reserve those sorts of actions for the most serious cases.
I think it’s fair to say that keeping our nation’s most sensitive secrets in your bathroom and trying to overturn a presidential election merit that kind of prosecutorial scrutiny, the fairly elaborate legal theory in the New York case doesn’t seem to meet the same burden of proof to me as an outsider. So I think that distinction is appropriate. When it comes to political opponents, yeah, again, I’m just going to say I think the threat is the most important factor. And what I worry about is people will say, these cases are rare, and therefore the worst case scenario hasn’t happened. But as we’ve discussed, the examples matter. The anticipatory actions take into account the threats. Vladimir Putin does not imprison all of his political opponents. He just imprisons enough to make everyone else fearful and take a step back from challenging him directly. Is Trump Putin? Of course not. Is that a direct analogy to the Department of Justice in 2025? No, I don’t think so. There’ll be lots of civil servants there doing normal Department of Justice stuff.
But again, even the threat of a prosecution can have a profound effect on people. And I’ll just even note that Trump could only bring successful prosecutions against his opponents, and that could still be an abuse of power because we know if prosecutors dig hard enough, they can often find things, cases that can be brought, and if that scrutiny is applied disproportionately, Trump could be, in one sense, carrying out the law, and another undermining it profoundly. So I think there’s real reason for concern. People just don’t… I don’t know that people have a sense of the scope of prosecutorial power and how hard it is to regulate a democratic society because those investigations take place in secret. It is profoundly vulnerable to political manipulation in a way that other kinds of state power and state action isn’t, and I worry that it’s one of the most vulnerable aspects of the government in that sense.
Matt Grossmann: So as you mentioned, you found that experts tend to overestimate or be too pessimistic about democratic threats, and you’ve now produced predictive estimates that adjust for that bias that you’ve found before. On the other hand, that seems to be a pretty fundamental error or a fundamental problem with the project if the experts on what you’re asking them about do have these systematic biases. So how should we think about that? What are you doing to adjust for it and how should we think about it?
Brendan Nyhan: Yeah, no, this is something I’m very excited about in the new Bright Line Watch report, and I would encourage listeners to check it out if it’s of interest. We’ve been eliciting expert forecasts for a long time and then evaluating them afterwards. And we have found experts rating a number of events as 50, 60, 70% likely to occur and those very often not taking place. And so working with Andrew Little at UC Berkeley, the new report examines that full set, the expert forecast from four separate waves of events that we think can be fairly characterized as negative for democracy. We don’t have enough events that aren’t safely characterized as negative to evaluate those separately. That’s something we hope to consider in the future. So in this set of negative events for democracy, we find the experts tend to be too pessimistic. In other words, they’re putting the probability of these negative events systematically too high.
However, that’s not an across the board pattern. We see that the events they rate as very likely to occur, actually occur at a very high rate. So there’s some real signal in there, and that’s different from the kind of simple story that research on expert forecasting has told in the past, most famously by Phil Tetlock, experts are no better than the monkey throwing darts. In this case, our experts are adding real value. They’re just too pessimistic. And so what we do in this most recent study is both we aggregate those forecasts to hopefully capture some kind of a wisdom of crowd effect, and then we net out that pessimism that’s been displayed in past forecasts, and in combination then, we’re able to generate a set of adjusted forecasts that we think better reflect the rate at which events are likely to occur given the performance of these expert forecasts in the past. And that’s something that we’re going to, I think, be applying and refining going forward.
And it’s important. I mean, I think it’s important to reflect on why that might be occurring. Our experts are human beings and not without bias. And actually, in some cases, again, research like Tetlock suggests that experts could be pretty good at coming up with stories in their heads about why things might happen. And, of course, there have been concerns that political scientists have been too concerned about the threats to democracy from people like Trump. So I think this is a nice corrective, and it’s one where we want to bring these expert evaluations to data and not just assume that they’re going to be correct. And if you take those new forecasts literally then, many of the events we surveyed about, most of them in fact are less than 50% likely to occur. But of course, events that are less than 50% likely to occur on that, some of them will occur. If you go through that list, there’s some pretty concerning things.
Matt Grossmann: So let’s talk a little bit more about this. Because if I’m reading the chart correctly, I mean, there’s events that people thought were 65% likely to occur that 20% of them occurred or so. So we got pretty high, and I know at the highest levels, they did occur, but we got pretty far into a consensus that didn’t materialize. So what are the reasons for that? Is there liberal bias in academia? Is there just a negativity bias when you’re asking people about a bunch of threats? Is there something that went wrong in the reasoning that we can tell?
Brendan Nyhan: We don’t have a lot of direct evidence on why this is taking place. It’s certainly possible that it’s political bias. I don’t think the totality of the evaluations offered by our experts across all their survey responses is consistent with that. I’ll say for instance, if you look at the overall evaluations of democracy during Trump’s term, the experts barely moved. They expressed concern. We asked people are these things threats to democracy? But they didn’t engage in the doomerism that we often saw online, and they consistently rated the US democracy as pretty stable, and while imperfect, significantly better than countries like Brazil and Russia and so forth. So I’m not convinced that the political bias story is the right one. I think negativity bias is an important idea to evaluate. And in our new future surveys, we hope to include negative events that aren’t related to democracy as well as events related to democracy that are neutral or positive to be able to better isolate how much that’s contributing.
The final factor is, I think it may simply be these events are ones that seem especially intuitive. They coincide with narratives about Trump and potential vulnerabilities. In part, that’s what we’re selecting in choosing the events to survey them about. They’re the ones that, to us, people with similar perspectives seem sufficiently plausible that it’s worth asking and sufficiently concerning that it’s worth asking experts about them. And in selecting them in that way, we may be drawing out their beliefs on the kinds of events that they’re most likely to override their prevalence of. So we’re going to have, I think, a broader set of events going forward and we’ll be working with… We’ve already drawn in another expert with expertise on the administrative state to help us develop a better set of measures going forward.
Matt Grossmann: So you’ve now been doing these surveys for quite a while, including through the Biden administration, and you mentioned earlier that there wasn’t much evidence that Biden was a big net positive or a negative for US democracy, but it was sort of also a test of your method. So I guess, I want you to reflect a little bit on to what extent can we use the surveys from the first Trump administration and the Biden administration to assess were there things that… Did we identify the bright lines and did we find the ones where even in the Biden cases, where there was sort of a real potential threat? And if not, why not?
Brendan Nyhan: Yeah, I do think the track record is important. We have these cross national surveys of the state of democracy, as you know, things like V-Dem and Freedom House, but they’re not as granular as the work we do, which of course has pros and cons. We may be focused too much on events that pass by quickly, but we hope we’re building a kind of evidence-based that helps us assess both whether something is truly unusual and concerning for democracy and whether a potential threat is likely to materialize. And first, we were just relying on expert judgment and experience from other countries and past history, but now, of course, we have this accumulating evidence base of the expert forecast themselves. So I think the hope is to build that kind of empirical baseline to help us better calibrate our understandings of threats and risk going forward.
The Biden administration, we continue to assess, I would say our focus shifted in part to the criminal cases and how to think about those as well as the persistence of public beliefs in widespread fraud and the rejection of Joe Biden as the rightful winner. So those were some of the most important narratives related to the state of US democracy during Biden’s term. That’s not to say that Biden’s actions were without concern. For instance, I’m sure if we survey our experts about the Hunter Biden pardon, which took place too late to include in our most recent survey, they would evaluate that very negatively. And it’s important to measure those kinds of actions too. So I’m hopeful we have a useful evidence base here, but we’re also limited by the tools we have. We’ve done some secondary research ourselves. We have new reports on…
Matt Grossmann: Sorry, let me challenge that a little bit.
Brendan Nyhan: Oh, please, please, please.
Matt Grossmann: It just seems like the Hunter Biden pardon as an example of like, well, why wasn’t that viewed as a big threat going forward and listed on a list of threats and it actually happened? Seems like we should give ourselves a chance to be wrong that the Biden administration will also make threats to democracy.
Brendan Nyhan: It simply was too late to be included on the survey when it comes to the Hunter Biden. So we were already out of the field. Sorry.
Matt Grossmann: Well, not, its potential.
Brendan Nyhan: Oh, the potential, yes.
Matt Grossmann: Yeah. Yeah.
Brendan Nyhan: Yes. No, it should have been on there. That’s totally a fair point. We should have asked them to forecast that. On that one, I think, views might differ. I mean, I’d be curious on your view, Matt, given the record of Obama is a kind of exception to this, but people pardoning relatives and allies on the way out is actually a time-worn tradition. So even though I think the manner in which the pardon was issued may raise further concerns and people would be normatively unhappy with it, I’m not sure it’d be seen as the kind of new threat if we went through and itemized that list. In terms of Biden, let me give you an example of something we investigated as a potential concern, and that’s the replacement of Biden as a nominee with Kamala Harris, what has now become a standard primary nominating process.
We asked both experts in the public about that process and how they evaluated whether she was the rightful nominee and so forth. And it was simply just not a matter of concern to anybody in appreciable numbers. So that was a kind of normative concern that we thought related to democracy that had been raised about the Biden administration. If there are other examples though, let me just say, our inboxes are open, so whether it’s Matt or anyone listening, please send ideas and suggestions for things we should be serving about. It’s really valuable.
Matt Grossmann: So part of this project came out of, I think, a realization after 2016 that American politics folks could learn more from comparative politics. And certainly there are international examples and proliferating ones that might help us evaluate the Trump administration. On the other hand, there were some things that maybe Americanists were more likely to notice, like problems in Congress that Trump had the differences in the party system in the US. So how should we evaluate our successful or not successful importation of comparativist ideas in evaluating Trump?
Brendan Nyhan: Yeah, that’s a great question, and I only feel partially qualified to speak to it as a non-comparativist. I’ll say the comparatives obviously were wrong about the threat, but the Americans were too. So I don’t think that’s distinctive. They had elaborate statistical models that forecast the risk of democratic breakdown in the US is almost impossibly low. We simply didn’t talk about it in those of us who study American politics. I don’t think the word democracy was uttered in a single minute of my grad school training when it came to Americanist work. So they were wrong. I think their model of democratic erosion and their understanding of how it takes place has been, I think, important and informative and provided some real guidance to us in learning about the manner in which democracy can be degraded. I’ve learned a lot from Levitsky and Ziblatt, for instance.
At the same time, there are these US factors as you identify, US political history is inexorably intertwined with the issue of race, of course. There’s the folks mostly at Cornell have a really excellent book on American political history and threats to democracy that goes through some of that and identify some of the key tensions. I think the two-party system obviously is really important. And I guess where I’ve come down on this is it’s less about a comparative US distinction because we’re seeing European systems coming under increasing pressure too, even though it seems that the two-party system made us more vulnerable, the same sorts of factors are pressing on their political systems too. So I’m concerned about explanations that put too heavy a weight on US specific factors other than two-party system, which I think is profoundly important.
The thing I would just add is I’ve really come back to the importance of institutions and party systems. I think a lot of the research has been on the public opinion side because it’s frankly easier to do. But to the extent that we can make democratic systems more resilient, I think that’s almost surely going to happen through reforms to the institutions that we have. And Rocío Titiunik and I have a policy forum piece in science that describes our argument that public opinion is not going to save democracy. We have to do the hard work of thinking about the electoral institutions that create the incentives for the politicians who are the engine of our politics for better or worse.
Matt Grossmann: So we’re entering the second Trump administration, and I want to get a sense of what you all are going to be doing, but we seem to be reprising a debate that we had a little bit last time where I am a little bit more willing to say, but does doing what scholars do work? And you’re more on the, well, we need the information. So that’s the first step. But it does seem like the first Trump administration was a arena to evaluate that a little bit and maybe raising lots of concerns about Democratic backsliding did not work in the way that scholars expected it to. So what will you be doing in the second Trump administration? Is there anything you can learn from the first Trump administration that will help it be differently impactful?
Brendan Nyhan: Yeah, fair point. I hope I’m appropriately realistic about the likelihood of the kind of scholarly analysis that we do affecting public opinion. I don’t think it’s what Bright Line Watch does is likely to be an effective strategy for resisting authoritarianism or illiberal leaders in general. That was never our intention or claim. It was to aid civil society folks and journalists in thinking about the risks and threats. And that remains our mission. Whether and how that translates into political messages, or if it should, I think it’s a far more open question. Again, the comparative evidence suggests that people just don’t prioritize democracy. People thought there might be blended messages that link democracy to policy issues of concern. We saw some attempts to try those kinds of messages in this last election. I don’t know of any particular evidence that they were effective. I just think that we don’t have to choose, this is not a simple binary here, and I’m not a democratic strategist in a capital D sense.
My job as a scientist is to report the empirical findings and discuss them in light of the broader normative concerns that I hope almost all of us share. And beyond that, it’s hard to know what’s most effective. But I do think having some way to calibrate on the nature of the threat is really critical because opposition by elites and civil society and critical coverage from journalists are a critical part of the mix of small D democratic resistance. Even if top line messages about democracy are not the most important in general elections at resisting authoritarianism. And so I think we can hold both those ideas in our head at the same time. That’s certainly been our approach at Bright Line. I’m really grateful to my colleagues for working with me on it. And our approach going forward is we’re going to try to do things differently as best we can a second term.
The first part of that is raising more funds. So your deep pocket of listeners should contact me immediately. Beyond that, we want to keep adapting. So we’re going to look, I think this forecasting work that we’ve been talking about is something we’ll be focusing on going forward. And we’re going to be thinking a lot about how to monitor and evaluate what Trump does with the administrative state. I think our first, the way we evaluated Trump’s first term really centered around the broader norm violations that he was constantly engaging in. But as Ben Wittes said in his post-election piece on Lawfare, the thing to be most concerned about in a second Trump term is what he does that’s legal, not what’s illegal. And I think Bright Line will definitely be keeping that lesson in mind.
Matt Grossmann: There’s a lot more to learn. The Science of Politics is available bi-weekly from the Niskanen Center, and I’m your host, Matt Grossman. If you like this discussion, here are the episodes you should check out next, all linked on our website. How much did Trump undermine US democracy? Does the public respond to threats to democracy? Is democracy declining in the American states? Who’s more afraid of democracy, the center or the right? And why do Americans accept Democratic backsliding? Thanks to Brendan Nyhan for joining me. Please check out Bright Line Watch, and then listen in next time.