Milan Svolik is the Elizabeth S. & A. Varick Stout Professor of Political Science at Yale University. He is the author of the 2012 book The Politics of Authoritarian Rule. He has also published a variety of influential journal articles such as “Polarization Versus Democracy”, “When Polarization Trumps Civic Virtue”, and “In Europe, Democracy Erodes from the Right.”
The Democracy Paradox is made in partnership with the Kellogg Institute of the Keough School of Global Affairs at the University of Notre Dame.
We are badly mismeasuring whether and how much people care about democracy.
Milan Svolik
Key Highlights
- Introduction – 0:20
- Measuring What Voters Really Believe – 3:33
- Militant Democracy and the Risks of Overcorrection – 16:51
- The Left, the Right, and Who Defends Democracy – 37:18
- The Voter as Democracy’s Last Gatekeeper – 52:13
Introduction
Today’s guest is Milan Svolik. He is the Elizabeth S. & A. Varick Stout Professor of Political Science at Yale University. He is the author of the 2012 book The Politics of Authoritarian Rule. But more recently he has published a series of journal articles on the relationship between democratic backsliding, polarization, and partisanship. A few of those articles include “Polarization Versus Democracy”, “When Polarization Trumps Civic Virtue”, and “In Europe, Democracy Erodes from the Right.”
People outside of political science probably have never heard of Milan Svolik. But for those who read political science regularly he is among the most influential thinkers in the discipline. What sets him apart is his willingness to combine rigorous social scientific methods with a creative outlook on politics.
This episode touches more on methodology than most in the past. We don’t cover research design or statistical methods but spend some time discussing ways to measure support for democracy. It’s an important question even for ordinary citizens because it helps us understand why people say they support democracy but support antidemocratic candidates. Are people insincere about their support for democracy or is their disagreement about the meaning of democracy?
The show has had many episodes on democratic backsliding and polarization. In many of those episodes, I brought up Milan’s research. This is an opportunity to hear from him directly. You’ll hear him explain his research and hopefully understand its implications. He also answers some of my critiques of his research. My challenge for you is to reflect on his ideas and consider whether it helps you better understand democracy. Let me know whether you agree or disagree with Milan Svolik. If you’re listening on Spotify, you can leave your thoughts with a comment on the episode. You can also send me an email to jkempf@democracyparadox.com. There is also a link to the complete transcript in the show notes.
The Democracy Paradox is made in partnership with the Kellogg Institute of the Keough School of Global Affairs at the University of Notre Dame. Make sure to mark your calendar for the upcoming Global Democracy Conference on May 19th and 20th. This year’s conference will take place at the University of Notre Dame. Check the link in the show notes to register today. But for now… here is my conversation with Milan Svolik …
Podcast Transcript
jmk: Milan, welcome to the Democracy Paradox.
Milan Svolik: Thank you.
jmk: Well, Milan, I’m so excited to have you here because I’ve mentioned you numerous times on this podcast, because there is a study that you’ve done, and actually you’ve done different versions of the study based on your publications, but it’s been really influential for me. I’d like to start out here and humor me to kind of dumb it down and describe it. It was this famous study where you test the revealed support for democracy. The study asks people about their support for democracy in the abstract, but then you create a scenario where they have to vote for a candidate and the one that they happen to support, their co-partisan, the one who reflects their ideological preferences, exhibits anti-democratic behavior.
The survey respondent has to make a decision between what they believe in terms of what they want policies to be and their preference for democracy. That’s the scenario that I’m describing. Tell me about the results from these studies that you’ve done.
Milan Svolik: There are many ways to view these studies, and you phrased it partly as a measurement problem, and it’s partly a measurement problem, but it’s also an issue about fundamentals of our reasoning about politics. So let me start with the measurement issue, which may sound technical, but I think it’s a very essential point and question that we have about our current moment, which is how much do people actually care about democracy. A conventional way of going about that question for a long time, and still in many instances, is to ask somebody how important is it for you to live in a country that’s governed democratically. Then the typical survey version continues. On a scale from one to ten, please give us your answer where one means not important at all and ten means absolute impact.
What you will see is that around the world, including in the United States, the vast majority of respondents will cluster around the answers 10, 9, 8, if you go almost anywhere in the advanced industrialized democratic world. That’s roughly what happens in the United States. The modal answer is either eight, nine, or ten and the average is somewhere around there. When we’ve looked at this around the world, the Swedes, for instance, give 10 out of 10 as the modal answer. There are a few places, for instance, Singapore stands out, where people, I think for reasons that are very interesting, give lower average reasons. The challenge there is, first of all, at the measurement level, can we believe this? Is there a credible way to measure how much somebody cares about democracy?
At the political substantive level, if somebody in the United States or in Hungary or in the many other places where significant fractions of electorates support candidates who show anywhere between, let’s call them illiberal, all the way to authoritarian tendencies. If they at the same time say democracy, 10 out of 10 for me, and then you vote for Orbán in Hungary, who is clearly attacking universities, dominating the press, making it very hard for anybody to challenge him politically, then I think the natural question is did they really mean it? Are we fundamentally misunderstanding something about their level of support? Are we mismeasuring something? What’s going on? I think at the measurement level, the answer certainly is yes. At least my proposition would be we are badly mismeasuring whether and how much people care about democracy.
So, the other approach that I’ve taken in a number of papers is to say instead of asking you about your support for democracy directly, let’s present a scenario that mirrors as much as possible the real world, and precisely the kind of setting in which your commitment to democracy is tested. Because presumably, if we ask about when somebody supports democracy, it’s because somewhere in their actions, usually political actions, that support for democracy will become relevant. In our contemporary era, when most democracies are threatened by candidates who propose institutional reforms, changes that in effect undermine democracy from within, that commitment to democracy is substantively politically tested.
When you as a voter face a choice, and now crucially, between a politician that you really like on substantive political grounds, but who also is proposing something undemocratic and a politician who is much further on substantive policies or partisan wise, but who is perfectly democratic… In other words, you have somebody in Hungary who likes the kind of policies that Orbán is for, but also sees that he has very anti-democratic illiberal tendencies. So they are faced in effect with a dilemma. Do I prioritize democracy over policy or do I prioritize policy over democracy?
My approach has been with a number of coauthors to basically say, and this is where you really find out how committed that individual is, and crucially, you don’t only find it on an abstract scale from one to ten, you can price out that commitment to democracy exactly in terms of that dilemma by the intensity of that trade off that you are faced with. If you anti-immigration in Hungary and you nonetheless are willing to go against Orbán, the most anti-immigration candidate that you can vote for, then you’re actually revealing through your choices that you care more about democracy than you care about immigration. So, in other words, it has real world relevance because this is precisely how a citizen’s commitment to democracy matters. And it has a politically meaningful interpretation. Now we know how much you care. You care so much that you’re willing to prioritize democracy over a policy you really care about.
jmk: Something I find fascinating about these studies is that it isolates the question enough that it takes away the context where somebody might justify their vote as defending democracy. For instance, Victor Orbán in his first election in 2010 when he’s running against the socialists, the socialists had discredited themselves by having this tape that had leaked just a few years earlier where they had badmouthed the people. They said that they were lying about everything all of the time and so part of the reason why Fidesz had such a massive win in 2010 was because their opposition party was effectively discredited. I can see how people might think of Orbán then as being more democratic, or rather the socialists being just as undemocratic, in that situation because you’re thinking about a comparison between the two.
But in your situation, you’re describing the situation that the candidate that they have to choose is the anti-democratic one. They literally are putting that into their decision. They’re having to make the decision between democracy or they’re revealed ideological preference or their inherent partisan preference and they’re having to make the choice between the two. Whereas, oftentimes, I see lots of people rationalize the decision as the other side’s just as anti-democratic. They did this. They did that. You don’t have that ability to rationalize that out when you have a survey question that provides limited information.
Milan Svolik: I would comment on maybe two aspects of what you said. The first one is that I have at least phrased a lot of these dilemmas in my previous answer as well. It’s often about positional issues where you are on economic issues, social issues, foreign policy, but I think in the Hungarian example that you outlined there are certainly positional differences between the Hungarian socialists and Fidesz. What you outlined is that the socialists in Hungary also failed on what you could call a valence dimension of politics. They basically admitted to be lying to voters and to deceiving them.
In other words, and in many instances, we have politicians who claim or whose main claim to electability is not because they’re closer to the electoral median or to where the majority is, but rather because they are more competent, because they are better on inflation or better on delivering economic growth or better on crime, better on many other issues that all voters would agree are a good thing.
So, I would basically say yes, and trade-offs can potentially occur across both dimensions, meaning sometimes voters are faced primarily with a trade off that’s spatial, positional in nature and sometimes the trade off is are you willing to take a more competent or less corrupt in the conventional sense of the word politician, but who’s also undemocratic? Or is it the other way around? For me, the prime example of this… I think there are many instances of this where this is a part of the political choice, but the prime example of this for me is Singapore. In Singapore you have a government that heavily restricts many, many civil liberties that is nonetheless elected and reelected repeatedly.
In spite of some challenges to measuring sincere preferences in Singapore, I am pretty confident that it’s fair to say that the majority of Singaporeans, if forced with a choice, would still go for the People’s Action Party. It’s partly because this is a party that has governed over one of the huge outliers in the world in terms of the level of economic growth, the level of GDP per capita, the degree of security that you have in that country, and crucially has also presided over a system that’s in principle undemocratic, that has strong authoritarian elements, even though they do have elections. This is an example of that. But now to the second part of your question, which is could it be that people rationalize things that are undemocratic as democratic?
I think here again, we have a very important measurement problem, which is that both academic and casual experience says if you ask somebody who just voted for a candidate that in our eyes is clearly undemocratic on some dimension – undermining civil liberties or electoral competitiveness, free speech – and if you asked him, so why did you vote for that politician? People will often say, no, there’s nothing undemocratic about what he or she’s doing. He’s way more democratic than the other side and here are the five reasons. The challenge there is that people are very good at giving self-serving answers for the choices they’ve already made.
I think people often are very good at pricing in whatever they find distasteful about the politician, including them being undemocratic into their decision and deciding that at the end of the day I still like this candidate or this party better than anyone else, and then offering justifications for why they did it. And so one way I try to confront this, and I think this is still an open research question, is before administering these experiments that you have described, ask people. The way I preface this question is I say countries around the world differ in how democratic they are so here is a sample of practices from around the world. Please tell us how democratic or not each of them is.
Then on purpose, I present people with statements that at the end of the day will correspond to something that the politician in the experiments that I faced them with later will actually propose. So, to give you an example, if in the experiment the candidate has suggested to prosecute journalists who accuse him of misconduct in office. Then I will at that early stage ask them, and in this country the prime minister has proposed in a debate that they prosecute a journalist who has recently accused him of corruption or something like that.
Because that question is framed as happening in a different country in a different context, I am able to check to what extent the respondent actually believes that that is undemocratic and how that same respondent 15 minutes later when faced with the choice between her favorite politician or whoever, also will say that statement about prosecuting journalists. If she votes for that politician, then we know that she knew that that statement was undemocratic because that’s what she earlier rated as undemocratic in a slightly abstract context. Then what she’s most likely doing is she’s trading off that aspect of the politician that she might not like for something else that that politician is proposing.
But I want to emphasize that this is still an important question. That is to what extent and what fractions of individuals are knowingly trading off democracy for other political priorities and what fraction genuinely have a misunderstanding of what democracy is. My empirical finding on this so far is that a majority of individuals are in the first camp. They have a fairly decent understanding of what counts as democratic versus not. So, when they’re making these choices, they’re in effect trading off democracy for competing political priorities.
jmk: So in reading your research and in terms of just talking to you right now, I mean, it’s very much the idea that people genuinely believe in democracy, but they’re making trade-offs, particularly in terms of partisanship, in terms of policy preference, ideological preferences. That those become more important than their support for democracy. What proportion of people do you think actually genuinely want anti-democratic actions by people that they support? That they see it as necessary to restrict the media, to limit opposition politicians from having a voice, from being able to run for office, because those journalists, those opposition politicians are dangerous. That by letting the opposition party have an opportunity to win, that it’s actually a bigger danger to the country and maybe even a bigger danger to quote unquote democracy to allow those candidates to compete than to take these actions.
I mean, that they actually want candidates to behave anti-democratic. What proportion of people do you think generally feel that way?
Milan Svolik: So based on these experiments. You’re facing two candidates. The one who you like better will be also described as undemocratic. So if we take the population of a country of which we have a random sample in these experiments, if we take that sample as an estimate of the population of that country, it is very hard to find a country… I’ve done this by now in about 20 countries around the world, including the United States. It is very hard to find a country in which our estimate of how much people care about democracy is either exactly zero or negative, meaning that people actually reward undemocratic behavior.
So, we are basically in between two worlds. People care about democracy, but they are willing to trade it off for other substantive political priorities, but they do care about it. Now crucially, this is the average, meaning this is the average within a society. The question now is, could it possibly be that there are pockets of authoritarianism? That maybe the majority values democracy, but there’s a small fraction that would go as close as to rewarding authoritarians. Even that is very hard to find. When I probe this through this experiment and try to slice those samples either by their substantive characteristics, so for instance, when I try to figure out what is it about an individual that we know about that best predicts whether in that trade off between democracy and the competing priority, they will side with democracy or not.
Some of the most predictive factors actually are, for instance, their knowledge of what counts as democratic. So those who more correctly or better pass this democracy knowledge test that I administer, they also tend to prioritize democracy in their choices. Another one is the so-called authoritarian personality. This is something from political psychology that’s often measured with the so-called child rearing battery. This is where I ask you, would you rather your child be curious or obedient? If you are on the obedient side as a parent, then you tend to have the so-called authoritarian personality. Indeed, individuals that exhibit this trait are more often willing to tolerate undemocratic behavior by candidates. But crucially, even for these individuals, they do care about democracy as revealed in their choices. It’s just that they care less than individuals on the other side of this kind of heterogeneity analysis.
Maybe going back to what we mentioned earlier, the question, how important is it for you to live in a country that’s governed democratically? I was critical about that question because I was effectively implying that it’s badly overstating how much people care about democracy. That 10 out of 10 may offer a very misleading impression. But crucially, the answers to that question do correlate with revealed preference commitment to democracy. So, in other words, 10 out of 10 does not mean that, yes, democracy is absolutely important for me, suggesting that you will always prioritize democracy over other considerations in your political choices. But it does mean that somebody who answers 10 out of 10 is more committed than somebody who answers two out of 10.
In fact, if you want to find individuals who don’t care much about democracy, it’s typically the individuals who are on these questions pick an answer that’s below the midpoint of the scale. That’s effectively their way of saying, I don’t care that much. But even they are at best indifferent, meaning they don’t reward authoritarian politicians. They just in these experiments behave as somebody who doesn’t care that much. So, to basically summarize my answer, it is very hard to find the slice of a sample of a population that actually rewards authoritarian tendencies in politicians.
jmk: What about overzealousness in favor of militant democracy. We see that right now in the United States on both sides of the aisle where people are supportive of gerrymandering when it’s in support of people that they believe in. There’s mixed results in terms of this. For instance, in my state, Indiana, we saw them vote against gerrymandering, but in California we saw it pass in a public referendum to gerrymander. Again, it was done in the context of militant democracy. They needed to defend democracy by gerrymandering their state to protect against gerrymandering efforts in other places. But both sides describe it the same way. They describe it as we need to protect our side because the other side’s so evil.
I don’t know that people are in favor of anti-democratic behavior for its own sake, but it seems that sometimes people are in favor of it for their side because they think the other side’s so evil that the only solution is anti-democratic solutions.
Milan Svolik: So one way of talking about commitment to democracy, especially in American politics, but partly in comparative politics, is the statement that people support democracy in the abstract, but when it comes to specific situations, they’re actually willing to tolerate undemocratic practices. I actually don’t think that statement is entirely correct. Because when it comes to, for instance, gerrymandering, if you ask people directly about one specific practice, gerrymandering, almost everyone says they hate it. For ordinary citizens in all kinds of surveys when you ask them, what do you think about gerrymandering, they’ll be very, very negative about it. So, we have a clear case where somebody says that gerrymandering as a very specific practice is something that I dislike as a voter.
Now we have situations throughout the country… and the United States is an outlier, because gerrymandering has been abandoned in almost any other democracy that has single member districts. People just don’t do this. It has been abandoned roughly in the way that California abandoned it, which is you delegate redistricting to a commission that’s either non-partisan or has some partisan balance, so it’s very hard to gerrymander on partisan grounds. The reason I made this remark is because that makes it clear that people are not fooled about gerrymandering because they will directly call it undesirable. So what are they doing? I think they are doing, in different terms, if you think of this as Texas versus California… They’re saying, I do dislike gerrymandering. But let’s think about it as an opportunity to obtain a partisan majority.
In other words, they’re faced with a tradeoff between a democratic commitment and gerrymandering as a means toward assuring some legislative majority or something like that. Throughout the United States, more Republicans than Democrats, but both parties, crucially, when elites have the opportunity to gerrymander go for this. You can even find statements by party activists or party elites saying, we’re doing this because it gives us an edge and the voters are willing to put up with it. Crucially, they understand that there might be a constraint on the voter side, but it’s not binding. Voters usually put up with this. So, in terms of the context of militant democracy, I think this presents an interesting normative dilemma. But also, I think it points to what ultimately is the mechanism that does enforce compliance with democratic principles.
That is when this kind of tit for tat escalates, the initial instigator comes to the conclusion that maybe this is not worth it. Because when you think about it at the normative level, what’s happening is, for instance, if you take the Texas versus California situation, is that Texas Republicans are knowingly disenfranchising in effect or underrepresenting Democrats in Texas. Somewhat interestingly, Democrats in California have decided that they’re going to underrepresent Republicans, not in Texas, but in California, in a national fight for congressional dominance.
So, both sides are engaging in a moral compromise. I would, at the end of the day, agree that somebody started it and that there is a certain normative justification to retaliatory measures. But from a Democratic perspective, this is a path of play that we would like to ideally avoid and maybe precisely because it’s happening, we’ll avoid it in the future. So, in some sense, this is the way I would offer a justification for what is happening and this is precisely what must happen so that in the future any party that would want to instigate this kind of fight understands that it’s ultimately not going to pay.
jmk: Well, part of the reason why I described it as overzealousness of militant democracy is to remove exact examples, because there’s going to be cases where we can parse out the exact situation and say it actually makes sense to adopt militant democracy to defend democracy. I mean, Germany does it in different ways. In the United States, the impeachment clause, you could argue is a form of militant democracy. You could think of militant democracy in lots of different ways that we find acceptable, but it can easily become brought to an extreme where it’s a justification for anti-democratic behavior.
I bring that up because in your papers there’s two of them specifically where I think this is very important, actually three of them that I’ve come across where you are identifying this phenomenon where people are trading off democratic sentiments for partisan preferences is exacerbated because of the state of polarization and the fear of the other party. That as polarization becomes more extreme and people find themselves further and further away from the opposition, from their political opponents, that it becomes easier and easier to justify anti-democratic behaviors or at least to ignore them or to support politicians that adopt them, because the idea of crossing the aisle is such a deep chasm that it becomes incomprehensible for the person. Can you talk a little bit more about how polarization affects this and how it deepens partisan behavior?
Milan Svolik: I think you outlined it very well, which is if we are in a world where individuals are willing to trade off democracy for competing priorities, then polarization becomes a systemic risk. It becomes a systemic risk because if we understand polarization as happening, and here, what I mean by polarization is if you think about a U-shaped distribution of voters rather than inverse U-shaped distribution of voters, then what we have is that a lot of individuals in terms of their substantive preferences are at the tails of this distribution and these are the individuals for whom the price of democracy in effect will be very high because in order to oppose a candidate from their own side who is undemocratic, they would have to move very far away from their preferred choice.
Trump versus Clinton in 2016 is a good example of two politicians who are very, very far from each other on any substantive ground. So, to ask the supporter of one to vote for the other, that’s a tall order. Polarization in effect means, societal political polarization in effect means, that the tails of that distribution are heavy and therefore the number of individuals who would rather put up with an undemocratic politician who endorses their policies rather than pay the price of voting for somebody who may be more democratic, but who is so far away in terms of policy is simply too high. So that’s what I mean by polarization being a systemic risk, because when polarization is high, we have a large fraction of the electorate who are simply willing to put up with somebody who is less democratic, but closer to them on policy.
This also relates to the other side of this logic, which is in a society that is consensual where a lot of voters are in the center, we in effect have a lot of voters who when faced with competitors who are spatially located around them, in effect, are not to ask to pay too much of a price to switch to the other side because at the end of the day, these voters are close to indifferent between the competing politicians.
When we have a bulk of voters like moderate centrists at the center, then we’re in a world where politicians will anticipate this kind of logic, will realize it doesn’t pay to do anything undemocratic, or even in a weaker sense simply unaccountable because my supporters are ready to abandon me because after all, it’s not too demanding to vote for my competitor. So that’s why both partisan and societal polarization, in effect, are a systemic threat to democracy.
jmk: So it sounds like two party systems would be much more dangerous than multi-party systems because in a multi-party system, you’re more likely to have somebody who agrees with you ideologically that would be less anti-democratic that you could shift your support towards. If you were on the right and you’re supporting a candidate, and you find that after he’s elected, that he’s acting very anti-democratic, such as maybe Victor Orbán in a more multi-party system. It’s going to be easier to find a candidate who is similar to Orbán that you could still support than in a two party system. Hungary is interesting because it was centralizing into a two party system up until around 2010 when the socialists fell apart. Now it’s got a variety of different parties.
It’s kind of centralizing into a two party system again. It feels like it’s a multi-party system, but it’s more like a hegemonic system where you’ve got Fidesz and then you have everybody else. So that’s kind of an outlier. But even in Israel, we see that Netanyahu continues to remain in the premiership despite exhibiting a lot of anti-democratic behaviors. I mean, how do we reconcile that? I would think that a presidential system would be more dangerous or not even a presidential system, but just a two party system’s more dangerous than a fragmented party system.
Milan Svolik: Yes, I would separate the two. So, whether it’s presidential or parliamentary, that’s one institutional axis of difference. Then whether it’s a system where maybe because of the features of the electoral system, there are strong pressures towards two party competition as opposed to multi-party competition. In principle, I don’t think there’s a reason why a two party system could not be one where that alternative is actually very close, meaning where the two parties converged toward the electoral median. It happens so that usually that convergence is limited and in the contemporary era, especially in American politics, it’s nowhere close to any kind of convergence.
So, your intuition that if we could infuse this system with a third party or fourth party and so on, which proportional systems allow because of the ease with which you can enter as a candidate or party and garner votes, I think the logic, if there is an alternative that is ideologically or policy-wise close to a party that has authoritarian tendencies, that the existence of that democratic, but ideologically proximate alternative is something that’s a pro-democratic force. I think one you mentioned Hungary. One interesting feature of the upcoming Hungarian election is that Orbán’s main competitor, Péter Magyar, is a conservative and a defector from Fidesz. He is not really more liberal than Orbán, but he says I will not be as corrupt or undemocratic as Fidesz and Orbán.
He’s basically trying to say you have got a former member of Fidesz but on this one dimension, corruption and democracy, I’m on the non-corrupt and the democratic side of things. So I think this is the advantage of a multi-party system, or better yet of the proportional system as an institutional setup. It allows challengers to emerge as a disciplining force.
jmk: But in the case of Israel, as I mentioned before, you have Netanyahu who’s exhibiting leadership with democratic backsliding tendencies and yet Israel is one of the most fragmented party systems and is also a parliamentary system. It’s not a presidential system either. I mean, we would’ve imagined that Israel would be a country that you would have plenty of options that if you saw Netanyahu behaving anti-democratically, that you’d find plenty of other options that could attract you. But that doesn’t seem to happen. Netanyahu’s been Prime Minister for an incredibly long period of time with very small windows where he’s actually lost power. Why is it that we’ve got an exception like that? I think if we think harder, we’ll probably come up with even more exceptions to that rule. It seems that people are still willing to vote for anti-democratic candidates, even within these multi-party systems.
Milan Svolik: You know, I don’t follow Israeli politics closely enough to give you a context specific answer. But one feature of the politicians who have been quite successful in, if not turning their countries in authoritarian directions in themselves, exhibiting undemocratic illiberal tendencies, is that they often are highly electable successful politicians. They’re good as politicians, right? I think one thing that’s hard to deny about anyone from Netanyahu to Orbán and Trump is that they are very good politicians. They’re very good at reading the electorate to have a huge electoral appeal that you could call anything from charisma to being able to sense where the median of the electorate is. That’s a separate axis. I think that is what’s notable about these politicians and that’s why I think a lot of challengers have a hard time mimicking them.
This is something separate from institutions in some sense. It says that when democracies are threatened from within, the mechanism itself is democratic, because to succeed in threatening, undermining democracy in that way, you have to get elected in the first place and if you are able to get elected in the first place, you’re not a mediocre politician. You’re somebody who can win. So, I think that’s the separate challenge there, which is that by virtue of already having won, these politicians in terms of their electability, are better than the average. It does represent a challenge that goes beyond institutions.
jmk: Now, you mentioned earlier about moderates really being central to protecting democracy because they’re most likely to cross lines because they don’t have as much holding them to one party over the other. Like in the United States, it would be that if you’re a moderate Democrat, there’s not a lot that would prevent you from voting for a Republican and meanwhile, a moderate Republican might find it easier to cross the line and vote for a Democrat again, particularly if the candidate who they’re supporting is also more of a moderate themselves. How has the change in terms of what it means to be on the right or the left, both in the United States, Europe, and throughout the world, really changed that because we see, for instance, in Europe, a lot of candidates on the right supporting welfare chauvinism, supporting higher degrees of social spending.
We see that Marine Le Pen over in France oftentimes attacking Macron almost from the left even though we think of her on the right. She’s attacking him for being too aggressive in terms of cutting social spending. It does feel that on a lot of issues, these candidates that we think of on the right and on the left, sometimes are not where we would expect. And so, I would imagine that it’d be easier for people if they have different ideological preferences to cross those lines, but it seems that there’s even greater partisan polarization, even though on issues, it feels like they’re starting to become more flexibility in terms of where candidates and where politicians can actually stand these days.
Milan Svolik: I think what you’re asking is something that I am just trying to make sense of myself. So, whatever I’m going to say is a speculation, but one way that I think about this is that the traditional alignment across economic and social issues that emerged after World War II and persisted for a very long time is increasingly being undone. What I mean by that is we have traditional, you could call them legacy, parties on the left or right. So, in Europe this would be the Social Democrats versus Christian Democrats or something like that where that legacy issue alignment was that if you are economically on the left, then you are also socially on what we would call being on the left or being liberal as opposed to conservative.
And if you are economically on the market side of things, then that’s where you should also be. They should be also conservative on social issues at the level of the voter. That doesn’t make sense to me. It doesn’t make sense because often the voters who need the state more, usually because they are poor or less educated and so on, also tend to have more conservative preferences rather than more liberal preferences on social issues. But this has been the traditional model of party competition.
It seems to me that those two issues are now increasingly becoming disconnected both at the level of the electorate and crucially, therefore, political entrepreneurs are able to propose and be viable with the platform that says, like the National Rally in France, and I think to a certain extent, both Orbán and PiS in Poland or possibly the Trump wing of the Republican Party, it will say, actually we’re social conservatives and that can take many forms, but one typically is we’re anti-immigration, for instance, or anti-globalization, but when it comes to economic issues, we are actually for supporting those who need the state for being pro-working class in the non-socialist version of that statement. So, PiS supports families with children economically, but only families with children and only native families with children.
This is a combination of the two dimensions. If you need economic support from the state, we’re all for it. But only if you live up to our conservative ideals and are a family with a lot of children, a lot of native children at the conceptual level. What I see happening here is that because the social and economic dimension is not as tightly aligned as it used to be after World War II, this opens economic space on both sides of that cross axis, but a lot of far right parties in Europe I think represent the one that combines social conservatism and especially anti-immigration attitudes with economic attitudes that would be more typical of being on the left rather than on the right. I really want to emphasize that this is more of a speculation rather than something that I would have any systematic evidence for.
jmk: Well, while we’re talking about the left and the right, particularly within Europe. Let me ask you about your paper “In Europe. Democracy Erodes from the Right.” It was co-authored with quite a few other authors, but it begs the question, why does democracy erode from the right in Europe? Why don’t we dive into that? Since we’re talking about the changing nature of politics within the US and Europe, why is it that Europe’s facing democratic challenges, specifically from the right rather than from both ends of the political spectrum?
Milan Svolik: So this is a paper that’s based on the kind of experiments that you mentioned earlier. That as we face people with a choice between two candidates, they describe the number of economic and social policies, but then at random experimentally, one of them is also assigned to say something undemocratic. Now let’s see how you as a voter are going to act when it’s the kind of politician that’s appealing to you on policy grounds that says something or does something undemocratic. Will you stick with them or will you be willing to effectively prioritize democracy over your policy considerations? We did this in seven European countries, anywhere from Sweden to Serbia.
So, one relationship we noticed in this data is if you try to count or quantify somehow the share of individuals who are willing to prioritize democracy over partisan interests, you subset it by party, there’s a clear pattern where the supporters of centrist or somewhat leftist parties, and this would also include green parties, for instance, are way more often willing to prioritize democracy over partisan considerations than supporters of parties to the right and especially of far right parties. This holds as much in Sweden as it holds in Germany, Poland, or Serbia. The fact that this is a general pattern across countries so different seems to indicate that there’s something going on here politically. For us it’s not so much as a finding as a puzzle that we then try to understand because we didn’t understand why that happened.
The first question you could ask is, could it be that supporters of far right parties put a much greater weight on their signature policy issues than supporters of leftist parties. Could it be that a supporter of the AFD in Germany cares so much about immigration, that it trumps any other consideration including democracy? Then it would be different from somebody who’s on the far left who cares about economic redistribution, but who is not as intensely committed to that policy issue to sacrifice democracy for it. But when we looked at this, the answer appears to be no. The left is just as committed to their policy priorities as the right is. So, it’s not the fact that you care more about your policy priorities.
So, then the second alternative is, could it be that individuals on the far political right care less about democracy. It’s not the policy side of the tradeoff. It’s just that when asked to trade off democracy for policy considerations, you don’t think you’re trading or sacrificing that much because you don’t care about democracy as much and that’s what appears to be happening. That’s what the data appears to be saying.
jmk: We can’t go back in time, but let’s just assume that we ran this same study during the Cold War in a place like Italy where its two main parties were a center right party and the Communist Party, do you think that it would’ve turned out differently where the left would have been okay with more antidemocratic behavior. I mean, this was a party that actually had ties to the Soviet Union, particularly when we’re talking about coming right out of World War II. The United States was very concerned about Italy falling into the Soviet orbit.
That would be a very different situation where democracy had the potential to erode from the left. Do you think that that would’ve been different? Or do you think it’s something about the ideological preferences of people who adopt ideas of the right, which would have very different implications for how we think about democracy?
Milan Svolik: I really can’t talk about the hypothetical of Italy during the post-war period. But the first place in which I attempted to do this kind of experiment was Venezuela. It was just after Chávez died and Maduro came to power. What I was going after was that we have a setting here where the situation is in some sense very simple. Electoral competition appears to be almost unidimensional, as in an economic left represented by Chavismo and at that time, Maduro. Then an economic right represented mostly by the opposition at the time. There was very little conflict in terms of social dimensions. It was the typical social dimensions of conflict in politics.
There it was the voters on the left who were willing to prioritize their economic interests over democracy in Venezuela, but crucially there at the time also, it was primarily the left that also in the real world was faced with that tradeoff. So, it was harder to study credibly for the political right in Venezuela and much easier to study for the political left in Venezuela. But another way to think about the question is what is it about these voters who vote for the far right?
And so, one way to address that question is to ask, could it be that our definitions of democracy, what we count as democratic, has leftist undertones or liberal undertones? Could it be maybe that, let’s say civil liberties as an aspect of democracy, which one way to think about civil liberties is that historically these liberties have been expanded over time and have been expanded primarily by activist parties and politicians coming from the political left, could it be that we have gone so far that those on the very far right don’t even recognize them as being part of democracy?
In the study that you mentioned, what we try to do is separate dimensions of democracy and these various positions, undemocratic positions, that the candidates in these experiments could adopt between those that attack electoral competition as opposed to those that attack constitutional checks and balances as opposed to those that attack civil liberties. Could it be that those on the very far right are just as committed to democracy as the left when it comes to electoral competition but show a very distinctive pattern when it comes to civil liberties? And the answer was no. So when it comes to democracy on every single dimension, those on the very far right were more willing to tolerate undemocratic behavior than their counterparts on the very far left and certainly than centrists.
That again, then begs the question, could it be that it’s something about the type of individual that votes for far right parties in Europe that makes them simply less committed to democracy in general. That is roughly where that analysis ends, because that’s what we could do with that data is that that seems to be the answer. I could offer speculations in the spirit of the kind of individual who votes for National Rally in France or Konfederacja in Poland, the very far right party in Poland. It’s a typical kind of individual that is usually somebody who is male and who holds very strong, not only anti-immigrant attitudes, but also simply illiberal attitudes. Very often those are attitudes that are not necessarily policy based, but that also, for instance, disagree with the many liberties that we associate with democratic competition.
jmk: I guess where my mind is going is some of the people voting on the far right are the former voters who used to vote for social democratic parties, at least in terms of their socioeconomic background. Britain is a great example of this where we have people who are Brexit voters today who’ve migrated to the Conservative party that used to be very strong Labour voters and Labour supporters. So, the question in the back of my mind is are these people that always had anti-democratic attitudes and have just migrated to parties that are on the far right or did those anti-democratic attitudes form because of their migration to the far right or even become exacerbated, because they’ve adopted a more illiberal party.
Maybe they were halfway there and now they’re a hundred percent there because of the type of party and the type of ideas that are reinforced from being on that side. I know that we don’t have clear answers from survey data. I know that the research isn’t there yet, but those are where some of the questions in the back of my mind are going in terms of whether or not the people have changed or whether or not it’s that the people have just changed who they support today.
Milan Svolik: I think you summarized it. We just don’t know. But I would maybe separate two issues. I would say from a democratic perspective, there is nothing illegitimate about attitudes that favor restrictive immigration or other conservative social policies. But it happens so that a lot of these parties that advocate for these policies are also illiberal in the undemocratic meaning of the word. Many other dimensions of social and political life and that transition from a former social democrat or labor supporter to these parties, I think implicitly reveals that you are the kind of individual who’s willing to put up with that. So it could be a sorting effect. It’s a speculation because I’m not aware of evidence that would clearly establish that.
jmk: So your earlier research touches on the politics of authoritarian role. I mean, that’s the title of your book from 2012. Your more recent research is very focused on the idea that democratic backsliding, democratic breakdown, today is different than it was in the past because rather than facing coups, consolidated democracies are in danger of backsliding that eventually leads to a breakdown. But it’s not a dramatic breakdown. It’s like it eventually crosses the line from democracy into autocracy.
But I want to bring these two lines of research together because in the politics of authoritarian rule, you argue effectively that modern authoritarianism doesn’t necessarily have different institutions than democracy. They might have legislatures. They might have elections and they might even have more than one political party, but that the politics itself becomes substantively different than in a democracy where those institutions work differently. When is it that we cross the line from one into the other? When is it that we cross the line in terms of democratic backsliding that politics moves from democratic politics into autocratic politics?
Milan Svolik: So I think to me, this is a question that’s relatively straightforward to answer at the conceptual level and very hard to answer at the measurement level, meaning at the conceptual level, the answer is when the opposition is no longer competing on an even playing field. But now how uneven? How we quantify and measure the unevenness of that playing field is a very tricky issue. So, I think for even many contemporary authoritarians… this is anywhere from the work of Levitsky and Way on competitive authoritarianism to the recent work by Guriev and Treisman on informational autocrats highlighting that if you have a new autocracy emerging today, it’s not going to have the institutional and often the repressive apparatus of what we’re used to during the Cold War. So, either single party communist regimes or reactionary military dictatorships, both of which were incredibly repressive and obviously undemocratic.
For me, this says democracy today, even in societies that ultimately for whatever reason become authoritarian, has strong enough appeal or at least there’s an absence of alternatives to it that those who want to govern nondemocratically must nonetheless pretend to be governing democratically. At the beginning of the 20th century neither the fascists nor the communists were pretending to be democratic in any way, and I think as we move 100 years forward what’s really striking is that it’s really hard to find a case for somebody advocating openly for some form of authoritarianism or dictatorship.
The modal scenario is somebody who is authoritarian claiming to be a democrat and I think that’s a very important difference in terms of how much legitimacy democracy has compared to its alternatives. This was very different a hundred years ago when the Marxist Communist alternative or the fascist Nazi alternative seemed to have popular appeal and in many quarters also ideological appeal.
jmk: So you mentioned Levitsky and Way. They had a paper that I think of as very controversial in Foreign Affairs recently where they said that the United States has crossed the line and that it’s now competitive authoritarian. It’s controversial because most Americans living in America today don’t feel like things have changed enough to become authoritarian. So the question I’ve got for you is not about the Levitsky and Way paper. It’s about whether there is a point when politics feels authoritarian and crosses the line where you go the way that we’re behaving in terms of politics no longer feels democratic, it now feels authoritarian, or is that not really relevant at all?
Milan Svolik: I think just like you mentioned or maybe as I mentioned in my previous answer this is a big measurement problem because when a transition from democracy to some form of authoritarianism in terms of its institutional forms implies minimal differences. That is you still keep a legislature. You keep elections. But what you’re doing is you are tilting the competition so much in favor of the incumbent that the opposition just doesn’t have a fair shot. Quantifying when that degree is large enough to call a country undemocratic I think becomes a challenge. It’s a challenge partly because even in perfectly democratic environments, incumbents do have an advantage. Usually, they’re more visible. They have greater resources available to them and crucially because they won in the first place, they also tend to be proven candidates.
There’s presumably some qualitative reason why they’re in office in the first place. So, in other words, what I’m trying to say is that some positive degree of incumbency advantage is in fact a feature of perfectly democratic politics. So the question becomes when that incumbency advantage takes on a level where we can no longer speak of fair electoral competition.
So at a categorical level I think I would disagree with the fact that the United States is now competitive authoritarian. I could go for a very long time enumerating the many undemocratic features especially at the level of the states of American politics. We mentioned gerrymandering. But I think the ultimate proof, and here I think I am certainly a student of Przeworski which is his alternation in power. It’s when that no longer happens, then we have a problem. Alternation in power seems to be happening still at the federal level. So I think the proof will come possibly as soon as this November or a test of that criteria.
jmk: We’ve covered a lot of topics, a lot of different papers that you’ve had and I’ve even mentioned your past book. What has the study of all of this taught you about, not just the definition of democracy, but the meaning and purpose of democracy itself?
Milan Svolik: I think maybe one point about democracy that I think these days is easy to forget when we speak of Orbán, Trump, Erdoğan. We often highlight these elites, these politicians. Sometimes we directly say Erdoğan has undermined democracy in Turkey, Trump is threatening American democracy and so on. We’re imputing so much agency to these politicians that I think it’s very easy to forget that these are politicians that came in as challengers and were chosen, elected, and in many cases repeatedly reelected by voters. So for me, the voter is the ultimate gatekeeper.
The voter is the one that decides whether she’s willing to put up with the politician who’s undemocratic or not. That centrality of the voter, I think is something that we really have to understand. When is it that that voter and under what circumstances can that voter be understood as a prodemocratic force? Under what circumstances is that voter a source of democratic vulnerability.
jmk: Well, Milans Svolik, thank you so much for joining me today. It’s been really an honor to speak to you. Thank you so much for joining me today. Thank you so much for contributing so much to democratic thought.
Milan Svolik: Thank you for having me.
Links
Learn more about Milan Svolik.
Learn more about his book The Politics of Authoritarian Rule (Cambridge University Press)
Learn more about the Kellogg Institute.
Register for the 2026 Global Democracy Conference at the University of Notre Dame.
Leave a Reply