Elite polarization may be helping young people sort out their politics but it could also make them misperceive how the other side lives. Delia Baldassarri finds that millennial and gen Z are now aligning their issue positions with partisanship as much as older generations. They are choosing their party more based on cultural and moral issues and, up to now, that’s been helping Democrats. But partisans assume that the other side has more stereotypical social groups and extreme views than they do. So polarization has confused us about how divided we are.
Guests: Delia Baldassarri, New York University
Study: “A Generational Shift”
Transcript
Matt Grossmann: Will young Americans become long-term Democrats? This week on The Science of Politics. For the Niskanen Center, I’m Matt Grossman. Elite polarization may be helping young people sort out their politics while also giving them misimpressions about how the other side lives. It used to take a while to come of age politically, but now young people see the party’s positions more clearly. That’s helping them align with the side they agree with, but it also may be making us look more socially and ideologically polarized than we are.
This week, I talked to Delia Baldassarri of New York University about her new Journal of Elections, Public Opinion and Parties article with Stuart Perrett, A Generational Shift. She finds that Millennial and Gen Z generations are now aligning their issue positions with partisanship as much as older generations. They are choosing their party more based on cultural and moral issues, and up to now that’s been helping Democrats. But her broader work finds a downside, partisans assume that the other side has more stereotypical social groups and extreme views than they do, and it’s harder than ever to get out of our bubbles. Elite polarization has meant issue alignment but confused us about how divided we are. I think you’ll enjoy our conversation.
So tell us about the main arguments and takeaways from your article on generational change and democratic voting?
Delia Baldassarri: Thank you. Thanks, Matt, for having me. So in general, in my work, I do study political beliefs and how they intersect with [inaudible 00:01:32] demographic traits and the social networks in which people are embedded in. And obviously these dynamics at the individual and group level have to do with what happens in the broader social context and political context of the nation. And what happened in the US over the past 40 years is a strong elite polarization. And there are two very distinctive aspect when it comes to political polarization of parties and Congress. One is the ideological alignment we are all familiar with, the fact that essentially parties have become much more consistent on most issues, but also another aspect has been that the number and type of issues have increased. So most of the politics till the ’80s was mainly about economic and civil right issues, but then this strong moral dimension was brought in and became very, very important.
And from my point of view, when I see economic, civil rights and moral issues bundled into distinctive parties, I ask how people fit in, how easy it is for them to line up with either party. And migration here is that for most people, it’s not necessarily easy to line up all their preferences because they have social demographic profiles or networks that actually don’t make it easy for them. And this is where the research on generational attitudes and shift comes in. In the sense that when you think about a group of people joining the political domain in the 80s, they’ll have just to figure out their position on economic issues, let’s say. But for younger generations, they do have to essentially respond to a package of policies that is much smoother.
So my question was essentially how has this characteristic polarization of elite impacted the younger generations and how do they actually construct their political belief systems in this context? And what we do see is that quite surprisingly in some ways, younger voters, when we join the political domain, when they start voting their late teenagers years or young 20s, they are actually much more consistent than most old generations when they enter politics. They actually bundle their political preferences in a quite coherent way and much more than in the past.
And the argument here is twofold is they came of age in terms of political beliefs in a moment in which the parties were much more distinct and so for them it was in some ways easy to learn how things go together. But also they were essentially pushed into a political arena that had essentially created different sets of incentives. And the millennials and Gen Z especially tend to be more progressive like any young generation is compared to the older generation. And their positions on liberal, moral and civil rights issues are not only predictive of their vote choice, but also they tend to offset conservative positions on economic issues in contrast to the way in which previous generations fought.
So essentially we have two dynamics happening with the younger generations that are unusual and might be considered as beneficial for the Democratic Party. One is that they are more progressive, which is not a new thing. But the second is that they actually tend to put more weight on non-economic issues. And in this domain, the Democratic Party obviously has an advantage.
Matt Grossmann: So let’s dig into each part of this. First, you said that you’re viewing polarization as kind of about alignment rather than about just the distance between the parties growing more liberal or conservative. So what in big picture happened on these dimensions aligning and what does that perspective bring versus seeing it as moving left and right?
Delia Baldassarri: So essentially people think of polarization as growing extremism as growing division along one or more issues. But I think that divisions per se is not a big issue for society. Conflict needs to be expressed. It’s good that actually there are arguments around issues and such a cohesion doesn’t depend on the absence of all these type of divisions. The type of polarization that threatens societies is the one in which these divides are actually consistent and divide people in opposite camps. Just to give you a very silly example, but a probably useful one, think about a society in which people divide by color and shape and you have 50% of yellows and 50% of green and 50% of squares and 50% of triangles. Now it makes a huge difference if yellows are also all squares and greens are also all triangles or if I cross-cut on the two dimensions.
And this is like the core intuition here. If political preferences or these social traits line up consistently, all Republicans are, for instance, rich, don’t want taxes, are religious, are anti-abortion, but if these traits don’t line up the same way, you have enough cross-cutting dynamics that can handle the fact that people are divided on single issues. That’s why I do think about polarization more in terms of alignment than in terms of division on single domains.
Matt Grossmann: So the other piece is the generational focus. So we have this elite level alignment. Why is it that younger voters are going to be most or differently affected by this? And does that change our perception of how ideological these younger voters are?
Delia Baldassarri: So one thing that surprised me about these younger voters is that they are very ideological, very organized in the way they think about politics the moment they join while what we see for all the previous generations is a process of learning. So they come in, they pick a party oftentimes, but they actually, it takes a bit of time to figure out how positions of different issues go together, make up their mind or just learn from the political discourse, how things go together. So they came in, the Gen X and some of the millennials very much knowing what goes [inaudible 00:09:16].
And then the other thing that was actually interesting and that’s why media also focused on them is that they actually joined in large numbers. We do know that the turnout among young voters is usually low and is still kind of low compared to other cohorts. But for this generation, at least for the Obama and Biden elections, they were turning out in large numbers. So the combination of their joining and the fact that they’re progressive and the fact that they already found their place in the political domain was a combination that was unusual. And in some ways we don’t know to what extent it will last.
Matt Grossmann: So generations are also a bit of a fiction with sort of lines based on birthdates. And originally with the baby boom, this was about what was the birth rate at the time were you born into an era when a lot of other people were born versus not a lot of other people were born in Generation X. So you look at traditionally defined generations, but you also look at political generations based on what was happening at the time people first started voting rather than what they were born. So tell us about how you’re thinking about generation in this and what the differences between thinking about it in terms of political versus demographic generations means.
Delia Baldassarri: Yeah, I think there wasn’t a lot of thought in this in the sense that we all know time is continuous, so we divide people in cohorts because it is useful and the way we break them down is somehow artificial. And obviously the most common approach is to think about generations as they have been classified by the way you described, like the silent generation, boomers, X and then millennials and Z. But at the same time, there was this subtle doubt, that actually we were building our own outcome because the way we classify these generations might actually in itself be in line with what is happening in politics. So the idea of just using cohorts the way demographers do by looking at decades was in order to check whether we would find the same pattern using a classification that cuts across a bit, the one that is most common and is neutral in terms of how we coded it and we find more or less the same results, which was reassuring.
Matt Grossmann: But I guess political scientists have found that it’s partly about who was popular when you first started voting and so it’s a little bit more micro, like if you first started voting in 1984, then you’re still more likely to be a Republican. So how does your story fit with that? I know it still allows for those kinds of factors, but is this just sort of a different type of that, that the people who are coming of age now instead of coming of age in a time when one party is dominant, they’re coming in an age of polarization and so they’re sort of affected by that.
Delia Baldassarri: This is a very good point. So there is one aspect about political socialization is important to highlight, which is that people make up their mind. There is a strong imprinting in how people think about politics throughout their life, but is actually done within printing in these late teenager years and early 20s. So it is true, but what’s happening around people in that age period of their life is strongly going to affect how they would think about politics and vote in future years.
And we say it in different ways. For instance, you mentioned the fact that if there was a Republican in charge, you’re more likely to continue that way or the other way around, but there are other things. So for instance, like boomers, as I mentioned, came of age in a moment in which economic issues were the most important thing and they still think only exclusively in terms of economic issues, but didn’t even line up with a moral issue type of divide.
For these more recent generations, I think that the most important element is the strong divide that they perceive from the elite. So they are in their own alignment, very, very divided, like whether they are on the left or the right, they are very consistent in doing that. So they picked up that cue from the political system. In terms of allegiances, I think there is an imprinting of a Democratic Party for those that we did study, but I don’t have data for the Trump people. So the people that in some ways were raised under Trump, so we will see what happens among them with future data if this is going to stick with them for the rest of their lives. For sure, the type of issues that were relevant is also something that would stick with them. And the last thing I’ll mention is in reading these generations, we really need to also factor in the role of turnout as I mentioned before and their general perception …
… and their general perception of politics. This is something that goes a bit beyond what the paper does, but these generations are also socialized in an environment where politics is not perceived as something positive in general. There is a lot of discourse about avoiding politics in order to survive in everyday life, in interactions. And somehow even when younger people really don’t want to interact with the out party is another way to avoid it, to avoid confrontation.
So these are traits that we will probably see stay with them. Dislike of politics, their level of trust in the institutions, including parties, obviously, is extremely low. So this would be something that leaves a sign on this group of people for the rest of their lives. And it shows also in their turnout rates which are going down again.
Matt Grossmann: So you have earlier work that had shown that people can reach partisanship without necessarily being constrained across issue areas. I know that there have been both kind of changes in the real world and changes in your research interest. So kind of walk us through from one to the other.
Delia Baldassarri: Yeah. So my first contribution to this debate was literally 20 years ago at this point with Andy Gelman. And the question then was simply we see that the parties and the elite is dividing. It’s dividing multiple issue domains, not only economic and civil rights, but also the moral issues. Do we observe the same alignment among regular people? And what we found was that… Like many other people in the debate, at the time, we found that very strong increase alignment in terms of issue domains and party identification, but we didn’t find alignment across domains.
So we did find that the consistent amount of Republican supporters, for instance, remained progressive on moral issues but were conservative and therefore voting a Republican Party for economic issues or the other way around. And that’s something that I still find and it’s a strong trait of American politics. That divide still exists in the population, and there is still a fair amount of people, even among these younger generation, that are cross-cut. They do pick a party, but they are cross-cut. They have preferences that pull them apart, and likely also have them shift party from one election to another or decide to sit out some elections. So this is one thing that comes from that work and is still kind of persistent.
Another thing that actually I added to my research using similar strategies, looking at how the relationship between political preferences and partisanship evolve over time was a study that did show how different moral issues are and change over time. So the classical story is that people have divided along partisan lines over time. And this is the case for classic economic and civil right issues. So essentially, we have a better sorting of Republicans and Democrats along partisan lines on these issues.
When it comes to moral issues, the major trend in all of them is a secularization trend where essentially what happens is that the population goes from being very conservative, thinking that women should work in the house, thinking that abortion is bad, thinking that you cannot smoke marijuana, or that the only way to manifest your sexuality is to be heterosexual, to actually being more accepting of women in the workplace or even encouraging it, being more accepting like gay, LGBT people and so on and so forth. And obviously, consumption of marijuana is a great example because it went from like 20% of people supporting it to 80% of people supporting it in 20 years. LGBT also is another great example of it. Really strong change in the broader population across partisan lines.
However, what’s happening in this moral domain is that essentially the Democrats, the people supporting the Democratic Party, are the people who pick up the more progressive views earlier. So at some moment in time, we see a big gap between Democrats and Republican on moral issues, similar to the gap we see in the economic and civil right domain. But that gap is actually a gap between two groups that are moving in the same direction, toward more progressive, not in opposite direction. And in fact, at some point, the Republicans catch up. And that’s why in the moral domain we keep seeing new issues coming in. Because in order to keep them… Like wedge issues. To keep them issues that actually divide the population, parties do need new topics. Because over a period of like 20, 30 years, people converge. So that made me think about that domain, not just an additional dimension, but as something that actually works in politics in a slightly different way.
Matt Grossmann: So that raises the issue of sort of, what are these alternative non-economic domains. You have treated them here, as most people do, like it’s another series of issues. It might be about abortion and gay rights instead of about taxes and spending, or it might be about race and gender instead of being about taxes and spending. But we have a lot of other constructs that sometimes people think might matter for vote choice and partisanship, things like core values and ideologies. And the point that you just made that on issue positions, these issue positions are public opinions moving leftward over time across these issue positions. And yet, the sort of people on each side of these cultural debates don’t seem to be matching that shift. They’re still people who view themselves as culturally conservative regardless of if the issues are changing.
So do you think about this, should we think about this is about policy issues or about some other construct that’s now kind of driving people to different partisanship or consistent voting than we had in the past?
Delia Baldassarri: I think this is a fair question. And indeed I believe that in general there was a shift in terms of how politics is played by elite parties and so on from a discussion and engagement with issues and eventually also actual real social problems to a more sort of symbolic conversation where people throw labels at each other, like conservative or woke, rather than getting into the details of what discriminate between them. And that’s an element that is real, that is probably traceable in how candidates talk about politics.
It is also mapping in some ways into what political scientists have been in love with in the past 20 years, which is all these literature on affective polarization, which has left issues in the corner and has started to really deal with emotions, like… Affective polarization in some sense is essentially how people feel about the opposite party. And I think there is something true to this, but it is too easy to think that that’s the full story.
I believe that when candidates bring issues back, they win, if they can relate to their constituents. So I believe that this shortcut of bypassing content and playing with labels, symbolism or emotions, negative emotions particularly, is a great strategy that parties, both parties, are using. But I wouldn’t say that people have changed. So I would say is another example of how changes in American politics are top-down. Elite polarization came before any form of public opinion polarization. And this change in the way we talk about politics is not something that comes from the bottom, from how people articulate their preferences, it is something that is imposed from the top. So I’m with you that that’s a possible explanation. But if fed the right content, people will react to it.
Matt Grossmann: So I’m not saying that these views are issueless labels. I guess I’m just interested in… Some people might have the correct perception that one of the parties is a culturally liberal party and one is a culturally conservative party. And that might mean that even if their opinions on gay rights and abortion change their view, that they are more affiliated with the culturally conservative party or that change is coming too fast or something along those lines might still put them on the right side of that.
Which leads into the next question, which is your story seems like it should be pessimistic for the Republican Party. There’s been increasing emphasis on social issues that might have backfired with younger generations. If anything, public opinion is moving leftward on these issues. There’s imprinting from this time forward. So all of that suggests that this should be beneficial to the Democratic Party. And yet, shortest term trends in the 2024 election weren’t in favor of Democrats, and there might be signs that this kind of won’t continue to benefit the Democrats even if the polarization is happening on these cultural issues. So what would you say in response to that?
Delia Baldassarri: I think that you are right that definitely the possible alignment of positive conditions for the Democratic Party, namely the fact that the younger generation was obviously more progressive on a set of issues, and that this generation was also paying more attention to non-economic issues is not what at least was picked up in the media in later elections.
And I think it is fair to say that that combination has not worked in favor of Democratic Party. And I think the major reason there is turnout, meaning the parties also need to activate these predispositions. The fact that actually the younger generations are more progressive, the fact that the younger generations are more diverse, so less white, is still there. They’re also more educated on average. All these preconditions are there, but you need to activate them.
And clearly, in different ways, Obama and Biden were able to tap into this potential. The Republican success in 2024 was actually able to tap into a different part of this community that was, of course, surprising to the media. And that’s why we are hearing all sorts of accounts about these young men, mostly white, who are embracing the Republican creed and turning out and even join churches as they say.
I think it was a thing that is real that happened, but it has to do with the capacity of mobilizing them. And even if you look at the data more concretely, take the example of young men, white men joining churches, what’s happening in general is that younger generations are showing up less in church than previous ones, this is happening, but the trend is different for men and women. So women are really not showing up, and men are showing up less. But compared to women, they are still going to church or rejoining it, whatever. So essentially, it’s the way you read it.
The underlying preconditions in terms of social demographics and their combination with party are still in favor with the Democrats. Whether you can activate that, it’s a completely different story.
Matt Grossmann: You also mentioned that younger generations are coming of age at a negative political time. And we also know that in aggregate, the number of people who say that they’re independents, when first asked if they’re Democrat or Republican, is going up, especially among the young. So is this younger generation becoming Democrats or are they just becoming anti-Republican? And what does that mean for their long-term likelihood of staying Democratic, especially if they have joined a party that they still disagree with on some issues?
Delia Baldassarri: Yeah. I do believe, first of all, that it is true that most people have become more negative about politics. And part of this effective polarization we are observing is really capturing greater disapproval or hate toward the out party. And this is clearly visible among younger generations-
And this is clearly visible among younger generations as well. Another thing that we are seeing but rarely consider is that the fastest-growing party identification in the US is independents. Now, there is a long story about interpreting people who identify as independents. One thing is independents are just people who don’t care about politics. They have no clue of what’s happening and oftentimes don’t even vote. Another is they are just close to partisans. They find it more acceptable, especially in a context in which politics is not a good thing to identify as independents instead of partisans, but they act like regular partisans.
But the other thing is that there is growing evidence and some of my work points in this direction, but independents is becoming really a label that identifies people that are cross-pressured or moderate or really don’t fit in in the party context that they are experiencing. And this makes a lot of sense. If you go back to the original observation I made, which is parties have become very divided on a large set of issues and taking very extreme positions. But people are in much difficult situation in terms of fitting in and where social demographics is one of the aspects that makes it hard to fit in.
So in this sense, people coming of age, younger Americans will also have to deal with a very difficult economic set of opportunities like lack of jobs or jobs that won’t give them the same type of economic stability that their parents have enjoyed, might actually not only come to these politics, but become also structurally more marginal in society. So these things could actually determine a lot of how they relate to politics. Whether in favor of one party or another, I don’t know.
Being European, being very familiar with what happens in Europe in multiparty systems, you have the possibility, lower cost for new parties to enter. And in those contexts, we have seen a lot of populist parties coming along. So that’s a classical way in which this combination of unhappiness with a party system and social marginality because you don’t find jobs or you’re losing with respect to your parents, convey support. We don’t have it. I don’t foresee a third party, a populist party, but one of the party might actually take on a bit more of that flavor in some historical moments, and I think Republicans have done it pretty well.
So there are ways in which this new generation might actually be attracted to the Republican Party, but we have not really fully seen what are possible, or that the Democrats become more populists and they attract them. So the opportunities are there for both parties. It’s unclear what they will do. I think they will probably continue to ignore the younger generations because they in general don’t show up.
Matt Grossmann: So you do observe and study politics comparatively as well. So that might bring some perspective on what is unique and similar about changes in the US. We do have multiparty systems elsewhere, but if you look generally on parties of the left and right and group them together, you sometimes see some pretty similar trends with gender, with education, with changes in the role of income in dividing the left and right that we have seen in the US and in the rise of social and cultural issues relative to economic issues. So what parts of this are just global changes that are just filtered through our two-party system versus unique circumstances having to do with our current levels of polarization?
Delia Baldassarri: Yeah, this is a very interesting question. I usually try to make a general point first before making any comparison between Western Europe and North America, which is that in most Western European countries, what we observed starting from the ’80s, let’s say, or even before, is that the dominant parties, traditional parties have moved toward the center. Whether it’s the Social Democratic Party and the Catholic Party or some version of it, we have seen a move toward the center, and that created space for populist parties to come along, both sides. In the US, the elite changes have been orthogonal topics. The elite changes have been parties have moved farther away. And also obviously as I mentioned before, there is no real opportunity for new parties to form.
With respect to general changes, there are several things that are common across the two worlds, and it’s because they were probably a response to broader forces, industrialization and globalization. And I just mention a couple. One, you brought it up, is gender. The gender divide, it’s always talked about, but it’s actually important to notice that til the ’60s, women were tendentially more conservative than men by a good margin, but this actually changed, changed quite quickly. And in the US, like ’60, ’70s, they reached party and then women became more progressive, more inclined to vote for the Democratic Party, and they also started to vote more and they are actually voting more than men these days.
And one thing that really seems to explain it is that in that period, women joined the workforce at much higher rates than before, and this is probably a story of emancipation. And we saw something similar happening in Europe, although in different historical moment a bit later, exactly because women were incorporating in the workforce at the same time.
Another trend that is actually common is very about education, and you’re an expert in this with your work on the diploma divide, and it is that education, traditionally highly educated people have been more in favor of conservative parties, while in more recent years, we see them supporting more progressive parties. And your explanation, which is that they actually give more weight to social considerations, moral issues, is definitely something that is happening across… I think that these trends need to be combined with something else, which is that income is still pulling people in a different direction, but Europe has seen something very similar. Education is, in Europe, one of the most predictive traits in terms of voting.
I would actually say a couple of other things. When it comes to the US, the most dividing cleavages are race, and this has always been since Second World War at least, religiosity. Religiosity had an alignment in the ’60s, ’70s where the most traditional denominations became more clearly politicized. But what actually is keeping religiosity high in terms of predicting partisanship these days, starting from the ’80s and very strongly in the ’90s, is the raise of nons, nons in terms of people who are non-affiliated, they don’t identify with a denomination, they don’t go to church, they define themselves as agnostics or apolitical. And they are essentially a growing component and they are disproportionately more likely to vote Democratic.
The interesting thing is that when you look at the combination of these cleavages, you see a lot of cross-cutting pressures. So when it comes to race and religiosity, for instance, you do see that being religious among whites makes them more Republican, but being religious, going to church, among Blacks makes them strongly more likely to support the Democratic Party. And religion does not do much in better, like racial categories, whereas the nons, the people who don’t go to church, don’t have any strong difference across racial categories.
So the point here is just to point out that we think of these cleavages as individually exerting a lot of pressure, but for many people, remember, people are not single traits. People are a bunch of stuff, they’re racial or poor, they’re men and women, they are Black or white, and oftentimes these things don’t align very nicely. And when it comes to the most predictive cleavages, race, religion, and education, what we find is that they also become more disaligned, meaning they have created greater pressure.
And another great example of this is education and income. Education and income are always positively correlated in almost all societies. However, in this historical moment, they are pressing people in opposite direction. If you’re rich and highly educated, your education brings you toward the Democratic Party, but your income still makes you support Republican Party. So this idea that cleavages need to be fought in combination is extremely important.
Matt Grossmann: So let’s connect that to the generational story because you do have this broader work on all of these different social cleavages that sometimes point in different directions. On the other hand, and you might see generation as just yet another, age is yet another characteristic that might divide people and might have cross-cutting. Cleavages, on the other hand, the younger generation is, as you mentioned, more racially diverse, less religious and more educated. So all of that would tend to actually maybe further align if they end up aligned with the Democrats to be extending and aligning these cleavages. So how do you see generation affecting this broader story of social cleavages?
Delia Baldassarri: Yeah. I mean, historically it’s funny, you never have generational effect that really become cleavages for the basic reason that people age. So in some ways you move out from that category in a way in which you don’t move out from your gender in [inaudible 00:41:46] or other things you can’t really change, race or just decide that you are a different class. So in some ways, generation in itself, the cohort might not do much, but you are right that we might be observing a generation that has a bunch of characteristics that should align with the Democratic Party. It is actually unclear how this can be exploited.
Another aspect of my research is quite, I think, telling here, which is that we think about the relationship between social divisions and politics in terms of cleavages, in terms of single traits or their alignment or disalignment and how they impact politics. But more and more, we might be moving toward a society in which your partisanship is determined by the specific intersection of characteristics you have. And these are very specific. So think about, let’s say, a highly educated woman who lives in suburban areas and is married. It’s hard to predict her partisanship unless you also know whether she goes to church and has a job or not.
And this might actually also be the case for younger cohorts, like how the combination of traits defines you might become much more individualized and specific. It might still map into politics in meaningful ways, but it’s actually less of a situation in which one trait pulls or pull you in a party direction and more of a specific identification. And this is interesting to think about because if you think about alignment and cleavages, you can create the conditions for two groups to divide.
But if you think about idiosyncratic combinations, as I mentioned before, you do actually think about this society that is fragmented because what makes you have stuff in common with some people might actually be a trait that is not in common with other people that might have the same interest. So the thing here is that from an expression of interest point of view, these increase intersectionality of, and here I use intersectionality only in terms of the combination of traits.
And if you think about a lot of evidence we have right now, like look at your book. You’re saying essentially education matters, but mostly for whites. We add other evidence in the religious domain, as I mentioned, like education, race and religiosity together do things, but separately it’s very hard to tell what they would do. So we do have an understanding of voting that is actually moving more toward this combination of traits. Also, if you look at the research on class or occupation and voting, like-
… on class or occupation and voting, you need to throw in race, you oftentimes need to throw in region, which has also to do with the industrialization process.
So, these actually suggest that we might actually start thinking about these processes in a slightly different way and model them less in terms of single variables because variables don’t vote. It’s likely combination of them that actually gets people to act one way or another.
Matt Grossmann: And one of the explanations for these differences is often that different sets of traits are associated with different social relationships and networks, maybe at church, maybe in schools, maybe in occupations. And you have some newer work that looks at the actual relationship between people’s perceptions of polarization and who’s in what party and how that affects what actions they actually take, who they become friends with, who they seek to avoid. So, tell us how well that fits this conventional story that polarization is also affecting our social relationships in a way that is going to reinforce them.
Delia Baldassarri: My starting point here is that, essentially, most people go about their life living politics out as much as possible in the background. And they do relate to each other on the basis of a variety of commonalities that might have little to do with politics.
So, when we look at social networks and how politically homogeneous they are, we always find some level of political homogeneity in networks. But oftentimes we tend to explain it in terms of the consequence of homogeneity along other lines. So, you are high income, you come from a high income family, you go to a certain type of clubs, you hang out with people who are from the same background and you might have the same political views. So, you’re similar, politically homogeneous, in your networks, but not because you chose each other on the basis of politics but because other social demographic preconditions have brought you to interact with these type of people.
And if a situation is like this, we never really end up isolating ourselves in an echo chamber in groups that are exclusively Democrat or exclusively Republicans because society is complex enough these days that you’ll always have ties with people that come from different backgrounds and have different political views. So, this is my premise.
Now, there is a lot of research in political science that is pointing strongly to the fact that people choose their friends, romantic partners, their economic partners on the basis of politics, that partisanship really drives decisions.
And I want to call attention to how these studies are conducted because it’s extremely important. What they usually do, in a survey setting or experimental setting, is to give very little information about other potential partners. But in this information, you have maybe the gender, age, race, and some traits that are relevant to the task, but then you have partisanship. And when you are given partisanship right away, given this very divided historical moment we are in, you infer almost everything about that person based on that little label, Republican or Democrat.
My question is like, people don’t do this in real life much, meaning they don’t go around with wearing partisanship on their sleeps. So, the experiments are very good at telling us what people would do when given limited information about somebody else and have to choose. And I do believe that partisanship drives the choice. However, if you think about how people interact in real life, in many cases, as I said, they don’t wear politics on their sleeves. They relate to each other. They find acquaintances through PTAs, leisure activities, their workplace.
So, my research in this domain is trying to show two things. One, that our actual real world networks are still quite heterogeneous, especially when it comes to acquaintances. And second, that this is due to the fact that in everyday life, people don’t start out with politics. And that by not doing it, they find the commonalities that they actually have and they also realize, in doing this, when they end up interacting with out-partisans, that out-partisans are not monsters. The one thing that is happening in surveys is that, when you elicit the label of Republican and Democrat, you are really thinking about the worst possible out-party person.
And the logic I bring to this is to say, don’t wear politics on your sleeves because the beauty of democratic regimes is that there are several parts of life, your private life, your civic life, your work life, where politics doesn’t need to be present. And that’s the beauty of democratic regimes compared to other types of regimes where politics tries to control everything. And you need to actually show your political allegiance in order to get jobs, in order to be credible, in order to have a role in life.
Matt Grossmann: So, on the one hand, you’re pushing back against some notions of extreme polarization. That is, you’re saying our social networks are more diverse than we think. There’s more people with some cross-cutting identities than we think. There’s still people with cross-cutting issue positions. And yet, a lot of this research, including yours, is showing a directional trend that’s still pretty consistent with polarization, right? We are increasingly aligned socially, aligned with demographic categories, although those categories are changing and especially across these issue positions that used to be separate and especially among the generations that are joining us in the political system.
So, it does seem like the long-term trajectory is still consistent with the story, even if the story is overstated now. Is that how you see it?
Delia Baldassarri: I would actually think of it as slightly different in the sense that I see the process as being a top-down process where actually people don’t fit in and actually do have to handle tensions in their everyday life, whether it’s relational or in terms of political representation, in order to still be participating in politics. So, the trend is not natural in the sense that it won’t occur just because of demographics or social networks or how people relate to each other.
Indeed, I do believe that if we model society without the two party system and the system of incentives that is leading to extreme polarization [inaudible 00:52:45], we would have a very different outcome because the level of social mix, the level of social differentiation, even economic heterogeneity within different social groups, are all increasing.
So, what I’m trying to point at is, we are going that direction of increased polarization because there is a constant push from the top. And it’s parties, but there’s also some aspects of economic system and mostly the media environment that are actually constantly reproducing this narrative of a divided country and there is very little that people can do in order to push back on this. And there are plenty of evidence that when there is no top-down pressure, people just don’t go along with that narrative because it’s not the dominant narrative they would adopt in a context in which it wasn’t constantly fed to them on a daily basis.
Matt Grossmann: So, I mentioned that the short-term trend in 2024 wasn’t wholly consistent with what you found. We did see some Republican movement among the youngest cohorts and not just among white voters or men. So, what should we be looking for in 2026 and 2028 to see whether this pattern is likely to continue or how it might be changing?
Delia Baldassarri: As I mentioned before, I think that the main determinant here would be turnout. If you have a candidate, likely on the Democratic side, that allows people to sort of identify with it…
I’m not actually thinking about identification here in the simple political identity way. Yes, it’s true that younger people identified in young Obama and young New Yorkers mobilized around Mamdani, but actually you also had the turnout for Biden, for Grandpa Biden, was huge, much higher than for Obama among the younger generation. So, I don’t think it’s exclusively some sort of identification dynamic or gender or whatever. But that is going to be relevant to see whether that group of young voters would actually show up for the Democratic Party.
I also think that there would be this sub-component of white men supporting the Republican candidate. But that’s a bit, again, dependent on the figure who is the candidate. Hegseth might do it, Vance might not. Trump does.
So, it’s hard to tell, but the demographics are in favor of the Democrats in that subgroup. Whether they leverage it or decide to go for it’s a different story.
Matt Grossmann: And anything we didn’t get to that you wanted to include or tout about what is next?
Delia Baldassarri: One thing I want to mention because it goes along with that idea I mentioned before, essentially in democratic regimes, you want to keep politics out from many spheres, is that I’m moving toward doing more research on how politics is entering private life, social life, economic life, the dynamics of the office. But again, from this perspective that, if they could, people would leave it out.
And we have a lot of examples of how churches or PTAs or the Girl Scouts have been hijacked by external forces that turned them into political battlegrounds. And I think that understanding these dynamics and giving these individuals and organizations some hints about how to defend themselves… And obviously, as you can imagine, the business world would be very receptive of this, how to protect themselves from politicization is an important step.
And I honestly think that after some effects on public opinion and the media representation of the nation, this sort of penetration in spaces that were generally non-political is the next step and it’s obviously already happening in front of our eyes. So, I’m going that direction.
Matt Grossmann: There’s a lot more to learn. The Science of Politics is available biweekly from the Niskanen Center, and I’m your host, Matt Grossman.
If you liked this discussion, here are the episodes you should check out next all linked on our website: “How Marriage and Inequality Reinforce Partisan Polarization”; “How Political Values and Social Influence Drive Polarization”; “Is Demographic and Geographic Polarization Overstated?”; “Are Claims that Social Media Polarizes Us Overblown?”; and “How Racial Realignment Ignited the Culture War.”
Thanks to Delia Baldassarri for joining me. Please check out “A Generational Shift” and then listen in next time.
