Voters now associate brands and lifestyle choices with each partisan side, meaning candidates can use ostensibly non-political signals like attending NASCAR and eating Chick-fil-a to show their loyalties. When do these candidate symbols translate into ideological reputations? Can candidates use the other parties’ symbols to attract their voters? Levi Allen finds that voters can intuit partisanship from symbols like cowboy boots and a truck but rely more on partisan and policy information when it is available. Yet a Republican candidate drinking Starbucks and driving a Prius can cost him with Republicans without helping him with Democrats.
Guest: Levi Allen, Indiana State University
Study: “Non-political Identity Signals as Information Shortcuts”
Transcript
Matt Grossmann: Do boots in a truck make candidates seem Republican? This week on the Science of Politics. For the Niskanen Center, I’m Matt Grossmann.
Candidates often use ostensibly non-political signals to show their loyalties, and voters now associate brands and lifestyle choices with each partisan side. When do candidate images translate into partisan and ideological reputations? Can Democrats attending NASCAR and eating Chick-fil-A help them relate to Republican voters? This week, I talked to Levi Allen of Indiana State University about his co-authored Public Opinion quarterly paper, Non-Political Identity Signals as Information Shortcuts and Related Research. He finds that voters can intuit partisanship from symbols like cowboy boots in a truck, but rely more on partisan and policy information when it is available. But there are costs, a Republican candidate drinking Starbucks and driving a Prius can cost him with Republicans without helping him with Democrats. We talk about these associations in the context of Allen’s other work on rural, racial, and evangelical identities. He studies those that don’t fit the partisan mold alongside the stereotypical candidates and voters, and he sees cultural tendencies as a source of cross pressure in our polarized age. I think you’ll enjoy our conversation.
So tell us about the findings from your recent article on non-political identity signals. What’d you find?
Levi Allen: Yeah, thanks for the question, Matt. So one of the things that I got really interested in graduate school was the idea that things that are ostensibly not political might be used in the political arena for political gain. And this was a lot of what I was seeing, sort of the contrast between being at Notre Dame and graduate school back with my rural community back home and how things were political that I didn’t think were political, but they did.
So we did three studies on these non-political identity signals. And the first one is a conjoint experiment where you’re faced with these candidate profiles and you make a choice, respondents make a choice. And we had something like 10 different traits and then a lot of attributes within those traits. And it sort of came out to a wash. We found those non-political identity signals, things like eating at a Chick-fil-A or doing yoga didn’t really matter.
And so we went to the other extreme and did a candidate vignette experiment where folks got these hypothetical descriptions of candidates. And what we found there is a couple of important things, which is, one, folks are really attuned to these cues and they seem to be making decisions based off of them. And the second finding that I think was really interesting, and we can talk about this throughout, is we gave folks a candidate who had sort of cross-cutting cues, these sort of what happens when they’re a Republican, but they do these democratic air quotes things like drink Starbucks and so on. And we found that that washed out, that was a null result. And that was really interesting to us. And the reviewers really pushed us to think deeply about what was going on there and that was really helpful.
And so then in the third study, we went back to the conjoint format, but we added in something that had mattered in the first study, which was policy information. And what we found is when we included policy information in the conjoint, that pretty much gobbled up any movement on the non-political identity signal. So there’s, I think, a normative good perhaps in a democratic sense that folks are using policy positions to make decisions. But then on the back end, if they don’t have those, they are sort of keying into wearing cowboy boots or going to the art festival.
Matt Grossmann: So we’ve got one experiment where there were lots of traits and nothing really stood out. We got another one where we were able to show that people can interpret these signals and decide based on them sometimes. And then a third one where we were able to eliminate it again. So walk us through the logic of each of those and kind of what remains unknown.
Levi Allen: Yeah. So one of the things we suspected may have happened in study one, and for those who view the paper you’ll see the conjoint results take up the entire page, it’s just a figure of things ranging from religious identification to party ID, of course, to occupation, to gender, to experience. We wanted to simulate what a campaign environment looks like, but we may have made things a little too noisy. And we were honest about that in the paper that we think that this is a pretty steep task to ask folks to parse all this information out.
And so then we wanted to strip it down and change formats and we went to the vignette approach and that really allowed us to causally isolate what’s going on here. And that’s where we see the movement because folks are only given partisanship in some instances, and then in other times these identity signals are both. So there’s no policy information, there’s nothing like that. And again, what really stood out to us is a candidate could say they’re a Republican or they could say, “I do these Republican coded things,” and it was pretty equal support among Republicans and the penalty exacted by Democrats.
And then in the third experiment, we wanted to go back to the conjoint because we were interested to see if policy information mattered, and it did. It was able to, when we had the hypothetical candidates take a position on the pipeline, a sort of hypothetical pro-pipeline stance, as well as the environment, I believe was the other issue. We saw movement there.
And one of the things we thought was important is not that folks don’t care about the environment, we know climate change is a very salient political issue, particularly I think on the political left. But we chose something that we felt like could engender bipartisan support potentially on a pipeline, the Keystone pipeline was a huge political issue in the Obama years. We didn’t choose something like abortion or guns, these really salient social issues, but even then we see movement. I think that’s a broader question of, is there an issue out there we could choose that folks know about that isn’t polarized or isn’t politicized, I think is pretty difficult.
Matt Grossmann: So, let’s back up and walk through some of these signals that you tested because you did try to figure out which ones would work and you have things like the pickup truck and Starbucks, Chick-fil-A following NASCAR. So walk us through your rationale and where’d you get these things from and which ones did it seem like had the most potential purchase that led you down the experiments that you did?
Levi Allen: Yeah, we wanted to take a systematic approach and ask folks, do some pretesting work on are these things actually politicized or is it just us as political nerds that see Chick-fil-A and Starbucks as politicized? One thing that was interesting in the pretest data we report in the appendix is that restaurants were sort of awash. We found a slight Republican lean for Chick-fil-A and Kentucky Fried Chicken. We found a slight democratic lean to Starbucks and somewhat surprisingly to us Popeye’s, but we think race could be playing a role there too where folks are inferring things about the perceived demographic that goes to those restaurants.
Cars and hobbies though were much more definitive. We found that a Ford F-150, a Chevy Silverado were by far viewed as Republican, whereas a Toyota Prius and a Tesla, although this was before Elon Musk sort of wading into politics. So we would expect that’s probably changed quite a bit there. And then on the hobbies, things like NASCAR being coded as Republican and then being a fan of the NBA being coded as Democratic. And so that was one approach.
We also sort of, as you know, looked at what do political elites say? What are the lines that they’re drawing? And we start the paper out with a quote from John Kennedy, the senator from Louisiana, where he invokes, I think it’s goat milk drinking, latte, avocado toast eating elites. And so we also sort of leaned on what did the elites say there?
And then finally, there’s a really great paper called Cued by Culture by Neuner, Soroka and Hiaeshutter-Rice, I hope I’m saying their names right, in political behavior where they really test a lot of these things. And I always show that to my students because they’re amazed to see that a tattoo, for example, is very democratically coded. And so that was how we tried to triangulate signals that would matter and that the public views as political.
Matt Grossmann: And were people asked to just talk about the stereotypes of these or estimate the proportion of actual people who were Democrats who did these things? I mean, what are we trying to get at here? Is this just like there is a symbolic attachment or are people trying to actually figure out who uses these products and stereotype from them, or what?
Levi Allen: Yeah, we asked folks, we would give them the name of the restaurant or the vehicle and we would ask them, do you associate this with the Republican Party, the Democratic Party, or neither? Now what’s interesting, and again, this is available in the appendix of the paper is we still have, even on things like a Ford F-150 or being a fan of NASCAR or the NBA, we still have decent minorities between a quarter and a third of Americans who say neither. So this is not a 90/10 sort of split that we wanted to give folks the option to say, “Well, I don’t view these as associated with either political party.”
And so I think what we’re trying to get at here is, and I think there’s a time series component here that I think is important, which is how did we get to this point in time and that driving a Ford F-150 is something that’s viewed as overtly Republican? And I think that’s complicated to parse out. But at least in the pretest questions, I think we were trying to understand better what linkages does the mass public make between the Republican Party and the Democratic Party and these objects.
Matt Grossmann: So let’s go to I guess the high end of the use case here, which is that you find that you get a candidate wearing cowboy boots and driving a Chevy pickup, and that’s as much as telling people that the candidate is a Republican. How should we interpret that? What do you think voters are doing there? Are they just saying this person’s probably a Republican? Are they imputing some kind of values associated with those purchases or that behavior? Or is it just like this person is someone like me or not someone like me?
Levi Allen: Let me take the graduate school answer, it’s all of the above in that I think partisanship’s obviously there. And one of the things that may be happening when we have that candidate who has those cross-cutting cues, who identifies as a Republican but then gives these sort of democratically or liberally coded signals washes out is one thing that may be happening is folks don’t know what to make of this candidate. What do they make of a Republican who has these identity signals? The other reading of that is that they may legitimately be more open to supporting that type of candidate as well. So I think partisanship is obviously a huge story there.
But I think your question about values is really important, and this ties into some of my work on rural politics as well, where especially for folks in rural areas, this sense of having a unique way of life that isn’t always respected. And as America becomes more geographically polarized, and understand that’s always existed, it’s not like Republicans began winning rural areas when Donald Trump ran for president. It’s existed long before then. But this sense of that it’s a candidate who is like me and who understands my way of life. And I think there’s an inverse there too.
Now, the rural politics literature would say that I think, and this is what my work has shown, is that the urban identity is not as crystallized as the rural identity, that folks in urban areas don’t really ascribe as much meaning, if you will, to identifying as an urban person. But I think when John Kennedy stands up there and says, “I know I’m going to Washington,” and he, by virtue of his background and who he is, has had a very elite life. When he says those things, he’s telling voters in his state, which in this case is Louisiana, that I am still one of you because I don’t put goat’s milk in my latte.
Matt Grossmann: But the practical advice for candidates here seems a little much-
… advice for candidates here seems a little muddled because you find that, for example, a Republican taking those Democratic cues can get a backlash among their own supporters without necessarily making up for it on the other side.
So, is there a playbook for candidates here or is this just telling us that these symbols make sense, but there’s not a good way for folks to get out of the stereotypes?
Levi Allen: I think it could be useful for candidates who are met with some skepticism as to whether or not they are truly a Republican or a Democrat. And we could get into what does that exactly mean, right? But I think you see this… We think of, and we mention in some places, candidates where there could be some crossover appeal. I think Joe Manchin was a great example of this, but I think it’s reasonable to assume that that might be very quickly going away or be practically gone.
And Manchin himself was unique in that he was a governor of the state, and we know gubernatorial elections are very different. I think the practical advice for candidates would be that these things, I think it would be difficult to defy the stereotypes that are associated with your party, that… Whereas we’ve had issue litmus tests, we can think of issues like abortion, where there was a real inability to cross over. And I think that’s become particularly true since the 2010 wave, where a lot of those pro-life blue dogs lost in the South.
I think where this can be useful is when a candidate is trying to present themselves as more genuine to the mass public, depending on which side of the aisle they’re on. Now, one thing we really hammer on in the paper and that we talk about is, is that this does seem to be more concentrated among Republicans, right? In that I’m not familiar with many Democratic politicians who will make a remark that’s something like, “Well, those people in flyover country, blah, blah, blah,” or, “Those people who wake up at 5:00 AM to milk the cows are these sorts of things.” And so I think there’s some asymmetry here on which group of partisans and which group of political elites this can benefit.
Matt Grossmann: So you mentioned that Joe Manchin was an example of this, but obviously is going away, and we have this broader pattern of the nationalization of our politics.
What is the scope for this, given our current politics? How much are candidates able to stand out, and are we about to reach the expiration of the capacity of these signals to differentiate anyone?
Levi Allen: Yeah, it’s interesting, Matt, thinking about where you’re at in your state that has a very contentious Senate race. And as you know, there’s many layers to this as far as the dynamics at play in the Senate race. But I was watching college football this last weekend and I had… That other school in Michigan was on the TV, and there was an ad by Mike Rogers that was attacking his opponent on football, right? And that he had once said football was all these sorts of things.
And Mike Rogers is looking at the camera, he’s holding a football and he’s saying, “Football’s part of the Michigan way of life” or something like that. I’m not sure the exact phrase, but that’s the gist of the ad. And I think in those ways it can be effective as an attack, right? And I frame this in how a candidate can use this more so to ingratiate themselves to their own voters, but I think it can also be used, and that’s kind of how John Kennedy uses it, to draw distinctions between the other side. And in that we are in this era of the nationalization of politics, we may find that’s what political elites do, is it’s just another weapon in their arsenal that they can pick up to try to say that the opponent is truly, quote, “Not one of us.”
Matt Grossmann: So, we talk about this as if it is a strategy that all candidates could use, but obviously there are real candidate characteristics, and this can obviously look quite inauthentic in some cases.
There’s actually risk of that in the Michigan race, since El-Sayed actually played football. But how do voters think about this with respect to authenticity? Does it matter if someone actually hunts or drives a pickup, or is this something you can fake?
Levi Allen: Yeah, I mean, I think it’s obviously determined by the district characteristics. And one of the things we really wrestled with is what is the logical progression here? Is it that voters have these characteristics and identities and elites pick up on them? Or is it that elites start talking about these things and voters notice?
I mean, that’s beyond the scope of the paper, but I think in this instance, I do think authenticity matters. And I think you see this play out in primaries a lot, where it seems more and more, especially with the rampant gerrymandering that exists, primaries are going to continue, I think, to be where even more of the action is at.
And you experienced this in the Michigan Senate race, where there’s a lot of questions about the genuine nature of candidates and who can win and so on. And I think when thinking about primary electorates as being some of the strongest and most engaged partisans, I think those electorates at least are almost to the point where they’re mandating these shows of… I think it’s identity in one way, but I think this authenticity matters.
I’m sitting in a rural Indiana district right now and we’re not in a competitive district. Our congressmen will win by, I don’t know, 25 or 30 points. But even the few ads that he is running are these shots of him at rural parts of the district and playing up this, “I am one of you.” And so I think it’s just become another layer of the ways that candidates are able to draw that distinction, that us versus them distinction.
Matt Grossmann: So you open with this John Kennedy quotation. Of course, he’s good for them. The latest funny one is he’s calling the DSA the “Lulu Lenins” to try to emphasize their cultural, I guess, class along with their politics. But what is the role of political rhetoric here?
Are these just being made fun of for sport? Is there a real role that politicians are trying to make these social differences into components of partisan identities? Or politicians also come from different places and so they just are reflecting backgrounds just like their voters might be?
Levi Allen: Yeah, I think what’s interesting about the Kennedy quote is he’s a couple sentences before he gets into these non-political identity signals, he has this line about, “Standing before the crowd as a proud deplorable.” And we can think of what is that signaling, right? It’s this sense, it’s this real populous sentiment, right? I think that, of course, playing into Hillary Clinton’s comments in the 2016 election, but he also calls these people, “Cultured and cosmopolitan.”
And so my rural politics brain gets going there when I hear those words, this notion that these urban centers of power are… That they look down, frankly, on the rural way of life. And I think in this instance, that’s what’s happening there. And so when he says, “I’m talking about these avocado-toast-eating insiders and elites,” I think that it’s more than just having fun with it or reinforcing it.
But I think in this time where we do have some political realigning going on… Now, we’ll have to see if this is durable or just sort of a Trump effect, and what this looks like in 10 or 20 years, but I think it is this wholesale movement of broadly the working class or those without a college degree.
And there’s, of course, some overlap there towards the Republican Party, whereas Democrats continue to really run up the margins in many suburban neighborhoods, where you do have… I mean, let’s be honest, in a suburb you’re going to have maybe a Sweetgreen eatery, you’re going to have multiple coffee shops, you can get a latte. It’s black coffee in my rural neighborhood, right? That’s the option.
And it’s egg bacon and cheese breakfast sandwich, not avocado or something like that. So, I think what’s dangerous here is it’s another cleavage, it’s another wedge that political elites can exploit for their own gain, whereas the loser in all of this seems to be the American public, because I think there is a question of do these differences actually matter? “Is it really that existential that I put avocado on my toast and now I’m this insert whatever stereotype?”
Matt Grossmann: So as you know, there’s a common view in social science that these are identity cues, not just for partisanship, but you mentioned that Popeyes might signal race. Some of these are signaling income, some of them are signaling urbanicity.
And so these might just function as alternatives to those or ways for you to signal your outsiderness or non-eliteness. How do you think about those roles of these consumer habits? Are they non-political but demographic identity cues?
Levi Allen: Yeah, it’s a great question. And I think a lot about the rich literature in race and ethnic politics about implicit cues, that you can’t, in theory… Although, there’s some great work that shows maybe we’re moving out of this where you can just be more explicit about these things.
But when I want to make these sort of claims and I want to use race as a cleavage that can be exploited for political gain, I can’t be overt about that. I’m going to use a lot of these dog whistles. And I think in some ways, and in some ways perhaps less insidious, given that I think implicit racial cues are particularly insidious in the way that they’re used, I think this is a way, as you mentioned, to talk about things that are associated with the non-political identity signals.
For example, I’m always amazed at my students and their elaborate Starbucks drinks. And sometimes I’ll privately ask what that cost, and maybe that’s too invasive, but they’ll tell me and I’m like, “Wow, that would get me four Cokes at McDonald’s or something.” And so I think when John Kennedy says these things, what he’s also saying is, is that it takes a certain socioeconomic status, a certain well-to-do-ness to be able to afford goat’s milk in your latte, or to even know that that’s an option, right?
And so then when one returns to their home district or state, particularly if it’s rural or largely uneducated, they’re able to make these claims that I think are more effective than just saying, “Well, you know the Washington insiders are rich.” “Well, how rich are they?” “They put avocado on their toast and they go to the Sweetgreen eatery because they can afford the organic foods.” And in that Cued by Culture’ paper, they find things like vegan and so on are very democratic-coded.
And I think obviously some of that is a sorting effect, where vegans are, I don’t have any public opinion data on vegans, but I would imagine they are democratic on average. But it also, I think, is inferring a certain status when it comes to income and education. And then as you note, on some of these, I think race would play a role as well.
Matt Grossmann: So the general finding in these literatures is that the stereotypes are generally directionally correct, but overstated. Does it matter how accurate these kinds of perceptions are? Do these things matter independently? Let’s say they don’t actually have an effect on vote or anything else, but we do start to develop these reputations…
Or anything else, but we do start to develop these reputations that affect where people go to eat or where they shop. Is that independently of interest or is that just kind of a side show?
Levi Allen: Yeah, it’s a great question. And one thing I probably should have started with is by saying that I think in our careers we have work that we really enjoy, and then we have work that we really enjoy and is really personal to us. And I think in one way this piece was personal to me is that I have a lot of these sort of misaligned cues, if you will. I very much love a good Chick-fil-A chicken sandwich, and I also love going to the art museum. And so I find myself in this sort of pull, if you will. And I of course have work on voters who defy these identities and that’s driven by that personal story.
I think it matters in two ways. I think one, the overstating of preferences, as you mentioned, the literature shows can be pretty damaging. And I do this exercise with my students in class where I’ll ask them, for example, what percent of Democrats do you think identify as a member of the LGBTQ community? And as you know, they sometimes will go as high as 50%, that one in two Democrats, which of course is preposterous. And then on the right, asking them about income or religion or race even, some of them will say 98% of Republicans are white, and that obviously is not true. And so I think this becomes yet again another place where we see this sort of amplified.
And again, going back to our supplementary data, our most coded non-political identity signal was that 55% of people though the Toyota Prius was coded democratically. That means that roughly one in two people don’t think that way. So I think that’s important. I think where the damage becomes is that it can come bottom up and top down in that the necessity in a pluralistic democratic society is that we have some of these overlap in non-political settings between Republicans and Democrats. That it’s really important that Republicans and Democrats go to the NASCAR race together or share a chicken sandwich, even if they’re not doing that together, they’re at the event together.
And as more of these things become politicized, I’m pretty concerned, and I see this in my students who will say things like, “Well, this person drives a Ford pickup truck and that must mean,” insert all of these things. And of course, in sometimes or maybe even plurality of the time, they’re going to be right about those things, but that still means there’s a huge side that isn’t. And so it’s important, I think, that you can be a Republican and put avocado on your toast. It doesn’t make you less of a Republican. And just like you can be a Democrat and you can wear cowboy boots, that doesn’t mean… And I think when we think 40 years ago, I think this was obviously much less common.
When we think about that, we obviously had more rural Democrats, we had Northeastern Republicans, we had California Republicans who would get to the Congress and have these debates within their own party about what does it mean to be a Republican? And those things I think were healthy for democracy. And as we lose those, I’m pretty concerned about what the culmination of that is.
Matt Grossmann: As you mentioned, you have this broader work on the relationship between different identities and partisanship, and you find that they are not always linear and two-sided and equal. So for example, you find that a rural identity is not quite as coded as an evangelical identity. But you also find that the urban identity matters more to repelling Republicans than the rural one to repelling Democrats. So talk through some of those kinds of differences and when we should expect a one-sided versus a no-sided versus a two-sided association.
Levi Allen: Yeah. I’m of the opinion that religion and race right now are the two most salient social identities beyond, of course, partisanship when it comes to determining political behavior. And so that’s sort of where I started is my dissertation and forthcoming book are about white evangelical or born again Christians who are Democrats and what exactly is going on there. And I think that the political elites are so powerful in this way in that when they begin, I think, to pick up these identities and view them as you’re either aligned with the party or not, that’s where we see real movement in the public.
And one of the things we’ve seen Republicans do, for example, is talk about being from a rural area. And so I don’t think it’s shocking then that we see folks pick that mantle up as well. And in some ways very legitimately believe that they are left behind and anyone who’s driven through rural areas can see the real decay that matters to folks and has negatively impacted their lives. I think when you think about identities then that are not as clear, something like urban, I don’t hear many candidates stand up and say, “I’m from urban Chicago,” or whatever as if Chicago has non-urban areas. What does that look like? What is an urban way of life?
And as you talked about earlier, of course, imbued in this is race, right? That when folks say urban, are they really talking about the fact that you carry your groceries on the subway to get back home, or that you take the public transit, or that there’s multiple orchestras you can go to that night? Or are they really talking about that, well, this is where non-white folks live? And I think that’s an empirical question that we’re tackling in other work to sort of parse that out.
Matt Grossmann: So as you mentioned, you’re working on a project or have long been working on a project about people who resist the common social sorting. So what can we learn from the exceptions?
Levi Allen: The literature does not assume a correlation of one. It doesn’t say that if you’re Black, you’re Democrat. If you’re evangelical, you’re Republican. And so I don’t want to hold that literature, which I really love and I think has moved the discipline forward in really important ways to a standard that it didn’t set. I do think on the other hand, there is a sort of deterministic nature that sort of comes about by the social sorting and political polarization literature that sometimes feels that we are nearing that perfect correlation of one. That all members of this group are Republicans, all members of this group are Democrats.
And so I think the first thing is just to say, wait a second. When I give talks on this, folks are always amazed. I ask them to estimate what percent of white Christians who say they’ve had a born again conversion experience related to Jesus Christ are Democrats. And I often get answers of 1%, 2%, 5%. It moves as public polling does and moves from election to election. But we’re looking at about a quarter, 25%. And when you add up and when you think about how many white born again Christians there are in America, that’s a huge group of people. And so I think that’s the first thing is I believe partisan stereotypes do matter and I think they can be really damaging. And so to push back and say, well, yes, of course the majority is, and I’m not pushing back on this notion that evangelicalism and the Republican Party are really enmeshed with one another, but that there are meaningful minorities.
I think the second thing is when it comes to the health of American democracy is that what I found is many of these, I call them political heretics, not in a sort of normative sense, but that they are defying political dogma. Which can, we’ve seen this in the parties, Mitt Romney at times viewed as this rhino, and we see this at times on the left, although I know John Fetterman, there’s other things going on there beyond his political moderation. Is that they are in these communities and they can actually create some connective tissue that create those cross-cutting relationships, which we know are really important to democracy and really healthy. And so I think that’s an important finding.
But I get the real sense, and this is sort of the third part of that, that they are under an immense amount of pressure. That at least in the context I study this within religion, that I have multiple people who are in these Republican majority churches, which I think we would expect given they’re in the Midwest and evangelical churches, who really do view political identity as a spiritual litmus test. And so then as a scholar of religion, I’m always concerned, and Tocqueville writes about this a lot in Democracy in America, that what happened in Europe is when these things got tied together, really bad things happened. And I think that’s what’s concerning in one pathway, one panacea in part to stopping this entanglement, which I think long-term is bad for not just the church, but also for the Republican Party.
I mean, I think evangelicals are going to find themselves in a captured state to think of things in the Paul Frymer way of talking about it, that they may be able to be the ones, these defectors, that can actually reduce the stigma of partisan stereotypes and the damage that it causes.
Matt Grossmann: So you said you have that hope, but what about the expectation? Is this something we should expect to just sort of more and more of cultural symbols and shopping behavior and everything else will start to get politicized? And as you mentioned, even in these more serious cases, there’s incentives to conform to group norms that you might expect a kind of a continuation of the sorting. But some of your findings suggest that maybe these are just substitutes for political information. So once we have that political information, we don’t really make that much out of them. So is there kind of an inherent kind of stopping point to this?
Levi Allen: Yeah. One thing that’s interesting in Study 2 is that we also tested a condition where our candidate identified as a Republican and had the non-political identity signals. And we were interested to see, is there an additive effect here? Does that really amplify things that I might be a Republican, but man, when I drive a Ford F-150, when I have my cowboy boots on, I’m really a true card carrying Republican. And we found that’s not the case at all, that it’s sort of an either or thing. And I think your point is a good one. And I think overall is a positive one, which is, well, candidates can get at this in different ways. And I think we provide compelling evidence these things are actually politicized. We don’t see null results. We see people move when they read these things, but when we put them together, there’s not a super boost or a super penalty.
We don’t find Democrats penalizing that Republican identity queue candidate more. And so I think right now I think a lot about what is the off ramp? What does this deescalation look like in American politics? Because elections are framed in, I believe, an existential sense that every election is for the battle and heart and soul of American democracy and the American Republic, and all the incentives are messed up. Our members of Congress are not incentivized to compromise. Our political leaders are not incentivized to not engage in this rhetoric that gets them on Fox or MS Now. And so I think we’re going to continue to see these dividing lines.
Now, I think where that goes next is interesting. And one of the things I’m interested in is there something on the political left that could be weaponized in this sense? And I don’t mean for that to sound as normatively grievous maybe as it comes across, but where is the wedge at? Where does the left have an advantage where they can start pushing on this? And I think Harris, in some ways, this campaigning on democracy was also about campaigning to a more educated and elite voter base. These sort of high-minded arguments of which I think there’s some merit to that democracy, as Reagan said, is a generation away from going extinct and that you have to fight for these things.
Generation away from going extinct and that you have to fight for these things and that’s important. But I don’t anticipate a Democratic candidate for Congress running in a competitive race and saying, “Well, my opponent, he doesn’t drink lattes or he or she doesn’t drive a Toyota Prius or doesn’t watch the NBA or whatever might be going on.” And so I think that asymmetry is problematic where it’s being used on one side and there’s no real counter. And I think that could make it quite damaging in the long run.
Matt Grossmann: So some of these associations have some capacity to change. You mentioned that Tesla has changed in the time that you’re doing the study. Obviously corporate actions and boycotts might bring attention to something. You also just might have expansion. Starbucks is in a lot more places than it used to be. So what do you learn from these trajectories where even if there’s kind of an overall status, some of these things can come and go depending on political events and some seem more kind of stable?
Levi Allen: Yeah. I think one of the things that’s interesting about Chick-fil-A, which remember the modal response for Chick-fil-A was neither, that was roughly 40%, but then 35% Republican, 25% Democrat. So a little bit of a lean there. Of course, Chick-fil-A’s leadership had taken a political position. It is by its nature a faith-based company. It’s closed on Sundays. If you dine in there, you will hear instrumental Christian music, all of those sorts of things. But it was because their leadership waded into the same-sex marriage debates that I think really amplified that. And I think the sames true of Tesla, right? When we fielded this study, we have 46% of folks saying it’s coded as Democratic, 37% say neither, and only about one in six say Republican. I think that’s going to be flipped at a minimum if we ran that again today. And so I think anytime private CEOs, private company CEOs get involved in politics, that’s going to shape the image.
I think also some of this is just more organic, that as we see again, the reddening of rural America where you’re going to by nature, I would imagine, see more pickup trucks on the road. That’s organic and that’s not all bad. I think we’ve talked a lot and I’ve talked a lot about how political elites sort of fan these flames, but sometimes this stuff is just happening organically in society because again, there’s not going to be a Starbucks in rural America. I mean, it’s just not happening. And so I think those are relatively stable. We talked, my co-author and I talked pretty deeply about what are the next things to be politicized and what are the next… And one of the things that we think has become, and again, I think in some ways I think there is a race component to this is the WNBA as being very politicized.
And of course that goes into not just gender equity, as we know there’s been this recent sort of bubbling up about transgender athletes, but also there’s a very clear, and at least in my view, there is a racial component to this where political elites talk about these things because, and I think this is something that’s important that comes up in the article a little bit, but I think could have been accentuated is if we think about Americans en masse as possessing relatively low levels of political knowledge, and I agree with the Achen and Bartels argument in some ways that folks are trying to put food on the table, raise kids, all of these things. We can’t expect them to be experts on what’s our foreign policy in Myanmar.
But I understand a Chick-fil-A sandwich and I understand a Ford F-150 and I don’t need to have a nuanced understanding of the child tax credit to understand that candidate drives a pickup truck and that candidate doesn’t. And then I get in my Ford F-150 to go to work every day, I know that candidate knows me because of those things. And so I think you’ll see elites continue to expand in their use of these cues.
Matt Grossmann: So you also obviously study the outliers because you believe that this cross pressure could be important. So at a time when a cross pressure from issue positions might be declining, we might expect that some of these cultural identities might have more cross pressure, not just in the group level ones, but there’s going to be hardly anyone who has zero cosmopolitan interests or tendencies or zero traditional ones. So maybe this is a way of potentially empathizing or having different views of the other cultural side. And could it be for candidates? I mean, could it be that candidates should start to think about the ways that they may not culturally fit their party as a way of reaching voters, not just instrumentally, but actually in understanding?
Levi Allen: Yeah. I think we see this as a line of attack often. We see that, again, this has sort of been framed as coming from the candidate to ingratiate themselves to the electorate. But I also think that, and we talk about this in the Michigan Senate race, we’re seeing this, I think in other competitive Senate races about… And I think in Georgia, for example, I think we’re seeing this in Kansas where Roger Marshall, the Republican candidate, his first ad is he’s walking, it looks like a cattle farm and he’s talking about being from Kansas because he’s come under fire that he lives in Florida. And so what does he do to show I’m a true Kansan, it’s he plays into these sorts of things associated with the state. And again, sometimes those aren’t divorced from the reality. As you noted, folks are right about getting the stereotype in the right direction. They just vastly overestimate that.
And so I think that when you have candidates who are struggling to make those connections with voters, this is, I think, a very effective way because it’s relatively low cost. Even if you’re on the other side of this, how do you attack a candidate when she says, “I drive a Ford F-150?” What’s the attack there? Are you going to make a nuanced climate change? I mean, you’re going to lose voters on that, so there’s a lot to be gained there. And again, it’s easily communicable to a mass public who, I think at this point too, is experiencing some fatigue. I notice this in my students of just… There’s not a nuanced discussion.
And I think in wrapping that point up, at least where I sit, it feels like this is happening in a backdrop of where we don’t really have serious policy debates anymore. And I think some of this has come through just the balkanization of the parties and that this is where they’re at and there’s no reason to talk about abortion because we know where the parties are and it’s so costly for them to even marginally move on either side. But we are at this time where we’re talking about things that don’t seem to be very serious. And so as politics in some ways I think has always been for entertainment, but I think definitely in the last 10 years is more and more…
I really love the way that Lilliana Mason puts it, the sort of sports game, if you will. There’s nothing more entertaining than attacking a political opponent, not because they took this nuanced vote on this piece of legislation you’ve never heard of. And let me explain to you in five minutes why that’s bad, but that, oh my gosh, they drive a pickup truck or they have a latte every once in a while. So I think that’s also sort of fanning the flames of these non-political identity signals.
Matt Grossmann: And anything we didn’t get to that you wanted to mention or tout about what’s next?
Levi Allen: Yeah, no, I appreciate it. I do think that those dual findings sort of pulled us when we thought about them in opposite directions. The positive finding we felt was as policy still matters and even policy that’s not super salient or high profile we felt still mattered. And I think that’s a good thing if we think of we want educated citizens who make decisions based on the issues. I think we were really surprised in two ways. One was the lack of an additive effect. I think we did have some reason to believe that you could really show your loyalty to the party, your shared identity by lumping together political partisan identity in these signals, and we didn’t find that. And then I think there’s work to be done on what’s going on when respondents get these cross-cutting candidates. Are they just confused? Are they more open legitimately to supporting them?
We find the penalty in both directions, right? So remember the Republican candidate who shows the non-political democratic signals said he was a Republican. It’s just that he does these non-political things that then blunt the support. And we thought that was interesting. I’ve talked about it a lot, but the project sort of occupying most of our time, most of my time right now is a project that really seeks to disentangle race in place. So I think there’s this argument in the literature and it varies in the intensity it’s made that folks in rural areas may not really have rural resentment that it’s just racial resentment repackaged. And how do we get at that and how do we parse those things out? And what we find is that it’s context dependent, right? That we don’t find that rural resentment is just boiled down racial resentment, but we also find that as that perceived racial threat moves closer, then folks, the importance of their rural resentment sort of goes down a little bit and racial attitudes are more important.
And I think that’s a meaningful contribution when it comes to sort of disentangling what I think is an emerging identity, the rural identity from of course a longstanding one.
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 Grossmann. If you like this discussion, here are the episodes you should check out next. All links on our website. How misperceptions and online norms drive cancel culture, how racial realignment ignited the culture war, how media incentives stoked the culture war, is demographic and geographic polarization overstated and explaining the urban rural divide. Thanks to Levi Allen for joining me. Please check out non-political identity signals as information shortcuts and then listen in next time.
