Despite calls for politics to return to kitchen table economic issues, the culture war rages on. That could be a product of the distinct incentives facing politicians, who have to win elections, and media actors, who just have to keep your attention. Aakaash Rao and Shakked Noy find that cable news outlets talk more about culture war issues while candidates favor economics. Every minute cable news spends covering the culture war, they gain audience from people who would otherwise prefer entertainment. When they talk economics, people switch channels. And where cable news penetrated more, people started seeing crime, immigration, and race and gender as more important than economics—and candidates eventually shifted their behavior to match. It’s not just about the part cable news played in the rise of the culture war but also about how actors seeking to mobilize rather than win converts might be the source of our wider polarizing shifts.
Guest: Aakash Rao, Harvard University; Shakked Noy, MIT
Study: The Business of the Culture War
We welcome your thoughts on this episode and the podcast as a whole. Please send feedback or suggestions to scienceofpolitics@niskanencenter.org
Transcript
Matt Grossmann: How media incentives stoked the culture war. This week on The Science of Politics. For the Niskanen Center, I’m Matt Grossmann. Despite calls for candidates to stick to kitchen table economic issues, the culture war rages on, and it might be the fault of the media rather than the candidates. While politicians have to win elections, new media actors just have to keep your attention. And it could be that their incentives differ, but we’re all now stuck in the world media built.
This week, I talked to Aakaash Rao of Harvard University and Shakked Noy of MIT about their new paper, The Business of the Culture War. They find that cable news outlets talk more about culture war issues while candidates favor economics. When cable news covers the culture war, they gain audience from people who would otherwise prefer entertainment. And when they talk economics, people switch channels. And where cable news penetrated more, Americans started seeing crime, immigration, and race and gender as more important than economics, and candidates eventually shifted their behavior accordingly. It’s not just about the part cable news played in the rise of the culture war, but also about how actors seeking to mobilize rather than to win converts might be the source of our wider polarizing shifts. I think you’ll enjoy our conversation.
Aakaash, tell us about the findings and takeaways from this new paper that is, I believe, your job market paper, The Business of the Culture War.
Aakaash Rao: Yeah, that’s right. So we studied the rise of the culture war in contemporary American politics, so since the 1980s. And by that, we mean the increase in political polarization on cultural issues. And so what the paper does is trace some of that rise of the culture war to changes in the media environment in the 1980s, in particular coming from the growth of cable television. And so kind of the key argument of the paper is that cable news producers like CNN and Fox News and MSNBC really developed a distinctive business strategy that made it profitable for them to lean into coverage of culture war issues. And so the first part of the paper is about trying to understand that business strategy. So why was it profitable for them to talk about the culture war so much and why is that distinctive to cable news and the cable news era?
And then the second part of the paper is about the consequences of that culture war centric business strategy. And so what we show is that exposure to cable news leads viewers to care more about cultural issues in politics, and then it leads politicians to respond to those changes and preferences by talking more about cultural issues in their campaign ads. And so kind of the big picture number we calculate is around a third of the increase, both among voters and among politicians in how important cultural issues are relative to economic issues. Around a third of that increase in cultural importance can be attributed to the effects of cable news.
Matt Grossmann: So Shakked, let’s start out with the basic differences here between the content of political ads and the content of cable news and these different corpuses that you put together. What were the big gaps? How consistent were they and how did you establish that?
Shakked Noy: Yeah. So what we do is we gather a large data set covering basically the campaign ads that politicians air on broadcast television in election years. And these are kind of the main expenditure item for most political campaigns or main mechanism of broad outreach to voters. And we basically take data sets that have been compiled by the Wesleyan Media Project and others to form a large data set of these campaign ads, which we classify using large language models. And what we basically see in this data is that politicians overwhelmingly are talking about bread and butter economic issues, kitchen table issues. They’re talking about how they’re going to create jobs. They’re talking about how they’re going to help the economy in their area grow, how they’re going to provide tax cuts for the middle class. So overwhelmingly focused on these kinds of kitchen table economic issues. And we see very little of the hot button culture war issues that tend to dominate the discourse and sort of news coverage and discussion of politics beyond what politicians are directly campaigning on.
And so in fact, when we then look directly at what kinds of issues TV news channels tend to cover, so we have data on the contents of the main cable and broadcast news channels, Fox News, MCNBC, CNN, ABC, NBC, and CBS back for a pretty long time period, we see that they talk about economic issues much less than politicians do. But they talk about culture war issues much more, particularly crime, racial and gender issues, terrorism, immigration, things like that.
Matt Grossmann: Aakaash, your theory relies on this difference between mobilization and poaching. That might be similar to what political scientists or political professionals think of as turnout versus vote choice, but you kind of apply this to media companies. So walk us through that comparison and why would it be that the cable networks have different incentives than politicians?
Aakaash Rao: Yeah, no, that’s exactly… So it really rests on the fact that you can think about a politician and a media company in a very similar way. Both of them are trying to attract people to choose them. So a politician is trying to get voters to choose them, media company is trying to get viewers to choose them. And so if I want to get one additional person, either as a politician or as a media outlet, I can get that person by poaching that person from one of my competitors, or I can get that person by mobilizing them from outside. And so what do we mean by that? Well, if I’m a politician, if I’m Trump in 2024, I can either try to get people who would otherwise vote for Harris to instead vote for me, that’s what we would call poaching, or I can try to turn out someone who otherwise wouldn’t vote. And so that’s what we would call mobilization.
And very similarly, if I’m Fox News, I can try to choose the type of content that appeals to someone who would otherwise watch MSNBC or ABC or some other competing news outlet, or I can choose the type of content that’ll get people who wouldn’t otherwise watch news at all, people who would watch sports or entertainment or sitcoms. And so that’s the sense in which these are really parallel problems, the politician’s problem and the media outlet’s problem. Both of them face the same trade-off between poaching and mobilization because the content that’s good at stealing people from your competitors might be very different from the types of content that’s good at getting people from outside of the market in the first place. The key difference between politicians and media outlets is that they are maximizing different things. So for a media outlet, what I really care about is just the number of people watching me. That’s what’s going to drive ad revenue.
Now, a politician doesn’t care about the number of people voting for them per se. They care about the number of people voting for them minus the number of people voting for their opponent. That’s how you win an election, you get more votes than your opponent. And so what that means is that media outlets are maximizing their audience size, but politicians are not maximizing their vote number or their vote size, they’re maximizing their vote share. And so what that means is that these two agents have different incentives. They want to lean into different types of content. Media outlets are going to disproportionately care about the content that’s good at getting people who would otherwise be disengaged, and politicians are going to disproportionately care about the type of content that’s good at poaching people from their opponent.
And so part of what we show in the paper is that economic content tends to be good for poaching, which politicians care more about, and cultural content tends to be good for mobilizing, which is what media outlets care about. And so that’s partly why you see these striking differences where media outlets are talking so much more about culture, politicians are talking so much more about economics.
Matt Grossmann: So again, one of the innovations here is this data from people’s actual TV watching habits. So we don’t have to rely on what people say they want to watch, but what they’re actually watching and how they respond to the content. And it looks like the data is sort of minute by minute, you can see changes in coverage and viewership, but give us a sense of what it looks like. Are most people watching sort of full shows here and where most of the variation is coming from that? Is it they do one culture story and one economic story and we can see the switch manifest? What does this data look like and what does it show?
Shakked Noy: Yeah. So the TV data we have is important for a couple of reasons. So one is that this distinction between poaching and mobilization that Aakaash described is important conceptually, but it’s really hard to get at empirically, because we might see that somebody voted for Trump, but we don’t know whether the Trump campaign poached or mobilized this person because we don’t know who this person would have or what this person would have counter-factually done had Trump run a different campaign. We don’t know whether this person counter-factually would have not voted or we don’t know whether this person would have voted for Kamala Harris. So that makes this distinction between content that’s good for mobilizing and content that’s good for poaching very hard to get at empirically. But what we have in our TV data is data from collected and aggregated from smart TVs and set top boxes, which basically measures for each second of the day whether the TV is on and what channel people are watching if the TV is on.
And what’s really important about this is that it actually lets us observe people switching between different channels. So we don’t just know that a person is watching CNN. We know, for example, that they switched from watching Fox News to watching CNN, or they switched from watching a reality TV show to watching CNN. So observing these flows between different channels is what lets us get at poaching versus mobilization empirically because we can say that flows from entertainment into news are successful mobilization, but flows from one news channel to another are examples of successful poaching. The other thing that’s really important about these data is, as you said, they’re extremely high frequency. They’re minute by minute. So minute by minute, we can see what people are watching and we can also see what the news channels that they might be watching are saying, which lets us exploit this kind of extremely high frequency variation in content, which is useful for causal identification.
We wouldn’t just want to use the fact that on average across the entirety of the year, this show tends to talk about economics and this show tends to talk about culture, because there might be various other differences between those two shows that are confounding that comparison. But if we have very high frequency variation where we can see that this show is talking about economics today, but culture tomorrow, or it’s talking about economics right now, but culture 15 minutes from now, that lets us make comparisons that get us closer to estimating the causal effect of cultural and economic content. So in terms of what these data look like, it is true that viewing tends to be highly persistent in general. So if somebody’s watching a show at one minute, they tend to be watching it the next. Most people do watch full episodes and so on. The TV tends to be on for hours at a time.
But on the margins, we do see a lot of people switching between channels and we do see them react to the content that they’re getting on news channels. And this kind of variation on the margins is of course also what news channels care about on the margins. On the margins, they want to retain people who might switch away. And so what we see with this data when we try to use this kind of high frequency variation in content is we see that when a news network switches from talking about culture to talking about economics, two things happen. So one thing that happens is that the flow of people who switch to competing news channels decreases. And what that suggests is that economics is better than culture at poaching, at retaining people who would otherwise switch to competing news channels.
But the other thing that happens when a show switches from cultural to economic coverage is that the flow of people leaving to alternative entertainment options or switching off the TV increases. And what that suggests is that cultural content is better for mobilization. It’s better for retaining people who otherwise wouldn’t watch the news. So what our TV data shows is that economic content is better for poaching, culture war content is better for mobilization. And if you’ll remember what Aakaash just described, politicians should prioritize poaching, media networks should prioritize mobilization. And so the fact that economics is better for poaching and culture is better for mobilization helps us explain why it’s the case that politicians focus so much on economics and media outlets focus so much on culture.
Matt Grossmann: Okay. But just to clarify, we don’t have that. It’s not like we have that data for the ads as well, right? So we’re just making the assumption that the same kinds of process would matter for the politicians and the news.
Shakked Noy: Exactly.
Matt Grossmann: So Aakaash, you make another jump here. We said that the overall politicians are talking more about economics and culture, but you’re also trying to explain why politicians are increasingly talking about culture, and so you try to use cable news to do that. So explain that analysis and kind of what magnitude of change we’re trying to explain here.
Aakaash Rao: Yeah. So here’s the basic idea. Imagine that you are a heavy cable news viewer. By what we’ve just described, a huge fraction of what you see, of the information that you consume about politics because you’re watching on cable news is about gun rights and abortion and free speech and crime and these other very cultural issues, and a much lower fraction of what you’re seeing is about healthcare and jobs and taxes and the more traditional economic political issues. And so over time, it’s pretty natural to think that you’re going to associate politics a lot more with these cultural issues than those economic issues. You’re going to come to see these cultural issues as a more important part of politics. And that’s exactly what you see in these surveys. So Gallup, the polling organization, has run this question for the last several decades. So going back to the 1930s and 40s where they just ask, what do you think is the most important problem facing the country? And viewers can respond with all sorts of things, they can say taxes or they can say terrorism or guns. And what we find is that in places that are more exposed to cable news, that is in places where people are watching more cable news, you see that a larger fraction of individuals are coming to say that these cultural issues are important. And a smaller fraction of individuals are saying that these economic issues are important. So it seems like cable news exposure is leading people to care more about these cultural issues and less about these economic issues.
And then similarly, we also see that politicians seem to be responding. So in places that are more exposed to cable news, politicians are talking more about cultural issues to try to meet these change voter preferences and they’re talking less about economic issues.
Now, a key part of what we do is we don’t want to just compare, say, a place in which a lot of people are watching Fox News to a place in which a lot of people are not, because there are many reasons that places might watch more or less cable news. And so if we really want to estimate the causal effect of cable news on people’s preferences, we need to kind of simulate an experiment of randomly assigning some places more cable news and randomly assigning some places less.
And so we’re going to borrow an approach from political scientist Greg Martin and economist Ali Yurukoglu. So they show in some previous work that there’s a nice experiment generated based on the channel number that cable news networks receive, which is different across different media markets. So Fox News maybe in one media market is channel 30, maybe in another it’s channel 60. And it turns out that in areas where Fox News was kind of randomly assigned a lower channel number, people tend to watch more Fox News. And this goes back to the older kind of TV logic. You turn on your TV, you’re flipping manually through channels and the lower channels tend to be the ones that are more heavily watched. And so if you’re kind of flipping through, you’re more likely to land on Fox News if Fox News is a lower channel number.
And so the upshot of this is that this kind of gives us a way to estimate the causal effect of cable news on voters’ priorities and on politicians’ campaign estimates. And what we find when we do that is that it does seem to be the case that cable news is leading voters to care more about cultural issues and less about economic issues. And in turn, it’s leading politicians to talk more about cultural issues and less about economic issues.
And so kind of the number we calculate is around a third of the increase since 1996 in the importance of cultural issues relative to economic issues can be attributed to cable news. So of course there’s a lot of other things going on. There’s social media, there’s nine eleven, there’s the great financial crisis. Our estimate suggests that around a third of this change we can attribute directly to cable news.
Matt Grossmann: And this is a third of what? Of the increase in the public prioritizing these issues over economic?
Aakaash Rao: Exactly. So an increase among voters and the extent to which they say cultural issues are more important than economic issues and around a third of the increase among politicians and the extent to which they’re talking about cultural issues relative to economic issues.
Matt Grossmann: So Shakked, we said this isn’t the only explanation for the increases in culture war issue discussion and prioritization. And obviously cable news is not the only change in the media that’s occurring. And I would assume some similar logic appears even before cable and things like talk radio or newsletters or other forms of mobilization, the sort of long history of conservative media. So give us some of that history and how you think your findings fit in. Is this kind of a mechanism for a wider change in the media and the rise of the culture war? Or to what extent is this a really a much more specific story about 24 hour cable news channels?
Shakked Noy: Yeah. So I think the first thing to make clear is that nothing about our kind of conceptual story is specific to cable news. So the idea that news outlets have different incentives than politicians because they care about audience size rather than vote share is common to all news outlets. In fact, it’s more broadly common to what we might call size maximizing actors, which might also include actors like interest groups who care about their membership or the amount of contributions they receive, pundits, authors, things like that. So we don’t mean to apply that anything about our broader argument is specific to cable news. It’s just that it happens to be empirically convenient because it’s easy to get data on. We have a source of exogenous variation, as Aakaash just described, and it is one of the important forces shaping American politics in the past 30 years.
More broadly, how important cable news is relative to other forces is a really hard question to answer. So the first thing that you might want to do when trying to answer this question is try to figure out the timing of when cultural polarization starts to accelerate and see what that timing coincides with institutionally. The problem is that it’s really hard to do that because you can try and think of a variety of different measures of cultural polarization and the answer that you come up with in terms of when cultural polarization starts tends to vary according to the measure that you use.
So you can use educational polarization as a measure, which is something that you have a fantastic book about that I have next to me right now. The idea there is that voters without college degrees tend to be culturally conservative. Voters with college degrees tend to be culturally liberal. So the more that non-college people vote Republican, the more that college educated people vote Democrat, the more important cultural issues must be to people’s voting behavior. You can try to use the predictiveness of people’s positions on economic and cultural issues for their voting behavior as a measure of how important cultural issues are.
You can basically use a variety of these different measures and they all kind of disagree on the timing, but broadly it seems like cultural polarization seems to begin around the 1980s or the 1990s, continue throughout the 2000s and accelerate quite sharply in the 2010s.
In terms of what was happening in the 1980s and 1990s, one thing is the entry of cable news, and in particular Fox News and MSNBC in the mid 1990s. There’s also though the rise of televangelism and the religious right as a movement that tends to raise the salience of issues like abortion and religiosity. There’s also the abolition of the Fairness Doctrine in 1987, which enabled the entry of partisan talk radio shows like Rush Limbaugh that also tended to be quite heavily focused on culture war issues.
So we think probably all of these were important and all of these examples can be kind of unified by the basic conceptual point we’re making about the difference in incentives between size maximizing media outlets and vote share maximizing politicians and the fact that size maximization tends to favor an embrace of culture war issues.
Matt Grossmann: So Aakaash, we said that this might have precursors in other forms of media, but it also has further changes. The cable news networks are now recategorized as old media and there’s all kinds of new online and social media factors that might have similar incentives, but there’s been quite a bit of work trying to sort of blame that more specifically for the extension of the culture war.
So how does your work fit into that? Do we see these trends accelerating? Do you believe the research on the expansion of broadband and others that you would find similar kinds of dynamics here or is there something that would make this different?
Aakaash Rao: I think really the key story is the fragmentation of attention. And so in the network era, the 1950s to the 1980s, you basically have these three big networks and you have viewers that are captive in the sense that there’s not that much else to watch on TV. And so the transition from that network era to the cable era in the late ’70s, early ’80s, that constitutes fragmentation of viewer attention. That’s really significant because now these news producers have to compete not only with one another, but also with movies on HBO and sports on ESPN and then these new cable news networks, so CNN and then Fox News and MSNBC. And so that’s one move toward greater viewer fragmentation.
Then the rise of the internet, so late 1990s, early 2000s, the rise of broadband that that political scientists have written about, this constitutes a second fragmentation of viewer attention, and then smartphones and social media that’s even further. And so this is just kind of one trend that I think has been accelerating. And so cable news was an important step, but it was by far not the only step. And I think smartphones and social media really supercharged this trend in around 2012.
I think another way in which you can think of social media as a deepening of some of these existing trends is in what’s possible to do with analytics. And so cable news much more than broadcast news, particularly Fox News, really started paying attention to the minute by minute ratings. And so every day, and there’s a great New York Times piece on this specifically about Tucker Carlson a couple of years ago, describing the process by which the hosts and producers would sit down with the quantitative rating analysts and they would go through minute by minute with the show of the previous day and look at what they were talking about and look at what causes viewers to turn on and what causes viewers to turn off. And so the ability to do this really fine grained targeting of content towards viewer preferences, that really became possible kind of in the late ’90s, early 2000s, the cable networks were the first to do this.
And so social media, you can think of as that process on steroids because not only do you not need to do it manually anymore, you can do it with algorithms, the capability to target content toward individual viewers is much greater than it was on cable news. There was already a fragmentation of the viewer base. So of course, Fox News knows that it’s going to target a narrower subset of the population than the older broadcasters and it can optimize its content accordingly, but that’s quantitatively much less significant than the ability of, for example, Instagram or TikTok or Facebook to quantitatively target news content to you based on your previous interests.
Another way that social media represents a deepening is that cable news cared a lot less than broadcast about getting people upset, right? Broadcast was trying to target a pretty broad base, and so they wanted to really avoid extreme content that would turn people off. Fox News or MSNBC, they know that they’re targeting a smaller base. And so Fox News doesn’t particularly care if they say things that are going to offend Democrats. And similarly, MSNBC doesn’t really care if they say things that are going to offend Republicans.
Now, social media is interesting in that in some ways it’s that even more in that you kind of want to offend the other side. That’s going to blow up your engagement. In fact, the Oxford word of the dictionary Word of the Year this past year was ragebait. And so that kind of is a nice characterization of the incentives of social media is that you want to engage people and you don’t really care if you engage them because they like your content or they hate your content. And so that’s kind of another way in which social media is really furthering some of these trends that we’re talking about.
There’s no real reason it has to be this way. There’s no reason that social media had to lead to increased partisan political divisions rather than, for example, micro targeting, meaning that we don’t need to use this kind of sensationalism in order to engage viewers. We can really target viewers’ non-political interests or we can mix politics with their non-political interests in a way that reduces polarization, but that seems to be the equilibrium we’ve been in since the rise of algorithmic media starting around 2012.
Matt Grossmann: Shakked, you’ve given us some concrete examples of the policy issues that politicians talk about for kitchen table issues versus the guns and abortion for culture war. But obviously the culture war is about a bunch of non-policy considerations as well. We talk about everything from movies to music to standup comedy. To what extent are we actually talking about policy issues here in this coverage versus just a broader cultural dynamic? And what kind of concrete stories are there here where you said, “Oh, okay, this is a broader dynamic here, but this person at this cable news channel really went all in on this and some politicians noticed.”?
Shakked Noy: Yeah, so politicians mostly are focused on policy. I mean, often they’re just saying, “I’m going to create jobs for this district without being particularly specific about how they’re going to do that,” but you can basically treat that as speaking about policy. But on the news, that’s not true at all. And as you mentioned, what we’ve increasingly seen over the past couple of decades is the sort of fusion of entertainment and politics where every facet of life becomes infected with these kinds of identitarian disputes over what it means to be American and what kinds of displays we should support or should oppose and so on and so forth. So we really see that on the news where there’s a lot of discussion of racial and gender issues, for example, or expressions of anti-elite sentiment in the context of entertainment and media and events that you would think have fairly little to do with policy. And that’s kind of one of the ways that we think that culture war content on the news is good at attracting people away from entertainment channels, mobilizing them away from watching entertainment into watching the news because culture war news can be quite similar to entertainment in many ways. So that’s certainly the case.
It’s hard to say the extent to which we see politicians leaning into this kind of thing. So in recent years, we’ve certainly seen a lot of examples on social media of politicians, particularly on the right, going very viral for leaning into these kind of performative culture war stances that are about the M&Ms not being sexy anymore or things like that. But that’s something we don’t really see in the campaign ads. So when the rubber hits the road and when a politician is running for reelection, they tend, even in 2022, which is the last year of our data, to talk mostly about policy issues, mostly about economic issues. And when they do talk about culture war issues, it tends to be through a policy lens. They’re talking about immigration policy or gun control or things like that, whereas that’s not true on the news at all.
Another thing that we do see changing over time is that the broadcast news networks, which predated the cable news networks and which tended historically to talk more about economics, have been converging towards the new strategy of the cable news networks and talking more about these kinds of culture war issues and entertainment inflected culture war issues. And we think this is probably because the role of these broadcast news divisions used to be to be as kind of loss leaders that generated prestige for the network without necessarily directly making money. But over time, the pressure on these news divisions to turn a profit and the economic pressures faced by these broadcast networks have increased. So the pressure on them to generate more viewership and to generate more profit directly has increased. And they’ve figured out, as the cable news networks did, that the way to do that is to talk about cultural issues.
Matt Grossmann: A bit of a tension with one previous finding in cable news topics, which is that they spend a lot of their time talking about mainstream media, that is they are trying to differentiate themselves. And that might suggest more of a poaching strategy and also might just suggest that there’s kind of a long-term strategy that might be different than the short-term strategy of get the viewers this minute is have people only trust the host and everyone else is kind of keeping the story from them. So what do you make of that, given your findings?
Shakked Noy: I think an alternative interpretation of that fact is that these cable networks are trying to attract populations who have historically felt alienated from mainstream news and have not engaged with mainstream news. And the way to attract those kinds of people is to talk about how the mainstream media is corrupt and lying to you. So that critique of mainstream media is not a way to actually persuade viewers of mainstream media to switch and watch you, but it’s a way to appeal to historically disaffected parts of the population who did not engage with mainstream media and a way to get them fired up. So that actually ultimately would be a strategy of mobilization through critique of the mainstream media rather than poaching.
Aakaash Rao: And one place, just to add, one place where we see that is if you look at consumer survey data, so if you look at who is taking up these new cable news networks in the 1990s when Fox News and MSNBC emerge, and you look at the types of consumers who start watching, say, between 1996 and 2000, so the very early adopters of Fox News and MSNBC, you can ask, who are these early adopters? Are they the kinds of demographic groups that were previously watching a lot of news?
And it turns out the answer is no. So cable news networks are disproportionately not poaching from the broadcast networks that existed before. They’re disproportionately drawing people who were previously not watching a lot of news and were instead watching entertainment like sports and sitcoms and things like that. So to us, that suggests that even from their very inception, cable news was really pursuing this mobilization based strategy.
One aspect of that mobilization based strategy was talking a lot about these cultural issues. Another aspect was kind of fusing entertainment with politics. And exactly as you’re alluding to, a third aspect was talking about how these groups were right all along to distrust mainstream outlets and they should trust these new cable networks that have emerged.
Matt Grossmann: Aakaash, we’ve been talking about this as a broad phenomenon of cable news and viewers, but I wonder what interesting variations you found. As you know, a lot of the previous literature found bigger or even the only effects for Fox News versus MSNBC. There’s also a lot of talk about the eras changing with the election of Donald Trump in 2016 or 2015. How much are we seeing changes over time in this process, in what kinds of viewers are mobilized? And is there really a symmetry between what’s happening on the right and the left here?
Aakaash Rao: Yeah, so one really striking difference that you see now is that when you look at what voters say is important, there are huge differences between Republicans and Democrats that did not exist before the 1990s. So if you look at, say, the 1950s, ’60s, ’70s, ’80s, the years covered by this Gallup, what is the most important problem facing in the country? In general, Democrats and Republicans tend to move pretty much in lockstep. In the late ’60s and early ’70s, everyone is kind of concerned about these disputes over race. In the ’90s, everyone is kind of concerned about this spike in crime. And starting around the entry of cable news, you see greater divergence between Republicans and Democrats in what they care about.
So you might think that that would translate also into what cable networks are talking about, and what’s surprising is that it doesn’t really. We went into this project expecting to find that Fox News is talking a huge amount more about immigration, and MSNBC is talking a huge amount more about climate change and racial disparities in policing. What it turns out is that these cable networks have turned not caring about an issue into an issue in and of itself.
And kind of taking a step back, it’s perhaps not that surprising. Fox News spends a tremendous amount of time talking about how there aren’t racial disparities in policing and how the police are being treated unfairly. And MSNBC spends a huge amount of time talking about how ridiculous it is that Fox News is spending so much time on the war on Christmas, things like this. So for that reason, if you just look kind of issue by issue and you compare Fox and MSNBC, you don’t see huge effects.
What does seem to be the case is that Fox News has larger effects on both voters’ beliefs about what is important and on politicians in terms of what politicians talk about. So this is something that, as you say, previous work has found as well. And again, it’s perhaps not that surprising because Fox News just has a much larger viewer base. And this is particularly true in the 2000s, mid and late 2000s, but it remains true now that Fox News has a much larger viewer base than MSNBC or CNN.
Part of that is we think that Fox News was just historically a lot more effective in fusing entertainment and politics. I think another part is that they kind of had an advantage coming in because conservatives in the 1990s, even as early as the 1980s, disproportionately distrusted mainstream outlets. And that’s kind of a divergence that maybe started around Nixon or maybe even earlier, but Fox News was really able to tap into that conservative distrust and kind of had a big latent demand for its product when it entered. And MSNBC just didn’t have that same kind of huge consumer base that it could tap into. So I think part of Fox News’ larger effects that we estimate is just that a lot more people are watching Fox News.
Shakked Noy: And just to add a little bit on the question of how symmetric the left and the right are in terms of the issues that we’re studying, if we look at politicians, as we’ve mentioned, we see that politicians start to talk more about culture war issues since about 2012. This increases a little bit larger for Republicans than it is for Democrats. And then obviously Donald Trump is kind of a singular figure in terms of his ability to attract a lot of attention to culture war issues. So there seems to be a little bit more happening on the Republican side on that front.
And then in terms of the news, as Aakaash just mentioned, Fox News is kind of a bit more of a monolith that dominates the right than MSNBC is on the left. But on the other hand, there are a variety of forces that we’re not directly studying, as you describe in your book, for example, in terms of the influence of left wing professionals who aren’t Democratic politicians, but who are often pushing left wing social and cultural ideas in their workplaces, on social media, in a variety of other venues. So it’s hard to say whether the kind of increase in the salience of cultural issues was driven more by the left or by the right.
Matt Grossmann: So you started out saying that this is an intervention in the debate about polarization and its rise, but we’ve kind of honed in on this more specific rise in the discussion of culture war issues relative to economic issues. Obviously that might be related or might not be related to kind of distancing between the parties or distancing between the voters in the parties. So fit this into the broader polarization literature for me. How big of a deal is the rise of cultural issues relative to economic issues in your thinking about polarization?
Shakked Noy: Yeah, so as you know, it’s important to distinguish exactly what we mean by polarization when we talk about it. So there are a few different kinds of polarization. There’s elite polarization in terms of how correlated Congressional voting is with political parties. That seems to start around the mid-century and seems to be kind of a top-down elite-driven phenomenon.
Then there’s affective polarization, which is what a lot of the polarization literature is interested in terms of basically how much voters of the two parties dislike each other. It seems plausible to us that this is connected to cultural polarization, meaning the rising influence of cultural issues relative to economic issues in politics. The reason is basically that cultural issues are kind of more deep-rooted in people’s identities and moral values. So it seems probably easier to hate somebody because you think they’re a racist or you think they’re a murderer because they support abortion than it is to hate somebody because they disagree with you about tax policy or optimal labor market regulations. So it seems plausible to us that there’s a connection between the rise of cultural issues in politics and the rise of affective polarization, but we don’t have any specific evidence establishing that connection.
More broadly in terms of how we think about debates about the rise of polarization, one kind of general insight that we think the distinction between poaching and mobilization offers is the fact that politicians are so incentivized to prioritize poaching suggests that they should really stick to the kind of median of whatever the status quo cleavage in politics is. So if politics is split on an economic axis, politicians should stick close to the median of that economic access because that’s where poaching can happen, where you can get close to the voters of the other party and try to poach them away and convince them to switch to you.
By contrast, the kinds of actors who should be more incentivized to stray away from the median of the status quo cleavage and try to bring in people who aren’t currently participating and mobilize new people into participating should be size maximizing actors who care less about poaching and more about mobilization than politicians. So one general implication of our theory is that polarization or issue realignment should be led by these kinds of size maximizing actors, whether they’re media outlets or activists or interest groups, because those are the groups who have more of an incentive to stray away from the median of the status quo cleavage and kind of go elsewhere in search of new people to mobilize and energize into participation.
Matt Grossmann: Aakaash, the relative polarization between the major political parties differs across the world, but it seems to me that this basic distinction that you’re making doesn’t differ a lot, that the trends across democracies are towards more culture war discussion relative to economic discussion. So on the one hand, we could use your findings to say, “If we think the US did this first or moreso, then maybe this helps to explain it.” On the other hand, maybe it suggests that this was a little more inevitable than linking it to the specific changes in the media environment around cable news. So how are you thinking about how the US, to the extent to which the US stands out internationally in this phenomenon and what that says about your story?
Aakaash Rao: Yeah. So it’s certainly the case that the culture war in the US started earlier and remains deeper than the culture war in other countries. So that’s absolutely true as you’re saying that the culture war in other countries has increased in recent years. It’s largely about immigration if you look at Western Europe. And there are other cultural issues that are important, but really immigration is the driving force behind the rise of far right parties in Germany and the UK and Sweden and other Western European countries. The culture war in the US, immigration is obviously important, but it extends to a whole host of other issues that just aren’t really on the political spectrum in other democracies.
Gun rights and abortion and a lot of aspects of religion and free speech, things that aren’t even in the political discourse in other countries. So the way we see it is that the transition to cable news and the unique media environment in the US that facilitated the rise and political importance of cable news, that kickstarted the culture war, made it a little earlier and broader in the US. And then the subsequent rise of social media and internet news, which was more common across countries and happened around 2010 or the early 2010s, that extended this cultural polarization to other countries.
Matt Grossmann: But doesn’t the poaching logic seems to apply less to politicians in other countries and multi-party systems, and there seem to be more size maximizing actors elsewhere. So it makes me think that we should have expected this to occur elsewhere earlier. I’m maybe asking you to explain too much here, but to me, it makes the likelihood that it’s about specific changes in our media and specific actions in cable news a little bit less likely.
Aakaash Rao: Yeah. I think so the mapping from poaching versus mobilization to two-party versus multi-party is something we’ve thought a lot about, and it turns out to be a lot more complicated than you might think at first. So for example, in a two-party system, suppose that everyone were able to vote or were forced to vote in a two-party system, then poaching and mobilization in this mandatory voting regime, those are essentially the same thing because everyone has to vote. And so either they vote for you or your opponent or not at all. Now, when you add multiple parties, then that complicates this, because now poaching and mobilization are no longer the same thing. And so getting people to not vote for another party in a multi-party system is less valuable than it is in a two-party system because you’re going to divert them away from voting for the other party, not only to you. And bring them not only to you, but also to a lot of other parties. So for a lot of different reasons, this poaching versus mobilization mapping to multi-versus two-party systems is not super clear.
Shakked Noy: Although-
Aakaash Rao: One thing that you … Go ahead.
Shakked Noy: I’ll just briefly intervene to note that it is the case that the parties and multi-party systems with an exposed flank, they’re just mobilizing on one side of themselves rather than poaching on both sides. They do tend to be the parties that are more focused on cultural issues. You think of far right parties very focused on immigration or green parties that are very focused on social and cultural issues and climate change.
Aakaash Rao: Yeah.
Shakked Noy: Sorry, continue.
Aakaash Rao: No, no, no. And I think just the other aspect of that is if you look cross country at the correlates of increases in affective polarization, one thing that you do find is that government support of media seems correlated with lower increases in affective polarization. And so of course, the US is the outlier in that and that we have very little government support of media, whereas countries like Germany or the UK have much greater government support of media. So that does seem to be a correlate. It’s not smoking gun evidence that the media environment is the key driver of changes in affective polarization, but at least the extent to which you have private ownership of 24-hour cable news channels does seem correlated with increases in affective polarization.
Shakked Noy: Another point to make about multiparty democracies, because I think it’s a very good question, is that I think that historically most of these countries were closer to two-party systems than they are now in the sense that there tended to be a dominant social Democratic Party and a dominant center right Christian Democratic Party. And so the rise of cultural war issues in these places has coincided with fragmentation into many smaller parties, and potentially that goes along with an increased focus on mobilization.
Matt Grossmann: I guess on the other side, there does seem to be evidence that American culture war issues spread, and obviously that the cable news infrastructure also has spread. Is there a story here that this really actually has wider international significance in that sort of figuring out what works here to build an audience helped spread issues internationally to the point that we had a weird phenomena where we had Black Lives Matter protests around the world in response to US dynamics, even when they weren’t as good a fit for the home country politics.
Shakked Noy: Exactly. I think the cultural dominance of the US globally is such that you get these weird things like a bunch of people on Twitter who live in foreign countries talking about US politics on Twitter all of the time. I grew up for about half of my childhood and adolescence in New Zealand. And my friends and I would talk much more about US politics than about New Zealand politics. And as you say, we do seem to see a lot of anecdotal evidence of this situation where culture war disputes that are incubated in American politics then get exported to other places.
And if you look at the culture war rhetoric that right-wing parties in Europe have recently adopted, much of it is modeled on rhetoric about trans people, for example, that was initially developed by Republican activists and politicians. And then you get other things like you get Republican political advisors, I think his name is Arthur Finkelstein, who worked for the Republican Party in the 1990s and then went to advise Viktor Orban and Benjamin Netanyahu and things like that. And his trips abroad to advise them seem to coincide with an increased focus on culture war and anti-elite rhetoric from those politicians. So there does seem to be kind of a variety of anecdotal evidence that the US is exporting the culture war in some sense, and that the rise of the culture war in other countries is downstream of the rise of the culture war in the US.
Matt Grossmann: How should we think about this normatively? You’ve used the negative valence term of the culture war to see what you’re trying to explain, but your words, mobilization actually has a valence than poaching. So there at least seems to be some case that we want people to be involved in politics. We used to complain about low levels of involvement in politics. Maybe be careful what you wish for, but to what extent have the cable news channels found a more effective way to engage people in politics that should be celebrated? And how are you thinking about the trade-offs here?
Shakked Noy: So I think certainly part of our story is that the cable news networks are responding to some latent demand that people have to consume this kind of entertaining culture war content. So if you were a neoclassical economist, you would stop there and say, “Oh, well, this must be good then because supply is responding to demand, people maximize their utility, so everything must be great.” I think there’s a few reasons to be suspicious of that kind of conclusion. One is that the kind of content that viscerally engages and activates people is maybe not the kind of content that’s actually good for them if they stepped back and made a more reflective decision about it. Another is that there might be externalities where everybody might individually love to listen to content about the culture war, but if all of politics becomes about this, the broader functionality of politics breaks down the ability to make effective policy to craft bipartisan legislation and so on breaks down.
And so even if everybody’s individually optimizing, this can lead to failures at a collective level. And then more specifically on the question of mobilization, an interesting fact is that this period in which the news seems to have leaned much more into mobilization has actually coincided with declining political participation, declining political knowledge, and just absolutely shocking increases in cynicism and disaffection. If you look at Gallup polls about how much people trust Congress or how much they trust institutions, things are really, really bad. And one potential explanation for that is that a kind of short-term policy of firing people up, inflaming them in order to mobilize them into engaging with the news in the short run is really good for engagement, but in the long run exhausts people, causes them to become alienated from and suspicious of politics. And so there’s this collective action problem where it’s best for each news network to try to maximize its own engagement by doing this super inflammatory stuff, but it’s bad for the whole ecosystem in the long run because it causes people to just burn out.
Aakaash Rao: And I think the FCC, their entire mission is predicated on this belief that at least in the media landscape, just letting the market regulate itself is not going to lead to an efficient outcome. And the FCC was obviously much more involved and active in the broadcast era. And part of that was because these broadcasters were using a public resource, which was the public spectrum, and so the FCC had a greater mandate to regulate them. But part of it was also a very intentional Reagan era turn away from regulation, culminating, as [inaudible 00:56:20] said in the repeal of the Fairness Doctrine in 1987. And so partly what we think has gone on is that there’s been a decrease in regulation coinciding with an increase in the amount of competition just generated by some of these technological changes. So the expansion of cable and then internet and smartphones and social media.
And so these two things together have meant that there’s much greater scope than before for unfettered and now algorithmic optimization for short-term viewer attention. And there’s no real reason to think that that’s the same as benevolent or omniscient optimization for a social optimum in politics.
Matt Grossmann: So I want to give each of you a chance to talk about what you’re working on now and how this fits in to your broader research, and anything else that we didn’t get to that you wanted to include.
Shakked Noy: Yeah. So we’re actually, we’re working together on another project. I think we’ve become interested in the rise of the attention economy, the fact that people in advanced democracies spend much of their days consuming content online. And it’s a very interesting space to study precisely because as I mentioned earlier, the kinds of content that engage and attract attention are not necessarily the same as the kinds of content that are good for people from a more reflective perspective. So we have a project now about social media that’s trying to dig into why social media seems to make people so unhappy. So we’ll report back when we’ve figured that out.
Matt Grossmann: There’s a lot more to learn. The Science of Politics is available biweekly from Niskanen Center, and I’m your host, Matt Grossman. If you like this discussion, here are the episodes I recommend checking out next. How Fox News Channel spreads its message and persuades viewers. Are claims that social media polarizes us overblown? How news and social media shape American voters. How the media economy drives local news. And how racial realignment ignited the culture war. Thanks to Aakaash Rao and Shakked Noy for joining me. Please check out the Business of the Culture War and then listen in next time.
