Shakked Noy
@shakkednoy.bsky.social
Paper here: https://www.dropbox.com/scl/fi/27v4xxenpo8wotkw1238e/rao-jmp.pdf?rlkey=x8rg0slqo9mwfhdlbqgftaw4b&e=1&dl=0
Comments/feedback welcome!
Comments/feedback welcome!
November 1, 2025 at 10:35 PM
Paper here: https://www.dropbox.com/scl/fi/27v4xxenpo8wotkw1238e/rao-jmp.pdf?rlkey=x8rg0slqo9mwfhdlbqgftaw4b&e=1&dl=0
Comments/feedback welcome!
Comments/feedback welcome!
Social media might explain the recent rise of (more immigration-focused) culture wars in Europe. We think the culture wars in other liberal democracies arrive later and are less intense than in the US, which we attribute to the (relatively unique to the US) phenomenon of partisan TV news.
November 1, 2025 at 10:35 PM
Social media might explain the recent rise of (more immigration-focused) culture wars in Europe. We think the culture wars in other liberal democracies arrive later and are less intense than in the US, which we attribute to the (relatively unique to the US) phenomenon of partisan TV news.
While our empirical focus is cable news, our poaching-vs-mobilization framework equally implies that social media content creators should have stronger incentives than politicians to lean into culture-war issues.
November 1, 2025 at 10:35 PM
While our empirical focus is cable news, our poaching-vs-mobilization framework equally implies that social media content creators should have stronger incentives than politicians to lean into culture-war issues.
A back-of-the-envelope extrapolation of our estimates suggest that Fox News and MSNBC alone can explain ~1/3rd of the increase in cultural polarization according to these metrics since 2000.
Of course, the fragmentation of the media environment continued with the Internet and social media.
Of course, the fragmentation of the media environment continued with the Internet and social media.
November 1, 2025 at 10:35 PM
A back-of-the-envelope extrapolation of our estimates suggest that Fox News and MSNBC alone can explain ~1/3rd of the increase in cultural polarization according to these metrics since 2000.
Of course, the fragmentation of the media environment continued with the Internet and social media.
Of course, the fragmentation of the media environment continued with the Internet and social media.
In constituencies more exposed to cable news, people report more culture-war issues and fewer economic issues in response to Gallup's "Most Important Problem" surveys.
Local politicians respond by airing more culture-war campaign ads.
Local politicians respond by airing more culture-war campaign ads.
November 1, 2025 at 10:35 PM
In constituencies more exposed to cable news, people report more culture-war issues and fewer economic issues in response to Gallup's "Most Important Problem" surveys.
Local politicians respond by airing more culture-war campaign ads.
Local politicians respond by airing more culture-war campaign ads.
In places where they were assigned a lower channel number, more people watch cable news (because of channel-surfing near the popular channels).
This lets us instrument for cable news viewership in an area with the local channel numbers of Fox News and MSNBC.
This lets us instrument for cable news viewership in an area with the local channel numbers of Fox News and MSNBC.
November 1, 2025 at 10:35 PM
In places where they were assigned a lower channel number, more people watch cable news (because of channel-surfing near the popular channels).
This lets us instrument for cable news viewership in an area with the local channel numbers of Fox News and MSNBC.
This lets us instrument for cable news viewership in an area with the local channel numbers of Fox News and MSNBC.
Question 2: did the entry of cable news, and the concomitant increase in culture-war coverage on TV, affect politics?
We borrow the instrument of Martin and Yurukoglu (2017). In different media markets, the cable networks were assigned different channel numbers for arbitrary reasons.
We borrow the instrument of Martin and Yurukoglu (2017). In different media markets, the cable networks were assigned different channel numbers for arbitrary reasons.
November 1, 2025 at 10:35 PM
Question 2: did the entry of cable news, and the concomitant increase in culture-war coverage on TV, affect politics?
We borrow the instrument of Martin and Yurukoglu (2017). In different media markets, the cable networks were assigned different channel numbers for arbitrary reasons.
We borrow the instrument of Martin and Yurukoglu (2017). In different media markets, the cable networks were assigned different channel numbers for arbitrary reasons.
The spread of cable TV in the 1980s and 1990s disrupted this equilibrium, introducing dozens of entertainment options that competed for attention and allowing the entry of 24-hour networks focused entirely on news.
We argue that more ruthless attention-maximization -> culture war coverage
We argue that more ruthless attention-maximization -> culture war coverage
November 1, 2025 at 10:35 PM
The spread of cable TV in the 1980s and 1990s disrupted this equilibrium, introducing dozens of entertainment options that competed for attention and allowing the entry of 24-hour networks focused entirely on news.
We argue that more ruthless attention-maximization -> culture war coverage
We argue that more ruthless attention-maximization -> culture war coverage
This arguably supported an equilibrium of vertical competition over dry, factual "hard news" focused more on economic issues and less on culture war issues.
(Of course, even then, the broadcast networks were criticized for their occasional sensationalism.)
(Of course, even then, the broadcast networks were criticized for their occasional sensationalism.)
November 1, 2025 at 10:35 PM
This arguably supported an equilibrium of vertical competition over dry, factual "hard news" focused more on economic issues and less on culture war issues.
(Of course, even then, the broadcast networks were criticized for their occasional sensationalism.)
(Of course, even then, the broadcast networks were criticized for their occasional sensationalism.)
... were loss-leaders focused on burnishing the network's prestige and satisfying regulators that the network was behaving in the public interest. (see quote below by Marc Gunther, 1999)
The broadcast networks also faced very little competition for attention. They were the only game in town.
The broadcast networks also faced very little competition for attention. They were the only game in town.
November 1, 2025 at 10:35 PM
... were loss-leaders focused on burnishing the network's prestige and satisfying regulators that the network was behaving in the public interest. (see quote below by Marc Gunther, 1999)
The broadcast networks also faced very little competition for attention. They were the only game in town.
The broadcast networks also faced very little competition for attention. They were the only game in town.
Cable news outlets' *size*-based objective functions favor mobilization more, so they talk about the culture war more than politicians.
But what about broadcast news outlets?
Prior to the 1990s, broadcast networks broadcast 1-2 hours of primetime news each day, and their news divisions...
But what about broadcast news outlets?
Prior to the 1990s, broadcast networks broadcast 1-2 hours of primetime news each day, and their news divisions...
November 1, 2025 at 10:35 PM
Cable news outlets' *size*-based objective functions favor mobilization more, so they talk about the culture war more than politicians.
But what about broadcast news outlets?
Prior to the 1990s, broadcast networks broadcast 1-2 hours of primetime news each day, and their news divisions...
But what about broadcast news outlets?
Prior to the 1990s, broadcast networks broadcast 1-2 hours of primetime news each day, and their news divisions...
We find no evidence that more-economic campaigns raise turnout, but they do seem to successfully poach voters from the opponent.
So our evidence suggests that economics is better for poaching, while culture is better for mobilization (at least on TV)
So our evidence suggests that economics is better for poaching, while culture is better for mobilization (at least on TV)
November 1, 2025 at 10:35 PM
We find no evidence that more-economic campaigns raise turnout, but they do seem to successfully poach voters from the opponent.
So our evidence suggests that economics is better for poaching, while culture is better for mobilization (at least on TV)
So our evidence suggests that economics is better for poaching, while culture is better for mobilization (at least on TV)
But we do our best, exploiting variation in campaign advertising across 3,000 House, Senate, and Governor candidates and interpreting results as suggestive (not necessarily causal).
In politics, economically-focused campaigns perform _better_ than culture-war focused ones.
In politics, economically-focused campaigns perform _better_ than culture-war focused ones.
November 1, 2025 at 10:35 PM
But we do our best, exploiting variation in campaign advertising across 3,000 House, Senate, and Governor candidates and interpreting results as suggestive (not necessarily causal).
In politics, economically-focused campaigns perform _better_ than culture-war focused ones.
In politics, economically-focused campaigns perform _better_ than culture-war focused ones.
If a similar tradeoff exists in politics, we should see the opposite performance effect: culture war issues should underperform economics one.
Identification in politics is difficult (due to absence of high-frequency variation), arguably impossible (candidates are bundled treatments).
Identification in politics is difficult (due to absence of high-frequency variation), arguably impossible (candidates are bundled treatments).
November 1, 2025 at 10:35 PM
If a similar tradeoff exists in politics, we should see the opposite performance effect: culture war issues should underperform economics one.
Identification in politics is difficult (due to absence of high-frequency variation), arguably impossible (candidates are bundled treatments).
Identification in politics is difficult (due to absence of high-frequency variation), arguably impossible (candidates are bundled treatments).
... from other news outlets.
The mobilization gain is sufficiently larger than the poaching loss that a *size*-maximizing news outlet prefers culture-war content on the margins.
A *share*-maximizing politician faced with the tradeoff graphed above, though, would pick economics
The mobilization gain is sufficiently larger than the poaching loss that a *size*-maximizing news outlet prefers culture-war content on the margins.
A *share*-maximizing politician faced with the tradeoff graphed above, though, would pick economics
November 1, 2025 at 10:35 PM
... from other news outlets.
The mobilization gain is sufficiently larger than the poaching loss that a *size*-maximizing news outlet prefers culture-war content on the margins.
A *share*-maximizing politician faced with the tradeoff graphed above, though, would pick economics
The mobilization gain is sufficiently larger than the poaching loss that a *size*-maximizing news outlet prefers culture-war content on the margins.
A *share*-maximizing politician faced with the tradeoff graphed above, though, would pick economics
... but it substantially _increases_ the flow of people switching to entertainment channels or turning off the TV (i.e., incurs a mobilization loss).
So, on TV, culture war coverage is better for mobilization (attracting people from entertainment channels), but worse for poaching...
So, on TV, culture war coverage is better for mobilization (attracting people from entertainment channels), but worse for poaching...
November 1, 2025 at 10:35 PM
... but it substantially _increases_ the flow of people switching to entertainment channels or turning off the TV (i.e., incurs a mobilization loss).
So, on TV, culture war coverage is better for mobilization (attracting people from entertainment channels), but worse for poaching...
So, on TV, culture war coverage is better for mobilization (attracting people from entertainment channels), but worse for poaching...
But crucially, our smart TV data lets us observe people as they *switch between* different channels.
Relative to cultural coverage, economic coverage _reduces_ the flow of people switching to competing news channels (i.e., it achieves a poaching gain)...
Relative to cultural coverage, economic coverage _reduces_ the flow of people switching to competing news channels (i.e., it achieves a poaching gain)...
November 1, 2025 at 10:35 PM
But crucially, our smart TV data lets us observe people as they *switch between* different channels.
Relative to cultural coverage, economic coverage _reduces_ the flow of people switching to competing news channels (i.e., it achieves a poaching gain)...
Relative to cultural coverage, economic coverage _reduces_ the flow of people switching to competing news channels (i.e., it achieves a poaching gain)...
(We can even look at variation in content _within_ a given one-hour episode!)
Shifting from an all-economic to all-cultural 15-minute segment reduces viewership by about 2.2%, roughly one-sixth the penalty associated with a commercial advertising break.
Shifting from an all-economic to all-cultural 15-minute segment reduces viewership by about 2.2%, roughly one-sixth the penalty associated with a commercial advertising break.
November 1, 2025 at 10:35 PM
(We can even look at variation in content _within_ a given one-hour episode!)
Shifting from an all-economic to all-cultural 15-minute segment reduces viewership by about 2.2%, roughly one-sixth the penalty associated with a commercial advertising break.
Shifting from an all-economic to all-cultural 15-minute segment reduces viewership by about 2.2%, roughly one-sixth the penalty associated with a commercial advertising break.
We link household-by-second level viewership data collected from smart TVs to the contemporaneous content airing on news channels and use high-frequency variation in content (at the 15- or even 5-minute level) to identify the effects of shifts from economic to cultural coverage.
November 1, 2025 at 10:35 PM
We link household-by-second level viewership data collected from smart TVs to the contemporaneous content airing on news channels and use high-frequency variation in content (at the 15- or even 5-minute level) to identify the effects of shifts from economic to cultural coverage.
So if culture-war issues are better for mobilization and economic issues better for poaching, this difference in objective functions could explain the observed differences in issue focus.
Is this the case? At least on TV, we show that it is.
Is this the case? At least on TV, we show that it is.
November 1, 2025 at 10:35 PM
So if culture-war issues are better for mobilization and economic issues better for poaching, this difference in objective functions could explain the observed differences in issue focus.
Is this the case? At least on TV, we show that it is.
Is this the case? At least on TV, we show that it is.
Broadly, news outlets are distinguished by their relatively greater incentive to mobilize *new attention*, whereas politicians are distinguished by their greater incentive to stick to the median of status quo cleavages in order to focus on poaching
November 1, 2025 at 10:35 PM
Broadly, news outlets are distinguished by their relatively greater incentive to mobilize *new attention*, whereas politicians are distinguished by their greater incentive to stick to the median of status quo cleavages in order to focus on poaching
... while a politician interested in maximizing their vote *share* values each poached voter twice as much as each mobilized voter.
In this example, a politician will prefer the poaching-focused strategy at the top, while a cable news outlet will prefer the mobilization-focused strategy below
In this example, a politician will prefer the poaching-focused strategy at the top, while a cable news outlet will prefer the mobilization-focused strategy below
November 1, 2025 at 10:35 PM
... while a politician interested in maximizing their vote *share* values each poached voter twice as much as each mobilized voter.
In this example, a politician will prefer the poaching-focused strategy at the top, while a cable news outlet will prefer the mobilization-focused strategy below
In this example, a politician will prefer the poaching-focused strategy at the top, while a cable news outlet will prefer the mobilization-focused strategy below
On the margins, they can get viewers/voters by "poaching" from competitors or by "mobilizing" those who otherwise wouldn't participate (wouldn't vote/watch news).
But a cable news outlet interested in maximizing its audience *size* is indifferent between a poached and mobilized viewer...
But a cable news outlet interested in maximizing its audience *size* is indifferent between a poached and mobilized viewer...
November 1, 2025 at 10:35 PM
On the margins, they can get viewers/voters by "poaching" from competitors or by "mobilizing" those who otherwise wouldn't participate (wouldn't vote/watch news).
But a cable news outlet interested in maximizing its audience *size* is indifferent between a poached and mobilized viewer...
But a cable news outlet interested in maximizing its audience *size* is indifferent between a poached and mobilized viewer...