Walter Cronkite
Fox News commentary host Tucker Carlson and CNN news anchor Don Lemon—both highly paid network stars for their audience ratings—were summarily fired during the same week in April 2023 for undisclosed reasons. Regardless of their professional and/or personal indiscretions (if any), their departures are notable because the journalists fell from grace in networks with polar-opposite politics, each with stridently contrasting ideological takes on the same news stories, as well as barbs about the other’s network.

The expanding ideological and partisan divide in television journalism has been largely explained by the growing political polarization of voters, which is, no doubt, an operative force I cover. But here, I explain the growing political divide among newscasts by drawing on the insights of the late public choice economist Gordon Tullock who suggested two additional forces at work: (1) growth in the range of substantive policy issues considered to be in the “public square” and (2) the expansion in the array of news outlets. These forces have pressed newscasts to diverge on their ideological and partisan slants on news stories.

The Homogenization of News in the 1960s

In the 1960s, network newscasts were only 15 minutes long and accessible by only 9 percent of homes. They covered similar stories, with all three broadcast networks—NBC, CBS, and ABC—having no overt political or ideological differences in their on-air treatments of news. In this earlier epoch, journalists—epitomized by CBS anchor Walter Cronkite—seemed genuinely committed to “reporting the facts.” If there was a partisan divide between the NBC anchors Chet Huntley and David Brinkley and CBS anchor Walter Cronkite, it was never so great that they traded on-air hostilities.

Why have the political and ideological divides among news programs emerged?

Explanations abound. One popular explanation, considered later, emphasizes the expanding political and partisan divisions tied to the growing diversity of racial, ethnic, and political groups in the electorate. Here, I seek to explain the growing divide by stressing economic forces, the dramatic expansions in the number of issues taken up in policy debates, and the growth in the array of news outlets.

Growth in Policy Issues

In the 1960s, strong political divisions among Americans were hardly nonexistent. The Vietnam War, civil rights protests, and President Lyndon Johnson’s Great Society policies were divisive major issues, and they dominated newscasts for the entire decade. In the following decades, these issues never faded, given the intervening spread of U.S. military operations around the world, racial tensions, and the growth in federal spending, among other events. Between 1980 and April 2023, federal outlays grew from a fifth to a fourth of gross domestic product, despite GDP rising almost ten-fold during the period. Without doubt, growth in federal spending has translated into growth in contentious policy issues (for example, student-loan forgiveness) debated in Congress and media circles. But few people in the 1960s (and before) ever thought about gay marriage, transgender bathroom rights, defunding the police, and late-term abortion—to name several contemporary hot-button policy issues.

The Growth in Media Outlets

Since the 1960s, the number of media outlets has grown substantially. In the 1960s, there were only two major television networks that divided most of the news audience, NBC and CBS, with ABC coming in a distant third. And these two (or three) networks broadcast their signals for news and entertainment programs, which meant competition for viewers’ attention was constrained and kept limited by FCC’s entry (and content) restrictions and by cable companies’ limited cable lines. (The percentage of households with cable didn’t reach 60 percent until 1989.)

Today, after the gradual breakdown of FCC entry barriers to cable television, Americans have access to hundreds of cable channels (as well as millions of news-related podcasts accessible worldwide via the internet). The contraction of cable subscriptions over the last decade or so with the advent of streaming services has likely intensified the competition among newscasts for viewers, likely pressing them to become ever-more divided.1

Median Voter Model: One Issue, Two Candidates

In 1948, Scottish economist Duncan Black laid the foundation for public economics by postulating that in a majoritarian democracy with only two candidates seeking election on a single issue, the candidate whose policy position best matches the policy preferences of the “median voter” will tend to win the election. This point is illustrated with the aid of Figure 1, which covers a normal distribution of voters on a left-right/liberal-conservative political spectrum.

Figure 1. Normal Distribution, Voters, Left-Right/Liberal-Conservative

If the two candidates—tag them D and R—stake out positions in the right wing of the distribution, say, at D’ and R’, they each will face pressure from the other candidate to move toward the center. At D’ and R’, R will receive all the votes to the right of R’ and half the votes between D’ and R’. D will also receive half of the votes between them D’ and R’, plus all the votes to the left of D’. D would then obviously win by a super majority.

R can change that outcome by moving just to the left of D’, a then-winning position, but only if D stays put, which is unlikely. D can also move to the left of R’s new position. The two candidates can be expected to hopscotch one another until they reach the middle, taking centrist positions such as D and R. Each can be expected to squeeze out as many votes as possible by moving even closer together (but only if they can do so and still give voters a reason to vote). If the candidates take identical positions, voters will have no reason to incur the costs of voting; the same partisan ideology (or the same police expenditures) would be adopted, no matter who wins.

The analysis can be applied with ease to the varying preferences of television viewers for, say, news reports on the contested size of the police budget. If there are two television newscasts with reports on police funding, competitive pressures for viewers will drive the newscasts toward representing the police-budget preferences of viewers in the middle of their distribution, at, for example, D and R. If the stations provide widely divergent representations, with D in the left wing of the distribution and R in the right wing, each can figure that a move toward the center can increase its share of viewers (and advertising revenue), with shareholders reaping benefits from rising share prices.

The Partisan Parting of Newscasts: Two Issues and Three News Outlets

Gordon Tullock had a knack for identifying issues in need of new twists on old arguments that changed settled conclusions in economics (consider his notable development of the now well-established theory of “rent-seeking”2). By the late-1960s, the median voter model was a staple of public choice economics (used most prominently in explaining why presidential candidates tended to adopt centrist policy positions). Tullock effectively asked, “What if there are three candidates—X, Y, and Z—who seek election concerned with two issues and who can win the election with a plurality?”

Tullock addressed the question with the aid of Figure 2, which has police expenditures on the horizontal axis and fire protection on the vertical axis. Voters are, again, assumed to be numerous and spread evenly within the space of the graph. Given the analytics of the previous median-voter model (one issue and two candidates), an initial presumption might be that the three candidates would be pressed to adopt similar positions on the two expenditure levels, in the center of the distribution, for example, at X, Y, and Z. But X has a clear incentive to move away from position X. X can lose votes among the “centrists,” but can pick up votes by more closely representing the interests of Y and Z voters.

Figure 2. Police Expenditures and Fire Protection

Candidates Y and Z have similar incentives to move away from the center, both to gain votes and to fend off the other two candidates’ movements to garner greater vote counts. By moving from their initial positions to Y’ and Z’, they can retake votes that X peeled off by moving from X to X’. After moving away from the center and apart, each has a strong disincentive to move back toward the center: they will lose votes on balance.

Of course, it follows that as the voters spread out in the two-dimensional graph, the greater the policy distances among the candidates. No one should be surprised if an increase in the number of candidates gives rise to a greater spread in the candidates’ positions. With a growing count of candidates, new entrants might find that taking more “extreme” (less-centrist) positions, say, closer to one of the four corners in the graph, improves their chances of being elected. The corners may have few votes, but entering candidates can reason that those corners still improve their election chances over what they would be if they moved closer to other candidates’ positions and divided the votes between them.

More Issues and More Media Competitors

Our foregoing discussion relating to politics has an obvious application to the evolving journalism market over the past five or more decades. With the dramatic growth in the number of issues of interest to news consumers and in the number of media competitors, news outlets should be expected to move apart on their coverage of policy issues, just as growth in candidates and policy issues should be expected to move the candidates apart in their efforts to win elections.

News competitors can be expected to part ways in their ideological and partisan coverage to “news” stories as Fox and CNN have clearly done. Fox now offers a daily menu of conservative takes on policy issues such as federal spending, border crossings, charter schools, critical race theory, climate change—and much more. CNN offers sharply contrasting liberal takes on many of these issues, with no less stridency. The major force moving newscasts further apart? Their competitive drive for viewers and ad revenues.

Key Changes in Media Markets

“The growing divisions among national news outlets over the last several decades could have been predicted, Tullock would surely have insisted, given the growth in the array of contested issues.”

The growing divisions among national news outlets over the last several decades could have been predicted, Tullock would surely have insisted, given the growth in the array of contested issues. In addition, cable and internet technologies have dramatically increased opportunities for alternative newscasts pressed to search out (profitable) segments of consumers.

During the early history of television news, diversity was throttled by government regulations (and limited competitors). In 1949, the Federal Communication Commission required news outlets to abide by its fairness doctrine, which meant that they had to provide a “balance” of views in their treatments of controversial issues to keep their FCC operating licenses. However, the FCC dropped its restriction in 1987. The relaxation of the fairness doctrine was all the media deregulation News Corp. head Rupert Murdock needed to extend his conservative-leaning media empire with the launch of Fox News in 1996, and with a decidedly conservative (and Republican) news slant, an obviously winning strategy. By 2019, Fox News had become the top-rated cable news network, attributable in part to other newscasts’ centrist (or even left of centrist) slants to news stories.

CNN’s initial 24/7-straight-news coverage (launched by Ted Turner in 1976) was challenged in 1996 by the entry of MSNBC, with the goal of providing an alternative 24/7 cable news source to CNN, but with contrasting liberal slants to Fox’s conservative news takes. To fend off the erosion of its viewership and ad revenues from Fox and MSNBC, CNN was pressed to move left, seeking to increase its appeal to Democrats with liberal leanings.

The Polarization of America

The growing divisions in news journalism have been compounded by the growing partisan divides among Americans, self-evident in polls, most notably those taken by the Pew Research Center. Consider the changing distribution of polled Democrats and Republicans from six Pew polls undertaken between 1994 and 2017 on Americans’ political leanings, with their responses distributed along a spectrum that runs from “consistently liberal” to “consistently conservative.” Pew found that between 1994 and 1999, the two parties’ bell-shaped distributions shifted somewhat left but notably, with the “median voters” of the two parties moving closer to one another. However, the Democrat/Republican distributions started moving apart sometime after Pew’s 2004 poll. The two distributions moved even further apart in Pew’s 2011 and 2017 polls.3

The growing political polarization of Americans is also reflected in their representatives’ votes in the U.S. Senate and House, as reflected in Pew’s assessments of congressional members’ “liberal” and “conservative” votes back to the early 1970s (92nd Congress) and forward to 2022 (117th Congress). Pew’s vote assessments show that in the Senate and House, both Democrat and Republican members’ votes diverged from centrist positions over the last half century, but Republicans have steadily become more conservative than Democrats have become more liberal. Pew analysts have concluded that “on average Democrats and Republicans are farther apart ideologically today than at any time in the past 50 years.”4

Concluding Comments

The growing divisions among television newscasts can be attributed partially to the growth in the array of policy issues subject to debate, as noted. The growth of issues provides increased opportunities for ideological and partisan divisions among television news audiences and their newscasters. The competition among outlets for viewers’ limited attention spans has encouraged media outlets to cultivate ideological and partisan identities, and brand loyalties that can expand their audiences and defend them against erosion by established competitors and new entrants, a point illuminated above with the aid of Tullock’s two-policies/three-candidates graph.

These divisions among Americans and their news sources are, understandably, interactive and self-perpetuating (at least up to a point). With many news consumers moving to newscasts that appeal to (and reinforce) their favored policy positions, and then blocking out competing newscasts with different, competing slants (to hold at bay the discomfort of cognitive dissonance), many viewers’ news biases have likely been reinforced and hardened, if not made more extreme.

For more on these topics, see

Of course, any strengthening of viewers’ ideological and partisan preferences will only encourage news sources to make their coverage more strident, peppered with derision for opposing views. In the process, viewers’ trust in the integrity of opposing newscasts can be progressively undermined. And, indeed, Pew has found a gradual dissipation of viewer trust in national news programs between 2016 and 2022. With open entry in news markets, newscasts can be expected to emerge that base their appeals on being “fair and balanced,” but without the partisan and ideological barbs.


[1] Bobby Chernev, “27 Curious Cable TV: Subscribers Statistics to Know in 2023.” TechJury, April 20, 2023.

[2] Gordon Tullock, 1967. “The Welfare Costs of Tariffs, Monopolies and Theft.” Western Economic Journal. 5: 224-232.

[3] See Pew Research Center, 2017. “The Shift in the American Public’s Political Values,” posted February 20.

[4] Drew Desilver, 2022. “The Polarization in Today’s Congress Has Roots that Go Back Decades,” Washington, D.C.: Pew Research Center, March 10.

*Richard B. McKenzie is a professor emeritus of economics at the University of California, Irvine and author of The Selfish Brain: A Layperson’s Guide to a New Way of Economic Thinking.

For more articles by Richard B. McKenzie, see the Archive.