Why Fox News in 1996 Proved There Was an Audience Nobody Was Serving

When a restaurant opens on a street where nobody eats and the line stretches around the block by Friday, there are two possible explanations. Either the owner is a genius who invented hunger, or the street was full of hungry people and every other restaurant had been serving something they did not want.

Economists prefer the second explanation, and so should you. That is the whole argument of this post, and I am going to make it in the open, with the evidence, and with the strongest objection on the table.

The restaurant in question is a cable news channel that launched on October 7, 1996, and found an audience so quickly that the rest of the industry spent years explaining it away. The channel's defenders call it a triumph. Its critics call it a partisan machine. I would rather ask a plainer question. What does a large, fast-arriving audience tell us about the product that was already on offer?

The claim

Here is the thesis, stated so it can be tested. The speed and size of the audience that found the Fox News Channel after its 1996 launch is strong market evidence that a large share of the country was not being served by the existing television news product.

Notice what the claim is not. It is not that the channel was balanced, or that any one network is the truth. It is not a claim about the channel's present ratings, which do not belong in an evergreen post. It is a claim about demand: that in 1996 there was an audience, that it was large, and that it had been waiting.

Let me define terms. Market evidence here means the behaviour of customers who vote with their remote controls, as distinct from the surveys in which people tell a pollster what they say they want. Underserved means that the product on offer did not match what a sizeable group of buyers would pay for, in money or in attention.

The landscape of 1996

To see what the channel's arrival meant, remember what television news looked like when it arrived.

For most of the postwar era the American public got its national news from three broadcast networks, ABC, CBS and NBC, which ran evening newscasts at roughly the same hour with roughly the same sensibility. The cable era began in 1980 with CNN, which was a technical breakthrough and an editorial continuation: it brought round-the-clock coverage to the same audience in the same register. A rival, MSNBC, launched in July 1996, backed by Microsoft and NBC, three months before the channel under discussion.

There was, in short, competition in every dimension except viewpoint. More channels, more hours, more anchors, a long menu, and a kitchen that cooked from one cookbook. That is the setting.

Two other developments mattered. In 1987 the FCC stopped enforcing the Fairness Doctrine, the rule that had required broadcasters to present contrasting views on controversial issues. With that rule gone, talk radio spread. A national radio programme went into syndication in 1988, and within a few years it had a very large audience, proving that a lot of people wanted to hear commentary they could not find on the news. Meanwhile, as we describe in Before Cable There Was Human Events, and Before That Almost Nothing, conservative journalism had spent decades working from the margins, in small magazines and newsletters.

By the mid-1990s, then, a market signal was already ringing. Talk radio had found an audience. Books by conservative authors were climbing the bestseller lists. The television screen was the last big medium without a conservative option.

What the launch looked like

The channel was founded by Rupert Murdoch's News Corporation, and its first chief executive was Roger Ailes, a veteran television producer and political media consultant. It launched on October 7, 1996, with a lineup built around a straightforward promise: news reported with a different set of instincts from the other networks.

Contemporary coverage reported what happened next. Distribution was the first hurdle, because cable operators control which channels reach homes, and a new channel had to be carried before anyone could watch it. Some large systems resisted: Time Warner chose to carry a rival instead, and the argument over carriage in New York became a local political story involving the mayor's office. Despite that, the channel built its audience fast. Within a few years it was competing seriously with CNN, and by early 2002 it had passed CNN in the cable news ratings, a result widely reported at the time.

I will not give you ratings figures for the present, and you should be suspicious of anyone who does without a citation. The point is historical, and it does not depend on a decimal. A channel that started from nothing in a market the incumbents had dominated since 1980 overtook the market's founder in about five years. That is not what happens when a market is well served.

The argument in numbered form

Here is the case, in premises.

1. If an existing product fully served a market, a new entrant would struggle. The incumbents would have the audience, the distribution and the brand, and customers would have little reason to switch.
2. A new entrant that grows quickly, against entrenched competition, in a mature market, implies unmet demand. People do not abandon a satisfactory product in large numbers for a product they did not know they wanted.
3. The channel grew quickly, against entrenched competition, in a mature market. CNN was sixteen years old, the broadcast networks were older still, and MSNBC arrived the same year.
4. Therefore there was unmet demand. A large audience existed whom the existing product did not satisfy.
5. The likeliest explanation of the unmet demand is a difference in outlook between the product and its audience. The channel's offering differed from its competitors' in the way it framed stories, and that difference is what the audience was buying.

You can attack the argument at any premise. I will take the attacks in order.

The strongest objection: a partisan market created, not found

The best answer to my argument is this, and I want to give it its due.

You are assuming the audience pre-existed. But media can create the audience that it serves. A channel that tells a segment of the public its fears and grievances will attract that segment, and the segment will then think of itself as a group that was being ignored. The business model, after all, is to sell attention to advertisers, and the most efficient way to hold attention is to confirm what people already believe. You have not found a hidden market. You have manufactured a partisan one. And the result has been a more polarised country in which each side watches its own news and distrusts the other's.

It is a serious argument. It contains a true observation: there is a commercial logic in telling audiences what they want to hear, and any outlet, on any side, can fall into it. I do not deny that. But as an answer to my thesis it fails, for four reasons.

One: the audience came first. The evidence is the timing. Talk radio had demonstrated demand for non-network commentary by the early 1990s. The audience existed before the channel did, and the channel was the television answer to a demand that had been visible on the radio dial for eight years. A manufacturer creates demand with a product, then waits. An entrant that arrives into existing, measured demand is doing something different.

Two: the entry argument cuts against the objector. If media create their audiences, the incumbents had every opportunity to do the same. They had the reach, the money and the talent. If a partisan audience could be conjured by any outlet willing to serve it, the established networks could have served it first. That they did not suggests either that they did not see the market or that they declined it. Both explanations fit my thesis.

Three: the "partisan market" was already partisan in the other direction. Here the objection tends to go quiet. A market is partisan when one side's perspective is treated as the neutral default and the other as a niche. The channel was called partisan because it started from different premises. The incumbents were rarely described that way, because their premises were the ones in the room. That is a claim about framing, and the evidence for it is the next section.

Four: customers choose. The audience was not conscripted. Viewers who wanted the older style of news could, and did, keep watching it. Nobody was forced to switch. A large number chose to, and they were free to switch back. That is how a market tells you what customers prefer.

The evidence of newsroom outlook

Now the hard part: the claim that the existing newsrooms did not share the outlook of a large part of their audience. That is a factual claim, and there is evidence for it in published surveys of journalists, which have been conducted for decades.

I will give you the sources and the shape of the findings, and tell you where to check the exact figures.

• The Indiana University studies of American journalists. Researchers David Weaver and G. Cleveland Wilhoit, and later colleagues, have surveyed working journalists repeatedly since 1971, with waves in 1982, 1992, 2002 and 2013. In the first wave, nearly 26 percent of full-time journalists identified as Republican. By the 2013 wave the figure had fallen to 7.1 percent, a small fraction of the profession. About half said they were independent, and roughly 28 percent said they were Democrats, several times the number identifying as Republicans. Check the exact figures in the published reports.
• The Lichter and Rothman surveys of the 1980s. Researchers Robert Lichter and Stanley Rothman published studies of what they called the media elite, surveying journalists at major national outlets. They found that those journalists were, on average, markedly more liberal than the general public on social issues, and later updates by the same researchers reported that large majorities of them voted for the Democratic presidential candidate.
• The 1992 survey of Washington reporters. A Freedom Forum survey of Washington bureau chiefs and correspondents after that year's election reported that 89 percent said they had voted for Bill Clinton. The sample was small, roughly 130 to 140 people, so treat the figure as a snapshot rather than a census, and read the methodology before you quote it.

I am deliberately not putting precise percentages in your mouth where the details of the survey wave matter, because the point stands without them and because the sources are public. What the surveys show, across different researchers, different decades and different methods, is that the people who staff national newsrooms hold political views that differ from those of the country at large.

It does not follow that every story is slanted, and I will not claim it. Journalists are professionals and many work hard to be fair. But there is a well-known and simple point about institutions: a group that shares a worldview does not notice its own assumptions. A fish, as they say, does not notice the water. A newsroom in which almost everyone holds the same view of a controversial question will tend to treat that view as the baseline, and will find the opposite view in need of an explanation. We catalogue the techniques that produce this effect in Seven Words Reporters Use to Tilt a Story Without Telling a Single Lie, and the habit of checking the checkers is covered in Why the Fact-Check Became Its Own Genre and How to Check the Checkers.

Why audiences reveal what newsrooms deny

This is the point of the post, so I will put it as plainly as I can.

Newsrooms deny bias in two ways. The first is sincerity: the journalists believe they are neutral. I take them at their word. The second is institutional: the organisation has a process, a code and an ethics statement, and so it concludes that bias cannot be present. Neither tells you what the audience experiences.

An audience reveals bias differently. It does not argue about intent. It does not hold a seminar. It simply compares the product to its own sense of the world, and when the two do not match, it looks elsewhere. The result is a signal that is not subject to the newsroom's self-image. If a large number of people leave one product for another, the first product has told you something about itself, whatever its editors say.

That is why a market test is better evidence than a debate. A debate can be won by the better speaker. A market is decided by the buyer. The buyers decided in 1996, and in the years that followed.

A word of humour, because this subject needs it. The industry's reaction to the audience it had not served was a masterpiece of an industry's capacity for denial. When a million people walk out of a restaurant, the owner of the restaurant is supposed to ask what was wrong with the food. Many in television news asked instead what was wrong with the diners. That is not market research. That is complaining to the weather. And if you follow that complaint one step further than its fans do, you reach the conclusion that millions of viewers had been perfectly well served all along and simply failed to notice, which is a strange thing to say about customers.

A Founding anchor

There is also a constitutional point, and it is about the structure of freedom rather than the content of any broadcast.

The press clause of the First Amendment was written in an age of many voices. Printers in the colonies had published partisan papers without apology, and the founding generation expected competing presses to argue with each other. The republic's idea of a free press was never a single objective voice. It was a noisy market of competing ones, with the public choosing among them. Hamilton backed the Federalist Gazette of the United States, and Jefferson and Madison backed the Republican National Gazette, and both papers were openly partisan. That is the tradition the 1996 launch fits into.

So when a critic complains that a channel is partisan, the right response is a shrug and a question: partisan compared to what? The ideal of a neutral media has always been a twentieth-century idea, and a short-lived one. The founders' ideal was a free press, meaning that anyone could start one. The Zenger case, told in John Peter Zenger and the Jury That Defended American Press Freedom in 1735, is a story about exactly that: a printer who started a partisan paper because the official one served the governor.

That is also why this post does not ask for any government remedy. No regulation is needed to fix what the market has fixed. The cure for a news product that does not serve its audience is a competitor, and the cure for a competitor you dislike is another competitor. Calls to regulate news content, to revive the Fairness Doctrine or to license the press, are calls to replace that process with an authority, and an authority picks winners.

What the argument does not show

I should say plainly what the evidence does not prove.

• It does not prove that the channel's coverage is accurate. A large audience is not a certificate of truth. Readers should judge each outlet on its record, including the channel in question.
• It does not prove that audiences always know what is good for them. People buy bad products. But on the narrow question of whether a market was being served, the buyers' behaviour is the best evidence available.
• It does not show that polarisation is imaginary. The objection about confirmation and tribal loyalty is real, and every outlet, left and right, is exposed to it. The remedy is to read widely. Our post on The Corrections Page Is the Most Honest Part of the Paper, So Read It is a good start.

What it does show is narrower and, I think, stronger: when an entrant arrives with a differently framed product and the audience comes in large numbers and quickly, the incumbents were not serving that audience. The rest is debate about why.

You can read more on the business of news, and on the question of where bias comes from, in Late Night Comedy Stopped Punching at Power and Started Lecturing, which makes a parallel point about another genre in which a single outlook has set the tone.

The case in brief

1. The Fox News Channel launched on October 7, 1996, into a market in which CNN had operated since 1980 and the broadcast networks for far longer, and it built a large audience quickly.
2. A fast-growing entrant in a mature market implies unmet demand, and that demand had already been visible in talk radio for most of a decade.
3. The "created a partisan market" objection fails because the audience came first, the incumbents had every chance to serve it, and viewers chose freely.
4. Published surveys of journalists across decades show newsroom politics diverging from the public's, and a group that shares a worldview tends to take its own assumptions for neutrality.
5. Audiences reveal what newsrooms deny, and the cure for a news product that misses its audience is competition, not regulation.

What you can do: for one week, read or watch two outlets that frame the same story differently, note every word they chose differently, and share the comparison with someone who reads only one.