When trying to understand the audience for cable news, these questions stand out:
- How many people watch cable news, and is that number growing?
- Who is really No. 1?
- How many people watch cable news versus broadcast?
The answers, we found, are not as simple as some people may think.
It has become fashionable of late to describe the current era as “the age of cable rising.” The ratings surge during the war in Iraq is usually cited.
The truth is much more complicated. If looked at clearly, the cable audience is really no larger today than it was two years ago.
The notion that cable is surging is based on numbers translated for journalists by the cable channels themselves, which convert the Nielsen ratings data into annual “averages.” By that accounting, thanks to an enormous but brief spike during the war, the cable news audience is described as surging. As reported in the press, Fox News in 2003 was up 53 percent overall (to 1 million viewers on average) and 45 percent in prime time (an average of 1.7 million viewers) over the year before. CNN had a 24 percent rise in all day viewing (665,000 at any given time) and a 22 percent jump for prime time (to 1.1 million).1
Yet such “averaging” tends to create a misimpression, the idea that the audience is spread relatively evenly through the year. In reality, cable ratings are among the most volatile in journalism, spiking and falling wildly with news events. A yearly average implies the typical cable audience is larger than it usually is. Instead, there is something closer to a normal base level cable audience most months punctuated by occasional spikes during rare major news events.
In mathematical terms, this would translate into looking at the median (defined as the middle value) rather than the average. For instance, taking the average viewership for 2003 and comparing it 2002 shows a large increase in the cable news audience – up 34 percent for the day and 32 percent for prime time. However, if we take the median, or the middle value of the 12 months, cable viewership was basically stagnant, showing no growth during the day and a gain of just 3 percent in prime time. Looking at the medians, CNN and MSNBC lost viewers in 2003, while Fox News saw an 18 percent rise in its median monthly audience.2
2003 Median and Average Audience Growth from 2002
Source: Nielsen Media Research
|Prime time Cable||+3%||+32%|
|Prime time CNN||-6%||+20%|
|Daytime Fox News||+18%||+53%|
|Prime time Fox News||+18%||+44%|
|Prime time MSNBC||-19%||+22%|
Going back even farther, a detailed month-by-month analysis of cable shows the following basic story line over the last six years. Cable audiences saw a gradual growth after the launch of Fox News and MSNBC in 1996, then a jump after September 11, 2001. Since, then, however, the cable news audience at any given time overall is probably most accurately described as flat except for the beginning of the war in Iraq in 2003.
1997 to 2003
|Design Your Own Chart|
Source: Nielsen Media Research unpublished data, www.nielsenmedia.com
1997 to 2003
|Design Your Own Chart|
Source: Nielsen Media Research unpublished data, www.nielsenmedia.com
Cable viewer data are generally broken down by daytime and prime time. Looking at audiences each month, fewer than 700,000 people watched cable during any given daytime moment between October 1997 and July 1998, and fewer than 1.2 million watched in prime time.3
That changed when the Clinton-Lewinsky scandal led to the Clinton impeachment proceedings in August 1998. The average daytime audience broke the 700,000 level six out of the next seven months. Prime time exceeded 1.2 million every month. But the audience began drifting downward again after the impeachment proceedings ended.4
The Florida election fiasco in November 2000 proved to be the next major boost for cable. Daytime audiences doubled and prime time audiences nearly did that month, and while audiences began to recede again, they never fell back all the way to their previous levels. For each of the first eight months of 2001, audiences were larger than they were in the same month the year earlier.
Then came September 11, which seemed to create a structural change in the appeal of cable news. The average daytime cable audience in September 2001 surged well above 2 million for the first time. It has not dropped below 1.3 million since, more than double its usual monthly total before then. The prime time audience surpassed 4 million viewers for the first time, and has not fallen below 2 million since. Cable had a new sea level – nearly double what it was in 1998.
Yet since September 2001, there has been no rise in that base level audience. Even the war in Iraq in 2003 did not have the kind of lasting impact on cable that September 11 did.
The war did represent a boon for cable news. When it began in March, cable ratings nearly tripled from the week before. For April overall, more than 6.9 million people watched cable news in prime time on average at any given time, a new monthly high (see Sidebar on this page).
Yet the gains were short-lived. In May the monthly average number of viewers watching in prime time plummeted back down to 2.7 million viewers, in June to 2.4 million and in July to 2.2 million.
The war, in other words, did not change the base level audience.
Another way of looking at this is to look number of people watching cable news each month year to year. From September 2000 to August 2002, cable had a positive year-over-year audience growth every month in daytime and every month but one in prime time.
Yet since then, that growth has stopped. In four of the last six months of 2003 fewer people were watching cable news in both prime time and daytime than the year before.
In other words, cable news has held on to basically none of the viewers it gained during the war.
The cable news world is certainly much bigger than it once was. The typical base level for audience cable news is more than double what it was six years ago. And the spikes during major news events may be higher, too. More people turned to cable during the war than ever before.
The Battle for the Top
Anyone who has followed the cable news industry over the two last years has surely heard that Fox News has overtaken CNN when it comes to audience. Or has it?
Looking only at Nielsen data, the most widely used source, the picture seems pretty clear. For 16 years, since its inception in 1980 until the launch of Fox News and MSNBC in 1996, CNN had a monopoly over cable news. Since late 1996, Fox News has grown rapidly. MSNBC, despite being carried on more cable systems initially than Fox News, never seemed to gain many viewers. Soon, it was largely a two-network race.
Starting in 2001, CNN began to lose viewers while Fox News slowly but steadily built an audience around shows such as”The O’Reilly Factor.”5 The election dispute in Florida represented a boon for the medium, with viewership at all three networks spiking significantly in November 2000 compared with the rest of 2000. Then, in the election aftermath, Fox News held on to more of its new viewers than CNN.
In January of 2002, Fox News for the first time surpassed CNN in total viewers and held its lead. This was due more to big CNN losses, however, than Fox News’s fairly steady but modest gains in viewers. (Fox News at the time averaged 1.1 million viewers in prime time versus 921,000 for CNN. MSNBC, a distant third, averaged 358,000 viewers in prime time.)6
A year later, in January 2003, Fox News had maintained its advantage (with 1,014,000 viewers on average, compared with 721,000 for CNN, and 252,000 for MSNBC).7 And immediately after the war in Iraq, it appeared in May that the network was possibly pulling farther ahead, holding onto more of its wartime audience than CNN.
In the months since, however, Fox News’s losses have actually accelerated, and its margin over CNN has narrowed slightly. Still, as of December 2003, Fox News has drawn better ratings than CNN in every month since January 2002 – 24 consecutive months as the cable news leader.
In 2003, the median monthly viewership of Fox News was 770,000 daytime viewers and 1.4 million in prime time, 52 and 62 percent more, respectively, than CNN. In December 2003, Fox News averaged 1.4 million viewers in prime time, and 961,000 in daytime, both roughly 60 percent more than CNN. In conventional ratings terms, Fox News is well ahead.
Two elements, however, need to be understood about the Nielsen data. First, Nielsen measures only the viewers in private homes. Thus, there are not reliable data on how many viewers tune in at work, the gym, airports or elsewhere. (About 18 million travelers are exposed to the CNN Airport Network each month, according to CNN).8
Second, Nielsen data measure only how many people are viewing a given program at a given time. This is what matters to advertisers. But the numbers do not tell us whether the people who are watching a given program at one time are different people or the same as are watching another program later on.
In other words, the ratings data do not tell us how many people watch cable news overall. There is no number here that would be analogous to newspaper readership or circulation, or the number of “unique visitors” to a Web site.
This may have worked fine for describing the appeal of broadcast television, where every show was a distinct product. But it misses something in capturing the scope of a medium like cable television, where much of the broadcast day is indistinct from another part.
As a result, CNN executives argue, the ratings numbers significantly undercount CNN’s real total viewership and may overstate Fox News’ appeal.
CNN executives argue that their internal research suggests that through the course of the day, more different people check in on their network for news updates. Fox News, they contend, has a smaller overall audience, but Fox News’s audience is more loyal and watches for longer periods of time, thus giving Fox a slightly bigger audience at any given moment.9
Is this just network PR spin?
Actually, there is some public research to suggest CNN may have something of a point.
The only way now to find out how many people overall are watching a given cable station, or even cable news generally, is through survey research rather than ratings. The survey work by the Pew Research Center has examined this over several years and finds that while Fox News is gaining, CNN actually is cited by more people as the source they turn to for most of their news.
In October 2003, in the latest data available, 17 percent of those surveyed cited Fox News as their primary news source, while 20 percent cited CNN. And in July, closer to the March/April war in Iraq, during which Fox News enjoyed a spike in ratings, CNN’s margin over Fox News was even bigger – five points rather than three (27 percent versus 22 percent).10
Thus, Fox News is widely understood in the general press as the cable news leader in viewers, and at any given moment, which is what advertisers care about, that seems true. In another sense, however, Fox News’ dominance is less clear. It is possible, but hard to pin down, that more Americans turn to CNN over time. But they are spread out over more of the broadcast day or even week.
Looking at the survey data, MSNBC remains a distant third, as is the case with ratings. Just 6 percent of respondents cited MSNBC as their primary source of news in October (and 9 percent in July).11
Cable Versus Network
The inevitable limitations of ratings raise another issue when it comes to trying to assess the reach of cable: How many people now turn in the course of a day or a week to cable instead of the older, traditional broadcast networks — ABC, CBS, and NBC?
For all that people like to note the rise in cable’s numbers, many critics are quick to point out that their ratings are nothing compared to those commanded by the old broadcasters. Indeed, if they aired on broadcast TV, even the cable network shows with the highest ratings — “The O’Reilly Factor” and “Larry King Live,” for instance — would be considered problems. Their 1 or 2 ratings points would probably get them canceled after a week.
Consider this: In June 2003, the CBS “Evening News” was watched by 6.5 million viewers, a recent low. Yet that was still three times higher than the average prime time viewership of all three cable news channels combined during the same week.12
November 1997 to November 2003
|Design Your Own Chart|
Source: Nielsen Media Research unpublished data
If the three nightly network newscast audiences in November 2003 were combined, a total of 29.3 million viewers, it would be more than 12 times the prime time audience for cable, 2.4 million viewers, during the same period.
What is more, this is comparing network shows that are on at a time when Americans increasingly aren’t even home – 6:30 p.m. and even earlier in some West Coast areas -to a cable average that includes programs that are on during the heaviest television watching period, prime time. When looked at this way, the cable numbers seem even lower.
Here, again, however, the limitations of traditional ratings in trying to reflect the full impact of cable may present a problem.
The ratings numbers do capture a sense of how many people are watching the three nightly network newscasts during the dinner hour versus any given cable program. And they show a remarkable, even underestimated vigor for the old nightly newscasts. Not only are these programs substantively different than cable in the nature of their news (see Network Content), but they are also vastly more popular as individual programs.
Yet these simple ratings comparisons do not fully capture how many people turn to cable news versus network news generally today to get their information.
For that, survey research again may tell us more. And the survey work suggests that cable has become more important than traditional ratings reflect. Cable may have even surpassed network as a source for news and information.
For example, in 2003, the household weekly average for time spent watching cable news was 3 hours and 6 minutes, significantly higher than the 2 hours and 19 minutes spent watching broadcast news (including network news magazine shows), according to an analysis by CNN of Nielsen data. This was an increase of 41 minutes a week for cable news in 2003 and a 2-minute drop for broadcast news.13
For some time, the Pew Research Center surveys have asked people where they go for most of their national and international news. Today, more people cite cable than network, and have for some time.
In January 2003, for instance, a Pew survey asking people simply to identify whether their favored news source was cable, network or local found cable held a 36-point advantage over network (49 percent cable, 13 percent network).14 Pew has asked that same question for 10 years, and has seen cable’s advantage increasing since 2001, though the gap narrowed slightly during the Iraq war.15
Do you get most of your news about national and international issues from network TV news, from local TV news, or from cable news networks such as CNN, MSNBC, and the Fox News Channel?
|Design Your Own Chart|
Source: Pew Research Center for the People and the Press, “Public Confidence in War Effort Falters,” www.people-press.org
When people are asked to name a specific channel where they go for “most” of their national and international news, CNN ranks No. 1 and Fox News No. 2, both ahead of any of the three broadcast networks.16
January 2002 – October 2003
|Design Your Own Chart|
Survey qu.: Do you get most of your news about national and international issues from (item)?
Source: Pew Research Center for the People and the Press, “President’s Criticism of Media Resonates, But Iraq Unease Grows,” www.people-press.org
These comparisons also, of course, reflect the vast differences in supply. Cable is on 24 hours a day. Network news is on, even if one were to include prime time magazines, at most three or four hours a day.
And, again, increasingly the nightly network newscasts are on at times when people are not home. People’s commutes have been getting longer, making it harder for them to be home when network newscasts are on, whether in the evening or morning. This is why in local news the early-morning programming, before 7 a.m., is the only growth area.
One has to ask about the level of commitment the networks really have to a signature evening newscast or to covering news in a systematic way. If one wanted to structurally limit a television program’s chances of success, airing it at 6:30 p.m. or even 7 p.m. would be a fair way of doing it, and that is when the network news is on. In many places in the West, moreover, such as San Francisco, the evening newscasts go on in an even less enviable time slot, as early, for some networks, as 5:30 p.m., when the number of television sets in use is much smaller.
Whatever the reasons, the disadvantages tonightly news have added up, and the audience is clearly shrinking. From 1993 to 2002, the percentage of Americans in the Pew surveys who said they regularly got their news from the networks steadily declined, from 58 to just 32 percent.17
Cable News’ Expanding Reach
One other factor has hastened cable’s growth, particularly Fox News’s. Over the last five years, Fox News and MSNBC have become available in more people’s homes.
Shortly after their launches in 1996, Fox News was available in 21 million homes, and MSNBC in 33 million. CNN, meanwhile, was already available in 72 million homes. Now, the numbers are much closer, with CNN available in 86 million homes, Fox News in 80 million, and MSNBC in 76 million.
Now Fox News could have difficulty, however, trying to build more audience. There are 90 million cable homes and that number is not likely to increase very much. Thus, most increases in Fox News ratings would depend on people choosing its programming rather than suddenly having Fox News available to them for the first time. By the same token, this added availability, or “carriage,” makes the lack of growth at MSNBC appear more pronounced. Even if only a steady but modest percentage of cable subscribers preferred MSNBC, it would have seen its ratings grow based on expanding the number of available households.
There is another facet of cable news usage that bears noting. The average age of those who watch the cable news channels is older than for other cable programming. While these may be affluent elites watching the news, they are older elites. This may be a sign of danger ahead for cable news.
The median age for CNN viewers is 59.6; for Fox News it is 58.3, and for MSNBC it is 52.4.18
This gives Fox News a slight advantage over CNN and has always been the key to MSNBC’s appeal – it has the youngest cable audience.
The age of its viewers offers some advantages to cable over broadcast networks, though again, the limits of when broadcast news shows air is probably a key factor. In fall 2003, the median age for the three network evening news programs ranged from 59.5 for ABC to 60.3 for NBC to 61.2 for CBS, slightly older than cable.
The limits of who is home when the broadcast news programs are shown is probably a factor. While retirees may be home at 6:30 or 7 pm, working people are more likely to be still on the job or in transit. The fact that these demographic figures for cable are based only on home viewership, not in hotels, airports or offices, where again many workers are likely to be in the evening, may also skew those numbers older as well.
1. Caroline Wilbert, “Fox fattens lead on CNN in peak year for cable news,” Atlanta Journal-Constitution, January 6, 2004, p. D1. MSNBC had an all-day viewing increase of 22 percent to 320,000 viewers and a prime time increase of 23 percent to 443,000 viewers.
2. The statistical methodologist Esther Thorson of the University of Missouri explains the choice of median rather than mean this way:
The median is a better indicator of central tendency than the mean when there are extremely high or extremely low observations in the distribution. These greatly influence the mean, but have little effect on the median. In other words, the median is the closest on the average to all of the scores in the distribution. Because there are some very high levels of cable viewing during the war, these observations pull the mean away from the average viewing scores too much. For that reason, the median is the better indicator of typical viewing levels.
3. Only one month, March 1998, when the relationship between Bill Clinton and Monica Lewinsky was revealed, proved to be an exception.
4. The only exceptions were the Columbine shooting and the death of John F. Kennedy Jr. in the spring and summer of 1999.
5. Phil Kloer, “Slow news period causes CNN’s ratings to drop,” Atlanta Journal-Constitution, June 29, 2000, p. D3.
6. “Fox News Channel overtakes CNN in ratings for the first time,” Associated Press, January 30, 2002.
7. “Ratings,” Cable World, January 13, 2003.
8. “CNN Nuts and Bolts,” available at http://www.cnn.com/SPECIALS/2000/CNN20/story/nutsbolts/facts. How many can actually be counted as “viewers” is subject to debate. Executives at CNBC, the financial network, say that their official numbers would be higher if workplace televisions were included; the other cable news networks would probably make the same argument. See David Bauder, “Bad business news boosts CNBC,” Associated Press, July 4, 2002.
9. According to Paul Farhi, Nielsen data show that in a three-month period, the average Fox News viewer tunes in for 8.9 hours while the average CNN viewer will watch 4.9 hours. See Paul Farhi, “Everybody Wins,” American Journalism Review, April 2003. Online: http://www.ajr.org/Article.asp?id=2875.
10. Pew Research Center for the People & The Press, “October 2003 News Interest Index,” Final Topline, Question 9. Available at: http://people-press.org/reports/print.php3?PageID=748.
11. Pew Research Center for the People & The Press, “October 2003 News Interest Index,” Final Topline, Question 9. Available at: http://people-press.org/reports/print.php3?PageID=748.
12. CBS “Evening News”: 6.5 million; cable news prime time viewers: 1.88 million. See David Bauder, “CBS News hits ratings low,” Associated Press, July 3, 2003, and “Ratings,” Cable World, June 30, 2003.
13. Caroline Wilbert, “Fox fattens lead on CNN in peak year for cable news,” Atlanta Journal-Constitution, January 6, 2004.
14. Pew Research Center, “January 2003 News Interest Index,” January 8-12, 2003. Available at:
15. Pew Research Center, “January 2003 News Interest Index,” January 8-12, 2003. Available at:
16. Pew Research Center for the People & The Press, “October 2003 News Interest Index,” Final Topline, Question 9. Available at: http://people-press.org/reports/print.php3?PageID=748.
17. Pew Research Center, “Public’s News Habits Little Changed by September 11,” June 9, 2002, Question 20. Available at: http://people-press.org/reports/print.php3?PageID=618.
18. See Allison Romano, “Wish you were younger? So do many cable nets.” Broadcasting & Cable, August 25, 2003.
19. “Ratings,” Cable World, March 31, 2003.
20. CNN was watched by 3.4 million viewers on average, while Fox News was seen by 1.9 million and MSNBC by 1.2 million. These figures come from Kevin Downey, “Viewers, not $s, drive cable news networks,” Media Life Magazine, September 19, 2001. Available at: http://www.medialifemagazine.com/news2001/sep01/sep17/3_wed/news1wednesday.html.
21. Richard Huff, “Even minus Brown, CNN led cable ratings Sat.,” New York Daily News, February 5, 2003.
22. Jesse Hamlin, “NBC declares victory in war’s TV ratings battle,” San Francisco Chronicle, March 22, 2003, p. W2.
23. Caroline Wilbert, “Saddam drama helps CNN, but Fox News carries day,” Atlanta Journal-Constitution, December 17, 2003
24. David Bauder, “For CNN, Fox’s open-and-shuttle case of piracy,” Associated Press, February 17, 2003.
25. Verne Gay, “High stakes in the war for cable news supremacy,” Cable World, March 24, 2003.
26. Mercedes M. Cardona, “TV ad spending rebounds during week three of war,” Advertising Age, April 23, 2003.
27. Kevin Downey, “Viewers, not $s, drive cable news networks,” Media Life Magazine, September 19, 2001.