THE BIG PICTURE

If one were to throw a dart at a map of the United States and pick up the local newspaper where the dart landed, chances are one would be reading a paper whose op-ed pages lean to the right. Putting aside for a moment the question of circulation, the data show unequivocally that most newspapers in America run more conservative syndicated columnists than progressive syndicated columnists.

In fact, there are fully three newspapers that run more conservatives than progressives for every one newspaper that runs more progressives than conservatives.

While it might be easy to bring to mind a few prominent newspapers (e.g. The New York Times) that run more progressives, looking across the data it becomes clear that at every circulation level, one finds more papers that skew to the right on the op-ed pages. This difference is modest within the largest papers -- the 103 papers with circulations over 100,000 -- but becomes an enormous gap that grows larger at each smaller level of circulation.

Obviously, larger newspapers tend to serve larger cities, which are not only more likely to have a progressive populace than smaller communities but also tend to be more demographically diverse in many ways. A small paper, on the other hand, may serve a local area that is relatively homogeneous. But without speculating too much about the ideological leanings of individual newspaper owners and the communities those papers serve, it can be said that smaller papers, at least on this measure, are more likely to lean right.

For instance, among the smallest daily newspapers -- those with circulations under 10,000 -- 64 percent run more regular conservative syndicated columnists than progressives, while only 16 percent run more progressives. Among papers with circulations between 10,000 and 25,000, the difference is similar: 62 percent run more conservatives, while only 18 percent run more progressives. Only among the largest papers were the two groups even somewhat close, with 50 percent running more conservatives and 35 percent running more progressives.

As interesting as these data are, they do not account completely for differences in the circulations of each paper. After all, the columnists printed in a paper with a circulation of 1 million will have greater impact than those printed in a paper with a circulation of 100,000. In order to more precisely account for circulation differences, we created a measure we call “relative ideological voice,” which compares the reach and influence of columnists within a newspaper, within a state, or within the country as a whole. It multiplies the number of columnists of each ideological stripe by the circulation of the papers in which they appear.

Because conservative columnists have their greatest advantage in small papers, the imbalance in voice is not as great as the overall newspaper-by-newspaper advantage, wherein 60 percent of newspapers in the country run more conservatives and only 20 percent run more progressives. Nonetheless, conservatives retain a clear advantage over their progressive counterparts on a national level. When American newspaper readers turn to the op-ed page of their local newspaper each day, the syndicated columnists they see there are more likely to be conservative.

THE CASE OF THE WALL STREET JOURNAL

This report concerns syndicated columnists; included in the data is each columnist’s “home” paper, where their column originates. This means that the data do not include the Wall Street Journal, which carries a number of columnists who write exclusively for that paper. By contrast, many of the columnists for other major newspapers, such as The New York Times and The Washington Post, are syndicated, which merits them automatic inclusion in the data. Considering that the Journal is both nationally distributed and widely circulated -- it has nearly 1.8 million readers, second only to USA Today -- we thought it would be worthwhile to see what impact its inclusion would have on the data.

The Journal regularly publishes the following columnists: Bret Stephens, Daniel Henninger, Holman Jenkins, John Fund, Kimberley Strassel, Mary Anastasia O’Grady, and Peggy Noonan (who, while published in the weekend section, could certainly be counted as an op-ed presence). All are conservative. Adding in the Journal’s lineup of columnists and its considerable circulation figures to the rest of the data, we found that the lead conservatives held nationally would widen even further. Whereas conservative syndicated columnists held a 46 percent to 38 percent lead over their progressive counterparts in relative voice in the original calculation, the disparity would have increased by 3 points, to 48 percent–37 percent, with the Journal’s inclusion.

TOP OF THE CHARTS

Every syndicated columnist holds a place within an elite stratum of our nation’s political debate. But there are some who constitute the elite of the elite -- and that group leans to the right.

Of the top ten columnists by number of papers, five are conservatives (George Will, Cal Thomas, Kathleen Parker, Morton Kondracke, and Thomas Sowell), two are centrists (David Broder and Cokie and Steve Roberts, who write a column together), and only three are progressives (Ellen Goodman, Leonard Pitts Jr., and Nat Hentoff⁶). As the nation’s most-read columnist, George Will appears in fully one out of every four daily newspapers in America.⁷

When the data are sorted by the circulation of the papers that run each columnist, the same pattern emerges: five conservatives, two centrists, and three progressives. Though a few of the names have changed, once again, George Will is clearly at the head of the pack, reaching over 6 million more readers than his closest rival.

Overall, there are 79 progressives whose columns are regularly carried in multiple papers. These progressives appear regularly in a total of 1,915 papers (counting each paper as many times as it has columnists) with a summed circulation of 125.2 million. By comparison, there are 74 conservatives whose columns are regularly carried in multiple papers. These conservatives appear regularly in a total of 3,076 papers (counting each paper as many times as it has columnists) with a summed circulation of 152.1 million.

At the top, the disparity is just as stark. The top 10 conservative columnists appear in 641 more papers than the top 10 progressive columnists; the total circulation for the top 10 conservatives exceeds that of the top 10 progressives by more than 20 million readers.

It is worth noting that on both of these lists, there are columnists syndicated in relatively small numbers of papers. For instance, David Ignatius of The Washington Post is carried regularly in only 22 papers, but reaches more than 3.6 million readers. As such, he has the highest average circulation of any syndicated columnist. Others -- Bob Herbert and Jonah Goldberg, for instance -- also are published in mostly high-circulation papers.

At the other end of the scale are columnists who are published mostly in smaller papers. Progressive columnist Gene Lyons appears regularly in 75 papers, but they have an average circulation of less than 12,000, meaning he reaches less than a million readers (see Appendix 2 for more information).

As for centrist columnists, when one moves past the top two -- David Broder and Thomas Friedman -- there is a steep drop-off in the number of papers in which the columnists appear and the total circulation they reach.

If the large circulation numbers involved here are overwhelming, another way of thinking about the top columnists is their “reach” -- how much of the American newspaper marketplace are they reaching? This can be expressed in percentage terms by dividing a columnist’s total circulation with the total circulation of all daily newspapers in America, about 50 million. The fact that the combined circulation of the papers regularly carrying George Will is above 21 million means that he reaches two out of every five newspaper readers in America, an extraordinary level of penetration.

These elite columnists dominate the syndicated marketplace to a dramatic degree. If we add up the combined circulation reached by all 201 syndicated columnists in this dataset, we see that the top 10 columnists alone account for more than 35 percent of the total syndicated market. The top 18 columnists reach as many readers as the other 183. George Will alone reaches as many readers as the bottom 80 columnists. The median columnist among the 201 in the dataset reaches 1 percent of the total American newspaper readership. In other words, this is an extraordinarily top-heavy list.

CONSERVATIVE ADVANTAGE ACROSS THE LAND

To this point, we have concerned ourselves with the national picture. But when one looks separately at regions and states, it becomes clear that the dominance of conservative syndicated columnists is spread across the nation. Once again, we are using the measure of relative ideological voice, the number of progressive, conservative, and centrist columns and how many readers they reach. We will start at the broadest level, then move closer to the ground.

It is often said that Americans are divided not only into red and blue states, but red and blue regions: the South and Midwest are more conservative, while the Northeast and West are more progressive. While most would agree that this is an oversimplified picture of Americans and their beliefs, we can say one thing about the broad regions of the country: In three of the four, conservatives have the advantage on the op-ed pages.

It is often said that Americans are divided not only into red and blue states, but red and blue regions: the South and Midwest are more conservative, while the Northeast and West are more progressive. While most would agree that this is an oversimplified picture of Americans and their beliefs, we can say one thing about the broad regions of the country: In three of the four, conservatives have the advantage on the op-ed pages.

Due to the influence of the Middle Atlantic states -- particularly New York, where progressives enjoy an advantage in some papers with very large circulations -- progressives do manage to hold a slight edge (44 percent to 42 percent) in the Northeast. But in all three of the country’s other large regions, conservative syndicated columnists reach more eyes more often than their progressive counterparts.

Furthermore, in every region of the country, the columnists appearing in the most papers are more likely to be conservative than progressive. Even in the Northeast, the region where progressives enjoy a small advantage in relative ideological voice, George Will appears in more papers than any other columnist. Will appears regularly in 63 papers in the Northeast, and occasionally in 8 more. The top 10 list for the Northeast shows five conservatives, four progressives, and one centrist.

The other region where progressives might hope to be at parity -- the West -- does show a slightly more balanced split among the top columnists. But here as well, George Will outranks all others, and the top 10 list includes more conservatives than progressives.

Given the extreme imbalance in relative ideological voice in the Midwest and South, it is no surprise that a similarly stark contrast emerges on the top 10 lists for those two regions.

In every region of the country, the columnist who appears in the most papers is a conservative: George Will in the Northeast and West, Kathleen Parker in the Midwest, and Cal Thomas in the South.

Next, using the nine areas into which the U.S. Census Bureau divides the country, the data show that in eight out of nine, the conservative voice outweighs the progressive voice.

The greatest advantage for conservatives -- a margin of 50 percent to 33 percent (with centrists making up the remainder) -- occurs in the South Atlantic, comprised of the Eastern Seaboard states running from Delaware south to Florida. Close behind, with a margin of 50 percent to 34 percent, is the West South Central, comprised of Texas, Oklahoma, Arkansas and Louisiana.

Only in the Middle Atlantic states -- New York, Pennsylvania, and New Jersey -- does the progressive voice loom larger than the conservative voice. Even in New England, the most progressive area of the country, conservative syndicated columnists have the advantage on the op-ed pages.

Finally, the relative balance of conservative and progressive syndicated columnists can be examined on the state level. Here, the data do not break down strictly on red-blue lines. There are some heavily Republican states where conservative columnists dominate as one might expect -- South Carolina, Oklahoma, and Georgia, for example -- and some heavily Democratic states where progressive columnists reach more readers. But there are also many “blue” states whose newspapers feature more conservative columnists. These include Illinois, Michigan, Connecticut, and California. Among the states where the progressive voice outweighs the conservative voice, there are Democratic states (Rhode Island, Vermont, Hawaii, New York) and swing states (Wisconsin, Arizona, and Tennessee), but no clearly “red” states.

But overall, the results are clear: Conservative syndicated columnists have a greater reach than their progressive columnists in 38 states, plus the District of Columbia. In only 12 states does the progressive voice outweigh the conservative voice.

CONCLUSION

Conservatives are often heard to complain about the “liberal media,” a nefarious cabal of journalists and media owners supposedly endeavoring to twist the news to serve their ideological agenda. Media Matters for America has shown in a variety of ways that the “liberal media” is a myth. Our two reports on the Sunday talk shows showed how those programs are dominated by conservative guests. Our analysis of the coverage of religion showed how that coverage favors conservatives. Analyses performed by other organizations have shown how conservatives dominate talk radio. And this study demonstrates that in yet another key portion of the news media, conservatives enjoy a structural advantage that gives them a leg up in influencing public opinion.

That structural advantage enables them to transmit an overarching narrative across the country, one that serves to convey the impression that conservative ideas that in many cases enjoy tiny support are actually the “reasonable center” in key debates. To take just one example, prominent conservative columnists who wrote about the topic were nearly unanimous in support of President Bush’s decision to commute Scooter Libby’s sentence, while some advocated pardoning him outright, despite the fact that polls indicated the decision had the support of only around one in five Americans.

In terms of the number of people reached by their ideas and opinions, of the authority they are granted, and of their prestige, there are few in the American news media who equal the lofty position held by the top syndicated columnists. Read by millions, even tens of millions, their opinions form the basis on which our democratic debate often proceeds. Because they have a national reach, they also have the power to advance ideas and narratives that local columnists simply do not have.

As this study has demonstrated, the landscape of syndicated columnists is dominated by conservatives. They reach considerably more readers than progressives. By a 3-to-1 margin, most American newspapers run more conservative syndicated columnists than progressives. In nearly every region of the country, the conservative voice on op-ed pages is louder than the progressive voice. And for every one state that has a greater progressive voice, there are three in which conservatives have more influence.

In short, while the right wing spends a great deal of time complaining about alleged bias in the media, when it comes to the nation’s op-ed pages, it is the progressives who are getting the short end of the stick.

METHODOLOGY

This project was designed and executed, and this report written, by Paul Waldman, Senior Fellow and Director of Special Projects; Elbert Ventura, Research Fellow; and Robert Savillo, Research Analyst. Further assistance was provided by Neal Fersko, Eliza Keller, Gerard Matthews and Greg Lewis. Special credit is due to Jennifer Hoffman, without whose extraordinary efforts the project would have been impossible.

For this study, we set out to determine the lineup of syndicated columnists in every daily newspaper across the country. We used the 2006 edition of the Editor & Publisher International Year Book: The Encyclopedia of the Newspaper Industry as our source for all of the nation’s dailies. According to the yearbook, there are 1,452 daily newspapers in America. For our purposes, we omitted foreign-language, business, or legal newspapers, and papers that had ceased publication since the publication of the 2006 yearbook, and combined the separate morning and evening editions of some newspapers. That gave us a total of 1,430 newspapers for our study.

DATA COLLECTION

To collect the columnist data for all 1,430 papers, we took the most direct approach -- we used the contact information in the E&P Year Book or on each newspaper’s website to contact editors. The vast majority of the data was collected in this fashion. In some instances, editors chose to respond to our survey via e-mail. In a handful of cases, editors could not be reached for an answer –- in some cases, after as many as 10 phone calls –- or refused to participate in our survey outright. In those few cases, the newspaper was left out of the study unless another means was found to discern the columnist line-up on their op-ed page.

Each paper was asked, over the phone or through email, the following two questions:

1. Which nationally syndicated columnists do you publish on a regular basis -- about once a week? (Regular)
2. Which nationally syndicated columnists do you publish intermittently but not every week -- about once a month? (Occasional)

We included only columnists who are published on the op-ed pages and who regularly write about public affairs. That meant that humor columnists, advice columnists, and lifestyle commentators were excluded, although there are some columnists who combine a number of genres. For instance, Garrison Keillor writes some slice-of-life columns, but also writes frequently on political issues from a progressive perspective, so he is included in our data.

For some newspapers, syndicated columnist data was collected from the Library of Congress. Those newspapers tended to be bigger-circulation newspapers that either refused to participate in the survey or could not be reached after repeated attempts. For data collected in this way, we examined one month’s worth of opinion pages and recorded each syndicated columnist who appeared. Columnists who appeared at least three times in a month were counted as regular, and those who appeared twice were counted as occasional. In many cases, the regular syndicated columnists were easily determined because they appeared on a set schedule in the paper’s op-ed pages.

In a few instances, newspaper columnist lineups were collected from a newspaper’s website. Major newspapers like The New York Times and The Boston Globe keep a fairly updated list online of their columnists, including their publication schedule. A handful of newspapers also print a comprehensive archive of syndicated columns that appeared in their paper’s print editions.

After we had contacted all 1,430 newspapers at least once and we had collected data for 95 percent of them, confirmation emails were sent to all newspapers asking editors to verify the syndicated columnist lineup that we collected from them. Each newspaper was contacted up to three times to confirm their data. All confirmation emails were sent between April and July of 2007.

The vast majority of newspaper editors we spoke with provided a detailed list of syndicated columnists that ran in their op-ed pages. In some cases, they provided the name of a syndicate or syndicates in addition to or in lieu of specific columnists. We recorded such syndicate data but did not 12 include them in the analysis, except for a few rare exceptions. We had considered listing all the columnists in a given syndicate on an occasional basis, but in the end we concluded that such a move would inflate the total number of papers for each columnist. We decided only to include columnist names which were directly stated by each paper’s editor. In rare instances, newspapers used syndicates with a clear ideological slant (e.g. Minuteman Media, a small progressive syndicate). In those cases, we included the syndicate in the data.

WEIGHING BY CIRCULATION

When recording the data collected from newspaper editors, we included other data as well. Each entry in our database contains the columnist’s name, his or her ideological alignment (either conservative, progressive, or centrist), the name of the newspaper, the state in which that newspaper is published, the Census-defined geographic division and region data, the frequency with which the columnist is published (regularly or occasionally), and the circulation figures for that paper.

The circulation numbers used for each paper were taken from the E&P Year Book. The yearbook provides a total of 17 circulation figures across all newspapers: a daily evening edition number (A); an evening edition number for Monday (B), Wednesday (C), Friday (D), and Saturday (E); a daily all-day edition number (F), an all-day edition number for Monday/Tuesday (G) and Saturday (H); a daily morning edition number (I), a morning edition number for Monday (J), Tuesday (K), Wednesday (L), Thursday (M), Friday (N), and Saturday (O); a Sunday edition number (P); and, finally, a weekend edition number (Q).

These circulation figures were then all weighted together by the number of days per week for which they apply (i.e., a daily morning edition number would be weighted by a factor of 5/7, while a Saturday morning edition number would be weighted by a factor of 1/7). This was done in order to create an average weekly circulation number for each individual paper. The entire equation used is as follows:

• (A(5/7))+(B(1/7))+(C(1/7))+(D(1/7))+(E(1/7))+(F(5/7))+(G(2/7))+(H(1/7))+(I(5/ 7))+(J(1/7))+(K(1/7))+(L(1/7))+(M(1/7))+(N(1/7))+(O(1/7))+(P(1/7))+(Q(2/7)) = Adjusted Circulation

One of our measures in this study is “relative voice.” Relative voice takes into account not just the numbers of newspapers conservatives and progressives appear in, but also the circulations of the newspapers that carried them. Such a comparison would yield a number that would show what proportion of a state’s total number of syndicated columns is conservative or progressive. The “voice” figure is a percentage of the total circulation of that state. “Voice” figures in the report include conservative, progressive, and centrist. The following equation represents how this number was calculated:

• (Number of Conservative Columnists)(Circulation) = Conservative Ideological Circulation
• (Number of Progressive Columnists)(Circulation) = Progressive Ideological Circulation
• (Number of Centrist Columnists)(Circulation) = Centrist Ideological Circulation
• Conservative Ideological Circulation + Progressive Ideological Circulation + Centrist Ideological Circulation = Total Ideological Circulation
• (Conservative Ideological Circulation)/(Total Ideological Circulation) = Conservative Voice
• (Progressive Ideological Circulation)/(Total Ideological Circulation) = Progressive Voice
• (Centrist Ideological Circulation)/(Total Ideological Circulation) = Centrist Voice

A sample will illustrate how this figure is calculated (this is repeated in a footnote in the body of the report). Imagine that a state has only two newspapers, Big City News and Small Town Post. Big City News has a circulation of 1,000,000, while Small Town Post’s circulation is 100,000. Each paper has five columnists. Big City News runs four conservatives and one progressive, while Small Town Post runs four progressives and one conservative. The total readership reached by the progressives is as follows:

1,000,000 x 1 (the one progressive from Big City News) = 1,000,000
+ 100,000 x 4 (the four progressives from Small Town Post) = 400,000
 Total progressive circulation: = 1,400,000

The total readership reached by the conservatives is as follows:

1,000,000 x 4 (the four conservatives from Big City News) = 4,000,000
+ 100,000 x 1 (the one conservative from Small Town Post) = 100,000
 Total conservative circulation: = 4,100,000

The relative voice in this state would be 25.5% progressive (1,400,000/5,500,000) and 74.5% conservative (4,100,000/5,500,000).

A real-world illustration of this approach: There are a total of 12 conservative columnists and 13 progressive columnists printed in Maine’s daily newspapers. Without weighing the circulation numbers, progressives appear to have a slightly larger presence than conservatives. However, once circulation numbers are taken into account, conservative syndicated columnists actually emerge with approximately 130,000 more readers than the progressive columnists. Measured by relative voice, Maine comes out with 46 percent conservative vs. 37 percent progressive (with the rest comprised of centrists).

FREQUENCY OF PUBLICATION Our analysis was also split between regular columnists and occasional columnists. Columnists who are published on a regular basis have a larger influence than those who are published occasionally. Based on this assumption, most of our analysis focuses on the regular columnist lineups, while the total columnist lineups (regular and occasional) are used only when comparing one columnist to another.