This is very interesting, the answers will likely surprise you. Forgot where I found this but H/T
Econometrica, Vol. 78, No. 1 (January, 2010), 35–71
WHAT DRIVES MEDIA SLANT?
EVIDENCE FROM U.S. DAILY NEWSPAPERS
BY MATTHEW GENTZKOW AND JESSE M. SHAPIRO1
Government Regulation Of News Media ownership in the United States is built on two propositions. The first is that news content has a powerful impact on politics, with ideologically diverse content producing socially desirable outcomes. According to the U.S. Supreme Court (1945), “One of the most vital of all general interests [is] the dissemination of news from as many different sources, and with as many different facets and colors as is possible. That interest presupposes that right conclusions are more likely to be gathered out of a multitude of tongues, than through any kind of authoritative selection.”
The second proposition is that unregulated markets will tend to produce too little ideological diversity. The highly influential Hutchins Commission report identified cross-market consolidation in newspaper ownership as a major obstacle to the emergence of truth in the press (Commission on Freedom of the Press (1947)). The Federal Communications Commission (FCC) “has traditionally assumed that there is a positive correlation between viewpoints expressed and ownership of an outlet. The Commission has sought, therefore, to diffuse ownership of media outlets among multiple firms in order to diversify the viewpoints available to the public” (FCC (2003)). This belief has justified significant controls on cross-market consolidation in broadcast media ownership, on foreign ownership of media, and on cross-media ownership within markets, and has motivated a sizable academic literature arguing that current media ownership is too concentrated (Bagdikian (2000)).
That news content can have significant effects on political attitudes and outcomes has been documented empirically by Strömberg (2004), Gentzkow and Shapiro (2004), Gentzkow (2006), Gerber, Karlan, and Bergan (2009), DellaVigna and Kaplan (2007), and others. In contrast, evidence on the incentives that shape ideological content and on the role of ownership, in particular, is limited. Existing studies have generally relied on hand collection and coding of news content, and so have been restricted to small numbers of sources (e.g., Glasser, Allen, and Blanks (1989), Pritchard (2002)). Groseclose and Milyo (2005) made an important contribution, proposing a new measure of ideological content based on counts of think-tank citations. However, their index was calculated only for a small number of outlets, and has not been used to analyze the determinants of slant.
In this paper, we propose a new index of ideological slant in news coverage and compute it for a large sample of U.S. daily newspapers. We estimate a model of newspaper demand that incorporates slant explicitly, estimate the slant that would be chosen if newspapers independently maximized their own profits, and compare these profit-maximizing points with firms’ actual choices. We estimate the contributions of consumer and owner heterogeneity to crossmarket diversity in slant and develop tentative implications for ownership regulation.
Our slant index measures the frequency with which newspapers use language that would tend to sway readers to the right or to the left on political issues. We focus on newspapers’ news (rather than opinion) content, because of its centrality to public policy debates and its importance as a source of information to consumers.2 To measure news slant, we examine the set of all phrases used by members of Congress in the 2005 Congressional Record, and identify those that are used much more frequently by one party than by another. We then index newspapers by the extent to which the use of politically charged phrases in their news coverage resembles the use of the same phrases in the speech of a congressional Democrat or Republican. The resulting index allows us to compare newspapers to one another, though not to a benchmark of “true” or “unbiased” reporting.
Two key pieces of evidence suggest that our methodology produces a meaningful measure of slant. First, many of the phrases that our automated procedure identifies are known from other sources to be chosen strategically by politicians for their persuasive impact. Examples include “death tax,” “tax relief,” “personal account,” and “war on terror” (which we identify as strongly
Republican), and “estate tax,” “tax break,” “private account,” and “war in Iraq,” (which we identify as strongly Democratic). Second, the index that we construct using counts of these phrases in news coverage is consistent with readers’ subjective evaluation of newspapers’ political leanings (data on which are available for several large papers in our sample).
See Full PDFhere: Media Bias