Viewpoint diversity and media ownership

Federal Communications Law Journal, June, 2009 by C. Edwin Baker

 I.  INTRODUCTION
II.  RATIONALES FOR OWNERSHIP DISPERSAL
     A. Democratic Distribution or Dispersal
     B. Democratic Safeguards
     C. Media Quality or the Undesirable Bottom-Line Focus.
III. EMPIRICAL STUDY OF VIEWPOINT DIVERSITY
     A. The Study
     B. Value of Viewpoint Diversity
     C. Reasons for Error

I. INTRODUCTION

The premise of Professors Daniel Ho and Kevin Quinn's "Viewpoint Diversity and Media Consolidation: An Empirical Study" is a repeated assertion. They believe that the claim that media consolidation reduces viewpoint diversity (the "convergence hypothesis") "forms the empirical bedrock" of federal regulation for restricting media consolidation (1) (presumably beyond what would be independently required by antitrust law). The FCC's ownership rules, they say, "[a]t heart.., rest on ... the 'convergence' assumption." (2) Given this premise, they apply innovative statistical techniques to a sample of five cases to show that mergers do not correlate with reductions of viewpoint diversity. (3) On this basis of having "challenge[d] long-held assumptions about viewpoint diversity," (4) they conclude that their findings justify cautious relaxation of existing ownership restrictions. (5) Specifically, Ho and Quinn use statistical techniques to categorize editorial positions on Supreme Court opinions as liberal or conservative. (6) They then analyze editorial positions about these Court decisions taken by papers before and after five mergers--for example, the merger of New York Times and the Boston Globe--and find no systematic reduction of viewpoint diversity.

Their study is subject to a number of obvious methodological criticisms, some of which Part III.A addresses. The primary problem, however, is that they are simply wrong in their basic assumption that the "convergence" hypothesis provides the main policy basis for ownership restraints. Three other concerns, which I have presented elsewhere (7) and summarize in Part II, provide the primary grounds to oppose media mergers. Their empirical study, consequently, is entirely irrelevant to appropriate reasons to oppose media concentration. Proper attention paid to the three most relevant concerns shows that any reliance on Ho and Quinn's study in policy debates would simply be perniciousness. To drive this point home, Part III.B explains why the quantitative amount of viewpoint diversity, which they purport to measure, is not even to be valued in itself--though how diversity or similarity of viewpoints develops is a proper policy concern relevant to why source (but not viewpoint) diversity matters. Part III.C concludes by speculating about causes of Ho and Quinn's mistake of focusing on viewpoint diversity--comments intended as a cautionary tale about the use and abuse of positivist empirical analyses.

II. RATIONALES FOR OWNERSHIP DISPERSAL

The three major reasons to oppose media concentration in general, and mergers in particular, can be labeled: (i) the democratic distribution value; (ii) the democratic safeguard value; and (iii) the media quality value, cashed out as an objection to a bottom-line focus. The first two reasons, I suspect, represented the primary--but usually unarticulated--concerns of the public when nearly two million people wrote to oppose the FCC's recent relaxation of concentration restraints, (8) while the third often finds expression, with various levels of articulation, among editors, journalists, artists and others in the media professions. I describe the logic of each in turn.

A. Democratic Distribution or Dispersal

A central premise of most normative theories of democracy is that democracy should constitute a wide, roughly egalitarian, sharing of political power. With a dire reference to the "unanimity of the graveyard," the Court asserts that here "[a]uthority ... is to be controlled by public opinion, not public opinion by authority." (9) This basic democratic premise leads to the formal equality embodied in the Court's "One Person, One Vote" requirement (10) Judicial resistance to a constitutional claim that political equality should be substantive and not merely formal does not reject the normative claim. Rather, the Court correctly recognizes that, because the proper form of substantive equality is democratically contestable, because substantive equality can never be fully realized, because moves in that direction necessarily involve institutionally complex trade-offs, and because some of the ways used to advance this value themselves create constitutional problems, (11) the claim should not have constitutional status. (12)

The egalitarian premise that justifies the formal one person, one vote requirement also applies to voice within the public sphere. Voice, more than vote, creates public opinion and provides the possibility of deliberation. It is likewise clear that the media is the central institution of a democratic public sphere. These observations lead inexorably to the recommendation of a maximum dispersal of media power, (13) power represented ultimately by ownership. (14)