Media Ownership and the Tone of News Coverage of Labour Unions: Evidence from a Natural Experiment

Despite the growing attention to the media’s role as a political actor, little research has examined the media’s conditioning of public opinion of other key socio-economic actors as a source of media influence. In particular, while it is clear that the decline in public support and positive attitudes towards labour unions both as a strong advocating presence in the workplace and as a broader political force has its roots in a variety of broad social and economic developments, the media have a unique position as a channel of information by which it can arguably serve to help accelerate or slow the rate of decline. Political economy scholars have long stressed the important role of labour unions in determining outcomes relating to inequality, redistribution and societal organisation yet a variety of media scholars have contended that commercial media organisations are likely to have policy preferences that conflict with the labour movement and this has tended to result in the production of negative coverage of organised labour with potential effects on public opinion and policy outcomes. My PhD project broadly examines the relationship public opinion regarding unions, the information about union activities and leadership they receive through news coverage and the variations in media structures at the firm and market level which influence the nature of union-related coverage.

As part of this larger project and building on theoretical innovations from the industrial organisation literature and new avenues in the media politics literature, I attempt to identify the factors that affect the incentives and capacity of commercial media owners to influence the coverage of labour unions by their outlets. In contrast to traditional critical media scholarship that presents commercial media preferences as largely monolithic, I identify key patterns in the development of commercial media organisations which tend to facilitate or inhibit the ability and incentives of media owners to portray unions negatively. In clear contrast with much previous work, I argue that the growing ‘managerial revolution’ in diversely-owned corporate organisations allows for greater freedom for journalists and editors who seek to present issues in a more objective or positive fashion while family-owned media firms have remained relatively constrained by the ideology of those ultimately in charge.

In the empirical literature, significant difficulties arise in identifying ownership effects on content arise as a result of variety of concerns including a relative lack of variation in the identity of media owners over time, potential endogeneity concerns when studying news outlets that are the sole focus of media takeovers, rational anticipation of desired changes in news content by journalists as well as additional measurement difficulties common to media research.

To address these sorts of concerns, this paper takes a relatively novel empirical approach by using a quasi-natural experimental design in which, the form of ownership of a news organization changes while, it is argued, additional potential explanatory variables are held constant or otherwise restricted to the greatest extent possible so as to constitute an example of “as if” random assignment. Specifically, I concentrate on the takeover of the Chandler family-controlled Baltimore Sun by the diversely-owned Tribune Corporation in 2000. A unique set of factors relating to the secrecy associated with the deal and the low-priority status of the Baltimore Sun within the larger merger allow us to assume that the takeover constituted an exogenous ‘shock’ to journalists and editors and changes in the tone of coverage of unions following this point can be traced specifically to the effect of change in the ownership and organisational structure.

Using a newly collected dataset of over 1400 stories across a 12 year period, changes in the tone of newspaper coverage of labour unions, measured using quantitative text analysis procedures, are therefore argued to causally emanate from the change in the concentration of firm-level ownership from family control to a more diffuse corporate structure and the transformation in the organisational structures and incentives and opportunities for interference that such an ownership change brings about. Preliminary results indicate that movement from concentrated family ownership to diversified corporate ownership results in a shift from relatively more negative coverage of unions to more positive coverage. This paper may hopefully contribute to scholarship and policy discussions regarding regulation of media markets, issues of cross-ownership and attempts to increase the diversity of outlets and news coverage available to citizens.