Media Briefings


  • Published Date: April 2009

Advertising can seriously damage the quality and accuracy of content in the print, broadcast and online media. According to research by Professors Matthew Ellman and Fabrizio Germano, biased content is most likely when either competition among media outlets is limited or when advertisers are large and can threaten to withdraw their advertising business from the media.

The results of the study, published in the April 2009 issue of the Economic Journal, also suggest a rationale for the ‘dumbing-down’ of programming in much of commercial TV, best exemplified in the words of Patrick Le Lay, president of the French TV channel TF1:

‘TF1's job is to help a company like Coca-Cola sell its products. For a TV commercial's message to get through, the viewer's brain must be receptive. Our programmes are there to make it receptive, that is to say to divert and relax viewers between two commercials. What we are selling to Coca-Cola is human brain time.’

Motivated by well-documented media failures, Ellman and Germano analyse the delicate interaction between advertisers, media outlets and their readers to
see how this affects the quality and accuracy of content.

They think of media markets as ‘two-sided markets’. The media sell content to audiences who value quality and accuracy and, simultaneously, sell space to advertisers who value advert-receptive audiences.

Even with competing media outlets, advertisers who can commit to withdrawing their business can force the content and quality of reporting down to the level that would arise with no competition – namely, the same level that would be provided by a monopoly media outlet.

Their analysis also explains biases by other bodies – such as governments, banks or other businesses – that directly or indirectly contribute to a media outlet’s revenues.

Truly independent media might have played a key role in mitigating the current housing and financial crises.

At the same time, advertising finances a substantial amount of media content.
Ellman and Germano find that when advertisers are small and diverse, advertising in newspapers and magazines or on TV and the internet can be largely beneficial and can actually promote independent and diverse content.

The analysis therefore accounts for two quite different sets of observations:

Advertising played a valuable role at the turn of the 20th century (for example, in Britain and the United States) in helping newspapers become less dependent on states and political parties. This was a time when advertisers were relatively small and diverse.

Advertising has caused some notable failures – such as the non-reporting of tobacco-related health hazards over decades by commercial media, and, more recently, the misreporting of anthropogenic climate change – particularly in the United States, where tobacco and car companies have been among the top advertisers since the 1950s.

Applied to recent media policy debates, the researchers show that advertising increases the need for policies that enhance competition in media markets – in sharp contrast to recent deregulatory trends.

But because regulating competition is not always sufficient, it can be necessary to
regulate advertisers’ (and other sponsors’) influence directly. They comment:

‘Subsidising non-commercial outlets, such as the BBC and not-for-profit media organisations can reduce advertiser influence on media content because this improves the competitive alternatives that evade advertiser-driven bias.’

Clearly, subsidies should be designed to minimise the risk of political influence.

Ellman and Germano’s hope is to encourage policy-makers to look carefully at the relevant market structures when comparing alternatives. Delays in public awareness of issues such as global warming, housing bubbles, and excessive financial deregulation prevent democracies from adopting adequate policy measures.

So reducing media competition (deregulating) in response to the current financial straits of our media is, they maintain, definitely the wrong answer.


Notes for editors: ‘What do the Papers Sell? A Model of Advertising and Media Bias’ by Matthew Ellman and Fabrizio Germano is published in the April 2009 issue of the Economic Journal.

Matthew Ellman is at the Universitat Autònoma de Barcelona; Fabrizio Germano is at the Universitat Pompeu Fabra.

For further information: contact Matthew Ellman on +34 93 581 1207 (email:; Fabrizio Germano on +34 93 542 2729 (email:; or Romesh Vaitilingam on 07768-661095 (email: