Media Concentration and 'share of revenue'
17 March 2014
There has been much discussion lately about media power and concentration of ownership, with a number of different proposals of how it should be measured and then limited.
Power and influence are nebulous things, and there will never be a metric that is better than vague. To make matters worse, any chosen measure will no doubt become victim to Goodhart’s Law: "When a measure becomes a target, it ceases to be a good measure."
The European Commission funded a research programme to design a survey tool to measure media plurality across the EU. The outcome included 120 different indicators, all of which need to have somewhat arbitrary weights attached to them so as to produce a single measure. Any attempt to base commercial laws on such a mechanism will invest extraordinary discretionary political power in the hands of the regulator entrusted with it. The regulator would no doubt become the focus of intense media attention. It seems unlikely to benefit anyone except perhaps the legal profession.
This begs the fundamental question: Is there any metric that is simple enough to be practicable, and yet will have any discernible success in reducing or preventing concentrations of media power?
One often flagged suggestion is the Enders Analysis proposal that media ownership limits should be based on share of revenue, the rationale being that revenue is simple, visible and a proxy for influence. But is it?
In political-economy terms, influence can be seen as a good that can either be consumed or put to productive use. But before that, it must be paid for. It doesn’t generate revenue automatically, but must be deployed in the service of some other commercial interest.
Take two examples: the Guardian and the Daily Mail. The Guardian and its parent the Scott Trust exist to ‘serve the liberal interest’. In other words it exists to influence. There is no profit to be made in serving the liberal interest, rather it costs a good deal of money, which is why the Scott Trust is there to make profitable investments to cover the costs of the influence that the Guardian consumes.
The Daily Mail is at first glance different, but is in some ways similar. As a business it is profitable, unlike the Guardian. However if, instead of just the Guardian, we compare it with The Scott Trust as a whole, we can ask whether the politically influential part of the Daily Mail makes a profit or a loss. When it publishes articles about for instance Ralph Miliband, does it really expect them to contribute to its sales, either of advertising or copies?
Theoretically, the news content of the Mail can be separated into influence and entertainment, where entertainment includes news about celebrities or the health risks of going abroad, and economically is equivalent to the profitable assets of the Scott Trust. Influence on the other hand is not inherently profitable because its purpose is not to attract readers and the revenue that comes with them, but rather to influence them.
Both the entertainment and the influence may be found in the same article, or even the same word, nevertheless, it is still distinct, because if the newspaper were run on an entirely commercial basis, there would be no room for influence, unless that is, it had some other commercial benefit, like for example the effect it might have on the tax status of its proprietor, or the loosening of media ownership restrictions.
Proprietors might argue that people enjoy having their beliefs reinforced and confirmed, or even antagonised, and they are simply meeting that demand. However, if that were the case, you might expect some roughly symmetric distribution of political leaning around the centre of public opinion. But when compared with election results, a big majority of the total circulations of newspapers are right of centre, and opinion polls show that their readers do not in fact share their ‘values’.
So the relation between influence and revenue is indirect at best. And if the influence does generate revenue then it could show up not in the newspaper’s accounts, but in some completely different financial interest. And in the case of a Berlusconi type proprietor, or a conglomerate, it may be something a long way removed.
A vast amount of news content is now available for free, and a great many news outlets do not seem to be financially viable, and yet they rarely close down, but seem to be able to lose money in perpetuity. Somebody somewhere always seems to see some value in them.
Revenue therefore is not a good proxy for scale and reach. To measure influence in terms of money, we need to measure not sales and advertising revenue but the opportunity cost of influence. If the Guardian exists for the sole purpose of influence, then this is straightforward. You can take the total cost of production and subtract sales and advertising revenue – in other words the total losses.
For the Mail and most other media organisations it is much more difficult, because you can’t separate the loss-making influence from the profit-making entertainment.
In any event, revenue is next to useless as an influence metric, particularly if you don’t include the revenue from the wider interests of the owner or organisation, which may be the beneficiary of the influence. Expenditure, which includes losses, is a better indicator, at least of scale and reach.
But rather than having thresholds and adjudicators, whose independence will always be at issue, wouldn’t it be better to have some automatic mechanism that realigns market incentives with the public interest? Currently the public interest is to have less media concentration, while the market incentive is to have more. To create diseconomies of scale to balance the drivers of concentration, we need some progressive taxes on scale. If there is no single indicator of scale then there should be no single tax, and for example there might be one tax on advertising and another on operational costs, both increasing with scale, like income tax and council tax.