Oliver Latham investigates whether a government’s popularity has an effect on the level of scrutiny it will face in the media – finding that a scandal hitting a government eight points behind in the polls will receive sixty per cent more coverage than an identical scandal hitting a government that is eight points ahead.
British celebrities and sportspeople have long complained that the tabloid press “builds them up and knocks them down”. Newspaper coverage is initially excessively positive, but, as the public begins to tire of the latest flavour of the month, it switches to being excessively negative, with the slightest misdemeanour enough to attract vitriolic coverage in the press.
Is there a similar phenomenon in political reporting? Recent experience suggests there might be: the highest-profile political scandals, such as the cash for questions affair and the MPs’ expenses scandal, seem to hit governments that are already unpopular. In contrast, when Tony Blair was at the height of his powers he earned the nickname “Teflon Tony” for his ability to emerge from sleaze allegations unscathed.
The recent furore over the relationship between the government and the Murdoch media empire is also a case in point. Many of the key elements of the allegations (the close relationship between senior politicians and the Murdochs, the possibility that Andy Coulson might have known about the phone-hacking while he was editor at the News of the World) have been known for some time. However, it is only now, after the coalition’s honeymoon period has ended, that the media storm has begun in earnest.
In my paper “Lame Ducks and the Media” I use statistical techniques to examine whether the press really does turn on unpopular governments. To do this I first used key-word searches to count the number of articles printed by British newspapers about each of around eighty investigations by the Standards and Privileges Committee into government MPs. I then used regression analysis to examine whether increasing the government’s lead in the polls results in less articles being printed about the investigations. (Actually things are a little more complicated than that, but I’ll leave that discussion to the end).
The results suggest that the press does indeed turn on unpopular governments and the effect is significant: the results suggest that a scandal hitting a government that is eight points behind in the polls will receive sixty per cent more coverage than an identical scandal hitting a government that is eight points ahead. This corresponds to dozens or even hundreds of additional articles across the eighteen newspapers in my sample.
Interestingly I also find that this effect depends on the ideology of the newspapers involved. For example, during the last Labour government, the effect of popularity was much stronger for right-wing newspapers (like the Telegraph or Mail) and nonexistent for left-wing ones (like the Guardian or Independent). The implication is that, while left-wing newspapers consistently gave less coverage to scandals affecting Labour MPs, right-wing newspapers were more opportunistic and altered their coverage depending on public opinion.
In some respects these results are good news for the coalition: if they believe that the main reason for their current poll slump is the state of the economy then, as long as the economy improves before the next election, they can rely on the support of a more favourable media. On the other hand, it implies that the on-going crisis in the Eurozone could have a doubly negative effect: not only will it directly harm their chances of re-election by hitting voters’ wallets, but this effect will also be amplified by attacks from an increasingly unfriendly media.
Methodology
The description above may have raised eyebrows. Correlation does not imply causation and so finding that scandals affecting unpopular governments get more coverage does not tell us anything about the media’s role. For example, it could be that particularly sleazy governments are both unpopular and more likely to be hit by serious scandals that attract more coverage. In addition, statistical work is complicated by the fact that government popularity is itself determined by media coverage.
In order to get round these problems I use a technique called “Instrumental Variables”. Essentially I look at how changes in popularity brought about by changes in unemployment affect the number of articles printed about each scandal. Because changes in unemployment directly affect government popularity, but should have no effect on the public’s interest in political scandals or the propensity of MPs to fiddle their expenses, indulge in extramarital affairs or otherwise engage in activities of interest to the press, I am able to uncover the causal effect of changes in government popularity on coverage.
The full paper on which this article is based is available here.
Note: This article gives the views of the author, and not the position of the British Politics and Policy blog, nor of the London School of Economics. Please read our comments policy before posting.
Oliver Latham is a PhD student in Economics at the University of Cambridge
