I read with interest a recent article by Greg Mankiw about media bias (HT Mark Thoma). It discussed research by Gentzkow and Shapiro on the US newspaper industry. They find that a newspaper’s political slant is governed by the politics of its readership rather than the politics of its owner. Of course television and radio are different from newspapers, and the UK newspaper industry is rather different from that in the US. But what interested me were Mankiw’s own closing remarks:
“These findings speak well of the marketplace. In the market for news, as in most other markets, Adam Smith’s invisible hand leads producers to cater to consumers. But the findings also raise a more troubling question about the media’s role as a democratic institution. How likely is it that we as citizens will change our minds, or reach compromise with those who have differing views, if all of us are getting our news from sources that reinforce the opinions we start with?”
Now even raising this question sounds rather radical, and potentially paternalistic. Should we stop people reading what they want to read, because we think it would be good for them to read something else?
The problem with political bias is that it covers a multitude of sins. At its most innocent, it can simply be presenting facts within a political or ideological context that matches that of the reader. In the UK, a lot of the reporting in the Guardian, or Telegraph, is of this kind. Where it gets more difficult is when this bias determines which facts get reported.
Take this report in the Daily Mail about the UK food banks run by the charity the Trussell Trust. (Or, as some government sources apparently like to describe the Trust, as a ‘business’ seeking publicity.) It is written as an expose, designed to give you the impression that the numbers of people using the Trust (almost a million) should not be taken at face value. It might just work, for someone who wants to avoid the implications of such a large increase in the use of UK food banks. Anyone with a different perspective would either laugh or get angry. (When the article was highlighted on twitter, donations to the charity increased.) You can spot one or two untruths: a headline says their reporter got three days of food ‘no questions asked’, when the article makes clear questions were asked. But mainly this article is biased because of its highly selective choice of facts presented.
(In reality a large part of the increase in demand at UK food banks seems to reflect mistakes made as a result of government changes to welfare, as this Economist article explains, or as John McDermott reports in this excellent piece of journalism. So the bias in this Mail report is not only in the ‘slant’, but the deliberate absence of key facts considered by the Economist and FT.)
At its most extreme bias can involve making things up. At the beginning of this year rules on Bulgarian and Romanian workers coming to work in the UK were relaxed. Here is a report in the Daily Mail online, indicating a rush of workers taking advantage of this relaxation. As this comprehensive discussion makes clear, the report was essentially a work of fiction. As this report shows, the costs to the newspaper of making things up in this case were minimal. (We have just had the figures for the first quarter: employment of Bulgarians and Romanians went down compared to the previous quarter.)
The traditional view is that slanting news and a highly selective choice of facts is tolerable, but making things up is not. The idea here is that people read newspapers on the understanding that they are based on facts rather than fiction, so including news stories that are made up is the equivalent of misreporting the sugar content on the packaging of some food. So the only thing that should worry us here is that whereas the manufacturer of a food product would presumably be subject to a large fine and damaging publicity if it got its food labelling wrong, a newspaper can effectively get away with it.
Yet this distinction does not address Greg Mankiw’s concern. If people are only told the facts that they are comfortable with, they will never change their minds. And as Paul Krugman observes (via Mark Thoma), if it is only the media read by your political opponents who will cry foul, politicians who just want to ‘play to the base’ are tempted to also distort or manufacture evidence, perhaps leading to descent into a world of fantasy.
Does this selection of facts actually influence people? Fewer people in the US think climate change is a major threat to their country than almost anywhere else (HT George Monbiot). The figure for the UK is also unusually low. This is particularly ironic as a good deal of the science telling us it is a major threat is done in these two countries. A major reason (pdf) why people in the US and UK think this way is that they are allowed the freedom to not to be told about the science, or to be given the opinions of skeptics as if they carried equal or more weight than the vast majority of scientists. (I look at other evidence of influence here.)
So this is an important issue. But can Mankiw be rescued from the charge of paternalism? (No, I never expected to write that line either.) I just want to raise two final thoughts. First, do we know that people are happy not to be told important facts that they might find challenging? If they had a choice between a newspaper that presented all the facts from their own ideological perspective, and another identical paper which only gave them the facts they wanted to hear, would they really choose the latter? Perhaps people do not get that choice. Second, even if they might sometimes choose the latter option, do they want that option? Perhaps it is a bit like being on a diet. We know we might be tempted never to be confronted with facts that challenge our priors, but we know we really should be. So just as those on a diet would rather not be constantly faced with the choice of lots of fattening foods, so perhaps people would not object if they did not have the option of just hearing what they wanted to hear.