Thursday, 14 September 2017

Economists show how Fox news changes votes

As I have noted before, economists are getting into media studies (we are natural imperialists) and beginning to provide empirical evidence on an age old, and critical, debate. Are some media outlets biased simply because their viewers or readers are partisan, or do these media outlets play a causal role in changing political views? And do readers/viewers discount the bias in media outlets, or does this influence how they vote?

We now have clear evidence on this in the case of US News Channels, and the answer is that Fox News changes votes in a big way: by magnitudes easily enough to swing elections. Those who argue that partisan right wing media does not matter very much now need to bring some counter-evidence to the table if they want to sustain that position.

The latest piece of evidence has just been published in the American Economic Review. Why is the study in a top-rank economics journal? One thing empirical economists are used to doing is looking for good ‘instrumental variables’, which in this particular case, means finding something that influences whether people watch Fox News that has nothing to do with their politics. If you then look at what these ‘accidental’ viewers do, that gives you a handle on the causal role of watching Fox News.

Dylan Matthews at Vox has an excellent explainer piece on the research, so I will not repeat the details here. Let me instead just list four key points.

  1. Fox News is the dominant (by primetime viewers) US news channel. A sizeable minority of Fox News viewers are not Republicans.

  2. The study’s data does not go beyond 2008, but in that year they estimate that if Fox News had not existed, the Republican vote share would have been over 6% points lower.

  3. In contrast the more left wing MSNBC was far less effective at gaining votes for Democrats.

  4. Fox News is not just setting its ideology to maximise viewers - it is much more Conservative than that. Instead it choice of ideology maximises its persuasive power.

The ‘headline figure’ of over 6% points in point (2) is an overestimate because other networks do shift their ideology to gain viewers. So if Fox News disappeared, other networks might have shifted right to capture ex-Fox viewers. But the key point is that Fox is acting in a way to maximise the propaganda power of its extremely right wing message, and it is successful in changing a significant number of voters minds. The apologist line that the media is ‘just reflecting the views of our readers/viewers’ does not hold for Fox News. In short, it is a propaganda organisation, not a ‘for profit’ news organisation. During one sample period, an analysis of the content of Fox suggested that over half the facts it reported were untrue.

One of the nice things about the study is that its results are reasonably consistent with earlier work based on the initial roll out of Fox News in 1996 to 2000 (which I referenced here). That earlier analysis used a different method to identify the causal impact of Fox on voters choices, so it is good that two different methods come to similar conclusions.

To some this will be no surprise (except perhaps that the research is in an economics journal). To paraphrase Obama, if he watched Fox News even he wouldn’t vote for himself. It fits my own account of how US politics has shifted so far to the right, and has become so partisan. To UK readers the obvious question is whether we can just substitute Mail/Sun/Express for Fox. In my view you can, and Brexit goes a long way to proving that. But explicitly or implicitly, the consensus view appears otherwise. Whether from left, right or centre, analysis of voter behaviour typically ignores or marginalises the role of the press. This and earlier studies suggest that is no longer good science.

Fox and their UK equivalents in the press have an importance way beyond academic studies of popular opinion. The story of Fox News and Trump are inextricably linked, as is Brexit and the right wing press. When you can serve large sections of the population real fake news - news that denigrates particular minorities or ‘outsiders’ and pushes the political views of their owners - then tyranny becomes quite compatible with democracy. Governments wield the ‘will of the people’ against pluralism and the rule of law, all enabled and even dictated by media as propaganda. We have examples within the EU and on its doorsteps. We are not there yet in the UK and US, but we are getting very close.



5 comments:

  1. May I please share this article on social media?

    ReplyDelete
  2. Sorry to be a wet flannel, but economists have also established that austerity in a recession is bad, and that Brexit is a terrible idea.

    Funny thing is, those policies still get implemented.

    I await with optimism the revolution in press regulation...

    ReplyDelete
  3. Last year sometime, Mika Brezinsky, joint presenter of NBC's Morning Joe programme blurted out words to the effect that "we tell people what to think". The discussion proceeded as if nothing had happened.

    All these news channels are after influence; Fox appears to be that much more successful.

    The more fundamental point is why Fox News succeeds and why there are so many who digest its message and for this you are better to look at Thomas Frank: "What's the Matter with Kansas" rather than the academic survey you quote to see how ordinary people can adopt a political position that is antithetical to their own interests.

    ReplyDelete
  4. You neglect the other side of the equation. Obama was popular because he was hopeful and promised change, running on a mildly populist program. Hillary did the opposite. Sanders and Corbyn did astonishingly well by putting forward bold programs.

    Centrist neoliberals have a material interest in downplaing that part of the equation. Their policies aren't popular except to their donors.

    ReplyDelete
  5. I found this thread on twitter (linked to initially by Paul Krugman) very instructive in the matter.

    It's central argument is that right wing US voters beliefs are shaped by what the right wing elites believe/tell them to believe. Ie. about climate change.

    https://twitter.com/drvox/status/906228932575408131

    16. To *very* briefly summarize: people don't know anything; they don't have strong opinions on political "issues"; they form opinions ...

    17. ... by following the cues of leaders in their various social tribes. We are social creatures; tribal ties (not "issues") are primary.

    18. So conservatives believe ... what conservatives believe. And they find out what conservatives believe from conservative elites.

    19. That means conservative pols, celebs, and local leaders, but especially, in US conservatism circa 2017, *media figures*.

    And seeing as Fox is owned by Murdoch, it means in the US Murdoch has the ability to (and in practice does) exercise pernicious influence over the electorate.

    An interesting thing is that the suggestion is that this is particularly the case for republican voters in contemporary US, but isn't the same for liberals and liberal media. And likewise, by extension, the effect isn't likely to be the same in the UK and UK media. Ie. the extent to which media will influence electorates depends on the extent to which these electorates are influenceable and also the extent to which those being influenced (presumably without realising it) believe that these news outlets represent them, share there values.

    I definitely think the Mail/Express have a pernicious effect on UK politics, but would question whether they are directly substitutable for Fox news. US society is a lot more split than UK politically and therefore less of an insider/outsider mentality in terms of consuming news only from outlets that you believe 'represent' you.


    ReplyDelete

Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.