Thursday, 24 April 2014

Gordon Tullock on the Clark Medal

Last week, I blogged about Chicago Booth Professor Matthew Gentzkow winning the Clark Medal. What I neglected to do was to highlight a part of the provenance of the idea of media slant and its relation to media consumers. To reiterate, according to research by Gentzkow and fellow Chicago economist Jesse Shapiro, newspapers cater to their customers, so that a left-of-centre audience will get mostly left-of-centre news, and so on. The point of this blog post is to relate this idea to work by Gordon Tullock.
I should say that I don't really know how far back this idea goes, but Tullock's enjoyable book Toward A Mathematics of Politics does provide a nice treatment. Tullock talks about many interesting things in this book, so I will restrict my focus to the parts which concern the research discussed above. Tullock's treatment of the issue is largely speculative and he has no empirical evidence the way Shapiro and Gentzkow have, but - as is Tullock's wont - he nevertheless manages to capture the essence of the issue. Also, Tullock is not cited by Gentzkow and Shapiro so this blog post might add some value for those curious about the underlying ideas. (I don't mean to call negligence on Gentzkow and Shapiro, who go well beyond Tullock's analysis, I only state a fact.)
Below are two graphs (click to enlarge) Tullock made for the book. The vertical lines show the position of a media consumer. The slanted lines capture willingess to pay (vertical axis) for given media positions (horizontal axis). Tullock suspects that willingness to pay should be highest for media in agreement with one's own politics  (panel a), although he also holds it possible that a media consumer would be prepared to pay a premium to have his position challenged (panel b). This fits well with the idea that markets provide the views which are sought.
An interesting aspect of Panel (b) is the possibility that newspapers and other media change the minds of its consumers. In Panel (b), this would mean that an individual starting out as a centrist (say), can drift left or right as he is influenced by the media he consumes. As he drifts, he exerts some pressure on the media to change their slant, causing further drifts. I don't know, but I would think that this situation is unlikely to apply in reality. My impression is that the media landscapes of many, many countries are rather stable. Since Panel (b) allows for very radical changes over time, we should have seen more of them if Panel (b) is the Truth. But these are only my vague impressions and I am not confident in their accuracy. However, Panel (b) does not have to be interpreted in that way. An alternative interpretation is that the preferences of the media consumer can remain the same; he merely enjoys being challenged, but he won't actually change his mind. I am insufficiently familiar with Gentzkow and Shapiro's research to know whether they can answer which type of panel is more likely to apply in the real world.
In addition to these insights on media slant, there is another implication by Tullock's graphs and by Gentzkow and Shapiro's research which says that advocacy is really difficult. Particularly if one tries to advocate radical change (i.e. movements towards positions which are unusual, such as open immigration or the abolition of mandated minimum wages). This is because it will not pay as much to publishers and other providers of platforms for reaching the public to spread unusual views as it will to spread common ones. Aggregate willingness to pay is greater for the latter. Of course, advocacy is no easier if Panel (b) is true but the interpretation is that no changes of mind occur. This brings to mind the related issue of the impotence of ideas.

No comments:

Post a Comment