The knowledge transmission mechanism

In a comment on my last post, Joseph Grossman asks “If the vast majority grasp and support the basic shape of the [fiscal] stimulus solution, and if we live in democracies, isn't it time to shift the analysis to expose the exact and precise mechanisms by which our electoral systems are failing miserably?” This is the question which, since 2010, I have asked myself almost every day. The question becomes even more relevant as the intellectual case for austerity crumbles, but the policy continues, and in some cases even appears to gain ground. There may be some answers that are specific to austerity: see Paul Krugman here or myself here. But in this post I want to use this example to look at the question of the transmission of economic ideas more generally. So let’s break the question down.

First, do the vast majority of economists agree? In the case of fiscal policy, I think the honest answer here is: majority, probably yes, vast, almost certainly no. For example, in this survey the vast majority did agree that the 2009 US stimulus did reduce unemployment. It would be very surprising if this were not the case - after all this is what we teach first year economics students, and it would be a very strange discipline indeed that taught its students something it also thought was wrong. (For this reason, I would love to know the results of a similar survey of German economists.)

Yet on the question of whether it was a good policy (benefits exceeded costs), only 46% agreed, and a large 40% were uncertain or did not answer. That is not a vast majority. I don’t think there is a single reason why so many economists are equivocal. Some worry about government debt levels, others have a deep distrust of government, still others have faith that this is better done by monetary policy, despite the ZLB. I suspect those numbers would be more favourable if the survey was restricted to macroeconomists, but there would still be a significant number who would be unsure.

However, what the majority - vast or not - of economists think would be irrelevant if no one listened to them. The transmission mechanism from economists to economic policy works along many channels. It may be direct. It may be mediated through the civil service. It may work through economists influencing popular opinion, which then influences policymakers. I think the last of these is the least important. In part this is because most people do not have the time and inclination to interest themselves in what economists think: I was going to say regrettably, and I am heartened and encouraged by those who do take an interest, but I doubt if it is reasonable to expect most people to try and find out directly what the arguments are on economic issues.

That is not to say that people do not have opinions. What this does mean is that their opinions come not directly from what economists think, but from how the media discuss economic issues. One way this could work is that people in the media consult enough economists to find out what the key issues are and what the balance of opinion is, and report that to the public. While some of this goes on, to suggest this is how the media generally works would be naive. To see why it is naive, it is useful to look at direct links between policymakers and economists.

Again, a naive view here is that politicians do what the ‘ideal’ media would do. But nearly all economic issues involve winners and losers, either directly or indirectly. Most politicians support the interests of particular groups. So politicians will select to talk to those economists who support policies that favour those interests. They will have little regard to whether those economists are in the majority or not. A classic example is the Laffer curve. Hardly any economists believe that tax cuts increase tax revenues, yet the Republican Party looked to the few who did, and it became a party line. (Paul Krugman, in his first book, talked about the ‘policy entrepreneurs’ who intermediated this process. Nowadays we have think tanks: some good, some propaganda factories.)

Now if we had a media that faithfully reported what the majority of economists thought, and questioned politicians on this basis, then this kind of self selection by politicians would happen a lot less, because it would run the risk of being exposed. This clearly does not happen, as Stevenson and Wolfers lament here. There seem to me to be three main reasons.

1) Those in the media react to incentives and pressure, like anyone else. So when Stephanie Flanders made the obviously correct comment that growth in UK employment despite sluggish output might not be good news because it meant productivity growth was low, she was jumped on by Conservative politicians shouting bias. Did the BBC ignore these complaints and tell the politicians to get real - like hell they did. Yet if someone in the media writes something that does not make sense in terms of what academic economists understand, do the massed ranks of professors stand up and loudly complain? It was a rhetorical question. Occasionally groups of economists write letters, although these are pretty ineffective, either because they get ridiculed, or because it provokes an apparently equivalent letter from the ‘other side’. (See Alan Manning here.) Which brings us to ...

2) For sections of the media that do have an interest in truth rather than propaganda, the perception of being unbiased is terribly important. So when an economic issue that is politically divisive comes up, the natural response is to report ‘both sides’. This is the origin of the ‘opinions on shape of the earth differ’ jibe first suggested by Krugman and immortalised by DeLong (e.g. an early example here). It is much more difficult, and risky, for the media to report which side is in the majority among the experts. So the public just end up thinking that economists disagree all the time, even when they do not.

3) A large section of the media in most countries is politically controlled. Just as it is unnecessary to own the majority of shares to control a company, it is not necessary to control all the media to have a pervasive and defining influence. Of course obtaining ‘control of the means of information’ costs money, so it is not something that ‘both sides do’ in equal measure. The fact that this political influence works more through TV in the US and through newspapers in the UK is an interesting difference, but that it works through some means is not an accident, but an entirely predictable feature of our modern democracy. What I find interesting is that many people, including some academics, appear to deny its importance.

I have become a bit obsessed by all this as a result of the widespread adoption of austerity policies, and their remarkable persistence despite apparent failure. It is interesting to try and assess how important each element is: for example would it make much difference if the vast majority of economists thought fiscal stimulus was both effective and a good policy? One way to judge this is to look to other areas where science and politics clash, like climate change, or even badgers (on rereading, one of my better posts). It is an issue that scientists in general have become increasingly concerned about: the introduction to this collection of essays from the American Academy covers many of the points raised here, and more. However there is a tendency to revert to the old line that academics must communicate more and better, and glide over some of the structural weaknesses in the transmission mechanism that mean it would make little difference if they did.

My own current view is that these structural weaknesses are to a large extent inherent in liberal democratic societies, where restrictions on what money can do are very limited. That has led me to be much more favourably disposed to the delegation of economic decisions, even though this appears less democratic, and can be seen as representing arrogance and self-interest by the academic community. Yet the problem is real enough. And it is personal: when you study, teach and research in a subject where some of its most basic findings - understood for more than half a century- can be brushed aside so easily, and millions of people are worse off as a result, you have to ask yourself what the point is.