Mark Buchanan on Wren-Lewis and alternatives to rational expectations

13 Nov, 2013 at 16:48 | Posted in Economics | 5 Comments

It seems to me that there are clear alternatives to rational expectations and I’m not sure why economists seem loath to use them. Simon Wren-Lewis gives one alternative as naive “adaptive expectations”, but this seems like a straw man. Here people seem to believe more or less that trends will continue. That is truly naive. Expectations are important and the psychological literature on learning suggests that people form them in many ways, with heuristic theories and rules of thumb, and then adjust their use of these heuristics through experience. This is the kind of adaptive expectations that ought to be used in macro models.

9781608198511From what I have read, however, the vast “learning literature” in macroeconomics that defenders of RE often refer to really doesn’t go very far in exploring learning. It seems afraid to go too far. A prominent review (by people Wren-Lewis mentioned) I read as recently as 2009 used learning algorithms which ASSUMED that people already know the right model of the economy and only need to learn the values of some parameters. I suspect this is done on purpose so that the learning process converges to RE — and an apparent defense of RE is therefore achieved. But this is only a trick. Use more realistic learning behavior to model expectations and you find little convergence at all — just ongoing learning as the economy itself keeps doing new things in ways the participants never quite manage to predict. As Simon Wren-Lewis himself notes, ” it is worth noting that a key organising device for much of the learning literature is the extent to which learning converges towards rational expectations.”

So again, it seems as if the purpose of the model is to see how we can get the conclusion we want, not to explore the kinds of things we might actually expect to see in the world. This is what makes people angry and I think rightfully about the RE idea.

I suspect that the REAL reason for this is that, if one uses more plausible learning behavior (not the silly naive kind of adaptive expectations), you find that your economy isn’t guaranteed to settle down to any kind of equilibrium, and you can’t say anything honestly about the welfare of any outcomes, and so most of what has been developed in economics turns out to be pretty useless. Economic theory loses its authority and most economists find that too much to stomach.

Mark Buchanan

Added November 14: Beating the dead horse of rational expectations …

5 Comments

  1. Dave,

    Levine is really just describing RE as economists use it. Buchanan et concortes keep on tearing down strawmen.

    RE never says that anything “has to be true”, but that forecasts are made as good as possible given the information available at the time. In fact, RE says that we are almost surely always wrong. But that we do not know ex ante whether we are wrong up or if we are wrong down.

    If a model is built on something other than RE, then the model will yield predictions which will suggest that the expectations used in the model was wrong up or wrong down. So if agents in the model would be informed of the model’s predictions, and these predictions were to be believed, the predictions would no longer be true. In RE models this contradiction does not arise.

    RE in its most naked form is indeed a bit stupid, and mainly used as a first-pass. Most important contributions deploy much more advanced versions of RE that carries names like “rational inattention”, “ambiguity aversion”, “robust control” and “global games”.

    Comment by Pontus— 14 Nov, 2013 #

    • Thanks. I really am trying to make sense of the article. It may be, as you say, that I misunderstood his technical work. I certainly recognize the general line of argumentation, but I still think that different economists have different views of RE, and my experience is that the nuanced view that I associated with Levine was not reflected in UK policy until too late – mores the pity.

      But the only way I can make sense of this Levine article is not that he has a stupid view of economics but that he has a very different view of policy making, and hence of what is required of economics to support policy. If policy makers are stupid, then RE is not the problem. Actually, too many UK policy-makers are less than totally rational, but even so I think the proper challenge for economists is to enlighten them. My experience of policy makers through the crisis supports this (hence my previous comments), but US experience may be very different.

      Levine’s main argument is essentially a ‘reflexive’ one. It is familiar and I will admit that it is logical in a sense, but perhaps not in an appropriate sense. (Whitehead / Russell.)

      Levine makes some interesting observations about artillery shells. He makes an analogy between predicting where the shell should go and predicting where economies should go. I have advised on the threat to civil aircraft from rogue artillery (missiles). The techniques used to consider what might go wrong are complimentary to those used to make central projections. If economic central projections are like predicting where artillery should go, then what we also need are ways of considering possibilities like those used for safety analysis. For example, if something goes wrong with a shell/rocket/missile, it is unlikely to go very much further, but it can drop very short. In the same way it seems to me that the possible deviations from routine ‘expectations’ are sometimes not symmetric, in which case some of the justification for RE looks shaky.

      I might agree that most policy-makers have certain desiderata for an acceptable economic theory, and that in this sense RE is the best that one can do. But if, as Levine says, policy-maker’s behaviour depends on what they believe, which beliefs need challenging? That would be an interesting article.

      Comment by Dave Marsay— 14 Nov, 2013 #

  2. Dave, that is a serious misreading you’re doing of Fudenberg and Levine. That article does not “undermine RE theoretically, and […] the slightest experience of booms and busts [does not] destroy[…] its empirical credibility”.

    For a more accessible article on Levine’s views on RE you can take a look here:

    http://www.huffingtonpost.com/david-k-levine/uncertainty-principle-economics_b_1220796.html

    As you can see, Levine is quite a vivid supporter of RE.

    Comment by pontus— 14 Nov, 2013 #

    • Thanks: a very important, and – for me – eye-opening link.

      It seems to me that Levine has gone beyond the notion of RE that is being criticised by Buchanan, and – in his papers – has a far superior version. The difference is what happens at the end of an epoch, such as the Great-Moderation. The view being criticised is that people just carry on with the old theories. Levine’s view can accommodate more general theorising, and hence is a different thing, generating qualitatively different dynamics.

      At first sight, the article appears to be a much-needed defence of rational expectations against behavioural economics, spoilt by the following mangled logic:

      In simple language what rational expectations means is “if people believe this forecast it will be true.” By contrast if a theory is not one of rational expectations it means “if people believe this forecast it will not be true.”

      Surely a theory that is not one of rational expectations could be one where one cannot forecast, or where even if people believe the forecast, it might not be true. (This seems to be the case.) And even the notion of RE is a little strange. Here in the UK people tend to be cynical about forecasts and they tend not to be true, and yet – usually – life goes on pretty much as if RE were true.

      BE is not the only alternative to RE, and critics of RE may also be critics of Be. (Their logic is also bad.)

      (As an aside, I wonder what Levine learned from his engagement with policy from 2006 on. Was he unconcerned? Frustrated as hell? He paints a very bleak picture.)

      Comment by Dave Marsay— 14 Nov, 2013 #

  3. It seems to me that http://scholar.harvard.edu/files/fudenberg/files/self-confirming_equilibrium_0.pdf undermines RE theoretically, and that the slightest experience of booms and busts destroys its empirical credibility.

    An alternative to RE would be to imagine that agents make decisions as best they can, based on the available evidence. But no better. And certainly without requiring them to predict the unpredictable.

    Comment by Dave Marsay— 13 Nov, 2013 #


Sorry, the comment form is closed at this time.

Blog at WordPress.com.
Entries and Comments feeds.