The use of models by policymakers

Martin Sandbu writes that disagreement throws light on some deep questions about how economic theorizing is and should be used to inform policymaking. DeLong expressed doubts that the Fed’s analysis was indeed compatible with existing models; Krugman asserted that conventional models straightforwardly showed the Fed to be in the wrong. Summers, on the other hand, considers that the Fed’s mistake comes down to its attachment to the standard Phillips curve mode of thought. In making his argument, Summers also showed a certain willingness to listen to those who may have an untheoretical “feel” for the market.

Paul Krugman writes that the Fed seems to hold beliefs that are very much at odds with Macroeconomics 101, whose basic Hicksian models do not at all support the Fed’s eagerness to hike rates. I think I understand how being an official, surrounded by men (and some but not many women) who seem knowledgeable in the ways of the world, can create a conviction that you and your colleagues know more than is in the textbooks. And that may even be true in normal times, when recent experience counts a lot. But in a world of zero-lower-bound macroeconomics, which is a world nobody not Japanese experienced for three generations, theory and history are much more important than market savvy. I would have expected current Fed management to understand that; but apparently not.

Larry Summers writes that the issue is more on the supply side than the demand side.  If I believed strongly in the vertical long run Phillips curve with a NAIRU around five percent and in inflation expectations responsiveness to a heated up labor market, I would see a reasonable case for the monetary tightening that has taken place. Since I am not sure of anything about the Phillips curve and inflation is well below target I come down against tightening.  The disagreement comes down to the Fed’s attachment to the standard Phillips curve mode of thought.  My disagreement is reinforced by other judgmental aspects that are outside of the standard model used within the Fed.  These include hysteresis effects, the possibility of secular stagnation, and the asymmetric consequences of policy errors.

Larry Summers finds it unhelpful and likely wrong to attribute this to a failure to understand and appreciate basic macroeconomics. People with substantial experience and even a track record of making money by predicting markets, have important insights even if they cannot speak the language of “models” in the way I teach in economics. Hyman Minsky is an example of a scholar whose warnings were ignored in part because they were not formalized not because they were incoherent or illogical.

Models as discipline tools

Paul Krugman writes that since 2008 we’ve repeatedly seen policymakers overrule or ignore the message of basic macro models in favor of instincts that, to the extent they reflect experience at all, reflect experience that comes from very different economic environments. And these instincts have, again and again, proved wrong – while the basic models have done well. The models aren’t sacred, but the discipline of thinking things through in terms of models is really important.

Paul Krugman writes that my point, which is not that existing models are always the right guide for policy, but that policy preferences should be disciplined by models. If you don’t believe the implications of the standard model in any area, OK; but then give me a model, or at least a sketch of a model, to justify your instincts. What, after all, are economic models for? They are definitely not Truth. They are, however, a way to make sure that the stories you tell hang together, that they involve some plausible combination of individual behavior and interaction of those plausibly behaving individuals.

Simon Wren-Lewis writes that if the economics point towards a conclusion, and people argue against it based on ‘confidence’, you should be very, very suspicious. You should ask where is the model (or at least a mutually consistent set of arguments), and where is the evidence that this model or set of arguments is applicable to this case? Policy makers who go with confidence-based arguments that fail these tests because it accords with their instincts are, perhaps knowingly, following the political agenda of someone else.

Brad DeLong writes that models are both filing systems and discovery mechanisms. Models-as-discovery-mechanisms suffer from the Polya-Robertson problem: It involves replacing what he calls "plausible reasoning", where models are there to assist thinking, with what he calls "demonstrative reasoning", in which the model itself becomes the object of analysis. The box that is the model is well described but, as Dennis Robertson warned, there is no reason to think that the box contains anything real. Models-as-filing-systems are often used like a drunk uses a lamp post: more for support than illumination.

The blogosphere is still alive

Martin Sandbu writes that one thing the internet has achieved is to open up to the public the discussions between academic experts. Paul Krugman writes that discussions like this are basically only a possibility thanks to the internet. Getting three well-known policy-oriented economists in the same room with time for a substantive discussion is very hard. And to-and-from discussions in the journals are (a) relatively stiff and formal, (b) v-e-r-y s-l-o-w. In effect, the web has recreated in a virtual way the kind of coffee house discussions out of which the modern scientific journals emerged, without the necessity of all of us being in London and drinking incredibly terrible coffee.