« links for 2009-04-11 | Main | Quick Note »

Saturday, April 11, 2009

Macroeconomic Meltdown?

The state of macroeconmics:

Are those who sweat the big stuff in meltdown?, by Tim Harford, Commentary, Financial Times: ...I am struck by the soul-searching that has gripped the [macroeconomics] profession in the face of the economic crisis. The worry is not so much that macroeconomists did not forecast the problem – bad forecasts are more a sign of a complex world than intellectual bankruptcy – but that macroeconomics seems unable to provide answers. Sometimes it cannot even ask the right questions.

Willem Buiter ... complains that macroeconomists have simply discarded the difficult stuff to make their models more elegant: “They took these non-linear stochastic dynamic general equilibrium models into the basement and beat them with a rubber hose until they behaved.”

He is not alone in his frustration. Paul Krugman ... thinks macroeconomics is in a dark age, in the sense that rather than discovering new insights, we are actually going backwards and forgetting what we used to know. Mark Thoma ... opines: “I think that the current crisis has dealt a bigger blow to macroeconomic theory and modelling than many of us realise.”

We shall see. While many commentators have reached for Keynes – or some caricature of Keynes – as a solution to this crisis, this is not because he is the fount of all knowledge, but because he was asking good questions about problems that now seem relevant again.

Economists now understand much more than Keynes ever could about networks and complex interactions (thanks to agent-based modelling), psychology (thanks to behavioural economics) and the real world (thanks to econometrics). In principle, these advances should inform our understanding of the crisis. An early attempt is Animal Spirits, a book by George Akerlof, a Nobel laureate, and Robert Shiller, who identified the housing bubble early. But macroeconomics has a lot of momentum and it will take time to turn the oil tanker around.

Justin Wolfers,... an unabashed microeconomist, says that, “formally elegant but empirically irrelevant macroeconomists had a much harder time getting hired this year,” while Buiter reckons that the central banks have already jettisoned conventional macroeconomics in favour of a pragmatic combination of hunches and judgment calls. If so, the market for macroeconomic ideas seems to be self-correcting – much like the market for financial weapons of mass destruction. It is just a shame, in both cases, that the correction did not come more smoothly and much, much earlier.

Let me a bit more specific, and add something more to problems with macroeconomics I discussed in The Great Multiplier Debate and "The Unfortunate Uselessness of Most 'State of the Art' Academic Monetary Economics". The main mechanism generating fluctuations and policy effects in modern New Keynesian models is Calvo type sluggish price adjustment. I think this model is useful for “normal” times as a way of understanding economic fluctuations, and for learning about optimal policy, and it represents a step forward in understanding monetary policy in particular. But do people really think that all would be fine right now if prices – and they must have housing prices in mind when they think about sticky prices as an explanation for the current episode – had only adjusted faster? If housing prices had dropped even faster than they have already, all would be well in the world?

Okay, so maybe they don’t have housing prices in mind. Still, do we really think that sluggish price adjustment is the main mechanism at work in the present crisis? If not, then what use is the evidence from those models? Why do we keep hearing about theoretical simulations that give values for the multiplier that are small, large, zero, less than one, whatever? Do we really think that sluggish price adjustment captures the essence of the factors driving the present crisis? I don't.

The fundamental mechanism driving the economic fluctuations is wrong, and people seem to be missing that when they try to use the evidence from this model to comment on the present situation. There are two big problems. First, we have very little data from episodes like the current one to calibrate these models. What should, say, the elasticity of labor supply be in a severe recession? Do we know? That's a key parameter in these models, and we can only guess what it's value ought to be. Second, even if we have good data from similar episodes, why run it through a model that does not capture the fundamental problem driving the downturn?

Again, I think the New Keynesian model is very useful for understanding "normal" business cycles, but I am very hesitant to use this model as the basis for policy advice in the present crisis.

So where does that leave us? We do not have either the theoretical models or the empirical evidence we need to understand this episode thoroughly and completely, and to provide the policy advice that will cure the problem with any degree of certainty. Without solid theoretical models and the associated empirical evidence, we really have no choice but to fall back upon older models that were "built to answer the questions that are important at the moment," i.e. the old-fashioned Keynesian model, and to rely upon loose, but solid theoretical principles rather than a tightly constructed model and vast amounts of empirical evidence. It's quite understandable that economists who have been striving to push the profession in a positive, scientific, solidly theoretical and evidence based direction would resist going backward, and resist strongly, but what choice do we have? Until we have a better mousetrap, the simple, old fashioned one will have to suffice.

    Posted by on Saturday, April 11, 2009 at 01:17 AM in Economics, Macroeconomics, Methodology | Permalink  TrackBack (0)  Comments (50)


    TrackBack URL for this entry:

    Listed below are links to weblogs that reference Macroeconomic Meltdown?:


    Feed You can follow this conversation by subscribing to the comment feed for this post.