Various comments have suggested that I’ve been too nice, or not nice enough, about Real Business Cycle theory, the main contribution that got Kydland and Prescott their Nobel prize. My remarks weren’t really meant as a careful evaluation, so I’ll offer a slightly more considered view now, with the warning that there are plenty of readers of this blog who know more about dynamic macroeconomic theory than me – they can set me straight in the comments section. Non-economists may want to skip on to the next post.
It seems to me that the version of RBCT put forward by Kydland and Prescott had three main claims
* the methodological claim that macroeconomic models should be based on neoclassical microeconomic foundations (this claim wasn’t new with RBCT, but RBCT was the first well-developed theory that actually had standard microeconomic foundations)
* the technical claim that the best way to do macroeconomics was through simulation models calibrated to be consistent with certain ‘stylised facts’ about the economy, rather than through econometrically estimated models
* the empirical claim that the observed behavior of the economy could be explained as the outcome of optimal individual choices in competitive markets subject to real shocks affecting technology and preferences
Of these, the empirical claim was the most interesting and the least successful. I don’t think anyone actually claimed that the Great Depression was the product of a sudden increase in preference for leisure, but there were claims almost as silly. I imagine commenters who refer to RBCT as ‘barmy’ have this kind of thing in mind.
The methodological claim that a macro model should incorporate reasonable micro foundations is pretty widely accepted by now, though I and many others would argue that such foundations should allow for imperfect competition and bounded rationality, among other deviations from the standard neoclassical model. Over time, applications of RBCT have tended to include a range of deviations from the standard neoclassical model, as these tend to improve the capacity of the model to produce plausible results.
Finally, for a lot of people these days, RBCT mainly means using calibrated simulation. I’m quite sympathetic to this – it has for example given a lot of insight into the equity premium problem
John Quiggin 
Business cycles are caused by several things. I believe that one influencing factor is ‘real’ factors (preferences, knowledge etc). Previously, business cycles had largely been studied as a monetary phenom. RBC showed us that real factors were important also.
People who attack RBC often do so by building the strawman that RBC theorists believe that only real factors matter. I don’t know of anybody who takes RBC to this extreme. Though, of course, positing extremes is a useful way of considering and testing theories.
“I don’t know of anybody who takes RBC to this extreme.”
John, go and pay a visit to the Federal Reserve Bank of Minneapolis, and you will meet many people who believe that only real factors matter”.
There is no such thing as a “Business Cycle”.
Cyclicity implies some sort of physical repetiveness of an economic factor based on a particular index.
What and which?
I don’t have a specific comment on RBC theory, so much as a cheesy joke.
A group of macroeconomists are sitting on a panel at a conference discussing developments in the discipline.
In a heated exchange, the New Keynesian says to the Real Business Cyclist: “You guys have put macroeconomics back twenty years with this nonsense!”
The Real Business Cyclist smiles and says “So you DO believe in negative productivity shocks after all.”
🙂
John, you comment that micro-foundations “should allow for imperfect competition and bounded rationality, among other deviations from the standard neoclassical model”. If we allow for such things – and for all of them – do we have formal micro-foundations or verbal descriptions of what we think people are like – ie Keynes’ micro-foundations?
Most of the arguments in defense of RBCT boil down to saying that numerical simulation models of RBCs are fun to play with. Well, I agree with that. But I don’t agree that a macro model can only be respectable if it starts with Robinson Crusoe planning his infinite future in the face of random productivity shocks.