The Vines and Wills article suggests two major changes in direction. The first, following Blanchard, is to partly reverse the microfoundations hegemony, by allowing models or parts of models that don’t include microfounded models routinely back into the top journals. The second is to focus on models that allow more than one long run equilibrium.
That’s Wren Lewis. About which the P³ tells us:
The two claims being discussed are so absurd it is staggering that they need be considered. What is being said is not that the macroeconomy is, in fact, simply the aggregation of the microeconomy, although that, of course, would be wrong in itself.
Err, no, that’s not what microfoundations says. It is obvious and clear that the total economy is indeed the aggregate of individual economic actions. Rather, microfoundations says that you should only be modelling the macro as that aggregation – this may or may not be true.
What is being said is that macroeconomics must be based on microeconomics or it is not considered suitable for discussion in peer-reviewed journals, whilst it also almost always assumes that there is one optimal solution to all economic problems.
Equilibrium and optimal are not the same thing. We know this from Keynes’ own macroeconomic ideas. That, for example, an economy can, via recession or an absence of demand, end up in a non-optimal equilibrium. If this were not true then there would be no point to boosting demand in recession. For the economy would not stay in that depressed position because there is only the one possible equilibrium, right? And we’re also stating, by recommending the demand stimulation, that at least one of those equilibria is non-optimal.
When Spud does things like this I do start wondering what people like Wren Lewis go on to think of him for doing so. Such immense missing of the point must grate at least at times, right?
Let me address the two issues, separately, and then together. Firstly, the presumption that there must be a microeconomic foundation to macroeconomic theory is simply to say that macro must be based on the critical assumptions that underpin micro that are intended to ensure that it can deliver mathematical solutions to problems too complex in reality to be subject to sensible mathematical analysis.
So, macro must presume perfect knowledge of both the present and future, perfectly competitive markets where no one participant can influence prices, perfect access to capital that is available to all who want access to it, and of course an absence of externalities.
No. You can, even using the microfoundations idea, still model those micro bits however you’d like. Which doesn’t, usually at least, assume perfect knowledge of the future but still. You can use oligopolistic competition in your micro if you like.
And before anyone says both macro and micro can manage deviance from these assumptions, of course I know that. But the point is it considers any situation where such things do not exist to be deviant.
Err, no, not deviant. It’s called relaxing an assumption in the model. You know, the very reason we use models. So that we can examine the impact of just the one change at a time, holding our other assumptions constant?
And the reason is linked to the second issue Simon challenges, which is the general requirement within macroeconomics that models be built where there is just one optimal or equilibrium scenario.
Optimal and equilibrium are not the same thing.
The question is what is more useful? Is it macro, based in micro, that is rigged to always suggest pure marker solutions to problems, because that is all it is programmed to deliver,
Except, of course, those entire libraries stuffed with books showing where pure markets don’t work and need steering, guidance and even entire replacement.