Microfoundations: A Critical Inquiry.

• Author: Froyen, Richard

Once, in the Maryland state legislature a member called for an end to a debate because "it had sunk to the level of principles." For economists the analogue is for a debate to become one about methodology. But sometimes it is useful to step back and ask just what it is that we are doing. This is what Professor Janssen does for the recent new classical and new Keynesian research in macroeconomics. Specifically, he considers the question of whether this research provides macroeconomics with individualistic foundations. In this sense, does recent research succeed in providing microfoundations for macroeconomics? For the most part his answer to this question is negative.

If such a negative conclusion is correct, it will certainly be a disappointment to those who carried out this research. The new classical economics of the 1970s was aimed at developing models where agents optimize (act rationally) and markets clear. The next generation of equilibrium business cycle models--real business cycle theories--sought to explain fluctuations as the responses of optimizing agents to changes in economic opportunities (e.g., productivity shocks) and shifts in individual preferences. The new Keynesians explicitly set out to provide rigorous microeconomic foundations for traditional elements of Keynesian economics. What (if anything) went wrong?

The best starting point for discussing Janssen's line of argument is his concept of methodological individualism (MI). Janssen defines MI as follows

. . . MI is interpreted as a view that allows only individual agents to be the decision makers in any explanation of aggregate economic regularities and aggregate economic phenomena. An explanation has to satisfy two conditions in order to conform with this definition: (i) it should be possible to represent it as a microreduction, i.e. an aggregation step has to be involved; (ii) the theory that is applied should be about the behaviour of individual decision makers.

The aggregation step is crucial. It is not just a step carried out after the individual behavior is specified, for the individual behavior depends on the way that behavior is aggregated. In particular, an individual's behavior depends on his or her beliefs about the behavior of other individuals.

Below I will give some examples to illustrate why Janssen concludes that much of the literature on microeconomic foundations fails to be consistent with MI, but first some general comments. Janssen's critique is wide...