Macroeconomic Theory: Thoughts About Its Future 2

In this column, I continue the discussion opened up yesterday by offering the reader the benefit (if benefit it is) of  insights derived during my lengthy career in economics.  I close the column with a statement of methododology:

3.  Do not expect too much of any single model

There are many macroeconomic models now available, most of which have something to offer in specific circumstances: classical, Keynesian, new classical, new Keynesian, monetarist, public choice and Austrian.  Environments vary from fixed exchange to floating exchange rate systems, capital mobility to exchange controls, etc. In my judgment, the loss of generality that results from picking and choosing among the models according to circumstance is worthwhile. Economists should be concerned more with explanation and prediction, and less with reputation-enhancement, than sometimes is the case. This is especially relevant when analyzing cyclical behavior, where specific triggers need to be addressed. 

4.  Do not hesitate to endogenize political actors in the macroeconomic model

Because many macroeconomic theorists are not schooled in public choice, they too often  inappropriately treat ‘government’ as an exogenous variable, an independent mover whose role is to stabilize the economy. This poses a number of problems. First, in the United States, at least, there is no such actor as ‘government’.  The separation of powers provides a competing set of political actors (President, House, Senate, Federal Reserve) whose interactions forge outcomes that no one initially had planned. Second, all political actors are motivated by objectives that require identification: personal wealth accumulation, vote maximization, campaign funding and ideology and,  in the case of the Fed, budget  rank high among such goals.  If one is modeling the macroeconomy for predictive purposes, all such actors should be incorporated into the model as self-interested players who respond to political market incentives.  The trick is to achieve this while still keeping the model as simple as possible. Public choice scholars have shown great ingenuity in developing such models. Mainstream macroeconomists remain largely unaware of such contributions.

5.  Always respect the logic of scientific discovery

Sir Karl Popper and his student Lakatos together provide an admirable logic for the evolution of deductive science.  For Popper, the most ruthless falsificationism is essential.  Every theory must be open to falsification, or it is not a scientific theory. Every theory must be tested as vigorously as possible. Any theory that fails the predictive test must be discarded, at least when an alternative better-performing theory is available.  Lakatos adds a dash of moderation to this ruthless formula. Theories that fail the predictive test must be recalibrated, but the core postulate (say rational expectations or adaptive expectations) may be retained. The research program remains progressive as models are recalibrated as long as the process of recalibration does not render them increasingly immune to the process of falsification. The Popper/ Lakatos criteria, in my judgment, should govern macroeconomic modeling.

In closing this column, I should like to refer the interested reader to my recent (September 2009)  book (co-authored with Nathanael Smith) and entitled: Economic Contractions in the United States: A Failure of Government.  This book is available from www.amazon.com and sells for $12 plus shipping. This short book draws informally upon a number of models (including public choice) that we deem to be relevant, calibrated against the experience of the Great Depression, and then applied to the several year  build-up to the 2008-2009 financial crisis and economic recession.  It has been well-received, so far at least, by  both  the  international media and the market-place.

Tags: adaptive expectations, Lakatos, logic of scientific discovery, Popper, public choice, rational expectations