"Macroeconomics as an Autonomous Discipline"

"Paradoxically, the main result obtained by the new classical economists is the demonstration - against their wishes and expectations - that a satisfactory synthesis of macroeconomics and microeconomics is not yet mature. As a matter of fact, the micro-foundations of macroeconomics which they suggest are by now far from satisfactory. They rely on the heroic assumption that the decision-makers of the models are representative agents, whose behavior fairly well approximates the aggregate behavior of the economy. Unfortunately this assumption surreptitiously eliminates the main object that should be studied by macroeconomics: aggregation problems and failures of coordination between the behavior of individuals. Even so, the suggested micro-foundations work only under very special assumptions which actually deny any importance to the main problems considered by Keynes's macroeconomics: uncertainty, disequilibrium, instability, structural change, etc. As we have seen, disequilibria are assumed to be non-intelligible and are therefore ignored; uncertainty is emasculated by the 'certainty equivalence' hypothesis; instability is defined away by arbitrarily restricting the analysis to stationary and ergodic processes and taking account only of the subset of stable solutions...

The failure of this reductionist research programme may be due to the immaturity of current macroeconomics, but it may also be due to weaknesses in existing microeconomic theory. Notwithstanding the widespread belief in its intrinsic solidity, the micro theory currently accepted by the new classical economists may prove on closer examination to be insufficiently powerful to provide solid foundations for a satisfactory macroeconomics. To take the preliminary steps towards a real synthesis between macro and micro theories, it is necessary to consider not only the micro-foundations of macroeconomics but also the macro-foundations of microeconomics (Hicks 1983).

The history of scientific thought shows that whenever a synthesis between different disciplines has been successfully accomplished, the result has been a new discipline with features not reducible to those of the original disciplines. Such a synthesis between micro and macroeconomics, if it is possible, is still far away. In the meantime the reciprocal autonomy of disciplines should be carefully safeguarded. It is particularly important to defend the autonomy of macroeconomics, as today this is greatly jeopardized by views like those mentioned above. Therefore we should revert to the original Keynesian concept of macroeconomics as an autonomous discipline. This does not imply that we should give up making serious efforts to provide rigorous micro-foundations for our macroeconomic statements, if that means searching for greater consistency between the two disciplines. In other words we should continue to pursue a full synthesis between microeconomics and macroeconomics. Many things have been learned from past attempts, unsuccessful as they were, and many others may be learned through future efforts.

But in the meantime one should not reject as non-scientific any contribution that lacks proper 'micro-foundations,' particularly in the restricted sense of a 'reduction to current Walrasian microeconomics.' As a matter of fact, though it may be found impossible to provide proper micro-foundations to a given macroeconomic statement, this might become possible in the future. Such developments have occurred many times in the past and it could happen again, especially if microeconomics extends its range well beyond its Walrasian bounds. To reject this view would be as irrational as to reject as non-scientific any biological statement not yet reducible to chemical statements. Unfortunately, as has been wisely remarked, the only known way to reduce biology to chemistry is murder." -- Alessandro Vercelli, Methodological Foundations of Macroeconomics: Keynes & Lucas, Cambridge University Press, 1991.