Good diagnosis requires good observations and the existence of good explanatory theories. Improving on observations and theories are the means by which one can really reach improvement in diagnosis. But for this progress to be most effective with respect to the needs of policy analysis, priorities must be identified correctly. They of course depend on the ultimate aims of the analysis, which ought never to be forgotten. They also depend on the type of situation one is facing; indeed, the economy does not function like a linear system in which the main endogenous variables exhibit constant degrees of sensitivity with respect to policy instruments – such a conception often provides a sufficient approximation for small variations with respect to the current situation, but the linear approximation then depends on what this situation really is. An important aspect of the diagnosis consists in so identifying the determinants to which the economy will particularly react at the time of study.

In the first part of this lecture I shall briefly outline the purposes assigned to macroeconomic diagnosis in our economies. The second part will be devoted to observations, not so much to the collection of data as to their first analysis. The third part will concern the theoretical framework or frameworks to be used in our explanations.

The purpose of diagnosis in market economies

The ideal vision of free market economies operating with perfect efficiency provides, in our case as in many others, the best starting point because it is the only well-articulated and well-understood reference at our disposal.