At the end of this brief survey of the role of measurement in economics I shall attempt some general remarks on the lessons that seem to me to emerge and on a few other matters which arise out of the recent growth of quantitative economics and its use in economic policy. I have tried to illustrate the varied needs for measurement in describing economic situations and processes, in testing theories, in estimating the influence of one factor on another and in prediction. The importance of measurement for all these purposes is now generally recognized and an unprecedented amount of quantitative work is now being done by economists. The eighteenth field of research listed by Newmarch nearly eighty years ago has truly come into its own as a subject of intellectual interest and practical importance.

The past has seen many a battle of giants over the respective merits of the theoretical and empirical approaches to economic problems. There are signs that this false dichotomy is becoming recognized more and more widely for what it is. Only recently ‘measurement without theory’ has been condemned in high places and if theory without measurement has not come in explicitly for similar treatment the shift of interest as evidenced in the literature leaves the unexpressed views little in doubt. These remarks must not of course be taken as a plea against specialization; obviously a man who is an acknowledged master at conducting economic and social surveys may be no better than a raw hand at constructing theories of international trade or imperfect competition. They are intended only to combat the notion that either theory by itself or empiricism by itself has exclusive charge of the key to useful knowledge. No doubt in some branches of economics what is most urgently needed is a better theoretical formulation, in others a better description of what actually happens. In much of economics, however, there exist tolerable theories and a great many facts which have a bearing on these theories. The great need is to get the specialists in different aspects of a given field to work together, show a reasonable understanding of and respect for one another's disciplines and advance to problems which none could tackle on his own.

Just as unsympathetic specialization is a bar to advance so also is the undue exclusiveness sometimes to be observed in economists.