The Teaching of Development Economics by Kurt Martin; John Knapp Reviewed by Stephen Hymer The Journal of Finance, Vol. 23, No. 4. (Sep., 1968), pp. 719-721.
When a student comes to us from an underdeveloped country to learn economics "we cannot tell him the answer to what he wants to know." This remark by Joan Robinson pinpoints the problem of the Conference held at Manchester in April, 1964, to discuss the teaching of development economics. It must have been an exciting conference, for it raises fundamental and provocative questions, and the thirty-three participants, to say the least, were well-informed and articulate. Yet, when all is said and done, the conference made little headway: There was a lively interchange of ideas, but little agreement on fundamentals. I suspect that few of the people who attended the conference changed either their research plans or their lectures as a result of its deliberations.
Of the eight articles contributed to the volume, those by Seers and Myint are the most important. Seers argued that economic science, as we now know it, was developed in the special case of the developed countries: and an economist, trained in our discipline, moving to an underdeveloped country has at least as much to learn and to unlearn as to teach. The disproportion between the questions raised by the problem of development and the answers we are able to give is so great that Seers feels a major revolution in economic thought is needed: He argues that it will come all the more quickly, the sooner we recognize the inadequacies of our existing body of knowledge and face up to the challenge of searching for new tools and theory. The attempt may even end by changing our attitude to industrial economies and, therefore, to the whole body of economic theory.
Myint, with great care and thoughtfulness, examines the uses of economic theory in underdeveloped countries and makes the important distinction between realism and relevance. He concludes that our current theory lacks realism when applied to underdeveloped countries, but is nonetheless highly relevant. In econometric language, this amounts to saying that the parameters of the model can be expected to vary from country to country but that the basic structure of the model is useful and appropriate for research and policy. Myint argues that the focus of neoclassical theory on scarcity and optimum allocation is, if anything, more important in poor countries than rich countries; and that the study of "how the market forces actually work or fail to work in the different types of underdeveloped country" ranks among the highest priorities for research and teaching. Moreover, since no satisfactory alternative to present theory has been proposed, ('what is really at stake is how far we should discard the existing static optimum theory before we have time or are clever enough to build up a satisfactory 'dynamic' approach."
The remainder of the book provides variations on these themes as the participants illuminate weaknesses and strengths of economic analysis, and discuss approaches to teaching development economics. The papers by Hagen, Streeten, and Balogh on the state of knowledge, and those by Ady, Zimmerman, and Martin on teaching economic development are often incisive, nearly always interesting, but on the whole less well thought-out, less even, and less to the point than the two lead articles. The oral discussion (there are almost one hundred pages of it) provides a forum for some leading economists (mostly English) to relate their experience in teaching and studying development economics. The emphasis varies from person to person, but nearly everyone notes that there were important problems in which economic theory "properly" applied is very useful, and many other important problems on which traditional economic analysis provides no help at all.
In general, the book supports Seers' argument that we do not have satisfactory answers to the pressing problem of economic development but neither rejects neoclassical analysis nor suggests an acceptable alternative. The conference, however, did not push the argument as far as it could have and, in a sense, skirted the main issue.
It is useful to evaluate the problem raised by Seers in particular, and the conference in general, in terms of Kuhn's theory of the structure of scientific revolution. {1. Kuhn, Thomas S., The Structure of Scientific Revolutions (Chicago: University of Chicago Press, 1962).}Although Kuhn is primarily concerned with physical sciences, his analysis is pertinent to economics as well. He notes that what we call a science is characterized or defined in terms of a paradigm: a basic theory which serves as a model for research. This model provides the questions to be asked by research and also indicates the procedures that should be used. Armed with a paradigm, a scientist has a viewpoint for observing the world and an instruction manual telling him what experiments he should carry out and the results he should expect. A paradigm is at once an aid and a prison, since it focuses on some facets of a problem and ignores others.
Paradigms do not last forever, but are periodically replaced by new paradigms which reinterpret the world in terms of a new perspective. The reasons for these scientific revolutions are not clear. They do not come about simply because the existing paradigm has certain weaknesses: Every paradigm has anomalies, but anomalies do not by themselves destroy paradigms, only a new paradigm can do that. Instead, an attempt is made to play down the anomalies, to explain them away by ad hoc constructions, or simply to ignore and forget them. The anomalies come to life only when and if a new paradigm focuses upon them.
Another feature of paradigms is that they always give the appearance of more strength and power than they actually possess. The reason is simple. Once adopted, they are used more extensively than rival theories, and elaborate experiments are conducted on their implications while the implications of other approaches are ignored. The quantity of information about the paradigm accumulates steadily while the body of knowledge relevant to other theories languishes.
The "neoclassical synthesis" may be termed the paradigm of modem economics. It defines our profession, tells us which questions and factors are relevant, and guides us in experimentation and policy. The question raised by the volume under review is whether a new paradigm is needed. One of the reasons the conference is unsatisfactory at a fundamental level, is that it tries to answer this compelling question in terms of the old paradigm. It points up anomalies in the neoclassical paradigm, but this is not fully convincing because all theories have things they cannot explain and we do not reject a theory because its power and scope is finite. The conference also points up some of the strengths of the paradigm "properly" stated. This too is somewhat beside the point, because a paradigm judged on its own terms is likely to have many successes.
The most important result of the conference is that it demonstrates forcefully that we do not have a paradigm for economic development. In short, we do not have a theory interesting enough to attract a large number of adherents, acceptable enough to form the basis for textbooks or graduate reading lists, or powerful enough to provide policy recommendations of the kind supplied by the "new economics" for monetary and fiscal policy.
A second point is that nearly everyone agrees that many of the problems associated with economic development are noneconomic. If these problems were dealt with adequately by political scientists, sociologists, anthropologists and psychologists, we would be quite right to remain in our shell and to deal with what falls within the purview of our expertise. In fact, however, other social sciences are in an equally unsatisfactory state with regard to economic development and their practitioners frequently argue that they cannot make headway because the problem is essentially an economic one. Each social science stakes out a small area and the combined claims fail to exhaust the product. We call a problem economic if it concerns a question raised by our paradigm and noneconomic if it does not. Since the important problems of economic development seem to lie outside everyone's province, there is a case for changing the questions.
The question now becomes what should we do next? Everyone would welcome a new theory that provides more answers than our present one and that is so exciting that it commands our enthusiastic attention. But participants in the conference argue that at present we must work within the one we have. The point is a powerful one, but it contains one danger. The chance of a new theory coming along is not independent of the number of people willing to strike out on new paths. The more the existing paradigm is accepted, the more likely it is to remain unchanged.
What does the book tell us about the extent to which we should use existing tools and the effort we should devote to searching for new tools? In other words, what relevant information did the conference provide on the opportunity cost of using existing theory? The book raises many points but, in the end, the reader feels as much at a loss as ever in deciding the best way to allocate his effort. It is on this point that we must conclude that the book as a whole is not a must. Those interested in the field would do well to look at it, but they can probably skim much of it. It is too scattered and diffuse, and many of its points are developed better elsewhere, especially by the participants themselves. Although I enjoyed reading the book, it does not seem to me to be the best place to look for new ideas. Moreover, since 1964 much has been done in this rapidly growing field, and the book is now out of date.
STEPHEN HYMER
Yale University