Douglass C. North
Beyond the New Economic History
The Journal of Economic History
Volume 34, Issue 1, The Tasks of Economic History
Mar. 1974, 1-7.
THE new economic history has been with us now for almost
a score of years. Its practitioners
have advanced from young revolutionaries to become a part of the middle-aged
establishment; and by all the criteria of publication and training of graduate
students, it has indeed transformed the discipline in the
What the new economic history contributed was the systematic use of theory and quantitative methods to history. The use of a scientific methodology has put a distinctive stamp on this approach, which clearly delineates it from the old economic history, but it is the theory that provides a particular cast to the contribution. It is the systematic use of standard neo-classical economic theory which both has provided the incisive new insights into man’s economic past and also serves to limit the range of enquiry.
I am indebted to Elisabeth Case and Robert Willis for helpful comments on an earlier version of this paper and to the John Simon Guggenheim Memorial Foundation which provided me with a fellowship and resultant time to reflect about the issues discussed herein.
I shall not dwell on the contributions; they have been thoroughly touted (and denounced). The limitations are:
(1) The research has been more destructive than constructive. We have destroyed a number of older explanations but we have not replaced them with an explanation of the way economic change has occurred in any systematic fashion. If we have found slavery profitable, railroads less than essential, and the net burden of the Navigation Acts “light,” we have not said what did make the system go - or what did change the distribution of income.
(2) The main emphasis of research has been on specific issues or institutions, but little light has been shed on the long-run transformation of economic systems - that is, long-run economic growth.
(3) There is no role for government in the analysis except as it is brought in in an ad hoc fashion.
(4) In fact, of the four sources of decision making in an economic system - the household, voluntary economic organizations, government, and the market - we have a sophisticated explanation for decision making in only the last of these (and then only a partial one), despite the obvious fact that a substantial if not overwhelming percentage of economic decisions has always been made outside the market place. Moreover, we have no explanation for why the mix among the four changes over time. How can one talk seriously about the economic past without an explanation for non-market decision making?
(5) Finally, I would add another limitation which stems from the first four and is of importance for the long-run future of our discipline: it is curiously unteachable at the undergraduate level. It leaves students frustrated because of its failure to come to grips with the above issues and to provide any integrated explanation of man’s economic past.
The limitations are those of the theory. Neo-classical economic theory has two major shortcomings for the economic historian. One, it was not designed to explain long-run economic change; and two, even within the context of the question it was designed to answer, it provides quite limited answers since it is immediately relevant to a world of perfect markets - that is, perfect in the sense of zero transaction costs: the costs of specifying and enforcing property rights. Yet we have come to realize that devising and enforcing a set of rules of the game is hardly ever costless and the nature of these costs is at the very roots of all economic system’s problems.
Accordingly, a theoretical analysis of the changing rules of the game is at the very core of the subject matter of economic history.
Let me emphasize that a study of the rights associated
with the use and transfer of resources is as relevant in socialist societies as
it is in capitalist ones. The rules
of the game determine efficiency and the distribution of income in any society:
In attempting to construct a broader analytical approach to history we have, it seems to me, two alternatives. We can throw out neo-classical theory and start all over again, or we can broaden the frame of reference to allow us to deal with the issues. In the latter case we accept the basic assumption of utility maximizing behavior (including the problem of specifying in operational terms a meaning for such behavior) and we see how far we can develop a theory of “the rules of the game.” The proof of the pudding is in the eating: if alternative frameworks provide better “fits” to the evidence, fine; but I am more than ever convinced that a theory of household economics and a theory of property rights, including a workable theory of the state - an essential prerequisite - are possible; and indeed, that we have made a promising start by expanding neo-classical theory. The approach I wish to suggest offers a common analytical framework to study the structure of economic systems. Standard micro-economic theory then becomes one part of a broader framework of analysis.
The controversy over the usefulness of neo-classical theory is an old one, and I certainly do not feel qualified to add anything to it on theoretical grounds; but as an economic historian I feel somewhat less diffident and would suggest to you the following:
(1) Neo-classical theory has been a powerful tool of analysis of the new economic history and has demonstrated repeatedly that it can shed light upon our economic past. In fact, I would put it stronger: A theory of choice - the self-conscious application of opportunity cost doctrine - is essential to the framing of meaningful questions in economic history.
(2) Transaction costs are the link between neo-classical theory and a broader theory of property rights. 1 The explicit historical study of transaction costs opens up new horizons for the economic historian. Much of the productivity change in past history has been a consequence of reduced transaction costs and their study suggests a quite radically different history than we read in the standard explanations.
(3) An equally promising extension of neo-classical theory is occurring in a more sophisticated approach to the household economy, with important implications for a theory of fertility .2 Demographic history has displayed much of the schizophrenia of the controversy between the old and the new economic history. It has been largely pursued outside the context of economic theory. Yet clearly, the essential requirement for the advancement of economic history is a wedding of economic and demographic theory. In effect, we need an economic theory of the family, and recent research offers the promise of providing such an analytical framework. Such an approach has two key assumptions: one, that some degree of control over fertility was possible, and two, that such considerations as the value of time and human capital investment - in effect, the opportunity costs of the parents - influenced fertility behavior. The first is not very controversial. Demographers have recognized that some degree of fertility control has existed since very early times. The second is open to all the attacks that historians have made against rational economic motivation as a behavioral assumption. The defense is the same. Let us see how well it tests as a working hypothesis.
(4) A major issue of economic history which has been completely neglected, at least in theoretical terms, is the logic of the mix among the four sources of decision making that occur in an economic system: households, voluntary organizations, government, and markets. We tend to treat explanation of this mix and changes in it over time as outside our explanatory system; but it seems to me that ongoing research building on a theory of household behave-
1. A convenient summary of the literature on property rights and transactions costs is contained in the December 1972 Journal of Economic Literature, “Property Rights and Economic Theory: A Survey of Recent Literature,” by Eirik C. Furubotn and Szetozar Pejovich.
2. “New Economic Approaches to Fertility,” Journal of Political Economy (March-April 1973), Part II. See particularly the essay by Theodore Schultz, “The Value of Children: an Economic Perspective.”
ior and transaction cost analysis offers the promise of providing a theoretical explanation for such issues as changing fertility behavior, the transformation of the economic role of the family, a manorial system, the rise of guilds, or the increase in the role of government in modern times. A theoretical explanation of the mix of economic organization opens the door to an explanation of much of the institutional structure of an economic system.
(5) It is surely a much simpler matter to explain why many economic decisions are internalized inside households, firms, guilds, or manors rather than made in markets than it is to explain why they are made by political units; but even here I believe we are making significant progress. The work of Baumol, Buchanan and Tullock, and Anthony Downs, as well as much ongoing research, provides us with a promising starting point.
If I am correct about the promise of this approach, then I suggest to you that the logical implications for future research are quite different from the directions we currently are pursuing. Specifically,
(1) Our emphasis on the last two hundred years, from the Industrial Revolution onward, is a misallocation of scholarly resources. We should spend much more time on the preceding 9800 years of man’s economic history than on the last 200. I am convinced that there were long periods in the past in which growth in economic well being occurred and that they have interesting implications for our understanding of economic history. In fact, the overriding issue of man’s economic history has been the relationship between population growth, diminishing returns to a relatively fixed factor, and man’s efforts to alter institutional arrangements to overcome this dilemma. Our emphasis on the present blinds us to the fact that few of man’s economic problems are new - that most have recurred endlessly in the past. Common property resource problems when man first developed settled agriculture in neolithic times; enclosed common pasture in medieval and early modern times - both are linked by the problems of changing relative scarcity to the modern dilemma of pollution and the quality of the environment. All equally entail modification of man’s institutional environment for solution.
(2) Any organized economic system involves not only the “team” production of goods and services, but equally the production of protection and justice. Both require the input of resources; and at least in principle we should be able to measure output and therefore productivity and changes in the productivity of each over time (that is,
changing output per unit of input). Moreover, both involve many similar ingredients in analysis and problems. In effect, I am saying that a theory of the firm that makes sense will also go a long way toward providing us with a theory of the state. In fact, historical study would suggest that economic organization is a continuum in which purely voluntary organizations or purely governmental ones are extremes and that such institutions as the medieval manor contain elements of both and require an analytical framework that encompasses a general theory of organizations.
(3) Just as technology has fundamentally influenced the size of the economic unit in the production of goods and services, military technology has influenced the size of the political unit and is worthy of equal study if we are going to deal with the efficient (that is, survival) size of political units.
(4) The study of the decline of political economic units or the failure of many to grow is more interesting than the study of successful ones. This is so because the logic of micro-economic theory and simple welfare economics suggests that growth should be inevitable. If any increase in productivity leads to a growth of income and the gainers compensate the losers, then economic growth is not an interesting issue (under some simple and not too controversial behavioral assumptions about present versus future goods). It is only when we introduce an economic theory of the family, transaction costs, and a theory of political decision making that we can explain decline or stagnation. 3
(5) The growth or decline of economic systems is clearly a function of increasing or decreasing productivity of the two sectors - goods and services and protection and justice (note that they are not synonymous with private and public) - taken together. Our examination of the goods and services sector only in explaining growth or decline has given us a misleading picture of the process of economic change. It is the interplay between the two sectors that is a key to an understanding of economic change. What leads to the development of “efficient” or “inefficient” property rights and how do these “rules of the game” influence the output of goods and services? Let me suggest three scenarios that I believe have been oft repeated in the history of the rise and decline of political-economic units:
3. This issue is discussed more fully in Douglass North and Robert Paul Thomas, The Rise of the Western World: A New Economic History (Cambridge: Cambridge University Press, 1973), Ch. 1.
(a) the rapid growth of an economy may be partially a consequence of substantial indivisibilities and resultant economies of scale in the production of protection and justice which only show up in our accounting as productivity change in the goods and services sector; (b) the relative retardation in one country’s growth (compared to others) may also be at least partially explained by that economy’s realization of all the scale consequences of productivity change in the protection and justice sector and therefore further productivity increase being limited to the goods and services sector alone; (c) the stagnation or decline of an economic system results from a rise in the costs of any given quantity of protection and justice leading to a search for new sources of fiscal revenues with adverse consequences for the efficiency of property rights in the goods and services sector so that declining productivity occurs in that sector as well.
It seems to me that these few modest suggestions could keep our profession fully and productively employed for a long time, and I commend them to your scholarly attention. I have no illusion that they lead to the promised land of ultimate truth and final explanation. I have too much respect for the complexity and contrariness of human behavior to believe that we can do much more than unravel a little more of an endless skein - but then, that’s enough to make it the most satisfying profession I know.
DOUGLASS C. NORTH,