The great attraction of the Keynesian system is its simplicity, which is, at the same time, its danger and its limitation. I propose to indicate how we may relax its cruder aggregative aspects without too hopelessly complicating matters. To accomplish this step we naturally turn to the Leontief matrix as an adequately simple representation of general equilibrium. Yet it is generically different from the Keynesian system by being homogeneous, i.e. the proportions are unique, but the scale of the whole system may be any multiple of the correct proportions. Only a small change is required to transform the one into the other, but it is just this small change which is necessary to study the short run generation and propagation of income.
The question is: does the multiplier process as worked out by Kahn and Clark, but expanded to include inter-firm and government payments, give rise to cycles? The answer is certainly yes, but, as Mr Chipman shows, there is a special case in which it does not, and it is exactly this special, and unrealistic case which multiplier theorists usually talk about. The correct statement is that there is latent in the transactions structure of society the possibility of oscillation, but that this may or may not be brought out according to the circumstances.
Keynes gave a clear and convincing analysis of how to ameliorate the endemic tendency of capitalist economics to engender periods of high unemployment. Partly as a result of his influence, the post-war history has been one of unparalleled and nearly continuous growth in output and employment. However for the past four or five years now this trend has been broken: everywhere there is unemployment and varying degrees of depressed trade. In spite of a (waning) commitment to Keynesianism, governments stand paralysed in this situation. It is generally agreed that this inaction has two principal sources: there is the effort to decelerate inflation (discussed in part II) and the fact that reflation of an economy leads directly to a worsening of the balance of payments. The enormous growth of international trade has meant that the scope for independent action of a nation has been seriously eroded. Not unconnected with this growth in trade has come a truly impressive accumulation of data, so that it becomes ever more feasible to attempt a quantitative analysis of the trade network of the many countries or regions of the world.
In recent years there have been at least two important contributions to our tools for the analysis of inflation. It is the aim of this paper to integrate these approaches with a simple dynamical version of linear general equilibrium theory. The resulting system is capable of empirical specification along the lines of Professor Leontief’s work,1 and contributes essential features lacking both in traditional quantity theory and in modern macro-dynamic models.
It is a melancholy reflection that, after about a century and a half of intensive effort by a distinguished line of investigators, it is still doubtful whether we can state any quantitative, empirical law in economics. A large part of the reason for this must lie in the nature of the problem, i.e. a very complex structure plus a state of continued violent change. These conditions make it an all-or-none affair — we solve the economic problem or none at all — and thus substantially vitiate the brilliant or ingenious assaults on its various parts. If the economy were stationary, or nearly so, we might fruitfully isolate particular sectors or types of problems. Or if the parts were not so intimately related, we might make simple dynamical analyses of their behaviour. It is the fact that both conditions are present that makes each analysis so intractable and yet so unavoidable.
The problem I propose to discuss in this paper is how to determine an economic plan for a growing economy in the medium run. The plan is required to be no more than a rough approximation and to give some detail but by no means complete detail. Within this range of precision, it is required to be consistent and optimal. Most economic discussions of planning are beside the point, since they are concerned only with small changes and with existing prices, equipment and technics. What is required is an analysis, however crude of the slowly cumulating changes bought by the expansion of output, employment and capital equipment.
The aim of this paper is not to treat input—output as such but rather to use it to discuss some old and some more recent economic issues. All the awkward empirical problems we intentionally put to one side and assume an economy completely and correctly characterised by a simple input—output system, with no joint products, only circulating capital, and only one factor, homogeneous human labour. The system is irreducible, primitive, and, being empirical, non-degenerate, that is, for n homogeneous commodities, the coefficient matrix has n distinct eigenvalues.1 All production is of the same unit duration, taking place in one period and available as output at the beginning of the next. Thus, in the spirit of the classical economists, and of Marx, reality is represented in a highly oversimplified form with the intention of illuminating certain central features, but at a sacrifice of detail.
As a student I found Wicksell more inspiring, more illuminating, more sympathetic to my own way of seeing things than any other economist. This symposium provided a happy occasion for a pilgrimage back to the works of that great man, to see how much of my earlier enthusiasm had survived forty years of being a practising teacher of economics. This effort has a certain poignancy for me because my other hero was Schumpeter, who, equally firmly, admired Walras above all others. Though we often argued the various aspects, neither of us was ostensibly swayed. I maintained Wicksell handled real issues with great insight and that Walras was empty of results; he argued that Walras showed the more profound method and the true path of progress. But it was an unequal contest, my arrogant ignorance against the skill and wisdom of that formidable intellect. Looking back I see that I was in fact rather subverted, in the sense that I have since spent more time on Walrasian that on Wicksellian issues. Therefore I would like to try a new look at Wicksell, from a perspective tinged with a minimal Walrasian colouring. Capital is the central and difficult concept in Wicksell’s work, and it is currently being once more seriously disputed. It is, of course, too large a subject for a brief essay, and I shall therefore proceed by rather compressed, unsubstantiated statements, with the request that it be judged as a whole, not by its parts.
