The Foundations of Modern Austrian Economics
In reconstructing Austrian economics, I have recommended the concept of market process as the foundation of all economic life (see my paper "On the Central Concept," included in this volume). In other words, we must start at the microlevel. How, then, shall we deal with such aggregates as income, consumption, and wages, which, according to the same terminology, must be regarded as macroeconomic magnitudes?
In my paper "On Austrian Capital Theory" (also included in this volume) I said that we may regard the relationship between the capital structure of society and its capital combinations as the prototype of the relationship between a macroeconomic magnitude and its microelements. In other words, the former must never be brought into an argument without giving a careful account of the latter; for changes in the constellation of the microelements will affect the macromagnitude itself.
However, in modern macroeconomic literature, whether neoclassical or neo-Ricardian, there is little awareness of this fundamental postulate. From the moment of inception, the macroeconomic aggregates in these writings seem to lead a life of their own, to be endowed with qualities sufficient to allow their adjustment to change in their environment, but change within them is ignored.
Among the macroeconomic aggregates, we have to distinguish between stocks and flows. While it is generally agreed that the former cannot in a world of uncertainty be measured, the reasons given are not satisfactory. Why, then, cannot capital be measured?
In stationary equilibrium capital can be measured; for here, but only here, the discounted present value of the highest income stream obtainable from a given capital resource must be exactly equal to its cost of reproduction if the same interest rate is used for both calculations. Otherwise there would be investment or disinvestment, neither of which is compatible with a stationary state.
Outside the equilibrium of the stationary state, replacement cost and present value of a discounted future income stream will diverge. If we regard the latter as the economically significant value—and few economists doubt that (bygones are bygones)—we still have no yardstick by which to measure the assets of different firms. The neo-Ricardians stress the role of the interest rate as a discount factor; every time it changes, so does capital value. True as this may be, the real reason for our inability to measure capital lies in the subjective nature of expectations concerning future income streams. If we tried to measure capital by asking each owner for his valuation of his capital stock (for example, fire insurance value), we would get a set of consistent answers, but every change in ownership would invalidate at least one answer. Such an attempt to extract measurable objective value from subjective valuation must fail. This does not mean, however, that in a more limited context and for more limited purposes similar attempts may not be more sucessful.
The second part of the argument is that, while stocks cannot be measured, flows can. All of modern macroeconomics—the theories about investment, income, exports, and wages, as well as the social accounting systems that are widely regarded as one of its triumphs—rests on the assumption that output and income streams can be measured. Can they? As we shall see, the main support for this assumption is superficial as well as highly misleading.
Here we have one of the most intricate problems of measurement we encounter in a multicommodity world. In a classical corn economy all stocks and flows have their physical measures, but in a multicommodity world gallons of brandy and of whiskey, surgical instruments and musical instruments cannot be added up. Like David Ricardo and Eugen von Böhm-Bawerk, we need a price system invariant to the very effects of the variations of the variables we wish to measure. Can such a system exist?
The market prices on which national income measurement is ostensibly based are presumably equilibrium prices. But what kind of equilibrium prices are they? Is consistent aggregation, which macroeconomic thought requires to make sense, possible on the basis of prices that may not be consistent? Walrasian-Paretian long-run equilibrium does provide a consistent price system, but the market prices at which output flows are evaluated in the national income statistics are no such long-run equilibrium prices. They are market-day-equilibrium prices at best. There is no reason why they should be consistent with each other. The whole argument for using average market prices over a year as a basis for the valuation of output flows appears to rest on a confusion between long-run, short-run, and market-day equilibrium. Only prices belonging to the first category are elements of a coherent price system and might be used for the purpose of consistent aggregation. All the others are not. An average of market-day-equilibrium prices over a long period is not the same thing as a Walrasian long-run equilibrium price. Yet the computation of the macroeconomic aggregates that are released daily by the media and are so prominent in the products of the textbook industry rests on such elementary confusion.
Even if such a coherent long-run equilibrium price system did exist and could be known, it would not last. Its data would not remain the same for long. The steady flow of knowledge will tomorrow produce a pattern of knowledge different from today's, and apparent changes in demand and supply will entail a new set of equilibrium prices. Not without reason are we compelled to look for a paradigm to replace the general-equilibrium model.
In a market economy, on the other hand, we have in the stock exchange a center for the consistent daily evaluation of all the more important capital combinations. This, to be sure, is not objective measurement. The measurement of capital is forever beyond our reach. But it is something more than mere subjective evaluation. Stock exchange prices of capital assets reflect a balance of expectations. There are two classes of traders in the market, and in an asset market the fundamentum divisionis is the optimism or pessimism of expectations. Hence, market prices of securities (and the capital assets they represent) have social significance, we might say a "social objectivity," which transcends the mere subjective expectations of buyers and sellers on which they are based. The objects of valuation are not individual capital goods but fractions of capital combinations that are the substrate of multiperiod plans. It is expectations about the success of multiperiod plans that bulls and bears are continuously expressing.
Stock exchange equilibrium is market-day equilibrium. Tomorrow's set of equilibrium prices will be different from today's. But as we are dealing with an exchange economy, not a production economy, the fact of change does not impair the consistency and significance of market valuation. Free access to all parts of the market, together with the speed at which brokers carry out their clients' instructions, makes it possible to be at once a buyer of one kind of security and a seller of another. Arbitrage does the rest to produce a valuation of all company-owned assets that is consistent in that it reflects a balance of expectations between bulls and bears. The fact that tomorrow's balance will be different from today's faithfully reflects the flow of knowledge.
The difference between the commodity market and the securities market is instructive. Today's potato price may not be consistent with today's prices for other vegetables. If so, equilibrating forces that require time will come into operation. But today's market equilibrium price may also be inconsistent with the long-run supply and demand for potatoes. The equilibrating forces released by the second disparity may impede or reinforce those released by the first. Certainly they require a different time dimension to be fully deployed. And the longer the time required by any force, the greater the probability that it will be affected by unexpected change. To opt for the market process against general equilibrium means to accept the implication that a fully coherent price system providing a basis for consistent aggregation can never exist.
We find here another reason why steady growth—uniform motion of that supermacroaggregate, the economic system—has to be regarded as absurd. Equilibrating forces in different markets, even if none is affected by unexpected change, require different time periods to do their work. This is obvious if we contrast agricultural produce markets with those for industrial goods. Steady growth, however, requires all equilibrating forces to operate within the same time period.
But the main argument against steady growth is a necessary consequence of the divergence of expectations. Equilibrium in a production economy requires an equilibrium composition of the capital stock.*42 With at least some capital goods durable and specific, can we conceive of such a state? A growing economy is a changing economy. It exists in an uncertain world in which men have to formulate expectations on which to base their plans. Different men will characteristically have different expectations about the same future event, and they cannot all be right. Some expectations will be disappointed, and the plans based upon them will have to be revised. The capital invested will turn out to have been malinvested. But the existence of malinvested capital is incompatible with the equilibrium composition of the capital stock. Hence steady growth is impossible. "Macroequilibrium in motion" is not an acceptable paradigm and has to be replaced by the market process.
As has been noted, the stock exchange, a fundamental institution of the market economy, imparts an element of social objectivity to individual stock valuations. This is by no means its only, or even its only significant, function. It facilitates the take-over bid by means of which capital resources get into the hands of those who can promise their owners a higher return. Without a stock exchange such bids would of course still be possible. shareholders could be notified that, if their capital resources were used in a different way (by producing a different output stream), better results could be obtained. But in the absence of market prices shareholders would be without a yardstick to measure the advantage offered them. For the optimal use of existing resources in a socialist economy, an elaborate bureaucratic organization would be needed to shift resources from points of lower to points of higher usefulness.
Perhaps the most important economic function of the stock exchange is the redistribution of wealth by means of the capital gains and losses it engenders in accordance with the market view about the probable success or failure of present multiperiod plans. In certain neoclassical writings the present distribution of resources is a datum of the set of equilibrium prices and quantities at each point of time. Over time, however, the mode of distribution of wealth changes as reflected in capital gains and losses. In fact, the present mode of distribution of wealth is nothing less than the cumulative result of the capital gains and losses of the past. Devotees of a redistribution of wealth in the name of social justice should be aware that, even if the state by the use of coercion were able to produce a supposedly socially desirable mode of distribution today, the market, if permitted to exist, via capital gains and losses would produce a different mode tomorrow.
I have tried to show that—in contrast to an opinion apparently widely held—flows in a multicommodity world are all too often inconsistent aggregates, whereas stocks, while unmeasurable in an uncertain world, may in favorable circumstances be subjected to consistent evaluation. The former is particularly true of the Keynesian I=S, current investment, the "net addition to the capital stock." If the stock is unmeasurable, how can we tell what is an addition to it? Gross investment is in principle measurable and would be in practice if we had a consistent price system for capital goods. But to divide this flow into net investment and replacement, we need an objective criterion.
Individual decisions by capital owners on such division may be consistent in the sense that the decisions of individual owners do not conflict, but not in the sense that the word measurement implies, that is, that different individuals measuring the same object get identical results. Each owner's judgment of his investment expenditure for maintenance and replacement of his existing wealth on the one hand and for a "net addition" on the other rests on a subjective expectation about the future, as nobody explained more forcefully than Keynes in his unjustly neglected "Appendix on User Cost."*43 It follows that a macroeconomic magnitude investment as an objectively measurable, that is, consistently ascertainable, entity does not exist. What does exist is an aggregate of subjective valuations dependent on owners' expectations and the distribution of ownership. An important part of what appears as investment in the official reports of government and business rests on subjective estimates of those who compile the statistics or of those who release the returns.
This imperfection in macroeconomics, like many others in an imperfect world, could be regarded as inherent in any application of abstract theory to a concrete situation. This defense, however, is not valid. We have to distinguish between defects due to the unsatisfactory nature of our material, that is, the state of our statistical sources, and defects due to muddled thinking. A cure for the former is no cure for the latter. Piero Sraffa on a famous occasion made this point with vigor and clarity. His position has been summarized as follows:
Mr. Sraffa thought one should emphasize the distinction between two types of measurement. First, there was the one in which the statisticians were mainly interested. Second, there was measurement in theory. The statisticians' measures were only approximate and provided a suitable field for work in solving index number problems. The theoretical measures required absolute precision. Any imperfections in these theoretical measures were not merely upsetting, but knocked down the whole theoretical basis.... Mr. Sraffa took the view that if one could not get the measures required by the theorists' definitions, this was a criticism of theory, which the theorists could not escape by saying that they hoped their theories would not often fail. If a theory failed to explain a situation, it was unsatisfactory.*44
Notes for this chapter
"An equilibrium path, let us remember, is a path that will (and can) be followed if expectations are appropriate to it, and if the initial capital stock is appropriate to it; both conditions are necessary" (John Hicks, Capital and Growth [Oxford: Oxford University Press, 1965], p. 116).
John Maynard Keynes, The General Theory of Employment, Interest, and Money (New York: Harcourt, Brace & World, 1936), pp. 66-73.
Douglas C. Hague, "Summary Record of the Debate," in The Theory of Capital, ed. F. A. Lutz and D. C. Hague (London: Macmillan & Co., 1961), pp. 305-6.
Part 3, Essay 6, The Austrian Theory of Money
End of Notes
Return to top