The Foundations of Modern Austrian Economics

Part 3, Essay 3

The Theory of Capital

by Israel M. Kirzner

It is not my purpose here to offer a concise Austrian theory of capital. Rather I shall present an Austrian perspective on several concepts fundamental to modern capital theory. I shall show how this perspective enables us to fit certain characteristically Austrian ideas about capital into the more comprehensive Austrian view of the way in which the market system operates. To clarify what constitutes the uniquely Austrian way of thinking about capital, I shall begin with a critical interpretation of a 1974 paper by Sir John Hicks. My use of Hicks’s work as a springboard also helps to explain why Hicks considered his book on capital theory to be “neo-Austrian” in character.
*14

HICKS, MATERIALISTS, AND FUNDISTS

Hicks grouped the numerous views on capital that have been expounded throughout the history of economic thought under two broad headings.
*15 By means of this classification Hicks called attention to the two basic ways economists conceptualize the notion of a stock of capital goods existing in the economy at a given moment in time. One is to see the stock as a collection of physical goods; by this approach aggregation into a stock provides us with a measure of the “volume of capital.” An important implication of this view is that “as between two economies which have capital stocks that are physically identical [the] Volume of Capital must be the same.”
*16 Those who share this view of the aggregate stock of capital, Hicks called “materialists.” The alternative view conceives of the aggregate stock of capital goods, not
as a volume of physical capital, but as a “sum of values which may conveniently be described as a Fund”—something evidently quite different from the physical goods themselves—with the values derived from the expected future output flows. Those holding this view, Hicks called “fundists.”
*17

This fundist-materialist dichotomy is an elaboration (and partial revision) of a comment that Hicks made in the 1963 edition of his
Theory of Wages.*18 In that discussion Hicks identified “the two basic ways in which economists have regarded the capital stock of an economy” as involving, first, a “physical concept,” which treats capital “as consisting of actual capital goods,” and, second, a “fund concept,” which reduces capital to the “consumption goods that are foregone to get it.”
*19 These are the same two concepts Hicks identified in his 1974 paper, except that there he preferred an interpretation of the fund view in terms of future outputs rather than in terms of opportunity cost.

FUNDISTS: A TERMINOLOGICAL PUZZLE

Now Hicks’s use of the label
fund to identify that notion of capital
not viewed as physical goods is puzzling and could cause much confusion about the meaning of his discussion. Hicks declared that he “of course” borrowed the term
fund from the history of economic thought; it is here that the confusion begins. The notion of capital as a fund is well known in the history of capital theory. It was carefully developed by John Bates Clark
*20 and repeatedly expounded by Frank H. Knight.
*21 Those who objected most vigorously to this view of capital were Eugen von Böhm-Bawerk in the first decade of the century and Friedrich A. Hayek in the fourth decade. Böhm-Bawerk declared that Clark’s concept of capital was mystical, and he insisted that to measure a stock of capital in value units in no way implies that what is being measured is an abstract quantity apart from the physical goods themselves.
*22 Hayek, in his debate with Knight, argued against the notion of capital as a fund of value, that is, a quantity apart from the particular goods making up the capital stock.
*23 Thus
Hicks’s list of fundists must include Clark and Knight, and Böhm-Bawerk and Hayek must be on his list of materialists (certainly not on the list of fundists).

However, this was not the way Hicks classified these economists. J. B. Clark was “clearly” a materialist, and Hayek “of course, was a fundist.”
*24 Also, Böhm-Bawerk. “kept the fundist flag flying!”
*25 A basic familiarity with the history of capital theory leaves one puzzled, if not startled, by Hicks’s choice of terminology. Let us look more closely at how he employs his definitions.

HICKSIAN FUNDISM

To Hicks, the term
fund apparently denotes a concept entirely different from that Clark and Knight had in mind when they used it. Knight’s capital-as-a-fund notion refers to a special way of viewing capital goods—as the temporary embodiment of a permanent store, or “fund,” of value. Hicks’s declaration that the Austrians are fundists has to do with their treatment of capital goods as essentially forward-looking components of multiperiod plans.
*26 The Clark-Knight view of capital as something other than the capital goods themselves is quite different from Hicks’s fundists’ view. Hicks himself—now a “neo-Austrian”—is on the side of his brand of fundists—the forward-looking kind. Without denying the propriety of a materialist, or backward-looking, perspective on capital, Hicks endorsed a “sophisticated” fundism in which the forward-looking character of capital goods is repeatedly emphasized. This made it easier for Hicks to recognize the importance of expectations in capital theory—a characteristically Austrian point of view.

Hicks apparently was inspired in his novel use of the term
fund by Böhm-Bawerk’s notion of a subsistence fund.
*27 Despite my criticism of Hicks’s usage of the term
fund, I must agree that Böhm-Bawerk did set a precedent for using it in this way. Moreover, by insisting that the notion of a subsistence fund is
characteristic of Böhm-Bawerk’s approach, Hicks preserved what we must consider to be the essentially pure Austrian element in Böhm-Bawerk’s theory. Some effort has been made in recent work to ignore the forward-looking, multiperiod-planning aspect of Böhm-Bawerk’s theory. The productivity side of his system has been emphasized, and the time-preference aspect has been either suppressed altogether or treated as an inessential encumbrance.
*28 However, as Hicks recognized, the notion of a subsistence fund is an essential element of Böhm-Bawerk’s thinking about capital. This notion embodies the insight that, in choosing between processes of production of different durations, men appraise the prospective sacrifices these processes call for in terms of abstaining from more immediate consumption. Crucial to such appraisals is the size of the available capital stock, because it influences the prospective disutility associated with each of the alternatively required periods of waiting. Not only is this notion of a subsistence fund central to Böhm-Bawerk’s theory of capital, but it also is—in spite of his disconcerting concessions to the productivity-interest theorists
*29—the essentially “Austrian” element in his thought. In the subsistence-fund concept is encapsulated Böhm-Bawerk’s concern for forward-looking, multiperiod human decision making; here the influence of subjective comparative evaluation of alternative future streams of income makes itself felt; and here there is room for a discussion of expectations and uncertainty. Hicks recognized all this when he identified the Austrians as fundists. Not only does the notion of the subsistence fund qualify Böhm-Bawerk as a Hicksian fundist, but it also, as Hicks implied, epitomizes what is “Austrian” in Böhm-Bawerk’s theory. I heartily agree with all this. At the same time, to emphasize the differences between the fund notion as used by Hicks and as usually associated with Clark and Knight, we may recall Hayek’s powerful criticisms of the subsistence-fund idea.
*30 Hayek’s extreme opposition to the Clark-Knight notion of a fund led him to point out some difficulties in Böhm-Bawerk’s original presentation of the subsistence-fund idea. While we can understand why Hicks labeled Böhm-Bawerk a fundist, we must still decide whether
Hicks was correct in naming Hayek and modern-day Austrians fundists as well.

AUSTRIANS, MATERIALISTS, AND FUNDISTS

Austrians reject the fundist-materialist dichotomy because of their special understanding of the role individual plans play in the market process. A capital good is
not merely a produced factor of production. Rather it is a good produced as part of a multiperiod plan in which it has been assigned a specific function in a projected process of production. A capital good is thus a physical good with an assigned productive purpose. To treat the stock of capital goods as a Hicksian materialist would is out of the question, for as Hicks recognized, this approach ignores the future streams of output that these capital goods are designed to help produce. But to treat the stock of capital goods as a Hicksian fundist might is also unacceptable, for it submerges the individualities of the various capital goods into a stock and replaces them with the sum of values supposed to represent the aggregate expected future value of the output imputed to these goods. To treat the stock of capital goods this way ignores the problem of the degree of consistency that prevails among the purposes assigned to each of the goods composing the capital stock. It also ignores questions of complementarity among goods, as well as the possibility that the productive purpose assigned by one producer to a particular capital good is unrealizable in the light of the plans of other producers or potential users of other capital
goods. But all this is precisely what, to an Austrian economist, cannot be ignored.

In somewhat different terms, the Hicksian materialist-fundist classification is objectionable not so much for these alternative formulations as for the incompatibility of the task they attempt to carry out with the Austrian approach. That task is to arrive at a single value aggregate to represent the
size of the stock of capital goods in an economy. It is this very attempt to collapse the multidimensional collection of capital goods into a single number that Austrian economists find unacceptable. If one wishes to talk about the total stock of all capital goods in an economy at a particular moment in time, one must not overlook the roles assigned to the various goods by individuals; in other words, one must treat the stock as consisting at all times of essentially heterogeneous items that defy aggregation. They defy aggregation not only because of physical heterogeneity but also, more important, because of the diversity of the purposes to which these goods have been assigned. Mises emphatically dismissed the very notion of an aggregate of physical capital goods as empty and useless. The “totality of the produced factors of production,” Mises wrote, “is merely an enumeration of physical quantities of thousands and thousands of various goods. Such an inventory is of no use to acting. It is a description of a part of the universe in terms of technology and topography and has no reference whatever to the problems raised by the endeavors to improve human well-being.”
*31

ON MEASURING CAPITAL: THE INDIVIDUAL

Other economists as well as the Austrians see the serious theoretical difficulties that face all attempts to measure real capital. It seems useful to review briefly where these difficulties lie, both in the attempt to measure a single individual’s stock of capital goods and the attempt to measure society’s stock of capital goods.
*32

Consider a list of all the physical items that make up a single
individual’s stock of capital goods. Their physical heterogeneity prohibits “adding them up”—there is no natural unit of measurement. Instead one may try to devise an index number that will permit the heterogeneous items to be treated as one dimensional. But if the measure thus obtained is to be interpreted as representing the quantity of physical goods in the stock, the attempt hardly seems worthwhile. Not only is the advantage in replacing a list by a number unclear, but such a replacement obviously hides an important aspect of economic reality. A man’s future plans depend not only on the aggregate size of his capital stock but also very crucially upon the particular properties of the various goods making up the stock. Goods that can be used in a complementary relationship permit certain plans that a purely physical measure necessarily suppresses.

A far more popular alternative is to measure the size of stocks of capital goods in terms of expected contribution to future streams of output. Attempts to arrive at so-called forward-looking measures of capital are, of course, the essence of Hicksian fundism. It is misleading to talk of a particular resource as being unambiguously associated with a definite stream of forthcoming output, in the sense that such an output stream flows
automatically from the resource itself. Decisions must be made as to how a resource is to be deployed before one can talk of its future contribution to output. Because there are alternative uses for a resource and alternative clusters of complementary inputs with which a resource may be used, it is confusing to see a resource as representing a definite future output flow before the necessary decisions on its behalf have been made.

MEASURING CAPITAL: THE ECONOMY

The difficulties involved in measuring the size of an individual’s stock of capital goods become exaggerated when it is the size of an entire nations’s stock of capital goods that one is measuring. The theoretical difficulties involved in aggregating physical items present formidable obstacles to thoroughgoing, “aggregative,” Hicksian materialism. As Hicks correctly pointed out, Austrian economists have little sympathy with such an approach. But measuring the size of the capital stock in economic rather than in physical terms is hardly more promising. We may dismiss backward-looking measures as having failed, even at the individual level. Forward-looking measures of an individual’s capital stock are impossible to achieve in an objective way but are being made all the time in the private dealings of individual agents. In considering forward-looking measures of a nation’s stock of capital goods—that is, Hicks’s fundism in the 1973 version—the possibility of subjective evaluation loses virtually all meaning (except in a sense to be discussed in the subsequent section); the
problems that prevent objective measurement at the level of the individual are intensely reinforced by additional problems. One difficulty is relating a given item (in the inventory of the social stock of capital goods) to its prospective output before the preliminary decisions, necessary for the identification of that output, have been made; another is that the prospective output associated with a capital good owned by Jones may be inconsistent with a capital good owned by Smith. A forward-looking measure of Jones’s capital stock and also of Smith’s must presume plans on the part of Jones and of Smith, but Jones’s plan and Smith’s plan may be, in whole or in part, mutually exclusive. Perhaps Jones expects rain and builds a factory to produce umbrellas, whereas Smith expects fine weather and builds a factory to produce tennis racquets. That one of the two has overestimated the possible output of his factory (and thus will, in the future, be seen to have incorrectly measured the size of his capital stock) is not significant. What is significant is that it is already
now meaningless to add a valuation of Jones’s factory to a valuation of Smith’s factory when each valuation depends on the expectation of one that the expectation of the other will prove erroneous. We are, therefore, forced as Austrian economists to decline Hicks’s invitation to join the fundist club. To view the aggregate capital stock as a fund requires us in the end to give up our concern with individual plans upon which forward-looking measures ultimately depend for their very meaning! By refusing to surrender our Austrian interest in individual plans, however, we face another difficulty.

ON THE NOTION OF THE QUANTITY OF

CAPITAL PER HEAD

We have seen that attempts to measure the quantity of capital goods in an economy must fail on theoretical grounds. Yet it is virtually impossible to avoid making statements in which such measurement is implied. In the writing of Mises, for example, we find frequent reference to the consequences of the fact that one
country possesses a greater quantity of capital (or a greater quantity of capital per worker) than a second country.
*33 Such statements imply the possibility that aggregate capital measurement has a rough meaningfulness. Yet how do we reconcile our theoretical rejection of aggregative capital measurement with a willingness to make statements of this kind?

The answer to this difficulty lies in the possibility, mentioned above, of subjective evaluation by an
individual of the aggregate worth of his own stock of capital goods. Individual forward-looking measurement is both possible and feasible, because the problem of possibly inconsistent plans does not arise. An individual evaluates each component of his capital stock in terms of the plans he has in mind;
he may have to take care to avoid possible inconsistencies, but in appraising his measurement of his capital we may assume that he
has successfully integrated his own plans. What cannot, except in the state of general equilibrium, be assumed for an economy as a whole is assumed as a matter of course for the individual.

Underlying statements that compare the quantity of capital in one country with that in another is a convenient and relatively harmless fiction. One imagines that one has complete control over all the items in a nation’s capital stock—that one is, in effect, the economic czar of a socialist economy. One is then in exactly the same position in relation to the nation’s stock of capital as an individual is in relation to his own stock. By the use of this fiction, problems of inconsistent plans have been simply imagined away.

In sidestepping in this way the theoretical difficulties that frustrate attempts to arrive at aggregate measures of capital, we have, it may be argued, endorsed Hicksian fundism after all. But this is by no means the case. In the context of a market economy the fiction that all inconsistencies among plans may be ignored is, for most purposes, highly hazardous and misleading. Although there are instances in which needed reference to the aggregate quantity of capital can be supported in the way described, to use this fiction to construct a general treatment of aggregated capital is entirely unacceptable to Austrian economists. It is the market
process that has the property of discovering inconsistencies among plans and of offering incentives for their elimination. To introduce capital into the analysis of the market process in a way that assumes that plan inconsistencies do not exist is to espouse decidedly non-Austrian assumptions and to become enmeshed in those insoluble contradictions characterizing orthodox microeconomic theory, the escape from which provides the strongest case for a return (or an advance) to the Austrian position.