3. Capital and Waiting

CHAPTER THREE

CAPITAL AND WAITING

We turn to take up the aspect of capital that has proved perhaps most fascinating to economists and yet seems to have provided them with their greatest difficulties and with fuel for their fiercest controversies,—the relation between capital and waiting. In general economists associate the “Austrian” approach to capital and interest theory with a major emphasis on the time lapse between inputs and outputs—the theory revolving around concepts such as the length of the period of production, or of investment, the amount of waiting thus involved, and the “productivity” of the time concepts so distinguished. On the other hand the Clark-Knight view of the capital and interest problem vigorously disputes the stress on waiting, casting its own theory in terms that not only do not depend on an integration of the waiting concept into the theory of production, but which even challenge the economic meaningfulness of the concept altogether.

The controversy has been alive now for over six decades, with now the one approach, now the other, seeming to hold the allegiance of the bulk of current theorists. In recent years both views have been reflected in the literature; in particular the “Austrian” approach has, as we shall see, been accorded a more respectful hearing than on many occasions in the past. Our task in the present chapter, will be to review the principal ideas to be found on the subject of waiting in this modern capital-and-interest literature. We will find much to criticize in this literature: while our greatest dissatisfaction must be with the Clark-Knight tradition, we find ourselves constrained to take sharp exception to many of the ideas used in recent attempts to rehabilitate the Böhm-Bawerkian approach. It will be our position that much of the confusion that has clouded the subject for so long can be avoided without difficulty by adhering resolutely to the planning approach—that is, by consistently seeking the explanation of all economic events by referring to the individual human plans to which these events can be traced back. What is valid in the “Austrian” approach, in particular, we will maintain, can be vastly improved by explicitly recasting Böhm-Bawerk’s seminal ideas within such a “planning”—rather than a technological—framework. It will prove helpful at this stage to preface our discussion with a brief review of our own position on the matter of capital and waiting as developed in the first chapter of this essay. It will be recalled that at the heart of this position there lay the idea of intertemporal exchanges; our view of capitalistic production sees it as a particular aspect of such an exchange across time.

Capitalistic Production as Exchange Across Time

Exchanges across time, it will be recalled, are viewed by the economic theorist as a special case of multi-period planning by forward-looking decision makers. The mutual coincidence of the multi-period plans made by two individuals may make it possible for each of them to gain by exchanging with the other across time: A giving up to B a quantity of a particular good at one date, and receiving in exchange a specified quantity of a particular good at another date. Even in a Crusoe economy however, it was noticed, it is convenient to view the plan to engage in a time-consuming process of production as a decision to engage in intertemporal exchange with nature, sacrificing inputs at one date in order to obtain output at a later date.[1]

Viewing production, in this way, as an example of multi-period planning embracing an exchange across time, we were able to proceed to recognize that the process of production may be so patterned as to afford opportunities for periodic review and possible revision of the original plans. While the original plans envisaged, perhaps, the final completion of a particular time consuming project, they were laid with the awareness that later events might make it advisable to alter the later stages of the original plan. It became of interest, then, in seeking the explanation of economic events as they unfold through time, to be able to take notice of the states of affairs as they appear at various stages of an as yet uncompleted project. Capitalistic production is initiated by plans in which producers find themselves, at the outset, half-way towards their final goals, as a result of the already partly completed projects upon which they are now able to work. What part does “waiting” play in this picture of the process of production over time?

We must at once distinguish between the prospective waiting (anticipated at the time when a time consuming process of production is being planned), on the one hand, and the actual waiting that occurs during the execution of the time consuming production process, on the other. In our analytical picture of the process of production only prospective waiting is taken notice of. Our approach, as always in micro-economic theory, is to trace back economic events to the decisions out of which they sprang. At the time when multi-period decisions are made, waiting is taken into account ex ante; the desirability of the output promised by any prospective production plan (as compared with the inputs whose sacrifice is required), is gauged, in part, by the length of time that must elapse before the output can be forthcoming. When a Crusoe initiates an entirely new project, starting without any partly completed production project to hand, he takes into account the prospective waiting that his plans calls for. When his project is already partly completed, and he is planning its later stages, then he is aware that the waiting time until his final goal is the shorter by virtue of his command over the capital goods produced during the already completed stages of the project. Where the later stages of the project are able to be carried out exactly as envisaged at the very outset of the entire project, then this means that the waiting that was originally undertaken in order to produce these capital goods, has fulfilled its ultimate purpose of making it possible to shorten correspondingly the waiting that now has to be undertaken during the later stages of the project as a whole. Where, in a market system based on division of labor, capital goods are purchased by the producers of consumer goods, the price paid for the capital goods will tend to reflect both of these aspects of waiting that are involved. On the one hand the price will tend to be high enough to cover whatever costs of waiting were undertaken by the capital good producer (i.e. the production of a capital good will be high enough to cover the anticipated costs of waiting required in its production). On the other hand the price will tend to be bid up to the level that reflects the productive usefulness of the capital good to the prospective producers of consumer goods, reflecting its ability to shorten the prospective waiting needed to be taken into account by these latter producers in making their decisions. The market process will in this way govern the volume and kind of capital goods produced and used in the economy.

It ought to be noticed moreover, that within the strictly formal context appropriate to micro-economics conceived as the pure logic of choice, it is not really necessary or even helpful to invoke the term “waiting” at all. All that is needed is the recognition that a production plan calls for the sacrifice of inputs at one date in order to obtain output at a later date. The choice made by the prospective producer will depend on the relative positions occupied on his value scale, of the inputs and outputs each with its relevant time subscript. That is all.

All this seems obvious enough and elementary enough. And yet we will discover that the dominant schools of capital theory have again and again displayed attitudes towards the role of “waiting” in relation to capital, that in one way or another ignore or deny the simple considerations here outlined. As we will see, this applies both to the current presentations of the Clark-Knight approach, and to recent attempts to present the “Austrian” approach in modern garb, (although our objections will of course have different applications to each of the approaches). We take up first that approach to capitalistic production which denies any role at all to “waiting” in the theory of production.

Lags vs. Simultaneity in Capitalistic Production

In the latter view of things, that of Professor Knight and his followers, no economic significance at all is to be attached to any lag between input and output. Whatever the time pattern that a process of production may follow from a technological point of view, for the economist the picture is one of an instantaneous emergence of product from each step taken in production: “taking production in the value sense, the result is instantaneous, whether in the form of a service (instantly consumed) or in that of a net addition to assets.”[2] First of all, it is pointed out, it is not necessary to see the result of a productive act only in the final product; “every activity has immediate results.”[3] Moreover, in the Knightian view production is appropriately defined as it was for J. B. Clark, only “in relation to economic equilibrium.” Production then “consists in using ‘productive agents’ of all kinds in a relationship of symmetrical cooperation, to provide an unvarying stream of consumable services or satisfactions as the ultimate product.” Within the context of this continuous equilibrium-flow of production, the “production of the services consumed in any period of time includes the maintenance of all productive agents and materials used in the economy, including in turn the replacement of any which are worn out or used up.” In other words production is defined in such a way that actions taken currently exclusively in order to ensure the continued flow during the future of the product service, are to be viewed as producing the flow of product service yielded during the current period. “Thus under equilibrium conditions production and consumption are simultaneous.”[4]

From this “synchronization” perspective the entire Austrian concept of “the gestation period of capital assets” must indeed appear a monumental exercise in futility. As one follower of Knight has recently expressed it, not only is the whole concept itself “fundamentally unimportant;” it has been chewed over to the point of becoming “time-consuming in the classroom and a source of confusion to the young.”[5] So long as the “equilibrium” concept of production is adhered to, there is no need to take notice of the ages of the individual components of the capital stock; the “time-structure” of this stock is of no significance. Even in the case of the fifty-tree forest, in which ten five-year old trees are cut each year and ten seedlings planted, there is no economic importance to the age-structure of the forest, so long as equilibrium conditions prevail.[6] In the Clarkian reservoir example “each drop of water coming in at the intake permitted the simultaneous release of a drop at the outlet;” the “average age” of the reservoir being of no economic moment. To point out, as Dorfman has recently done, that “if water improves with age (it sometimes does . . .) then there is an advantage in having a large reservoir,” is, technologically, perfectly correct. But, so long as equilibrium conditions are maintained, the desired superior quality of water, is secured merely through the size of the reservoir, without explicit attention having to be paid to age as such. To go on to argue, as Dorfman has, that Knight is “irrelevant” in stating that in the “stationary” view want-satisfying services are consumed as they are rendered, because Knight ignores “the fact that detention per se can be valuable,”[7] is to fail to understand the economic aspect of equilibrium production as Knight wishes it to be understood.

Professor Knight has perhaps best illustrated his view of the economic irrelevance of time lags in production, by his picture of Crusonia, an imaginary land where such lags are also technologically irrelevant. In Crusonia, as has been noted earlier in this essay, sustenance is provided by “the natural growth of some perennial which grows indefinitely at a constant (geometric) rate, except as new tissue is cut away for consumption.”[8] In such an economy, as Lerner has pointed out, “the period of production is completely arbitrary. The current growth may just as well be considered as due to the stock not being consumed five years as to its not being consumed a hundred years ago or to its not being consumed one second ago, so that the period of production concept is quite meaningless.”[9] As cited in the previous chapter, the Knightian view of the economic process is one in which the capital of the economy (its stock of productive agents) is permanent, and production consists in drawing forth the steady flow of service which these productive agents are capable of yielding indefinitely. No economic significance can, in this view of things, be attached to technological chains of causation operating over determinate time spans. To the Knightian economist production and consumption are simultaneous.

A Criticism of the Synchronization View

If the above outline of the Clark-Knight denial of significance to “waiting” in the theory of capitalistic production be accepted as a fair one, then it appears clearly vulnerable to serious objections from the point of view adopted in this essay. The cardinal objection must be that the “waiting,” the time-lag, that is denied significance by Professor Knight and his followers, is not at all the waiting that is of importance for a theory of capital. Even the acceptance of the relevance of a concept of production based on the stationary state—overcoming one’s sympathies for Hayek as he brands Knight’s use of this concept “an absurd abuse of words”[10]—does not in any way entail, (as the Knightians clearly believe it does entail), the extrusion of waiting from capital theory.

As was pointed out earlier in this chapter, the time lag that a theory of capital must take into account is never anything but a prospective, ex ante, notion. The time lags that the synchronization school seeks to persuade us to ignore can only be ex post lags. So long as planning for the future does in fact take note of prospective waiting, then a theory of capital based on the decisions made by forward planning individuals can never avoid coming to grips with “waiting,” and with the prospective time lags that make it necessary to anticipate waiting for the output technologically dependent upon contemplated inputs.

Now, there can be no doubt that, whatever Böhm-Bawerk and Wicksell may have meant, their expositions of what is usually considered the “productivity” side of their theory, fail completely to emphasize the ex ante character of waiting periods in production. It is with justification that critics who wished to build a theory of capital without the “period of production,” trained their offensive against the technological time lag involved in production, the actual time that in fact elapses, during a process of production, between the application of input and the emergence of the desired product. And it is not impossible to understand how an economist might come to consider the specific relationship between the productivity of inputs and the time-length of the production process to be unimportant for the formal, analytical notion of production. But, we must here insist, so long as our theory is to be based on the analysis of decision-making, and so long as we wish our theory to have reference to a world in which multi-period plans are made, we are precluded from doing violence to the time-aspect of these decisions. We may, for the purpose of an introductory analysis of the price system, indeed abstract from the multi-period aspects of the market, treating all decisions in the system as having to do with a single instant. But if we wish to understand the course of economic events over time, including those events that are the outcome of multi-period plans, we must be prepared to deal with these plans in the way in which they are constructed. Our analytical apparatus must be able to handle decisions in which it is contemplated to sacrifice inputs today in order to obtain output tomorrow; and our explanation of states of affairs and of chains of events in the real world, must be able to trace them back to such multi-period plans.[11]

It follows that the arguments employed by the Knightian capital theorist against the concept of the Austrian period of production, cannot possibly carry weight in connection with the waiting concept here held to be of such importance.[12] Even in an equilibrium context, with agents of production ceaselessly yielding a smooth steady flow of net output service, we are not exempt from examining the decisions being continually made to prolong in this way the currently prevailing course of events. The allocation of a portion of current gross productive capacity for the maintenance and replacement of the productive agents is the expression of a decision that this allocation be made. While it may be analytically convenient for some of the purposes which the Knightians have in mind to treat the fabrication of such replacements as part of the productive activity yielding the output service forthcoming in the current period, such a convention cannot be adopted insofar as concerns the decision to allocate current productive capacity for replacement purposes.

It is of service to notice that even in Knight’s Crusonia model, which exemplified most powerfully his idea of the arbitrariness of the period of production concept, we may nonetheless still distinguish the anticipated waiting periods here held to be the only ones worthy of economic interest. We saw that in the Crusonia perennial “the current growth may just as well be considered as due to the stock not being consumed five years ago as to its not being consumed a hundred years ago. . . .” But, clearly, this is true only with regard to the elapsed waiting to which one seeks to ascribe a specific portion of current growth. In no respect does the Crusonia model exhibit arbitrariness insofar as concerns the anticipated waiting that, at each date of decision-making in the past, was necessarily taken into account. At each date in the past, five years ago, a hundred years ago, or a second ago, a decision was made as to how much of the existing stock of the perennial should be consumed, and how much of it should be left to grow. Each such decision was a multi-period one, in which the sacrifice of current enjoyment was weighed against the future enjoyment promised by the conservation and growth of the plant. So long as we do not consider the prospective dates at which enjoyment is anticipated as matters of indifference to the decision maker, we are forced to recognize that, at each past decision making date, he took into consideration the anticipated waiting associated with each bit of dated future plant-enjoyment, that he was able to contemplate as part of one of the alternative future consumption streams with respect to which he was about to choose the first steps.

The Period of Production Concept in the Contemporary Literature

Having reviewed critically the school of thought whose aim is to eliminate the “waiting” concept from the theory of capital, we turn to examine the recent literature in which the “Austrian” approach has, after some refurbishment, been accepted into analytical respectability. In this literature the contributions of Professor Dorfman have commanded the principal interest:[13] one writer has credited Dorf man with “rescuing the concept of the period of production from obscurity and breathing life into it.”[14] Much of our discussion will revolve around Dorfman’s attempt to provide a modern exposition of the Böhm-Bawerkian theory and a defense of it against the objections raised by Clark, Knight and their followers. As was indicated at the beginning of this chapter, we will discover that our own criticisms of the Dorfman-Austrian approach are at bottom similar to those raised in the preceding section against that of Clark and Knight, viz. that grossly insufficient attention is paid to the plans, by which capital is accumulated and utilized. Once again, we will find, much of our discussion will have to do with the question of how to view “waiting”—as elapsed waiting or as anticipated waiting.

The aspect of the neo-Austrian position as exemplified by Dorfman,[15] to which we wish to draw critical attention, can perhaps be brought into sharpest focus by referring to what Dorfman calls the “bathtub theorem.” This states that, in a tank in which a constant volume of water is maintained, (while water flows into and out of the tank), the average period of detention of water equals the given volume of water divided by the rate of flow. If K is the volume of water in the tank, I the rate of flow into (or out of) the tank, and T the average period of detention of water then T = K/I. The bathtub theorem is applied directly to the case of capitalistic production in a stationary economy. The analogue of the given volume of water is now the stock of capital; corresponding to the rate of flow we have now the level of investment (which in a stationary economy equals the rate of capital consumption); T is the average period of investment.[16] A major purpose of Dorfman’s work is to apply the bathtub theorem to a two-sector economy (one in which labor and “machines” are used in one process of production to produce a consumer good, and in a different production process to produce “machines”) and to demonstrate the meaningfulness of T in this context. By this demonstration Dorfman is able to refute a major objection to the Böhm-Bawerkian theory that has been raised by Knight and others. This objection rests on the claim that since the existing capital goods were produced with the help of earlier capital goods (which in turn were produced with the help of still earlier capital goods, and so on), the period of production cannot be defined. Dorfman refutes this objection by actually calculating T for a simple two-sector economy.

In addition Dorfman is able to develop, through use of the bathtub theorem, the Böhm-Bawerkian relationships between the stock of capital, the real wage, the size of the labor force and the average period of production, in a stationary (one sector) economy. Given the first three of these quantities, the average period of production is given by the equation T = K/Nw where N is the size of the labor force, w the real wage, and K is measured in terms of the final product it will ultimately become. (Nw is thus equal to the rate of consumption flow, it being assumed that only workers engage in consumption, doing so without any saving).[17]

It is in the interpretation of this average period of production T, that Dorfman’s emphasis on the bathtub-like character of capital in a stationary economy finds its clearest expression. Alluding to Böhm-Bawerk’s interpretation[18] of this period as measuring the length of time for which society can be supported out of its “subsistence fund” (the capital stock) while waiting for the output that will result from currently invested inputs, Dorfman describes this as an “inessential” way of looking at the matter. The bathtub-view of things enables us, Dorfman points out, to see T as simply measuring the average length of time for which units of inputs are locked up in the form of unfinished products. Böhm-Bawerk saw the period of production as measuring the delay that the given stock of capital permits to be suffered between the present and the time when the results of current inputs will be available (given N and w). For Dorfman this is an incidental implication of the system. The essence of the matter for Dorfman is that in order for stationary conditions to be fulfilled, the size of the available capital stock determines the detention time of invested inputs (given N and w).

Now, it is by no means our wish to claim that Böhm-Bawerk’s understanding of the period of production was primarily an ex ante conception. This, regrettably, is not the case.[19] But it is clear that Böhm-Bawerk’s view of capital as possessing, as one of its attributes, the character of a subsistence fund (permitting the economy to bridge the time gap between the present and the date when currently invested inputs will finally yield consumption goods), does involve a forward looking perspective. What the assumption of stationary conditions enables Böhm-Bawerk to do is to show how one particular unchanging attitude with respect to the future (on the part of consumers and capitalists), is able, in equilibrium, to result in the continuous maintenance of the economy’s stock of capital, together with its (ex post) period of production. For Dorfman this forward-looking perspective on the capitalistic process of production is carefully suppressed. The same seems to hold true of Boulding’s “bathtub” presentation of the Austrian concept of the period or production, in which capital is represented as viewed as a “population of ‘value’.” “Whenever value is added to [goods] a new value is ‘born,’ and whenever value is subtracted from them by consumption a value item ‘dies.’” It is shown that in equilibrium the average “life” of these “embodied values” is given, as it is for biological populations in general, by the ratio of the size of the “population” to the rate of “births.”[20]

Our objection may be phrased in a different way. Böhm-Bawerk’s own understanding of capital as a subsistence fund enables his system to incorporate, at least implicitly, the time preferences of the individuals in the model. For Dorfman time preferences play no role whatsoever.[21] For Böhm-Bawerk the subsistence fund notion helps us to recognize that the average period of production—the delay which the system can afford between the date of input and the date of output—depends on how urgently consumption goods are needed for the near future. If w rises, so that T (as given by the formula T = K/Nw) falls, this, for Böhm-Bawerk, is a result of the fact that on average, the consumption needs of the community for the near future have become more urgent. (This is the case because we are assuming, in the Dorfman-Böhm-Bawerk model, that all consumption is out of wages, and that all wages are consumed.) As a result of the increased relative urgency of the need for consumption for the near future (as against the far future), the given subsistence fund now suffices for only a shorter period of delay. For Dorfman, if w rises then T falls simply because the average period of detention of water in a tank of given volume falls when the rate of inflow (and outflow) increases. If the flow of consumption goods provided by the capital stock is drawn down at a greater rate, then, if the size of the stock is being maintained, the average time for which inputs are tied up in the stock must be shorter, the superior productivity technologically possible through longer periods of production cannot be exploited; that is all.

There can be no doubt that the extrusion of time preference considerations from Dorfman’s model (and from the similar models of other recent expositors of the Austrian theory) is a concomitant of the refusal to view the period of production as anything but a segment of elapsed clock-time. Time preference enters into economic analysis only when multi-period decisions are being examined. When constructing a multi-period plan the individual is comparing now the desirability of consuming in the near future with the desirability of consuming in the far future. For the period of production concept to be integrated into a system of multi-period decisions, it must be envisaged ex ante.

For a theory of interest based, on the other hand, exclusively on productivity considerations—and Dorfman recognizes his aim as such[22]—ex ante aspects can be dispensed with without strain. Interest emerges in each period simply as the surplus flow of product remaining after wages have been paid; it emerges because the stock of capital has permitted a production period long enough to increase the productivity of current flow inputs so as to yield such a surplus (after paying the flow inputs a return based on their marginal product). No attention at all needs to be paid to the fact that interest payments emerged from the intertemporal exchanges made in the past, as a result of the mutual coincidence of the multi-period plans made at that time by the market participants. All that is necessary is the technological relationship existing between the length of the production period and the productivity of flow inputs. This technological relationship in no way requires an ex ante view of the period of production. If water improves with aging, then the degree of improvement depends on the elapsed detention time of water in the reservoir. The size of the reservoir is important because, with given rate of inflow and outflow, the reservoir size determines the detention period. In this way the improvement in the water yields a return to the owner of the capital stock (reservoir). All else, Dorfman declares, is inessential.

From the point of view of the position taken up in this essay, it ought to be pointed out that the recognition of a technological relationship between the length of the production process, and the productivity of flow inputs is, of course, fully taken into account in an ex ante approach. In constructing his multi-period plan an individual takes very carefully into consideration the technological terms of exchange upon which he is choosing between immediate consumption and deferred consumption. The bathtub model of the stationary state can easily be recast into an analysis based on multi-period decision-making. But in such an analysis it will be impossible to ignore the relative positions taken up, on the value scale of an individual at a particular date, by the prospective consumption of given commodities at different specified future dates. It will be impossible to conceive of the flow of interest payments being mechanically yielded as a result of the superior productivity conferred by possession of a stock of capital.

It is worthwhile to observe that it is only the ex post conception of the production period that makes it necessary for Dorfman to defend the concept against the Knightian charge, that, in a world in which capital goods have always been produced through the employment of other capital goods, the period of production cannot be defined. We have seen, of course, that Knight’s own understanding of the Austrian period of production was (with complete justification) indeed an ex post version. So that Dorfman’s refutation of the Knightian objection (by, as described earlier in this chapter, actually calculating the production period for a two sector economy in which machines are used to produce machines) by no means constitutes a case of tilting against windmills. And yet, the task of calculating such a period, as Dorfman had done, or even of demonstrating its meaning, can be avoided by simply denying the relevance of such a period to the theory of capital. As soon as one recognizes that periods of production are of importance only insofar as they are taken into account in the forward looking multi-period plans of acting individuals, the entire Knightian objection falls away.[23] If the operator of a Roman iron mine did not introduce into his planning any of the implications of the fact that some of the ore may find its way into the pocket knife of a twentieth-century school boy,[24] then no production period extending from Roman times to those of the schoolboy is relevant for the theory of capital, either as applied to Roman industry or to that of the twentieth century. The only periods of production that have relevance to the emergence of interest payments and to the organization of capitalistic production, are the periods which prospective decision makers are able and forced to take into account either directly or indirectly; any periods which cannot be defined are not so taken into account and do not affect decision making. Such periods of production can, then, hardly be of interest for capital theory.

Capital and Waiting

Our emphasis on the analysis of the multi-period decisions to which capitalistic production can be traced back, can be further exploited to clarify the role of “waiting” in relation to capital goods, and to throw light on the debate concerning the expediency of considering “waiting” to be a factor of production in the capitalistic process of production. Both questions have given rise to considerable confusion.

In the previous section it was pointed out that the time lag separating the date of input application from the date of the emergency of the corresponding output has been variously treated in the theory of capitalistic production. The case was presented in favor of a treatment that would introduce the time lag into analysis exclusively on an ex ante basis. In the present section we examine the relationship between “waiting”, and capital as such; we are directly concerned here not with the period of production as an aspect of the production process, but with the ways in which capital goods have been considered as themselves “representing” waiting periods variously conceived. The various views that we will notice are all in some way derived from the Böhm-Bawerkian emphasis on the role of time in the theory of production.

(a) Capital as “congealed waiting”: Böhm-Bawerk’s stress on the fact that the only “original” factors are land and labor, and that capital goods are the intermediate goods produced by the original factors in the course of time consuming, roundabout, processes of production, has led to the view that capital goods are “stored-up” labor, land and waiting. Capital goods “embody” the past investment of original factors as well as the “waiting” performed in order to obtain these capital goods.[25] In this view of things the productivity of capital goods in the later stages of the production process is to be traced back to the labor, the land and the waiting which these goods represent. This view of capital as a stock of past waiting is (as we will see in a subsequent section of this chapter) closely related to the attempt to treat “waiting” as an original factor of production, and also (as we will see in the following chapter) to the attempt to measure capital by the amount of waiting which is represents.

The fundamental objection against using terms such as “congealed waiting” and the like is that this cast of language conduces to a “cost-of-production” way of thinking about capital goods. The truth is that, as Hayek has reminded us, bygones are bygones in the theory of capital no less than elsewhere in economics.[26] A theory of production which commences with a given stock of capital goods need pay no attention whatsoever to the genesis of these goods.[27] This objection applies, of course, no less to descriptions of capital as “stored up labor and land” (which Böhm-Bawerk himself endorsed[28]) than to the descriptions which see capital as also stored-up “waiting” (which, as we will see in the following section Böhm-Bawerk rejected because he did not recognize “waiting” as a factor of production).

On the other hand, the very attempt—central to Böhm-Bawerkian thought—to trace back capital to the “original” factors which were invested in the past to produce it, gives a certain value to the “stored-up-factors” description of capital. The position which the Austrians were primarily concerned to demote, was that which sees capital and capitalistic production in a purely “timeless” context; in this “timeless” view nothing that is of importance for the theory of capital and interest needs to be related to time-lags in production. To emphasize the opposite view, which sees the time-dimension of the production process as the cardinal element in the theory of capital and interest, it is of value to be able to stress the fact that capital goods are themselves the results of processes of production begun in the past. This is the rationale of the “stored-up-factors” description.

From the point of view adopted in this essay, the issues seem extremely simple. Looking forward to capital goods that have not yet been produced, the multi-period planner knows what current inputs must be applied, and the length of time for which the capital goods must be waited. Again, looking still further ahead in time, the planner sees the final consumer products able to be produced with the help of the prospective capital goods. If he wishes he may, looking forward in this way, view the prospective capital goods as the concrete vehicles by means of which his current inputs can be “stored-up” in order to yield their final product, consumer goods, far in the future. But clearly there seems little analytical convenience in looking at things in this way, from the point of view of the date at which the decision is made to construct the capital goods. When the capital goods have already been constructed, and the economic historian wishes to explain analytically how these particular capital goods came to be produced, he must of course do so by referring to the past inputs of “original” factors which the decision maker long ago decided to invest. From the point of view of the decision maker himself, with the capital goods at his disposal, the past is of no consequence whatsoever. In his capacity of decision maker the presently available capital goods are no different, in this respect, than any other factors of production over which he has command. Because to the decision maker bygones ought to be bygones, the analysis of the decisions made at this point, need likewise pay no attention to the history of the capital goods. To refer to them as stored-up land and labor or waiting adds nothing to the explanation of subsequent events. (Of course, once the capital goods have been used to produce final output, then the economic historian who wishes to explain analytically the course of past events, must do so by referring to the original decisions which gave rise to the particular capital goods produced, the availability of which determines, in part, the decisions taken during the final stages of production.)

In an explicitly ex ante approach such as this, there seems little to be gained from the use of “stored-up-factors” descriptions of capital goods. Where such descriptions are used, the appropriateness of including “waiting” as one of the factor services “congealed” in the capital goods, depends entirely on one’s view of whether to treat “waiting” as a factor of production in the capitalistic production process. This will be taken up later in this chapter.

(b) Capital as a Store of Potential Waiting: We have already seen that Böhm-Bawerk described capital as a “subsistence fund” that enables society to bridge the gap in time that separates the date of input from the date of output. In this sense the length of the production process that society can afford is determined by the size of this subsistence fund. Capital enables society to wait the time required to take advantage of the superior productivity of the more roundabout methods of production. Capital has come, in this way, to be viewed as constituting a fund of “waiting.”

Professor Hayek has extensively discussed the errors to which this kind of view can lead.[29] This view sees a given capital stock as making possible a definite waiting period. This notion could be acceptable only if the capital stock consisted of a stock of consumers’ goods (or at least consisted of “inchoate” consumers’ goods in the sense that the capital stock is able to make possible only one particular future pattern of consumers’ goods) and if the given pattern of consumption can be assumed. Only in such an oversimplified world can one talk of a capital stock as a definite store of potential waiting. In less unrealistic worlds a “given stock of capital goods does not represent one single stream of potential output of definite size and time shape; it represents a great number of alternatively possible streams of different time shapes and magnitudes.”[30] The consequence of the “fund of waiting” way of looking at capital has been to tend to avert attention from the concrete items composing a particular stock of capital, and to treat the stock as if it were homogeneous. And from the notion of capital as a homogeneous fund, it is not far to the idea that this abstract fund somehow has an existence apart from the concrete items of which the stock is composed, an existence which is “permanent” in the sense that the fund “is maintained intact though the things in which it is invested may go and come to any extent.”[31]

It may be added that emphasis on the ability of the capital stock to permit a period of waiting in the above sense, makes it more difficult to accept any distinction that might be made between capital resources and non-capital resources. It would seem that the larger the command possessed over resources of all kinds, the longer would be the processes of production upon which one could enter (because of one’s ability to allocate a sufficient portion of one’s resources as a whole, towards the satisfaction of one’s wants for the nearer future).

Despite these criticisms of the view which sees capital as a store of potential waiting, we ought not refuse to recognize the valuable insights that this view undoubtedly does incorporate.[32] The lengths of the production process upon which one will choose to embark at any time certainly are governed by one’s command over the resources capable of yielding products in the relatively near future. And to the extent that one’s provision for the future inclines one to invest now in order to be able, in future years, to engage in those longer, more productive, production processes which one cannot now afford one is in fact accumulating the resources that will in those future years enable one to “wait” for lengths of time longer than one can now afford to do. These resources will certainly be capital resources. From the point of view of economic history too, there is unquestioned merit in the recognition that one’s present ability to undertake lengthy processes of production is to be ascribed to the stock of capital goods built-up during earlier years. While the “store of waiting” concept of capital turns out to be of little service for the theory of the capitalistic process of production; it does represent an application of that theory that is of considerable practical significance.

(c) Capital as a means of shortening the waiting period: Yet another relationship has been emphasized in the literature as existing between capital and “waiting.” In this view capital goods are seen as instruments whose function it is to shorten the waiting time necessary before the emergence of the final consumption goods.[33] The possession of capital makes one nearer in time to the final attainment of one’s goal than one would be without it. Böhm-Bawerk himself did not emphasize this aspect of capitalistic production. (In fact, Böhm-Bawerk actually found it necessary, in discussing the notion of capitalistic production as a more time-consuming way to produce, to defend it against the objection that a tailor can produce a coat in a shorter time with the help of a sewing machine than without it).[34]

The emphasis on the shortening of the waiting period made possible by capital goods, arises out of a wish to avoid a misunderstanding for which the Böhm-Bawerkian terminology must bear some responsibility. The notion of “roundabout” processes of production has sometimes helped writers to overlook the fact that, wherever undertaken, such processes represent the shortest possible route to the finally desired goal. Especially when taken in conjunction with the above discussed views of capital as congealed waiting and as a fund of waiting, the “roundabout” aspect of capitalistic production has been interpreted as implying that the possession of capital requires one to wait for the output to emerge.[35] It is the wish to stress the calculated purposefulness and the temporal efficiency of capitalistic production that has given rise to the above description of capital goods as instruments to shorten the time for which the final goal must be waited.

Of course a statement that the possession of capital makes one nearer in time to one’s goal is open to the objection that this is merely a way of saying that the use of capital goods is able to increase output. Surely, it will be asked, the control over resource services of any kind similarly able to increase output can be described as placing the producer within closer reach of his final goal? From the point of view adopted in this essay some light can perhaps be thrown on the matter.

We have seen that capital goods represent the as yet only partly completed projects initiated in the past by multi-period decision makers. The time lag originally incorporated in the multi-period plan was the shortest waiting period necessary to achieve the planned final output and all things considered represented the most desirable production plan available. If this project is able to be carried forward to full completion as planned, then, of course, the fact that the waiting time during the closing stages of the project is shorter than the time lag originally incorporated in the entire multi-period plan, is merely the result of the fact that the earlier time-consuming stages of the plan have already been completed. It is this aspect of the matter that is noticed when capital goods—the result of the earlier stages—are described as “shortening the period of waiting.” Without these goods the multi-period project would have to be commenced from the very beginning. Of course, when one considers the final stages of the plan in isolation, this aspect of the matter loses significance. Bygones are bygones; with given capital goods at hand one is, it is true, nearer to the final goal, but the same holds with respect to command over all kinds of productive services. But we have already noticed several times that the significance of capital goods qua capital goods emerges only when the entire multi-period production project is considered as an integrated plan. And from this viewpoint it is indeed useful to recognize that the intermediate tools or other goods that such a plan calls for, will represent the completed earlier stages of a longer plan. The possession of an acre of farmland renders the owner nearer in time, no doubt, to a bushel of wheat. But insofar as no multi-period plan was involved in making this resource available, there is no particular analytical insight to such a description of the farmland. But the possession of a tractor makes the farmer nearer to the wheat in the sense that earlier periods of waiting were undertaken in order to place the farmer in this favored position. Where other views saw the tractor as a store of past waiting, the view here under discussion sees the tractor (because of the past waiting that was deliberately undertaken), as the means whereby the farmer is spared further waiting.

Is “Waiting” a Factor of Production?

Our discussion of the various ways in which capital goods have been conceived of as related to “waiting,” makes it of interest to take up at this point the question of whether to consider “waiting” as a factor of production. Of course, for a theory of production in which no significance at all is ascribed to time lags between the date of inputs and the date of output, there can be no question of ascribing productivity to time as such,[36] or to waiting. The debate concerning waiting as an independent, “original” factor of production has had relevance only in the context of the Böhm-Bawerkian-type theory of capitalistic production.

Böhm-Bawerk himself vigorously denied the possibility of distinguishing any independent third factor of production to stand side by side with labor and nature. Capital itself is not such a factor, it is merely the intermediate product of nature and labor.[37] All production achieved with the help of capital must be ascribed to the nature and labor which, some time in the past, cooperated in the production of capital. No portion of the value of the final product can be imputed back to any other original factor. This was the basis for the Böhm-Bawerkian rejection of theories of interest based on the productivity of capital itself.

On the other hand it was one of the cardinal features of the Böhm-Bawerkian view of capitalistic production, that capital cannot come into existence without prior saving, or abstinence from consumption. “Production and saving constitute equally indispensable conditions of the formation of capital.”[38] The question then arises whether saving, or “abstinence,” or “waiting” should not be recognized as an independent factor, coordinate with labor and nature, responsible for the productivity later to be displayed by the capital goods thus accumulated. Followers of Böhm-Bawerk who emphasized that aspect of his system which depended on the productivity of time-consuming methods of production (as compared with the time-preference aspect), have indeed tended to treat time, or waiting, in this way as a factor of production, measuring the marginal product of waiting and viewing interest as the return to those responsible for the waiting represented by the capital employed.[39]

But Böhm-Bawerk himself refused to treat waiting as a third independent factor. The function of saving, Böhm-Bawerk insists, “is not in the nature of performing part of the work which is peculiarly its own province. Instead, its function is to see to it that the productive forces nature and labor, which, in any event, must perform the entire work of production themselves, shall not direct their efforts toward any goal but production, that is to say, shall produce capital goods and not consumption goods.”[40]

More recent writers have remarked on the “philosophical” aspects of the debate. Haavelmo refers to the alternatives of regarding time itself as an active productive agent or only as a medium permitting the flows of inputs to cumulate into the product.[41] (On the other hand Haavelmo has elsewhere objected without reservation to treating “waiting” itself as productive. “Certainly it is not the delay in output which itself is ‘productive’, but the fact that the more time we spend, the more capital we can construct.”[42]) Fraser has discussed at length the appropriateness of treating passive conditions of productive processes (among which he includes “waiting”) as factors of production side by side with those resources which are more “intimately bound up with the actual technical processes whereby goods are made.”[43] Hayek has remarked that time cannot be treated as a factor of production “since no definite ‘quantity’ of time is given in a way which would enable us to distribute this ‘fund’ of time in alternative ways between the different lines of production so that the total of time used will always be the same.”[44]

Among recent writers it is Professor Dorf man who (taking up a posture of originality that does not appear to be entirely justified) has been most emphatic in treating waiting as an independent, original factor of production. He makes this way of handling the matter—which, as we have seen, Böhm-Bawerk himself rejected—the cornerstone of his rehabilitation of Böhm-Bawerkian capital theory.[45] Clearly we have an unusual kind of divergence of opinion on this question. Some clarification of the issues involved can, we believe, be obtained with the help of the point of view upon capitalistic production that has been developed in this essay. Part of the confusion has to do with the two views concerning the production function (and hence the significance of productive factors) that were discussed in the previous chapter.

It will be recalled that the one view treats the production function in a purely “positive” manner; it is seen as representing simply a set of technological relationships. It answers the question: what combinations of ingredients are able to bring into existence the various quantities of the product? The second (“economic”) view of the production function, on the other hand, looked upon it as presenting the alternative opportunities afforded by technology from among which the human producer can choose. While for most purposes little substantive difference arises as a result of these two different views, we have already noticed in the previous chapter how these views do importantly affect the treatment of capital. We now have before us yet another instance where the two views on the interpretation of the production function do make a difference. Once again we will discover that it is the second of these two views which proves most helpful, and most appropriate for economic analysis.

For, so long as the production function is seen merely as a statement of technological relationships it is difficult for anyone conscious of the time-dimension of production processes to avoid looking upon time, or waiting, as a productive factor, i.e. as one of the ingredients necessary before the final output is available. Whether as an “active” or a “passive” ingredient, the lapse of time, or waiting, must precede the emergence of the product of a capitalistic process begun with only labor and nature. Labor and nature alone cannot yield output; labor plus nature plus waiting time yield output.

But, when the production function is looked upon from the economic point of view, that is, when the inputs and outputs are viewed as components of alternative production plans, the matter looks quite different. Now the production function comprises alternative multi-period production plans, each viewed from a point in time before it has been adopted. Contemplating prospectively one such multi-period plan, the decision maker finds, say, that the investment of a certain quantity of land and labor now, will yield a definite quantity of grain at a specified future date. Contemplating a second plan, he finds that the investment now of similar quantities of land and labor, in a different process of production, will yield a smaller quantity of grain (at a date so near as to appear for all practical purposes) in the immediate future. It is now clearly possible to maintain that the greater quantity of grain promised by the first plan does not call for any resources above those required by the second plan. The first plan merely requires that the decision maker decide to adopt the first production process rather than the second, this decision being taken with the awareness that the output is to be available only at the specified future date; once the futurity of the grain has been taken into account as qualifying the utility promised by the investment, no further input “sacrifice” in the form of waiting need be taken into consideration in the production plan. It is as if a boy were to stand with a dime before two vending machines, one of which yields its contents several seconds earlier than the second. For each machine a dime is all that is required. Looking prospectively at both alternatives the boy is merely aware that the contents of the second machine seem further away in time than do those of the first.

Looked at ex post, or from a purely technological point of view, such would not seem to be the case. In the technological sense grain is grain. Even if time subscripts be carefully appended, by itself this does not indicate the relative desirabilities of grain at various dates. For a complete technological description of the time consuming process it is therefore necessary to say that the investment of land and labor at one date, plus the lapse of a specified time interval, yields grain at a future date. It is only in the context of the plan, contemplated ex ante by the decision maker, that it is superfluous to say that, in addition to the investment of the labor and land* he must also invest waiting in order to obtain grain at the future date. But in the planning context such a statement would indeed be superfluous; once the planner is aware of the date of the future output, once he takes into account the relative position on his present scale of values of grain-in-the-specified-future, he merely has to recognize that in order to obtain this grain (rather than the smaller quantity of immediate grain) he must invest the labor and land in process one (rather than process two). No further sacrifices need be taken into account. In this sense Böhm-Bawerk was on very firm ground indeed when he pointed out that the decision to save is not a decision to apply additional input of some kind to the production process, but merely a decision to guide the process towards future goals rather than immediate goals.[46]

It turns out then, not entirely to one’s surprise, that (a) when the time-consuming process of production is viewed prospectively, it is perfectly feasible to ignore “waiting” altogether when discussing inputs, and to take notice of all the implications of “waiting” that need to be considered, by having due regard to the effect of time preference upon the present desirability of the prospective future outputs; and that (b) when the time-consuming process of production is viewed “positively,” on the other hand, there is every justification to notice “waiting” as one of the inputs, and to proceed to discuss the superior productivity of the more time-consuming processes of production selected, as attributable to the productivity of time. Given the strictly ex ante point of view adopted in this essay, it should be clear that it is the former of these alternatives that is to be endorsed. Böhm-Bawerk’s refusal to recognize waiting as an input—a matter that has frequently occasioned mild surprise—follows consistently from his fundamental insight that interest is a phenomenon that emerges from intertemporal exchange, in which the time preferences of the exchanging individuals are of the essence. It is only in that line of thought (exemplified, say, by Wicksell and by Dorfman) in which Böhm-Bawerkian doctrine has been developed almost exclusively along productivity lines, with almost no attention to time preferences at all,[47] in which the ex ante concept of waiting periods is unnecessary, and in which, as a consequence, waiting came to appear naturally as a productive factor with its own marginal productivity and earning its own productivity return.

One further observation on the question of the productivity of waiting is in order. The Böhm-Bawerkian thesis that more roundabout methods of production are more productive was the cause of a great amount of confusion in the subsequent literature. So long as the “productivity” side of the Austrian theory is treated as in a vacuum, it is indeed not an easy matter to establish the universal validity of the thesis. But as soon as Böhm-Bawerk’s own rejection of waiting as a factor of production is recalled, and when it is borne in mind that the basis for this rejection exists in the appropriateness of the “plan” approach to multi-period decision-making in production, the matter becomes much clearer. It is by no means necessary to postulate that more waiting means more product (as must at least be postulated by the purely “productivity” versions of the Austrian theory, in which waiting is a factor of production). All that is needed is the recognition that “unless waiting is assumed to be pleasant” the only situations which are economically interesting are those marginal ones in which “the utility of extra products has to be balanced against the disutility of extra waiting.”[48] At the “margin of roundaboutness” the decision maker is always careful to make sure that his multi-period plan does not call for sacrifices of satisfactions in the near future which, in his estimation are not warranted by the yield promised for the more distant future. With more and more resources at his disposal, so that the former sacrifices no longer have to be made, he will be able to exploit the desirable, high-productivity processes which had earlier to be rejected only because of the length of waiting time involved. That such processes can nearly always be assumed to exist has been explained by Hayek.[49]

---

[1] For a discussion of the insight that Crusoe’s activities consist of acts of “exchange”, see Mises (1953) pp. 38–39.

[2] Knight (1936) p. 456.

[3] Rolph (1939) p. 280.

[4] Knight (1946) p. 387; cf. also Stigler (1941) pp. 313–314.

[5] Dewey (1963) p. 134n; see also Dewey (1965) p. 80 and pp. 201ff.

[6] Cf. Lerner (1953) p. 544.

[7] Dorfman (1959b) pp. 354–355. See also Schumpeter (1954) p. 907.

[8] Knight (1944) p. 30; see also Dewey (1963) and Dewey (1965) pp. 80ff.

[9] Lerner (1953) pp. 543–544.

[10] Hayek (1936) p. 370, n.25.

[11] It is somewhat surprising that in the literature defending the “Austrian” view from the attacks of Professor Knight, the distinction between “ex ante waiting” and “elapsed waiting,” made here in the text, was not used with much greater emphasis and with much greater effect. Hayek (1936, p. 364) clearly spells out the distinction; see also Saulnier (1938, p. 273) and Hayek (1941) p. 90. The most vigorous insistence on the ex ante character of waiting is to be found in Mises (1949) pp. 477ff; see also Rothbard (1962) p. 45, Blaug (1962) p. 466. One possibility of defending the “Austrian” view from the criticisms of Professor Knight, that is somewhat parallel to that developed here in the text, has been noticed in the literature. This has to do with Knight’s insistence that, no matter how far back one goes into history, one never discovers processes of production in which only pure “original” factors of production are used, so that the distinction between “original” and “produced” factors cannot be made. It has been pointed out that “the services of the resources accruing at the present moment might be regarded as ‘original factors’ as against the services of resources accruing at any subsequent moment.” (Kaldor, 1937, p. 158); see especially Schumpeter (1954) p. 908, see also Conard (1959) pp. 21, 112–113.

[12] The remarks in the text do not deal with one major difficulty associated with the waiting concept, which critics of the “Austrian” approach have emphasized. This has to do with the difficulty of identifying the date of the output that corresponds to a given input, resulting from the “jointness” present in typical processes of production. This difficulty exists for the ex ante notion of waiting as it does for the ex post concept. See Kaldor (1937) p. 159n.; Hayek (1941) pp. 67, 205f.

[13] Dorfman (1959a; 1959b). See also Blaug (1962) pp. 459–478; Uhr (1960) Chapter 5. For further signs of the renewed fashionability of Austrian-type capital theory concepts, see Lutz (1961) and Lindahl (1961).

[14] Neuberger (1960) p. 150.

[15] Besides the contribution by Dorfman (1959a) cited above, presentations of the Austrian position on somewhat similar lines are to be found in Stigler (1941) pp. 278–285, Brems (1959) Chapter 18.

[16] For earlier similar treatments of capital in terms of bathtubs or like analogies see e.g. Lerner (1944) p. 325, Boulding (1950) p. 193.

[17] Cf. Böhm-Bawerk (1921) p. 365; see also e.g. Dorfman (1959b) p. 354. On the assumption that labor is expended on each unit of output at an even rate throughout the period, the period of production T is one half the absolute length of a production process in which a unit of final output emerges.

[18] Böhm-Bawerk (1921) pp. 312ff.

[19] See below in this chapter for a critique of the subsistence fund idea.

[20] Boulding (1950) p. 193.

[21] See Neuberger (1960) for a criticism of Dorfman’s excessive emphasis on productivity as an explanation of interest, to the complete exclusion of time-preference. Dorfman’s refusal to consider the effects of alternative patterns of time preference vitiates also his attempt to prove that present goods are more valuable than future goods of like kind and number. The refutation of Dorfman’s proof, (like that of Böhm-Bawerk’s “Third Ground”) was provided long ago by Fisher. For a recent careful exposition of the relevant fallacy in the Böhm-Bawerk-Dorfman proof, see Conard (1959) p. 43. See however Kuenne (1962a, 1962b).

[22] Dorfman (1959b) p. 360.

[23] See further above in this chapter footnote 11.

[24] Cf. Böhm-Bawerk (1921) p. 86; Stigler (1941) p. 205; Schumpeter (1954) p. 908. See also Blaug (1962) p. 470.

[25] See Dorfman (1959b) p. 363; Blaug (1962) p. 471. Cf. also Mises (1949) p. 490.

[26] Hayek (1941) p. 89,

[27] Hayek (ibid.) uses this consideration to reject the “produced means of production” definition of capital goods. See also Hayek (1936) p. 364, n.19.

[28] See e.g. Böhm-Bawerk (1921) p. 100.

[29] Hayek (1936) pp. 374ff; (1941) pp. 93, 146ff, 189–190.

[30] Hayek (1941) p. 147.

[31] Knight (1935) p. 57.

[32] Even Hayek (1936) p. 374, concedes some justification to this view “for didactic purposes.”

[33] See Mises (1949) p. 490, Rothbard (1962) p. 45.

[34] Böhm-Bawerk (1921) p. 83.

[35] For an apparent expression of this kind of interpretation, notice the following: “Why is a long lived consumption good regarded as a ‘capital’ good? Because it is full enjoyment necessitates an appreciable degree of ‘waiting’.” Fraser (1937) pp. 266–267.

[36] For Knight’s earlier views on this point see Knight (1931) p. 198.

[37] Böhm-Bawerk (1921) pp. 97–98. Kaldor’s rather surprising statement on this point (1937, p. 192) has already been sharply criticized by Schumpeter (1954) p. 901, n.26.

[38] Böhm-Bawerk (1921) p. 116.

[39] Wicksell (1934) Volume I, p. 177.

[40] Böhm-Bawerk (1921) p. 117.

[41] Haavelmo (1960) p. 47.

[42] Haavelmo (1960) p. 40.

[43] Fraser (1937) pp. 208–209.

[44] Hayek (1936) p. 377.

[45] Dorfman (1959b) pp. 359ff.

[46] A few additional words of clarification may be helpful. It may be objected that even in the “planning” context, it is of course possible for the decision maker to view the first plan as calling for additional input in the form of waiting, in order to obtain the greater quantity of grain. This is perfectly true. Analogously if one pays fifty cents for the privilege of sitting on a hard chair (with the hardness viewed as a necessary evil), one may indeed alternatively say that one is sacrificing fifty cents plus the discomfort of sitting on a hard chair, for the privilege of sitting on a chair at all. Clearly it may not always be possible to declare without arbitrariness that a certain disutility associated with a particular enjoyment is to be viewed as lowering the relative position of this enjoyment on one’s value scale, rather than as an independent sacrifice required as an “input” for the purchase of the enjoyment. The point made in the text is simply that in the planning context it is possible (and sometimes even natural) to view the futurity of output as qualifying its desirability (from the point of view of the present)—in which case no separate sacrifice of waiting need be taken into account. This is in contrast to the “positive,” technological, description of a process, in which, even when the date of output has been specified, it is not necessarily superfluous to notice that a time interval elapses between the investment of inputs and the final yield.

[47] Cf. Hayek (1941) p. 420.

[48] Fraser (1937) p. 387. This formulation of the “law of roundaboutness” has become quite common in recent literature, see e.g. Hayek (1941) p. 60, Mises (1949) p. 492, Hazlitt (1959) pp. 212, 230, Haavelmo (1960) p. 92.

[49] Hayek ibid.