Books / Digital Text
4. Capital Accumulation and the Length of the Structure of Production
We have been demonstrating that investment lengthens the structure of production. Now we may consider some criticisms of this approach.
Böhm-Bawerk is the great founder of production-structure analysis, but unfortunately he left room for misinterpretation by identifying capital accumulation with adopting “more roundabout” methods of production. Thus, consider his famous example of the Crusoe who must first construct (and then maintain) a net if he wishes to catch more than the number of fish he can catch without any capital. Böhm-Bawerk stated: “The roundabout ways of capital are fruitful but long; they procure us more or better consumption goods, but only at a later period of time.”24 Calling these methods “roundabout” is definitely paradoxical; for do we not know that men strive always to achieve their ends in the most direct and shortest manner possible? As Mises demonstrates, rather than speak of the higher productivity of roundabout methods of production, “it is more appropriate to speak of the higher physical productivity of production processes requiring more time” (longer processes).25
Now let us suppose that we are confronted with an array of possible production processes, based on their physical productivities. We may also rank the processes in accordance with their length, i.e., in terms of the waiting time between the input of the resources and the yielding of the final product. The longer the waiting period between first input and final output, the greater the disutility, ceteris paribus, since more time must elapse before the satisfaction is attained.
The first processes to be used will be those most productive (in value and physically) and the shortest. No one has maintained that all long processes are more productive than all short processes.26 The point is, however, that all short and ultraproductive processes will be the first ones to be invested in and established. Given any present structure of production, a new investment will not be in a shorter process because the shorter, more productive process would have been chosen first.
As we have seen, there is only one way by which man can rise from the ultraprimitive level: through investment in capital. But this cannot be accomplished through short processes, since the short processes for producing the most valuable goods will be the ones first adopted. Any increase in capital goods can serve only to lengthen the structure, i.e., to enable the adoption of longer and longer productive processes. Men will invest in longer processes more productive than the ones previously adopted. They will be more productive in two ways: (1) by producing more of a previously produced good, and/or (2) by producing a new good that could not have been produced at all by the shorter processes. Within this framework these longer processes are the most direct that must be used to attain the goal—not more roundabout. Thus, if Crusoe can catch 10 fish per day directly without capital and can catch 100 fish per day with a net, building a net should not be considered as a “more roundabout method of catching fish,” but as the “most direct method for catching 100 fish a day.” Furthermore, no amount of labor and land without capital could enable a man to produce an automobile; for this a certain amount of capital is required. The production of the requisite amount of capital is the shortest and most direct method of obtaining an automobile.
Any new investment will therefore be in a longer and more productive method of production. Yet, if there were no time preference, the most productive methods would be invested in first, regardless of time, and an increase in capital would not cause more productive methods to be used. The existence of time preference acts as a brake on the use of the more productive but longer processes. Any state of equilibrium will be based on the time-preference, or pure interest, rate, and this rate will determine the amount of savings and capital invested. It determines capital by imposing a limit on the length of the production processes and therefore on the maximum amount produced. A lowering of time preference, therefore, and a consequent lowering of the pure rate of interest signify that people are now more willing to wait for any given amount of future output, i.e., to invest more proportionately and in longer processes than heretofore. A rise in time preference and in the pure interest rate means that people are less willing to wait and will spend proportionately more on consumers’ goods and less on the longer production processes, so that investments in the longest processes will have to be abandoned.27
One qualification to the law that increased investment lengthens production processes appears when investment turns to a type of good which is less useful than the goods previously acquired, yet which has a shorter process of production than some of the others. Here the investment in this process was checked, not by the length of the process, but by its inferior (value) productivity. Yet even here the structure of production was lengthened, since people have to wait longer for the new and the old goods than they previously did for the old good. New capital investment always lengthens the overall structure of production.
What of the case where a technological invention permits a more productive process with a lesser amount of capital investment? Is this not a case in which increased investment shortens the production structure? Up to this point we have been assuming technological knowledge as given. Yet it is not given in the dynamic world. Technological advance is one of the most dramatic features of the world of change. What then of these “capital-saving” inventions? One interesting example was cited by Horace White in a criticism of Böhm-Bawerk.28 Oil was produced first by ships hunting in the Arctic for whales, the whale oil being processed from the whales, etc., an obviously lengthy production process. Later an invention permitted people to bore for oil in the ground, thereby immeasurably shortening the production period.
Aside from the fact that, empirically, most inventions do not shorten physical production processes, we must reply that the limits at any time on investment and productivity are a scarcity of saved capital, not the state of technological knowledge. In other words, there is always an unused shelf of technological projects available and idle. This is demonstrable by the fact that a new invention is not immediately and instantaneously adopted by all firms in the society. Therefore, any further investment will lengthen production processes, many of them more productive because of superior technique. A new invention does not automatically impel itself into production, but first joins the unused array. Further, in order for the new invention to be used, more capital must be invested. The ships for whaling have already been built; the oil wells and machinery, etc., must be created anew. Even the newly invented method will yield a greater product only through further investment in longer processes. In other words, the only way to obtain more oil now is to invest more capital in more machinery and lengthier production periods in the oil-drilling business. As Böhm-Bawerk pointed out, White's criticism would apply only if the invention were progressively capital-saving, so that the product would always increase with the shortening of the process. But in that case, boring for oil with one's bare hands, unaided by capital, would have to be more productive than drilling for oil with machinery.29
Böhm-Bawerk drew the analogy of an agricultural invention applied to two grades of land, one grade previously yielding a marginal product of 100 bushels of wheat, the lower grade yielding 80 bushels. Now suppose use of the invention raises the marginal product of the lower-grade land to 110 bushels. Does this mean that the poorer land now yields more than the fertile land and that the effect of agricultural inventions is to make poorer lands more productive than fertile ones? Yet this is precisely analogous to White's position, which maintains that inventions may cause shorter production processes to be more productive! As Böhm-Bawerk pointed out, it is obvious that the source of the error is this: inventions increase the physical productivity of both grades of land. The better land becomes still better. Similarly, perhaps it is true that an invention will cause a shorter process to be more productive now than a longer process was previously. But this does not mean that it is superior to all longer processes; longer processes using the invention will still be more productive than the shorter ones. (Boring for oil with machinery is more productive than boring for oil without machinery.)
Technological inventions have received a far more important place than they deserve in economic theory. It has often been assumed that production is limited by the “state of the arts”—by technological knowledge—and therefore that any improvement in technology will immediately show itself in production. Technology does, of course, set a limit on production; no production process could be used at all without the technological knowledge of how to put it into operation. But while knowledge is a limit, capital is a narrower limit. It is logically obvious that while capital cannot engage in production beyond the limits of existing available knowledge, knowledge can and does exist without the capital necessary to put it to use. Technology and its improvement, therefore, play no direct role in the investment and production process; technology, while important, must always work through an investment of capital. As was stated above, even the most dramatic capital-saving invention, such as oil-drilling, can be put to use only by saving and investing capital.
The relative unimportance of technology in production as compared to the supply of saved capital becomes evident, as Mises points out, simply by looking at the “backward” or “underdeveloped” countries.30 What is lacking in these countries is not knowledge of Western technological methods (“know-how”); that is learned easily enough. The service of imparting knowledge, in person or in book form, can be paid for readily. What is lacking is the supply of saved capital needed to put the advanced methods into effect. The African peasant will gain little from looking at pictures of American tractors; what he lacks is the saved capital needed to purchase them. That is the important limit on his investment and on his production.31
A businessman's new investment in a longer and more physically productive process will therefore be made from a sheaf of processes previously known but unusable because of the time-preference limitation. A lowering of time preferences and of the pure interest rate will signify an expansion of saved capital at the disposal of investors and therefore an expansion of the longer processes, the time limitation on investment having been weakened.
Some critics charge that not all net investment goes to lengthening the structure—that new investments might duplicate pre-existing processes. This criticism misfires, however, because our theory does not assume that net saving must be invested in an actually longer process in some specific line of production. A longer production structure can just as well be achieved by a shift from consumption to investment that will lengthen the aggregate production structure by greater investment in already existing longer processes, accompanied by less investment in existing shorter processes. Thus, in the case of Crusoe mentioned above, suppose that Crusoe now invests in a second net, which will permit him to catch a total of 150 fish a day. The structure of production is now lengthened even though the second net may be no more productive than the first. For the total period of production, from the time he must build and rebuild his total capital until his product arrives, is now considerably longer. He must now cut down again on present consumption (including leisure) and work on his second net.32
- 24. Böhm-Bawerk, Positive Theory of Capital, p. 82.
- 25. Mises, Human Action, pp. 478–79.
- 26. See Hayek, Pure Theory of Capital, pp. 60 ff. Similarly, there are numerous long processes which are not productive at all or which are less productive than shorter processes. These longer processes will obviously not be chosen at all. In sum, while all new investment will be in longer processes, it certainly does not follow that all longer processes are more productive and therefore worthy of investment. For Böhm-Bawerk's strictures on this point, see Eugen von Böhm-Bawerk, Capital and Interest, Vol. 3: Further Essays on Capital and Interest (South Holland, Ill.: Libertarian Press, 1959), p. 2.
- 27. It should be clear that, as Mises lucidly put it,
Originary [pure] interest is not a price determined on the market by the interplay of the demand for and the supply of capital or capital goods. Its height does not depend on the extent of this demand and supply. It is rather the rate of originary interest that determines both the demand for and the supply of capital and capital goods. It determines how much of the available supply of goods is to be devoted to consumption in the immediate future and how much to provision for remoter periods of the future. (Mises, Human Action, pp. 523–24)
- 28. Eugen von Böhm-Bawerk, “The Positive Theory of Capital and Its Critics, Part III,” Quarterly Journal of Economics, January, 1896, pp. 121–35. See also idem, Further Essays on Capital and Interest, pp. 31ff.
- 29. Böhm-Bawerk, “The Positive Theory of Capital and Its Critics, Part III,” pp. 128 ff.
- 30. Mises, Human Action, pp. 492ff.
- 31. The futility of “Point 4” and “technical assistance” in furthering production in the backward countries should be evident from this discussion. As Böhm-Bawerk commented, in discussing advanced techniques: “There are always thousands of persons who know of the existence of the machines, who would be glad to secure the advantage of their use, but who do not dispose of the capital necessary for their purchase.” Böhm-Bawerk, “The Positive Theory of Capital and Its Critics, Part III,” p. 127. See also idem, Further Essays on Capital and Interest, pp. 4–10.
- 32. As Hayek states:
It is frequently supposed that all increases in the quantity of capital per head ... must mean that some commodities will now be produced by longer processes than before. But so long as the processes used in different industries are of different lengths, this is by no means a necessary consequence. ... If input is transferred from industries using shorter processes to industries using longer processes, there will be no change in the length of the period of production in any industry, nor any change in the methods of production of any particular commodity, but merely an increase in the periods for which particular units of input are invested. The significance of these changes in the investment periods of particular units of input will, however, be exactly the same as it would be if they were the consequence of a change in the length of particular processes of production. (Hayek, Pure Theory of Capital, pp. 77–78)Also see Hayek, Prices and Production, p. 77, and Böhm-Bawerk, Further Essays on Capital and Interest, pp. 57–71.