vol. 12, nos. 2/3 (June-September), 2002 


Capital, Interest, and Professor Kirzner

Roger W. Garrison

1. Introduction
Inspired by the teachings and writings of Ludwig von Mises, Israel Kirzner has devoted his career to the exploration and development of Misesian themes. Mises's very conception of the nature of economic science—as the study of market processes anchored to a theory of human action—underlies the whole of Professor Kirzner's work. A scholar of the first order in his own right, Kirzner has pushed beyond Mises in several instances to the point of making the Misesian theories his own. We owe our understanding of competition and entrepreneurship and of capital and interest as much to Kirzner himself as to his mentor.
          In the half-dozen sections to follow, the focus is on the Misesian/Kirznerian theory of capital and interest with special attention to the pure time preference theory. It seems fitting that this paper draws liberally from earlier writings of my own, going back almost to the year (1974) when I first heard Professor Kirzner lecture at South Royalton, Vermont—at the conference now taken to mark the beginning of the modern Austrian resurgence. Section 2 draws from a 1997 review article, (1) which deals with Kirzner's 1996 Essays on Capital and interest, a book that compiles more than a quarter of a century of his theorizing. Several of the remaining sections draw to one extent or another from a formal comment on Peter Lewin's "Rothbard and Mises on Interest: An Exercise in Theoretical Purity." (2) Lewin presented this paper at the Southern Economic Association meetings in New Orleans in 1995. My comment was never published in any form, but the discussion below (Section 6) on the parallels between a time-preference theory of interest and a leisure-preference theory of wages is adapted from it. Also included in the present paper (and in the 1995 comment) is an edited excerpt from an ancient unpublished paper, "Reflections on Misesian Time Preference." which was presented in 1975 at the Conference on Austrian Economics at the University of Hartford, Hartford, Connecticut. (3) Comparing Mises's interest-rate theory to that of Frank Knight, this old conference paper and Section 5 below suggest that the Mises/Knight contrast in the reckoning of time preference is analogous to the Kelvin/Fahrenheit contrast in the reckoning of temperature. Although the pairing of Mises and Kelvin in this regard resonated among the participants, the idea never found its way into print until now. 

2. Capital and Interest (1966-1996)
Eugen von Böhm-Bawerk, whose name has come to be virtually synonymous with "roundaboutness" (of capital-using production processes), penned the original Austrian perspective on capital and interest. (4) He wrote three volumes (History and Critique, Positive Theory, and Further Essays) over a span of a quarter of a century (1884-1909). In 1959 the twelve-hundred-plus pages of Capital and Interest were translated into English by Hans Sennholz and George Huncke. In his spirited review of the new translation, Mises described this "monumental work" as "the most eminent contribution to modern economic theory." (5) Mises went so far as to suggest—as only Mises could—that no citizen who takes his civic duties seriously should exercise his right to vote until he has read Böhm-Bawerk! 
          With a lag of more than eight decades, Kirzner's writings echo Bohm-Bawerk's. The years over which Kirzner gave attention to the economics of a capital-using economy are at par with those of his early Austrian predecessor, spanning from his 1966 Essay on Capital to his 1996 Essays on Capital and Interest: An Austrian Perspective. (6) There is no intent here to insert these volumes between the voter and the voting booth. However, the position that these writings occupy on the Austro-neoclassical landscape is an eminently imposing one—so imposing as to warrant our issuing a Mises-style taboo, not to voters, but to all economists who adopt the Austrian perspective. But first we must put into perspective Kirzner's insights into the nature of capital and interest, focusing for this purpose on his 1996 Essays and saving our taboo for the final section.
          These Essays provide a virtual history—and pre-history—of the modern Austrian resurgence, which dates to that 1974 South Royalton conference. The four parts of Kirzner's 1966 book (on "Unfinished Plans," "Stocks and Flows," "Capital and Waiting," and "Measuring Capital") read like the work of a lone scholar trying—succeeding in most instances—to satisfy himself. This early book, possibly the most underrated of all his contributions, appears anew as the longest of the 1996 Essays
          In late 1974, Kirzner presented a paper titled "Ludwig von Mises and the Theory of Capital and Interest" in a special symposium at the Southern Economic Association meetings in Atlanta. At that time, a year after Mises's death and a few months after South Royalton, there was a small but eager audience for the Austrian perspective. Kirzner shows how Mises's theory differs from Böhm-Bawerk's and how it compares favorably to the theories of J. B. Clark and F. H. Knight. In part a stocktaking, in part a research agenda for himself and for others, this paper was first published in 1976 along with other papers at the SEA symposium (7) and published again in 1979 along with other writings by Kirzner. (8) Its appearance in the 1996 volume provides a perfect segue between his early work on the theory of capital and his later work on the theory of interest. 
          The final essay, "The Pure Time-Preference Theory of Interest," first published in 1993 (9), is clearly the work of a well-seasoned scholar. Kirzner's good scholarship shines though in all his writings, but here we see him as a veteran of many symposia and conferences complete with their unyielding question-and-answer sessions. His work had attracted a growing and challenging audience. He responds to criticisms as if he has heard those criticisms many times before and in many different forms—because he has. The exchanges with allies and critics over the years have allowed him to clarify his own ideas and to offer them in the most rhetorically effective ways. 
          Ralph W. Pfouts, who offered a generally favorable assessment of the 1966 Essay in the American Economic Review, suggested that the Austrian perspective is not "the magic wand that makes all mysteries disappear." (10) Kirzner, with his 1996 Essays, and especially with his essay on the theory of interest, shows that it is very nearly just that. Interest is always and everywhere a matter of time preferences. The primordial preference for the sooner over the later is the basis for a unified treatment of intertemporal exchange—whether the exchange is with nature in a Robinson Crusoe setting or with other economic actors and whether or not it involves the use of capital. Tracing out the consequences of the systemic discounting of the future provides us with a Copernican account of an economic phenomenon that otherwise would have to be explained by what we might justifiably call Ptolemaic interest-rate theory. Capital-productivity theories and waiting-as-a-factor theories appear strained, partial, and oblique in comparison to the pure time-preference theory—that magic wand so skillfully wielded by Professor Kirzner. 
          If interest is to be understood in terms of time preferences, capital is to be understood in terms of multiperiod plans. An Austrian subjectivist perspective on capital features the plans of individual entrepreneurs—plans that are subject to revision as the attempts to carry them out reveal conflicts with reality and with the unfolding plans of other entrepreneurs. As Kirzner demonstrates time and again, the forward-looking, plan-oriented account of a capital-using economy wins out over the alternative accounts that focus on some isolated slice of time or on the physical productivity of the produced means of production.

3. Sooner or Later; More or Less
On the key questions of interest (Is the interest rate necessarily positive? How is the market rate of interest determined?) Kirzner is a thoroughgoing Misesian: The phenomenon of interest—the very existence of it—is virtually synonymous with the notion of time preference. The particular rate of interest is a reflection of a systematic time discount—a discount that manifests itself overtly in the market for loanable funds and that permeates the entire market system in the form of actual and implied intertemporal price relationships. However complex the relationships may be between the spot price of copper and the price of a copper-futures contract or between the spot price of grapes and the price of (future) wine, the market registers a systematic and premordial preference for sooner as compared to later.
          In the Austrian tradition, the notion that sooner is preferred to later has the same status as the more conventional and widely accepted notion that more is preferred to less. Both notions are applicable only under ceteris paribus conditions, and exceptions abound, as when a specific quantity (one lump of sugar and not two in your coffee) or specific timing (coffee in the morning and not tonight) is an integral part of what gives the good its subjective value or when the object being valued is actually a "bad" instead of a "good" (coffee as served in England).
          In his most recent account of Mises's theory, Kirzner deals with these sorts of exceptions in terms of an example provided by Mises himself: The fact that people prefer ice next summer to ice this winter provides no evidence against the time preference theory of interest. Rather, in this and many other cases, the timing of the good's availability should be considered as an integral part of the definition of the good. (11)
          Still, Kirzner notes, Mises's theory of capital and interest, especially when compared to the alternative theories advanced by Irving Fisher or Frank Knight, presents a puzzle to the modern mainstream and even to fellow Austrians. (12) What accounts for the radical difference between the Austrian theory and the neoclassical theory? Recognizing that the seemingly similar terminology belies the radical difference, Kirzner hints at a resolution to the puzzle: "For Mises, time preference refers not to dates [as it does in modern mathematical formulations], but to future distances from the moment of evaluation." (13) The resolution he has in mind may well be consistent with—and elaborated by—the arguments in the following two sections.

4. Time Preference Theory: Formulation and Application
Peter Lewin reminds us that in Mises's Human Action and in its 1940 predecessor, Nationalökonomie, there is some variation in his statements as to what in particular is preferred in the present (14) The terms "consumption," "satisfaction," "gratification," and "enjoyment" are each used separately or in combination with no apparent effect on the intended thrust of the statement. While Lewin sees the variations on a theme as indication of "potential inconsistencies," (15) there is a strong basis for seeing the theme itself as giving unity to Mises's various arguments. Each of these terms refers to some action or to the hoped-for result of some action. In keeping with Mises's "praxeology" (which means, literally, "action logic"), the most basic and most general proposition that expresses Misesian time preference is one that refers to action per se rather than some particular category of action, such as consumption, or some anticipated result of action, such as satisfaction. In this respect, the heading of the section containing Mises's various renditions of the time-preference theorem may be more instructive than any or all of the various statements quoted by Lewin. The section heading reads: "Time Preference As an Essential Requisite for Action." (16)
          That action per se is the ultimate basis for Mises's time preference theory is even more evident in Mises's earlier reckoning: "Whoever uses or consumes anything, whoever seeks by acting to relieve to a greater or lesser extent a felt uneasiness is expressing a preference for an earlier over a later satisfaction." (17) The focus here is clearly on "acting." And even the phrase "uses or consumes" can be taken as all inclusive; it excludes no conceivable (purposeful) action. Whoever uses or consumes anything simply means whoever behaves purposefully or simply whoever acts. By acting now the individual reveals that such is preferred to deferring action. And since all acts are directed at achieving satisfaction (or gratification or enjoyment), the individual reveals a preference for consumption in the nearer future to consumption in the more remote future. This is the Misesian time preference theorem. It is formulated in terms of action and then applied (variously) to the anticipated results of the action.
          Mises's formulation differs fundamentally from those of Fisher and Knight. While Fisher's formulation has come to dominate modern textbooks, Knight's formulation, similar in all relevant respects to Fisher's, allows for the most direct and telling comparison with Mises's. The salient difference between Knight and Mises concerns the "reference pattern" of consumption that serves as a basis for gauging time preferences. According to Mises, any action, including the act of consuming, serves as evidence of (positive) time preference. The absence of time preference, or zero time preference, could be associated only with inaction. Apart from this praxeological benchmark, however, no particular temporal pattern of consumption would be inconsistent with the general preference for sooner over later.
          For Knight, the reference pattern is not a pattern of zero consumption (or zero action) but one of uniform consumption, a notion found in Knight's earliest writing: "[A] zero time preference obviously means a uniform distribution (of consumption) over time." (18) This stating of the "obvious" underlies Knight's simple taxonomy of time preferences. Having established a reference pattern, actual time preferences can be categorized as positive, zero, or negative. Individuals who prefer a uniform pattern of consumption over all other possible patterns exhibit, by definition, a zero time preference. Individuals who prefer a relatively high rate of consumption in the nearer future at the expense of a relatively low rate in the more remote future are characterized as having positive time preference, while those who prefer a relatively low rate of consumption in the near future coupled with a relatively high rate in the more remote future are characterized as having a negative time preference.

5. Mises is to Knight as Kelvin is to Fahrenheit
Which of the two formulations, Knight's or Mises's, is the more substantive or meaningful? Ultimately, Knight's criticism of the Misesian formulation was not that it is wrong or illogical but that it is "without meaning," that a meaningful preference of present over future consumption would have to be measured with respect to a uniform rate of consumption over time. (19) But Mises, having been critical of Bohm-Bawerk's psychology-based theory, was intent on basing his own theory squarely on the choices and actions of individuals. This essential praxeological grounding is the Austrian criterion for meaningfulness, and it is in this respect that Knight's formulation, based as it is on an arbitrary (though Knight said "obvious") reference pattern of consumption, falls short.
          It can be hazardous to point out parallels between developments in the social sciences and developments in the physical sciences—precisely because of the inapplicability of praxeology to physical phenomena. But it is instructive, particularly in regard to the question of meaning, to liken the two different "measures" of time preference to two different measures of temperature. References to temperature scales are frequently made in discussions of the cardinal and ordinal measures of utility—the cardinal measure, which admits of all linear transformations, being likened to the measure of temperature. For our purposes, however, the focus is not on the transformability but rather on the reference point, or zero point, that serves as a basis for the measure.
          The measure of time preference in the Misesian formulation corresponds to the measure of temperature in degrees Kelvin. William Kelvin based his scale on the relationship between temperature and (molecular) motion in the same sense that Mises based his interest-rate theory on the relationship between time preference and (purposive) action. That is, a total absence of molecular motion corresponds to a temperature of zero degrees Kelvin. Any motion whatever is evidence of a temperature greater than zero. Similarly, any purposive action whatever is evidence of non-zero time preference, while negative time preference is as unthinkable as negative action. Just as a positive temperature is a universal fact in a world of motion, so positive time preference is a universal fact in a world of action. The impossibility of negative temperature and negative time preference does not render the formulations meaningless. Quite to the contrary, these impossibilities derive from the fact that temperature is based on motion, time preference on action.
          The measure of time preference in the Knightian formulation corresponds to the measure of temperature in degrees Fahrenheit. The construction of the respective scales for both Knight and for Fahrenheit began not from an awareness of any particular theoretical relationship, but from the awareness that "there is something there to be measured." In order to establish a basis for measurement, a zero point had to be chosen. There are conflicting reports about just how Gabriel Fahrenheit chose his zero point (Amsterdam's record low for the winter of 1714?), but the choice is generally recognized to be an arbitrary one. A temperature of zero degrees Fahrenheit has no particular significance.
          Superficially, the zero point of the Knightian time-preference scale seems to have at least some significance. As noted above, Knight considered the uniform rate of consumption to be the obvious choice for a reference pattern. But upon further reflection, the arbitrariness of Knight's reference pattern becomes apparent. First, outside the artificial context of a one-good world, the notion of a uniform rate of consumption of a heterogeneous collection of goods and services has no unambiguous meaning. (20) Second, the idea of a uniform rate fails to link up in any theoretical way with the actual choices and actions of market participants. It is neither an objective nor a constraint. And finally, except for simple temporal patterns of consumption (monotonically increasing or monotonically decreasing over time), the ranking of patterns in accordance with the implied time preferences or even the identification of a pattern as involving positive, zero, or negative time preference may not be possible. Here the switching and re-switching of the Cambridge capital controversy come to haunt.
          Having chosen an arbitrary zero point, both positive and negative values of Fahrenheit temperature and of Knightian time preference are (equally?) possible. Whether actual temperatures or time preferences are positive or negative is simply an empirical question. It is the mere fact that this question can be answered sometimes one way and sometimes another that seems to give meaning to both the question and the answer. But in relation to individual choices and actions, the contrasting answers (negative and positive) are as meaningless as their dividing line is arbitrary.
          The Farhenheit and Knight scales give no hint of the lower limits to the things being measured, as do the Kelvin and Mises reckonings. The fact that temperatures and time preferences actually observed cannot drop below absolute zero—and in most applications do not even approach absolute zero—does not deprive Kelvin's and Mises's theory of meaning. The meanings, rather, derive directly from these essential facts of physical and economic reality.

6. Time Preference and Leisure Preference
Pre-Austrian economists commonly divided the factors of production into the classical trilogy: land, labor, and capital. Mises's insights into the nature of capital suggest that a more appropriate categorization would be land, labor, and time—with land and labor (or, equivalently, nonhuman and human factors) designated as the "original" or "primary" factors of production. (21) In the imaginary construct of the evenly rotating economy, where there are no profits and losses, all income can be either attributed directly to these original factors of production or, using Kirzner's phraseology, "swept back" to them through the market's process of value imputation and time discounting. Drawing from the Austrian trilogy, we can say that it is labor and time that are the most inextricable tied up with human action. In fact, praxeology as conceived by Mises is defined as logic in the context of purposive action and the passage of time. (22)
          Given this construction articulated by Mises and ably defended by Kirzner, we might wonder why there isn't an analytical symmetry between the theory of interest and the theory of wages. We have a time-preference theory of interest. Why do we not have a leisure-preference theory of wages. Resources which fall into the category of land are converted into consumable output with the aid of labor and time—or, to emphasize the notion of process, through working and waiting. If working and waiting were ends rather than means, then people would not have to be paid to perform these functions. It is precisely because of their status as means that, in a market setting, people do have to be paid a positive wage rate and a positive rate of interest.
          So, why did Mises not deduce from the action axiom that the (originary) wage rate is necessarily positive? It would seem that the case for a primordial leisure preference and a positive wage rate is a direct parallel to—and is as strong as—the case for a primordial time preference and a positive rate of interest: According to the action axiom, people employ means (labor) to achieve ends. If labor, as contrasted to the forgone leisure, were itself a consumption activity, an end, then the whole means-ends framework would lose its relevance. In reality, however, labor is a means. Mises would have had to deal with the apparent exceptions in the forms of labor of love and escape from boredom, but these exceptions would strengthen rather than weaken the parallels.
          Instead of maintaining the parallel in this way, Mises started with the action axiom and then added the apparently supplementary proposition that "Labor is a means, not an end in itself." (23) Posing our question about possible parallels an alternative way, we might ask: Why, then, did Mises not add a second supplementary proposition to the effect that people put a premium on the present? The parallel could have been maintained in this alternative way. With this construction, the rest of the Misesian system would remain intact—and the lingering "puzzle" about Mises time-preference theory of interest would never have confronted its critics and admirers.
          It is worth noting that Alfred Marshall did maintain a parallel between the payment of interest and the payment of wages. According to him, our understanding that "waiting" must be rewarded rather than penalized, i.e., that the rate of interest is positive, is (at least on an intuitive level) a close parallel to our understanding that working has to be rewarded rather than penalized. Marshall claims that a state of affairs in which the interest rate is negative and hence waiting penalized is conceivable. He immediately puts this claim into perspective by pointing out that "it is also conceivable, and almost equally probable, that people may be so anxious to work that they will undergo some penalty as a condition of obtaining leave to work"(24) He mentions prisoners eager to be assigned to unpaid work details. Marshall argues with his characteristic rhetorical effectiveness and maintains a perfect parallel, in this respect, between interest and wages: Under exceptional conditions, people might be willing to pay to work or pay to wait, but under more normal conditions, they would have to be paid to work and be paid to wait.
          Mises actually offers an answer to the question: "Why no leisure-preference theory of wages to complement the time-preference theory of interest?" He argues—though questionably, as pointed out below—that a zero or negative time preference confronts us with a contradiction, while a zero or negative leisure preference does not. A primordial preference for achieving ends later rather than sooner (or an indifference as between sooner and later) is, in Mises's construction, contradictory to the action axiom. But no analogous contradiction arises in the case of a negative or zero leisure preference.
          The possibility and implications of a zero leisure preference can be pieced together from his Chapter II "The Epistemological Problems of the Science of Human Action" and his Chapter VII "Action within the World." In a section of Chapter II entitled "The Procedure of Economics," Mises writes:

The disutility of labor is not of a categorical and a prioristic character. We can without contradiction think of a world in which labor does not cause uneasiness, and we can depict the state of affairs prevailing in such a world. But the real world is conditioned by the disutiltiy of labor. Only theorems based on the assumption that labor is a source of uneasiness are applicable for the comprehension of what is going on in this world. (p. 65)
          In a section of Chapter VII entitled "Human Labor as a Means," Mises returns to the issue of leisure preferences:
In a world in which labor is economized only on account of its being available in a quantity insufficient to attain all ends for which it can be used as a means, the supply of labor available would be equal to the whole quantity of labor which all men together are able to expend. In such a world everybody would be eager to work until he had completely exhausted his momentary capacity to work. The time which is not required for recreation and restoration of the capacity to work used up by previous working, would be entirely devoted to work (p. 131).
This passage seems to suggest that in the absence of a leisure preference, the Misesian means-ends framework, so closely associated with the action axiom, degenerates into a Knightian circular flow, in which individuals work to eat and eat to work. It is devoid of any genuine acts of consumption that are not at the same time a prerequisite for expending still more labor. But Mises' praxeology is based firmly on a means-ends framework that exhibits an essential temporal linearity: the employment of means precedes and is distinct from the corresponding achievement of ends. To recognize the means-ends linearity as essential is to reject the notion of the Knightian circular flow. Kirzner's criticism of the Clark-Knight synchronization view is to the point here. (25) The necessity of avoiding the eclipse of the means-ends framework that the circular flow entails would seem to give an a prioristic character to the notions of a (positive) leisure preference, which recognizes that means and ends are distinct and a (positive) time preference, which recognizes that means precede ends.
          Further, a leisure-preference theory of wages, paralleling the time-preference theory of interest, would concern itself with the existence of positive wage rate. Issues of wage-rate determination as actually brought about by the market process could admit of all sorts of considerations, such as specific complementary relationships between leisure time and consumption goods. Insights into the allocation of time as spelled out by Gary Becker and others would fit into such an understanding of wages at both a theoretical and an applied level.

7. Concluding Observations and a Mises-Style Taboo
I have not actually constructed—let alone recommended—a leisure-preference theory of wages. Rather, I have hinted how such a theory might be constructed and suggested that virtually the same consideration (the need to avoid contradicting or eclipsing the action axiom or the temporally linear mean-ends framework) that led Mises to construct a time-preference theory of interest might well have led him to construct a leisure-preference theory of wages.
          But there is a greater point. Replacing Mises's supplementary proposition that leisure is a consumer good with a more thoroughgoing praxeological treatment of labor and wages is not apt to have much of an effect on the rest of the Misesian system. Alternatively, replacing Mises's thoroughgoing praxeological treatment of waiting and interest with an supplementary proposition that individuals tend to discount the future is not apt to have much of an effect on the rest of the Misesian system.
          The robustness of Austrian economics with respect to its epistomological underpinnings has been demonstrated in a more general sense by Murray Rothbard. In an early article, Rothbard questioned the epistemological status of Mises's action axiom. Arguing "In Defense of Extreme Apriorism," (26) Rothbard took an Aristotilian stance, claiming that our understanding that individuals engage in purposeful action—that they employ means to achieve ends—is ultimately a broad empirical observation, which includes, of course, introspection. Mises, by contrast, had taken a Kantian stance, claiming that human action was an a priori category.
          Modern practitioners of Austrian economics can hardly help but notice how little this basic disagreement over the underlying epistemology affected the development and application of economic theory. My discussion of the alternative roles of praxeological truths and supplementary propositions suggests that Austrian economics may also be robust with respect to the differing treatments of the issues of interest and wages—and that recognizing this robustness might be a way of putting to rest the "puzzle" of Mises time-preference theory of interest.
          Whatever the ultimate resolution of the puzzles with which theories of capital and interest confront us, the writings of Professor Kirzner are bound to figure in importantly. In my 1997 review article, I issued a Mises-style taboo specifically aimed at Austrian macroeconomists: No Austrian economist who takes his subjectivism seriously should erect macroeconomic models until his has read Kirzner's Essays. Here, the taboo can be reissued in a more general form: No future academic who takes the issues of capital and interest seriously should attempt to make his own contribution to that literature until he has mastered Professor Kirzner's.


Böhm-Bawerk, E. von (1959) Capital and Interest, translated by George D. Huncke and Hans F. Sennholz, 3 Vols. South Holland, IL: Libertarian Press.

Dolan, E. (1976) Foundations of Modern Austrian Economics, Kansas City: Sheed and Ward.

Garrison, R. (1975) "Reflections on Misesian Time Preference," unpublished paper. Presented at the Austrian Economics Conference, University of Hartford, Hartford, Connecticut.

Garrison, R. (1990) "Austrian Capital Theory: the Early Controversies," History of Political Economy, supplement to Vol. 22, pp. 133-154. Published in hardcover as B. Caldwell, ed., Carl Menger and his Legacy in Economics, Durham, NC: Duke University Press.

Garrison, R. (1995) "The Robustness of the Mises-Rothbard Theory of Interest: Comment on Lewin," unpublished comment. Presented at the Southern Economic Association meetings, New Orleans.

Garrison, R. (1997) "The Undiscountable Professor Kirzner," The Freeman, Vol. 47, No 8 (August), pp. 511-13.

Greaves, P. (1974)Mises Made Easier, Dobbs Ferry, NY: Free Market Books.

Herbener, J., ed. (1993)The Meaning of Ludwig von Mises, Norwell, MA: Kluwer Academic Publishers.

Mises, L. von (1959) "Capital and Interest: Eugen von Böhm-Bawerk and the Discriminating Reader," Freeman, Vol. 9, No 8 (August).

Mises, L. von (1966) Human Action: A Treatise on Economics, Chicago: Henry Regnery Company.

Kirzner, I. (1966)An Essay on Capital. New York: Augustus M. Kelley.

Kirzner, I. (1979)Perception, Opportunity and Profit, Chicago: Chicago University Press.

Kirzner, I. (1996)Essays on Capital and Interest: An Austrian Perspective. Brookfield, MA: Edward Elgar.

Kirzner, I. (2001) Ludwig von Mises: The Man and His Economics, Wilmington, DE: ISI Books.

Knight, F. (1921) Risk, Uncertainty, and Profit, Boston: Houghton Mifflin, Co.

Knight, F. (1934) "Capital, Time and the Interest Rate," Economica, New Series, Vol. 1 (August), pp. 257-286.

Knight, F. (1941) "Professor Mises and the Theory of Capital," Economica, Vol. 8 (November), pp. 409-427.

Lewin, P. (1997) "Rothbard and Mises on Interest: An Exercise in Theoretical Purity," Journal of the History of Economic Thought, Vol. 19 (Spring).

Marshall, A. (1920) Principles of Economics, 8th ed., London: Macmillan and Co.

Moss, L. (1976) The Economics of Ludwig von Mises: A Critical Reappraisal, Kansas City: Sheed and Ward.

Pfouts, R, (1968) Review of An Essay on Capital, American Economic Review, Vol. 58, No 1 (March).

Rothbard, M. (1962)Man, Economy, and State: A Treatise on Economic Principles, 2 Vols. Los Angeles: Nash Publishing Co.


1. Garrison-1997.

2. Lewin-1997

3. Garrison-1975. The 1975 conference at Hartford was the follow-on conference to South Royalton. At South Royalton, participants listened to lectures by Israel Kirzner, Ludwig Lachmann and Murray Rothbard. Other senior Austrian economists present included Henry Hazlitt and Bill Hutt. The lectures were published in Dolan-1976. At the University of Hartford, the South Royalton listeners presented their own papers, which were then discussed by the seniors. Senior discussants included two of the lecturers of the year before (Kirzner and Rothbard) plus Friedrich Hayek Bill Hutt, Emil Kauder and Leland Yeager. No conference volume was ever published.

4. Böhm-Bawerk-1959.

5. Mises-1959, p. 52.

6. Kirzner-1966 and Kirzner-1996. 

7. Moss-1976.

8. Kirzner-1979.

9. Herbener-1993.

10. Pfouts-1968, p. 98. 

11. Kirzner-2001, p. 158; Mises-1966, pp. 489-490.

12. Kirzner-2001, p. 149.

13. Ibid., p. 159.

14. Lewin-1997, pp. 146-150.

15. Ibid., p. 148.

16. Mises-1966, p. 483.

17. Lewin-1997, p. 150. Lewin is quoting from a translation of a section of Nationalökonomie that Mises specifically identifies as relevant (Mises-1966, p. 488). The translation of the entire section, titled "A Critique of Bohm-Bawerk's Reasoning in Support of His Time Preference Theory," is included in Greaves-1974, pp. 150-157.

18. Knight-1921, p. 132.

19. Knight-1941, pp. 412-13

20. Kirzner-1996, pp. 67-69. Even Knight himself recognizes the element of arbitrariness when he shifts the focus from a temporal pattern of consumption to the choice during each income period between consuming and accumulating wealth: "The only possible basis for interest theory is simply to assume some indifference curve between current income as consumption and as increase in wealth" (Knight-1934, p. 272n).

21. Mises-1966, pp. 637-640.

22. Mises-1966, pp. 99-100.

23. Mises-1966, p. 131. Murray Rothbard is more explicit about adding the supplementary proposition. His labor economics is "derived from the implications of the action axiom and the assumption of leisure as a consumers' good" (Rothbard-1962, p. 40). In a related footnote, he continues: "This [proposition that leisure is a desirable good] is the first proposition in this chapter that has not been deduced from the axiom of action. It is a subsidiary assumption based on empirical observation of actual human behavior. It is not deducible from human action because its contrary is conceivable, although not generally existing" (p. 437, n.27). 

24. Marshall-1920, p. 232.

25. Kirzner-1996, pp. 75-77. See also, Garrison-1990, pp. 141-44.

26. Rothbard-1957.