CHAPTER 12 Human Capital, the Nature of Knowledge, and the Value of Knowledge Introduction: The Implicit Tension

To create knowledge is not to fathom or command it or own it. All knowledge is born, and forever dwells, behind a veil that is never shed. After their birth, bodies of knowledge remain forever unfathomed and unfathomable. They remain forever pregnant with consequences that are unintended and cannot be anticipated.

(Bartley 1990:32)

Having reviewed some of the implications of the human capital approach, we may now return to an unfinished theme started in the introduction to the previous chapter. Human capital refers to the perceived value of accumulated knowledge in various contexts. Yet, as has been noted at various points in this work, knowledge is a very difficult phenomenon to analyze. The term “knowledge” encompasses an enormous amount. It underlies every action, every thought. It is riddled with paradoxes and self-references. On the one hand, it is certainly clear, as suggested above, that an essential aspect of economic development and progress lies in the existence and growth of vast bodies of knowledge. On the other hand, although knowledge is often consciously and purposefully acquired (“produced”), in recognition of its necessary role as a facilitator of growth and development, there are aspects of knowledge that cannot be purposefully acquired because they cannot be perceived ahead of time. There is a real sense in which “knowledge of knowledge” is an impossibility. In this section we attempt a resolution of this implicit tension. How can knowledge at once be consciously produced and an unfathomable phenomenon?

Part of the resolution lies in recognizing that the phenomenon we refer to as “knowledge” is jam-packed with many diverse subphenomena. There are many different dimensions that emerge when knowledge is “unpacked.” Along some of these dimensions knowledge can be shown to be fallible, unfathomable, inexpressible, and tacit.

Knowledge is Fallible

One dimension along which one can think about knowledge is its truthfulness, its fallibility. Following the work of Karl Popper, it is widely (though perhaps not universally) held that knowledge is and must always be fallible. Today’s valid theories may be refuted tomorrow. Although he was talking primarily about scientific knowledge, Popper’s work can be (and has been) applied to knowledge in general with science seen as a metaphor for all knowledge.

My interest is not merely in the theory of scientific knowledge, but rather in the theory of knowledge in general. Yet the study of the growth of scientific knowledge is, I believe, the most fruitful way of studying the growth of knowledge in general. For the growth of scientific knowledge may be said to be the growth of ordinary human knowledge writ large.

(Popper 1989:216)

All knowledge, according to Popper, is accumulated by trial-and-error elimination of error. In scientific endeavors the elimination of error is more conscious than in everyday life, but the process is essentially the same. An implication is that all bits of knowledge, all theories, are at base “conjectures” held with a greater or lesser degree of confidence. And conjectures can never be positively justified or verified, although they can (ideally) be refuted. All knowledge is thus tentative to some degree. There can be no ultimate and absolute knowledge. Knowledge is always fallible.¹

A model in which human capital is purposefully accumulated in response to the accurate perception of its benefits is thus, if we follow Popper, an inadequate depiction of the truth. More importantly, it would seem to be inappropriate as an aid to understanding how and why knowledge is accumulated. For such a model presupposes that a particular and certain (infallible) stock of knowledge exists (or could be produced) and is available to be replicated and distributed. Yet if, on the contrary, all knowledge is tentative, how is it that people are able to make decisions regarding the acquisition of something whose properties are inherently and radically (as opposed to probabilistically) uncertain?

Knowledge is Unfathomable

A related but distinct and more subtle dimension that can be used to think about knowledge is its understandability, its fathomability. The unfathomability of knowledge transcends its fallibility. “Understanding” is related to but is different from knowledge. It is not easy to explain what understanding is (one suspects, to anticipate, that there is an element of tacitness to it). Various aspects of cognition defy verbal expression. Bartley hazards:

To understand what a theory asserts will require, among other things, understanding its logical implications, its content and its context, what historical problem situations it addresses, what problems it can solve, and how it interconnects logically with other theories. Among these preconditions for understanding, the hardest to understand (sic), and to characterize and convey adequately, is the idea of content, even though, or perhaps because, we all have an intuitive idea of what must be involved.

(Bartley 1990:34)

In exploring ways to characterize content, Bartley defines two related concepts, logical content and informative content. Together they make up the objective content of a theory or body of knowledge.

1. The logical content of a theory (or a body of knowledge) is all the non-tautological consequences that can be logically derived from it. All statements that follow logically from it as well as further implications that result from combining this theory with other theories proposed or assumed are part of its logical content. “It is well known that the logical content of any theory must be infinite” (ibid.:35).

2. The informative content of a theory is the set of statements that are incompatible with it. It is also infinite. The more a theory forbids the more it says. To say that it will either rain today or it will not is not very informative and does not exclude very much of interest.

The logical and informative contents are distinct but related. The elements of the informative content of a theory stand in a one-to-one correspondence to the elements of the logical content of a theory. This can be seen by realizing that for every element in either set there is, in the other set, its negation. “Hence the logical and informative content of a theory increase and decrease together” (ibid.:36). To explain the infinite size of the informative content of any theoretical system, and therefore of its logical content, Bartley uses the well-known example of the relationship between Newton’s and Einstein’s theories of gravitation.

Call Newton’s theory N and Einstein’s E. Any statement or theory that is incompatible with N will belong to the informative content of N, and similarly, any statement or theory that is incompatible with E will belong to the informative content of E. Since Newton’s and Einstein’s theories are mutually incompatible, each belongs to the informative content of the other. [But] Einstein’s theory is not simply incompatible with Newton’s, it is historically connected with it in the important sense of having superseded it. It has superseded it in the sense that Newton’s theory is inadequate to the facts, and that Einstein’s theory appears to come closer to the truth. This illustrates that any new theory that supersedes a reigning theory . . . has to belong to the informative content of the superseded theory. [Thus we may say] any existing theory includes in its informative content any theory that will eventually supersede it, and in its logical content the denial of any theory that will eventually supersede it.

(ibid.:36–37)

This should be sufficient to illustrate the sense in which it can be claimed that knowledge is a product not fully known (or knowable) to its creator. Theoretical systems have objective content beyond the accessibility of any individual mind. The full extent of a theory can never be fathomed. The content of Newton’s theory is not identical with Newton’s thoughts or opinions about it.

Bartley points out that the meaning and relevance of any theoretical system, in the sense of the common understanding of it at any particular time, is in an important way a historical matter, as well as partly a matter of logic. The significance of a theory “depends on what has been discovered, at a certain time, in the light of the prevailing situation, about the theory’s content; it is, as it were, a projection of this historical problem situation upon the logical content of the theory” (Popper 1976:28). Bartley calls this projection the “accessed slice” of the objective content of a theory.

The historical relevance of a theory suggests that its meaning has little to do with the particular words and terms used in it. Almost any statement may be rendered in different terms without change in meaning. In fact, meaning and objective content are not identical. The meaning, relevance, and common understanding, and thus also the economic value, of a theory (or a body of knowledge) “shifts as we uncover more of, or gain access to a larger slice of, its objective logical and informative content” (Bartley 1990:38).

Knowledge existing at any point in time is reflected in the “knowledge objects” to which it gives rise. Books, tapes, and computer programs are obvious examples. But in an important sense, a pharmaceutical drug, a machine, or a bunch of keys are also “knowledge objects.”

What is crucial [in our present context] about an item of objective knowledge—a book, or a pill, for instance—is its potential for being understood, or being utilized in some way that has not yet been imagined, a potential that may exist without ever being realized.

(ibid.:44)

Bartley is here (unconsciously) pointing toward the essential and sequential complementarity of knowledge elements and an important and unnoted implication of this. The value of any item of knowledge depends crucially on what is already known. Expectations of the prospective value of any knowledge object depend on what has already been learnt about it in existing applications. So, for example, the valuable use of many medical drugs arises out of imaginative conjectures and serendipitous extensions to applications other than those for which they were created (like the use of blood pressure medicine to aid in the growth of hair). So the creation of the objects or, indeed, the teaching of any theory, could not have been completely connected to and could not have taken full account of their prospective uses (because of the impossibility of acquiring the knowledge on which those uses will depend); although retrospectively one can (in principle) trace out the changing influences on their values. Since knowledge as a product is sequentially complementary, and since future knowledge is unavailable, the productivity of knowledge can never really be known ahead of time. One may again question the sense in which individual motivations can be captured in observed (calculated) rates of return to the accumulation of human capital.

The fact that knowledge is unfathomable, that its objective content is beyond the reach of any human mind, is another aspect of the difference between information and knowledge. As we have noted, information may be traded although knowledge may not. We see now that information is pregnant with meaning but that meaning may not always be, and frequently will not be, available to any individual, because its objective content is infinite. It is the content of knowledge that, in a sense, extends beyond the individual knower, not the knowledge itself, which is subjective and contextual. What is valued in the marketplace is the accessed slice of any particular body of knowledge.

Demand for an item of knowledge, to the extent that it is available in the marketplace, will be based on dispersed subjective understandings or estimates of, or preferences for, its accessed slice, and on speculations about its potentialities. It will not, and cannot be based on its objective—unfathomable and autonomous—logical and informative content. This is not available or readily accessible for anyone to value.

(Bartley 1990:48)

Knowledge is Tacit

Knowledge can also be thought about in terms of its explicitness. It has been recently acknowledged in a number of different contexts—in the philosophy of science (Polanyi 1958),² economics and political economy (Hayek 1979), and even recently in the theory of business management and organization (Nonaka and Takeuchi 1995)—that knowledge is, at least in large part, tacit in nature. This distinction between tacit and explicit knowledge is related to the dimensions of fallibility and fathomability. Our acceptance of and our understanding of bodies of knowledge rest necessarily on tacit (unarticulated and frequently inarticulable) rules, conventions, and inclinations. The question raised here is, if tacit knowledge is, as it must be, part of human capital, how is it valued and how is it produced? If we do not even “know” what it really is, how is it that we are able to “teach” it? If it cannot be taught, how is it transmitted and acquired?

A particularly relevant and well-known application of the importance of tacit knowledge emerged from the “second round” of the socialist calculation debate, particularly in the work of Hayek (and the seminal article that followed it, surely one of the most widely quoted articles in economics (Hayek 1945)). It has since permeated (as we saw above) into the theory of the business organization, particularly in the prolific work of Herbert Simon and of Oliver Williamson.

Hayek argues that the economic problem does not consist in achieving the best allocation of resources among the various means available, as though the means and ends were somehow given and known.

The economic problem of society is . . . not merely a problem of how to allocate “given” resources—if “given” is taken to mean given to a single mind which deliberately solves the problem set by these “data.” It is rather a problem of how to secure the best use of resources known to any of the members of society, for ends whose relative importance only these individuals know. Or to put it briefly, it is a problem of the utilization of knowledge which is not given to anyone in its totality.

(Hayek 1945:77, italics added)

Knowledge is dispersed and it is specialized. Individuals have special knowledge of “time and place” and they are often not even aware of the knowledge that they have. The implications of this for the impossibility of central planning have been well covered (see, for example, Lavoie 1985a,b; Kirzner 1992:ch. 6). The fact that much of our knowledge is tacit and inarticulate and unconscious, places insurmountable barriers in the way of would-be central planners trying to duplicate the achievements of decentralized market systems. It is not simply a problem of the technical difficulty of collecting and processing all of the relevant information. It is, furthermore, the impossibility of being able to:

1. extract from this information the necessary knowledge;

2. know what needs to be known when the individuals in possession of the necessary knowledge might not even be aware that they have it;

3. gather the necessary information when much of it has yet to be generated by the market process.

The market, indeed, is a process that coordinates inarticulate knowledge by continual, and implicit, trial and error. The “knowledge” possessed by different individuals is often inconsistent. How, then, could it possibily be centralized?

The tacitness of knowledge is thus of singular importance in understanding the market process. Its importance is further enhanced by a realization that all articulated knowledge rests on unarticulated foundations. The component parts of any statement accepted as true (one plus one equals two) include undefined words which cannot be completely defined. Requiring a complete articulation of all terms pushes us into either an infinite regress or into circular reasoning.³ And, moreover, the very act of formulating any general statement necessarily requires using rules of proper statement formulation which are themselves inarticulate. For example, in describing real situations we must always select certain abstract qualities to which we wish to draw attention. This implies that nothing can be completely articulated, since the very process of articulation involves abstracting from real features of the phenomenon being described (Lavoie 1985b:60; also Hayek 1967 and 1973). Perhaps the most graphic illustration of the necessity of tacit knowledge is the way that children learn language. A child, who usually does not learn the rules of grammar early in life, if ever, can nevertheless construct grammatically correct sentences. This is evidence of the kind of tacit knowledge that underlies all articulated knowledge. “It is only because our minds are capable of operating according to effective rules of which we are unaware that we are able to learn to speak a language” (Lavoie 1985b:61). Another example is the act of riding a bicycle and other acquired skills. “Articulation, then, is just one kind of skill that we learn without knowing precisely how we accomplish it” (ibid.:62).

The fact that all knowledge is in part tacit and that all explicit knowledge rests on tacit foundations, implying that knowledge can never be made fully explicit, raises interesting questions for the economics of information in general and human capital in particular.

Equilibrium and the Nature of Knowledge

Many aspects of the human capital model suggest the assumption that agents operate in a context of equilibrium. In particular, there is the suggestion that the data historically observed in the labor market on earnings in relation to education and training are to be thought of as valid and accurate guides to the decisions that brought the data about (and, for that matter, to decisions relating to the current future). This implies that the projections that motivated the decisions are borne out by the outcomes that resulted and that there are no inconsistencies in the perceptions and plans of the various decision-makers. It implies, in short, a Hayekian equilibrium (see Chapter 3).

If this were true, we would have to conclude that the human capital approach not only abstracts from important aspects of time and change, it also begs certain important questions raised by the nature of knowledge discussed above. If knowledge is fallible, unfathomable, and tacit, it must be a product whose value cannot be fully known ahead of time and whose value is continually changing. The analogizing of knowledge to a consciously produced product is fraught with pitfalls and paradoxes arising out of the fact that it is a product of unknown quality and form. So we must ask, how do the seemingly valuable insights of the human capital approach apply in a disequilibrium world?

The world as we know it is, and must be, in disequilibrium, in the sense that individuals at any given moment are operating under different and often inconsistent expectations and theories. The outcomes that we observe, in the labor market and elsewhere, are the results of individual decisions that have been, and are bound to be, to a greater or lesser extent, mistaken. Just as observed financial portfolios do not represent, except in a trivial sense, optimal portfolios (i.e., those that would have been chosen if all the knowledge now available were available at the time of decision), so portfolios of human capital cannot be optimal. Capital gains and losses are, and must be, experienced as part of the market process. The market value of accumulated knowledge is an outcome that is (with rare exceptions) different from its anticipated value, to the extent that it can be anticipated.

I would suggest, however, that in order to make sense of our (tacit?) understanding of the market process, of our very real conviction that people do accumulate education and training in the reasonable expectation of economic (and other) gain, we re-examine our understanding of equilibrium and the nature of knowledge. As was suggested in Chapter 3, rather than seeing equilibrium as an all-or-nothing proposition, we should realize that we are, in relation to the various aspects of our lives and the decisions we take, simultaneously in both equilibrium and disequilibrium. It will be recalled that I suggested a simple tripartite taxonomy of knowledge. Knowledge type 1 (K1) is knowledge of the laws of nature; knowledge type 2 (K2) is knowledge of “social laws,” of conventions, routines, customs, etc.; and knowledge type 3 (K3) is knowledge of specific and unique events that have occurred (history) or will occur. Knowledge types 1 and 2 are knowledge of an abstract kind, knowledge of general principles (related to the natural world—apples fall from trees to the ground; or related to the social world—people stop at red lights, dollar notes are a generally accepted means of payment), whereas knowledge type 3—historical knowledge and expectations or anticipations—is knowledge of specific unique events.

It should be clear that K1 and K2 are necessarily fallible, unfathomable, and involve tacit elements, many of which are the evolved result of spontaneous social interaction. Even knowledge of the natural world is fallible, as we have seen following Popper. At any point of time there will, however, be a (more or less widely) shared understanding of “how the world works,” in Bartley’s terms the “accessed slices” of the various bodies of knowledge, and this will inform and help coordinate individual actions in important ways. Similarly, with “social laws” of behavior, many of which will be tacit, to the extent that they are widely “understood” they will serve to harmonize important aspects of individual plans. With regard to K3 we expect there to be inconstancy and disequilibrium. Since individuals’ plans are multilayered they could be, and, we contend, are, in equilibrium with respect to some aspects (like what they will do or not do if profits are not earned—they will not go to war, or resort to theft) while they are in disequilibrium with regard to other aspects (like being unable to anticipate accurately the level of profits, or, indeed, the returns to investment in education). As we have seen, however, the disequilibrium aspects do not incapacitate the decision-making abilities of economic agents precisely because such decisions are taken within an institutional environment that provides K1 and K2 with meaningful content.

Calculated rates of return on investments in human capital represent historical outcomes, but these historical outcomes will inform individual decisions to the extent that such outcomes are regarded as reliable guides to the future. A priori, there is very little that one can say in general about how likely this is. It is clear that individuals do look at educational opportunities as opportunities for economic advancement. Some of these opportunities will be regarded as more speculative than others. For example, investments in general human capital—generalized training that is not specific to any firm or industry, or generalized education (such as high school)—will generally be regarded as being more secure than an investment in more specialized, specific training. Investments in general training may be regarded as investments in K1 and K2 (their accessed slices), as may some investments in certain types of specific training. The uncertainty that attaches to the value of acquiring different types of knowledge is a function not only of the changing fortunes of different products, services, and the technologies associated with them, but also to the shifting sands of the validity of different principles and theories as new slices are accessed. The above discussion relating to general and specific training certainly continues to apply. One should emphasize, however, that cost and benefit calculations should be interpreted from a prospective perspective in so far as they are motivators of decisions. Every such decision is an entrepreneurial decision, with a greater or lesser degree of speculation involved. In this regard, human capital is no different from physical capital.

There remains the question of how tacit knowledge may be purposefully acquired. Clearly not all such knowledge can be. To the extent that we have knowledge of which we are not aware, we cannot have been aware of acquiring it. But there are types of tacit knowledge of which we are aware. Although we may not know exactly (from the point of view of physics) how it is that we learn to ride a bicycle, we do know what we have to do in order to learn. The connection between practice (experience) and performance provides a pragmatic or practical guide for very accurate decision-making, even though we may not know why. Through trial and error we discover better or worse ways of teaching language. Apprenticeships and internships of a very informal nature are often effective even if they mean simply providing the right environment for the spontaneous emergence of certain skills.

---

¹For an interesting application of Popper’s ideas to the understanding of entrepreneurship see Harper (1996).

²Polanyi (1958) (and related work) is the best known, but the work of many diverse and often seemingly inconsistent philosophers of science like Popper, Kuhn, and Lakatos all arguably imply a similar conclusion with regard to the question of tacit knowledge. For a remarkable—though necessarily incomplete—survey that motivates this position, see Lavoie (1985b:appendix).

³“It was . . . Alfred North Whitehead (1968:95) who so concisely pointed out . . . ‘there is not a sentence which adequately states its own meaning. There is always a background of presupposition which defies analysis by reason of its infinitude’ ” (Lavoie 1985b:60).