Organizational Economics: A Novel Perspective on Knowledge Management
Overall
Despite beginning as a theory of the existence and optimal scope of the firm (Coase, 1937; Williamson, 1985), during the last twenty years or so organizational economics has increasingly been applied to internal organization issues. In particular, it has directed attention to the coordination and incentive problems that are caused by the pathologies that unavoidably accompany an internal division of labor, such as asymmetric information, diluted performance incentives, measurement difficulties, bargaining problems, moral hazard, duplicative (redundant) efforts, etc. In turn, organizational economists have explained how a host of organizational arrangements, such as various kinds of authority, payment schemes, delegation of decision rights, etc. serve to alleviate the severity of such problems.
Beginning our brief sampling of organizational economics perspectives, agency theory perspectives have predominantly addressed issues related to payment schemes (Holmström, 1979, 1989) delegation of decision rights (Fama and Jensen, 1983; Jensen and Meckling, 1992; Aghion and Tirole, 1997), multitasking (Holmström and Milgrom, 1994), and managerial commitment (Baker et al., 1999) under assumptions of moral hazard and asymmetric information. Transaction cost economics (Williamson, 1985, 1996) and property rights insights (Hart, 1995) have been brought to bear on issues related to allocation of rights and design of contracts when investments in human capital are firm-specific, agents may behave in an opportunistic manner, and contracts are incomplete. Team theory (Marschak and Radner, 1972; Casson, 1994; Carter, 1995) has addressed the optimal design of organizational structures, given the bounded rationality of individuals (but absent conflicts of interest). Finally, work on complementarities between organizational elements (e.g. payment schemes, delegation of rights, supervision methods, etc.) (Milgrom and Roberts, 1990, 1995) lends strong formal support to the traditional notion that there are stable, discrete governance structures that combine organizational elements in predictable ways (Thompson, 1967; Williamson, 1996). It is fair to say that the empirical base of organizational economics, in terms of the number of corroborations of predictions of these theories, is fairly strong (Shelanski and Klein, 1995; Prendergast, 2002).
Although organizational economics is constituted by a number of different theories, nevertheless there are a number of common threads in the literature (cf. Foss, 2000). On the method level, all of organizational economics is unabashedly ‘individualistic’ in the sense that all organizational phenomena should be explained as the outcome of the choice behavior of individual agents. At the theoretical base, the whole literature is concerned with ‘efficiency,’ that is to say, how resources are allocated so that they yield the maximum possible value. Two closely related implications follow immediately. First, the organizational economics perspective is intimately taken up with value creation; as noted, maximizing the value that can be created is the meaning of economic efficiency. Second, since the allocation of resources is (also) a matter of how the relevant resources are governed and organized, and since value-creation is dependent upon governance and organization, it follows that an efficiency perspective allows one to discriminate between alternative forms of economic organization in terms of efficiency. Rational actors will choose those organizational forms, contracts, and governance structures that maximize their joint surplus and will find ways to split this surplus among them.
In turn, the influence of alternative organizational arrangements on value creation may be analyzed in terms of motivation, knowledge, information, and complementarity—and how alternative arrangements embody different ways of influencing these variables (cf. also Buckley and Carter, 1996). These are all in different ways related to those ‘transaction costs’ that (in various guises) are central in all organizational economics theories, and the size of which influences the value that may be created from organizing and governing scarce resources in particular ways. The value that can be created, in the presence of transaction costs, falls short of what may be created in a world with no problems of motivation, knowledge, information, and complementarity (a ‘first-best’situation), and, hence, no transaction costs. While such a world may be imagined, it is not the world of managers and other inhabitants of organizations. However, the above may be manipulated so that the organization approaches it. We discuss motivation, knowledge and information, and the coordination of complementary actions seriatim in the following.
Motivation
The motivational assumptions of organizational economics have been subject to a good deal of scrutiny and critical discussion. Many scholars in, for example, organizational behavior, have been critical of the seemingly cynical assumptions with respect to human nature that drive much of organizational economics analysis. To these critics, opportunism (‘self-interest seeking with guile,’ Williamson, 1996) and moral hazard (i.e. using asymmetric information to one’s advantage and the other party’s disadvantage after a contract has been concluded) are not descriptively accurate. They may furthermore be ‘bad for practice’ to the extent that managerial action based on prescriptions from these theories may, by treating people as would-be opportunists, lead to self-fulfilling prophecies (Ghoshal and Moran, 1996). However, such motivational assumptions fundamentally serve to highlight the—presumably undisputed—fact that actors often have very different interests; opportunism and similar assumptions are stark ways of highlighting this. Moreover, the motivational assumptions serve to emphasize that economic organizations need to be designed with an eye to the possibility that some (by no means all) actors may act in a morally hazardous or opportunistic manner.
In the context of internal organization, the largest effort so far may well have been devoted to exploring how various aspects of internal organization—from accounting principles over payment methods to the nature and function of hierarchy itself—may be explained as efficient responses to various principal–agent problems. Thus, particular attention has been paid to differences between input and output-based payment, and how the choice between these is determined by the observability of effort and states of nature; the role of monitoring and of subjective and objective performance measurement (Prendergast, 2002); and of how a hierarchical structure may constrain ‘rent-seeking,’ that is, attempt to influence superiors to one’s own advantage (Milgrom, 1988).
One perspective on all this is that various aspects of internal organization arise to curb the resource costs of agents pursuing their own interests in a way that is harmful to the organization. Under an organizational division of labor, management (and the owners of the firm) delegates some rights to employees, ranging from the trivial (the right to work with the company’s vacuum cleaner) to the all-important (the right to make decisions on major investment projects). Management wishes these delegated rights to be exercised in an optimal manner. However, since the right holders cannot be constantly monitored, and since performance pay schemes trade-off incentives and risk, some losses (compared to a full-information situation) are usually unavoidable. Internal organization arises as a trade-off between these losses and the costs of designing monitoring schemes, incentive contracts, etc.
A particular set of incentive problems is caused by problems of managerial commitment. For example, often employees wish to specialize their human capital to the firm, thus becoming more productive and hoping to capture some of the marginal productivity created. In other words, they expect to be compensated for their investment. However, by specializing in this way, employees become subject to a potential hold-up problem (Williamson, 1985, 1996; Hart, 1995). To be sure, the possession of specialized knowledge may be a strong bargaining lever. However, there is another strong party to the bargain situation, namely the firm to which the employee specializes. The implication is that employees cannot expect to capture all or even most of the quasi-rent from their specialized human capital investments, which harms incentives to undertake the investments (Hart, 1995). Strong and credible managerial commitment to not using the hold-up option may solve the problem (Kreps, 1990). Another way of solving the problem is to allocate (more) decision rights to employees who undertake human capital investments (Rajan and Zingales, 1998). Thus, in professional service firms, often employees with a long tenure and good demonstrated performance become partners. A final managerial problem has to do with managerial interference in the business of agents to whom the same management have delegated rights (e.g. to run their own projects). This ‘problem of selective intervention’ (Williamson, 1985) arises because it is often hard for management to commit to not interfering. For example, it is not possible to make a court enforceable contract to prevent managerial interference once decision rights have been delegated. However, arbitrary intervention, the breaking of promises to not intervene, etc., all of which will often be very tempting for management, are very destructive for motivation (Baker et al., 1999; Foss, 2002).
These incentive problems are clearly relevant to the understanding of the costs of knowledge management practices. To the extent that agents’ human capital investments consist in the gathering and building-up of specialized knowledge and skills, they are not likely to be willing to share the relevant knowledge and skills with other agents, unless they are properly compensated. They are not going to give up a strong bargaining lever without compensation. However, it is often difficult to contract over knowledge and skills. Moreover, there is a fundamental problem of managerial commitment: since it is difficult to write and enforce contracts on the sharing of the knowledge and the compensation to the employees between those employees who possess important specialized knowledge and the firm, it is tempting for management to renege on the promise after the sharing of knowledge has actually taken place. Two implications of direct relevance for knowledge management follow. First, forced knowledge management initiatives may well be experienced as hold-ups by those agents inside the firm who control specialized knowledge and skills. Their future investment incentives are harmed accordingly. Second, unless these agents can expect to be compensated they are unlikely to share their knowledge at all. It is likely that the best way to handle this (i.e. to invest in human capital and to share knowledge embodied in this capital) is by giving the relevant employees appropriate incentives, perhaps even making them partners through providing ownership rights.
Asymmetric knowledge and information
Even if agents can be motivated to take actions (i.e. exploit their decision rights) that are ‘incentive-compatible’ with those of other agents or principals, there is still no guarantee that they will also make optimal (i.e. value maximizing) choices. Willingness is not the same as ability. To some extent this is a problem of information transmission: under an organizational division of labor, no agent inside the firm is likely to have all the information needed for making an optimal choice, and transmitting all of this information to him or her is prohibitively costly. Delegation may arise as a cost economizing response to this. However, it is also a matter of the often fleeting, subjective, and tacit character of knowledge—a favorite theme of the knowledge management literature. As Hayek famously argued:
The peculiar character of the problem of a rational economic order is determined precisely by the fact that the knowledge of the circumstances of which we must make use never exists in concentrated or integrated form but solely as the dispersed bits of incomplete and frequently contradictory knowledge which all the separate individuals possess. The economic problem of society is thus 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)
Arguably, firms face this problem of dispersed knowledge to a smaller extent than societies do; however, it is still relevant to them. Firms may cope with the problem in different ways. Again, they may delegate decision rights so that these rights are co-aligned with those who possess the relevant knowledge, balancing the attendant benefits with the agency costs that are caused by delegation (Jensen and Meckling, 1992). However, knowledge sharing is an alternative to this. Thus, rather than delegating decision rights in order to better utilize local knowledge, the existing rights structure (i.e. existing authority relations, payment schemes, organizational structures, etc.) remains unchanged and the relevant knowledge is gathered and shared among those who can make profitable use of this knowledge. Such knowledge sharing is, of course, a key focus of knowledge management.
However, in the knowledge management literature, knowledge sharing is often discussed and endorsed without any examination of the alternative of delegating rights so that knowledge is better utilized in this way. An organizational economics perspective not only identifies the relevant (organizational) alternatives, but also allows us to say something about the costs and benefits of these alternatives. Thus, one obvious advantage of the knowledge-sharing alternative is that it does not necessarily involve any delegation of decision rights. Knowledge sharing, as portrayed in the knowledge management literature, may therefore impose smaller agency costs on an organization than the alternative of delegating decision rights. However, there are other costs to consider when the choice has to be made between the two alternatives of knowledge sharing and delegating decision rights. For whereas knowledge sharing that takes place within an existing organizational structure may not impose the same agency costs as delegating decision rights does, knowledge sharing is likely to impose higher costs of, for example, communicating, storing, and retrieving knowledge than the delegation alternative. The point is not here that specialized IT systems have to be set up in order to reach the goal of knowledge sharing. Rather, the point is that knowledge sharing may introduce costs that are caused by the bounded rationality of individuals, that is, their limited ability to identify, absorb, process, remember, and so on, knowledge. And, of course, there are costs associated with trying to transform knowledge that only exists in a tacit form into an articulate form. As Hayek (1945) argued, decentralization economizes on these costs. In firms, delegation may be an attractive means of economizing on the costs associated with bounded rationality and tacit knowledge (Jensen and Meckling, 1992). The bottom line is that a full assessment of what alternative is superior in a specific situation—the improved utilization of knowledge by means of knowledge sharing or delegation of decision rights—generates a number of costs that have to be balanced against the relevant benefits. In its present manifestation, the knowledge management literature identifies neither the relevant alternatives nor the relevant net benefits.
The coordination of complementary actions
Even if agents can be motivated to take incentive-compatible actions and even if they possess the right information or knowledge (because they are specialists or because this information or knowledge is somehow transmitted to them), there is still a problem of coordinating actions inside the firm. In particular, the more complementary actions are, the more closely they need to be coordinated. Through the use of the price mechanism, markets cope well with the coordination problem (Hayek, 1945). However, the more complementary actions are, the more necessary is it to supplement the use of the price mechanism with other mechanisms, such as communication (Richardson, 1972). Firms have only limited access to the price mechanism, but they may have privileged access to the mechanism of communication (relative to markets). In this perspective, one advantage of knowledge management may actually be that it assists the coordination of complementary actions by spreading knowledge, effectively bringing about common knowledge conditions (see Foss, 2001 for such an argument). Knowledge management thus reduces what Koopmans referred to as ‘secondary uncertainty:’
In a rough and intuitive judgment the secondary uncertainty arising from a lack of communication, that is, from one decision maker having no way of finding out the concurrent decisions and plans made by others . . . is quantitatively at least as important as the primary uncertainty arising from random acts of nature and unpredictable changes in consumers’ preferences.
Koopmans (1957: 162–163)
When the acquisition (creation, sourcing) of knowledge in a firm is delegated to specialist knowledge workers, the firm faces this kind of secondary uncertainty (cf. Buckley and Carter, 1999). One possible function of knowledge management is thus to reduce secondary uncertainty, although this is not one that is identified in the knowledge management literature.
Summing up: Organizational economics aspects of knowledge management
In the frictionless world that dominated microeconomics textbooks before the revolution in information, property rights, and transaction costs economics about three decades ago, there are no problems of motivation, knowledge, information, and coordination. In this nirvana, resources, including knowledge resources, are allocated in the best possible way (‘first-best’). Contracts can be written and enforced without cost and information is free. Therefore, there are no losses from lacking motivation, defective or missing knowledge, or coordination that goes wrong. There are no problems of exchanging knowledge either, so that markets are as efficient for this purpose as firms are. However, in a more realistic world, contracts are imperfect so that, for example, it is hard and perhaps impossible to write contracts that compensate those who ‘give up’ (i.e. share) valuable knowledge; commitment (including managerial commitment) may be broken; employees may be held-up by management so that their incentives to invest in and share knowledge are harmed, etc. Lest managers live in a paradise or nirvana, knowledge management practices are subject to these incentive costs.
The argument so far is therefore that organizational economics is able to illuminate the practice of knowledge management in important ways. In particular, by focusing on incentive compatibility problems, particularly as these relate to issues of investing in the production and sharing of knowledge, organizational economics identifies important, but hitherto neglected, incentive costs and benefits of knowledge management practices. This is the reason why organizational economics should be seen as an indispensable part of the disciplinary foundation of knowledge management. In the following section we deal further with processes of knowledge creation and integration in an organizational economics perspective.