"New business procedures would then be analogous to new mutations in nature. Of a number of procedures, none of which can be shown either at the time or subsequently to be truly rational, some may supplant others because they do in fact lead to better results. Thus while they may have originated by accident, it would not be by accident that they are still used. For this reason, if an economist finds a procedure widely established in fact, he ought to regard it with more respect than he would be inclined to give in the light of his own analytic method."
(Roy F. Harrod, 1939, Oxford EP)
(A) The Old Marginalism Debate
(B) The New Institutionalist Theories
(C) The Evolutionary Theory of the Firm
(A) The Old Marginalism Debate
The "Marginalism" debate emerged in the late 1930s after several studies - one by the Oxford Research Group (see Hall and Hitch, 1939) and another by R.A. Lester (1946) had argued that, empirically, it was not apparent that entrepreneurs followed the marginalist principles of profit- maximization/cost-minimizations in running their firms. In particular, they found that many firms conducting "full cost" pricing rules and routines and that predicted falls in employment as a result of higher wages were not evident. They consequently questioned the relevance of the profit-maximization assumption in Neoclassical theories of the firm.
The Oxford results were accompanied by a paper from Harrod (1939) which argued that perhaps profit- maximization was not observed in many firms partly because the necessary information - marginal revenue and costs - for such calculations was hard to obtain. Yet, he added, decisions would nonetheless be made and that there may be a "natural selection" process at work whereby, because of its superior results, the profit-maximizing decision would be selected for and other decisions selected against. Thus, Harrod argued, entrepreneurs may not get profit-maximization exactly, but they would in actuality "grope" for it.
The studies provoked an immediate response on the part of Fritz Machlup (1946, 1947) and George J. Stigler (1946, 1947). They defended the marginalist principle assiduously, but on different lines than Harrod. In particular, Machlup argued that the Oxford Group's find that firms often follow established "routines" rather than make deliberate, careful decisions could mean that, in earlier times when the routines were established, they were the profit-maximizing procedure and had simply not been updated in light of new circumstances. Furthermore, even if they did not have their roots in conscious decisions, one must be careful about the meaning of marginalist analysis: as Machlup notes, there are many factors which may affect profits and thus, as many traditional relations implied by profit maximization (e.g. rise in wages leading to less employment) are often considered in theoretical isolation, they are rarely isolated in fact.
As such, some of these individual relations may not be observed in empirical studies, but it does not mean they do not exist. Using his famous analogy of the decision of a driver to overtake another car on a road, Machlup (1946: p.534) argued that particular considerations such as the speed of the other car, the speed of one's own, distance, etc. certainly have individual impact on the decision, but all these factors operate at once and the car-driver is often forced to "size-up" the situation as a whole and make a decision. Similarly, entrepreneurs are forced to "size up" a total situation and thus empirical researchers might not be able to differentiate and trace the impact of different, isolated components on business decisions. But the theory of profit-maximization - like the "theory of overtaking" - must write down all the components separately.
Armen Alchian (1950, 1953) took issue with both Machlup and the anti-Marginalists. His argument was that, quite simply, the Neoclassical theory of the firm is not about firms as such but industries. Individual firms, Alchian argued, essentially followed the routines claimed by the researchers, but it would be the industry which adheres to the marginalist principles. Thus, if wages rise, we should see the average industry labor employment fall, but this is not because firms change their techniques of production (i.e. they don't hire more labor), but rather because those firms which happen to have the routine which corresponds to the new optimal labor employment will be more profitable than those that do not. In what is considered today a classical paper on evolutionary theory, Alchian (1950) proceeded to make a "natural selection" argument (although substantially different from that of Harrod (1939)): the profitable firms (those which, by luck(?), had the optimal routine) would last longer while those which were less profitable (did not have the optimal routine) would soon go out of business. Thus, the industry as a whole would move towards the "optimal" decision, not because firms change their behavior, but rather because those firms lucky enough to have the optimal routine will be selected for by this process and those which do not will be selected against. Thus, profit-maximization, Alchian argued, is not a result of firm decisions but rather a result of an evolutionary process being conducted at an industry wide-level. The Marginalist principle, thus, is fully applicable to the industry, albeit not the firm.
Penrose (1952, 1953) took issue with Alchian. If "market forces", as Alchian argued, were the natural selectors, then there is a presupposition that competition exists. What she goes on to ask is what generates competition if not the very profit-seeking motives of entrepreneurs? Alchian (1953) responds quite clearly that Penrose confused intended profits and realized profits: the selection procedure is simply on the basis of realized profits that permit a firm to continue operating; it is not necesary that they actually search for them.
Alchian's argument that profit-maximization behavior on the part of entrepreneurs is irrelevant for the result that industry will attain the profit-maximizing solution was taken by Milton Friedman to a ridiculous extreme in his 1953 paper on "positivist" methodology. He, however, seems to forget Alchian's argument half-way and proceeds to make his famous "billiards" analogy which attempted to argue that agents who do not profit-maximize will be selected against. This is again a confusion of intended profits and realized profits that is necessary for Alchian's arguments.
Gary Becker (1962) follows up on Alchian and refines his argument somewhat: the selection procedure Becker envisions is that of collapsing budgetary constraints for non- profit-maximizing behavior. Firms, Becker argued, cannot survive for long with the wrong routine regardless of what their motives are. If firms are resource-constrained (!), they will shrink and then go out of business if they do not have the optimal routine wheras those who do and are reaping profits will grow. Thus, Becker demolishes Friedman while restoring the original meaning of Alchian's theory.
Israel Kirzner (1962), however, takes issue with Alchian and Becker and asks the simple question: are not prices the outcomes of behavior in equilibrium? Alchian and Becker argued on the basis that if, say, prices change, the budget constraints of firms with the optimal routine would expand wheras those without it would contract. But are not these changes in price, indeed the very phenomenon of price, a result of interacting agents (and thus firms)? Why should they act as price- takers? Would not irrational decisions implicate the resulting prices and thus the degree and the manner by which budgets contract? Prices, Kirzner argues, are outcomes and not data of economic processes. It is possible, Kirzner's view, for the "wrong" firms to continue making profits and survive.
(B) New Institutionalist Theories
The Marginalist debate changed tone in the 1970s with the emergence of the theories of agency costs, property rights, transactions costs theories of the firm. The new theories of the firm attempted to prove a variant of Friedman's (1953) conjecture that firms which had profit-maximizing behavior would be selected for. More precisely, they argued that firms with "efficient" organization would be selected for whereas those which were run inefficiently would be selected against. Here the object of study changed from how to reconcile firm behavior with marginalist principles to how to reconcile firm structure with marginalist principles. This concern was first taken up by R.H. Coase (1937) when he noted the apparent disparity between the marginalist notion that markets were efficient organizers of resources with the existence of highly hierarchical, control structures within firms. If the Neoclassicals were at all right in the Socialist Calculation debate, it would seem at least somewhat peculiar to find "command economies" operating internally within capitalist firms. Why not use the price mechanism within firms as well? Coase's (1937) conclusion was that transactions costs were high, or rather that within firms, command structures had lower transactions costs than using price mechanisms. (it does not take much imagination to perceive the similarity between Coase's (1937) arguments for the firm and Oskar Lange's (1936, 1938) for the economy in the Socialist Calculation debates!)
The challenge to Coase was set up by Alchian and Demsetz (1972). What they essentially argued was that it was not that command structures minimized transactions costs, but rather that they minimized "monitoring costs". As many of the firm's processes involve team production, then there will necessarily be monitoring costs to assess individual efforts within a team. In order to overcome shirking in team production, firms assign a central agent the monitoring role - and that this central agent is himself discouraged to shirk by making him the "residual claimant" of the earnings of the team. As a result, the residual claimant must be given "property rights" over the output. In short, Alchian and Demsetz (1972) justify the role of the profit-earning entrepreneur and a command structure - i.e. the capitalist firm - as a result of what can be argued to be a series of market-based contracts.
Alchian and Demsetz's idea that the behavior of the firm is in fact not too different from the behavior of the market was also famously proposed by Jensen and Meckling (1976) who purse the idea of "agency costs" as the source of firm structure. Even in modern firms, where management is divorced from residual claims, external equity and bond markets in effect play the role of the monitoring agency on managers. Fama (1980) has gone on to claim that even if management truly does not care about financial markets, they still care about the market for managerial labor which can induce management to perform their monitor role more seriously. Fama and Jensen (1983), indeed, pursue factor market arguments in general in this regard. Of course, if all fails in the end to overcome agency costs, there is always that ultimate market disciplining device: the hostile takeover. Thus, at least since Alchian and Demsetz (1972), the internal organization of firms has been argued, contra Coase, to be dominated and explained by market relationships in the presence of agency costs. Firms are not independent of market disciplinary devices.
A form of the Coase position was famously resurrected by Oliver E. Wiliamson (1985, 1987, 1991). invoking the Herbert Simon's theory of bounded rationality and opportunism. "Opportunism" is simply self- interested behavior, "bounded rationality" implies however that full agency contracts cannot be written up because not all contingencies can be forseen. Furthermore, when there is "asset specificity" (i.e. factors particular to the firm), then factor market disciplinary devices simply will not be available. Thus, Williamson informs us, while there will be high ex ante market-influenced competitive bidding before contracts between firm and outside agents are signed, this contract does not fully account for all problems that may arise in the future. The "opportunistic" character of the agents imply that they will diverge from their contract-assigned purpose if they can and find it profitable to do so; bounded rationality implies that the original contracts could not account for all the possible ways agents could diverge. They could bring the contracts back to market every time they find divergence, but this would be costly. Thus, in sum, Williamson argues that ex ante bidding is competitive, but ex post firms face "sunk costs" in this respect and thus to modify contracts in response to new behavioral incentives would involve heavy transactions costs (i.e. opening up bidding in factor markets again to replace the shirking or irresponsible manager or worker). Thus, firms do exhibit a hierarchical non-market structure internally which is independent of market disciplining devices largely because of transactions costs. This has been the heart of the New Institutional economics.
(C) The Evolutionary Theory of the Firm
The old marginalist debate re-emerged in the 1980s with the evolutionary theory of Richard Nelson and Sidney Winter (1982) which combined the earlier Alchian-Becker arguments with the theories of the firm of the New Institionalists.
R.L. Hall and C.J. Hitch, 1939, "Price Theory and Business Behavior", Oxford EP