What is a firm?

Firms are an essential part of the economy. Although during the past seven decades, economists have made huge progress in moving beyond the firm as a black box to incorporate incentives, internal organizations and firm boundaries, the very fundamental question of "What is a firm" is still far from uncontroversial. The recent book by Daniel Spulber -- The Theory of the Firm: Microeconomics with Endogenous Entrepreneurs, Firms, Markets and Organizations (Cambridge University Press, 2009) -- has caused yet another ripple among the institutional economists on debating over the very basic concepts of this great human invention and institution -- the firm. One major critic (gated version) of Spulber's book is a guru in modern economic theories of organizations, Harvard professor, Oliver Hart.

To accomplish his goal, Spulber clearly defines a firm in the following sense: "The firm is defined to be a transaction insitution whose objectives differ from those of its owners. The [Fisher] separation is the key difference between the firm and direct exchange between consumers" (p. 63). Then he uses this definition to argue that consumer organizations such as clubs, and basic partnerships are not firms, nor are worker cooperatives, many family businesses, nonpublic organizations, or public enterprises. However, clubs (and worker cooperatives and partnerships) could become firms if and when a market is created in memberships. He uses an example -- a worker cooperative -- to illustrate his basic idea. Put it simply, a "basic" worker cooperative (without membership fee) chooses an employment level to maximize each member's share of surplus. Since there is no separation between the cooperative and its owners, this basic worker cooperative is not a "firm". Moreover, in this case, the optimal employment level turns out to be inefficient in the sense that it is not profit-maximizing as if the cooperative is a standard "firm". When certain level of membership fee is charged, however, by maximizing the objective of the initial owner's payoff, the profit-maximizing employment level can be achieved. Now the objective of the initial owner and that of the "new" cooperative can be separated, so by definition, this worker cooperative with a membership fee is a firm. Although the logic here is perfectly coherent, the definition seems too restrictive -- I'll explain later. For now, I think there is another possible caveat with this definition even in this very illustrative example. Intuitively, these two cooperatives differ only in terms of scale, why would we on earth call a more efficient one a "firm" and not the other one? Hart's treatment makes more sense, that is "both forms of worker cooperative (as well, of course, as a profit-maximizing firm) are firms, but one form is more efficient than the other. If the world is as described above then in equilibrium we would expect to see the basic worker cooperative being replaced by one with membership fees or by a profit-maximizing firm."(p. 11)

Why is this definition too restrictive? Hart contends that apart from the fact that many natural firms, like Microsoft, Google, News Corporation, Berkshire Hathaway, CBS, and the New York Times, all seem to fail the test, there is also a question of how to empirically apply this defintion. In his words, "How can we say empirically whether an entity has an objective function that differs from that of its owners? How do we learn what the objective function of a firm is?...More generally, does it even make sense to talk about the objective function of an oganization?". As Jensen and Meckling argued in their famous 1976 article, "...the personalization of the firm implied by asking questions such as 'what should be the objective function of the firm'... is seriously misleading. The firm is not an individual. It is a legal fiction which serves as a focus for a complex process in which the conflicting objectives of individuals...are brought into equilibrium within a framework of contractual relations". Indeed, a large body of literature in the principal-agent theory since the early 1970s is trying to formalize these conflicts within firms and find optimal contracts to overcome incentive problems. Although this stream of literature has achieved limited progress on how to essentially characterize the firm, more great insights have been drawn by other organizational economists studying the internal organizations as well as the boundaries of the firm. But still, neither the incentive schemes nor the internal organization or the boundary issues seems to be the central focus of Spulber. To me, as well as pointed out by Hart, Spulber's notion of firm is more or less a redux to the neoclassical paradigm of treating firms as production function rather than opening it and examining what is going on inside.

For spulber, his goal in this book is to explain why firms exist, how they are established, and what they contribute to the economy. Considering this, his definition and the implication that firms may behave as if they have different objective functions are plausible and of practical importance. However, as pointed out by Hart, "...this raises the question: should we take a firm's objective functions as the starting point of our analysis, or do we need to dig deeper and derive the behavior of the firm from its governance structure, incentives of managers, culture, etc.?" For a prominent strategy scholar like Spulber who is passionate to understand the performance heterogeneity among firms as well as their contributions to the whole economy, it's difficult for me to understand why he doesn't put enough emphasis on the crucial issues like the incentives, internal organizations, and boundaries of the firm.