Organization Form: Selection and Consequences

Contributed by: Nick Argyres and Rachelle Sampson

This reading list traces research on organizational form that has built upon organizational economics. It also includes readings that challenge this economic approach from a sociological perspective, as well as readings that seek to reconcile the two approaches. In all of these readings, “organizational form” refers to the allocation of decision-making authority, coordination and adaptation mechanisms and the incentive structure within organizations and between them. Note that we refer to organizational form and governance structure interchangeably. The main objective of the readings in this section is to identify organizational forms that have the ideal balance of coordination, control and incentive power, given firm (or inter-firm, in the case of alliances) performance objectives.

The selection of organizational form and its consequences for firm performance were arguably the very first phenomena to be studied in the strategy field. Many scholars trace the origins of the field to Chandler’s (1962) Strategy and Structure. Chandler’s hypothesis -- that the multidivisional form (M-form) was an efficiency-enhancing, organizational innovation adopted to accommodate new strategies of diversification -- began a central stream in strategy research known by the title of the 1962 book. The first statistical test of this hypothesis, Armour and Teece (1978), is listed among the foundational readings in this section.

Chandler’s (1977) second book, The Visible Hand, emphasized the role of a different aspect of organizational form, vertical integration, in facilitating the growth and profitability of large U.S. firms during the mid-to-late 19th century. This work helped inspire the development of the transaction cost economics of Williamson (1975, 1985). Williamson re-introduced the twin fundamental questions first posed by Coase (1937): why do firms exist, and what limits their scope? While Coase’s arguments have been most often applied to explain vertical integration, the fundamental cost comparison of different organizational forms envisioned by Coase applies to horizontal integration as well.

Along with Klein, Crawford and Alchian (1978), Williamson developed transaction cost theory to help operationalize the theory first posed by Coase (1937). Williamson (1991) represents a relatively comprehensive statement of transaction cost theory, although interested readers should also refer to Williamson’s (1985) book, The Economic Institutions of Capitalism (New York, The Free Press) for a full treatment and many additional applications to various organizational phenomena. Transaction cost theory holds that firms, markets, and intermediate organizational forms are discretely different governance structures, and that efficiency requires that they be matched with different kinds of transactions in a discriminating way. Readers should also be aware that the historical accuracy of Klein, Crawford and Alchian’s (1978) central example of GM-Fischer Body has been questioned by several scholars (e.g., R. Casadesus-Masanell & D. Spulber (2000) “The Fable of Fisher-Body.” Journal of Law and Economics 43, no. 1: 67-104]. Notwithstanding the challenges to the accuracy of this example, the theoretical argument in the Klein et al. paper remains valid and has substantial empirical backing in other contexts.

Grossman & Hart (1986) propose a closely related theory, that formalizes some, but not all, of the insights of transaction cost theory. The key implication of this theory is that residual rights of control over an asset should be allocated to the party whose contribution to the transaction is relatively more important to the performance of that transaction. This “property rights” theory offers greater precision than Williamson’s verbal theory, particularly in defining ownership and in predicting the directions of integration. However, while property rights theory emphasizes selection of organizational form based on minimizing ex ante bargaining problems, it does not capture the problem of ex post adaptation that Williamson emphasized, and has also proven more difficult to test due to difficulties in developing appropriate measures for its key variables. Notwithstanding this, property rights theory has spawned a large theoretical literature and some empirical studies (e.g., Baker and Hubbard (2001) “Make Versus Buy in Trucking: Asset Ownership, Job Design and Information.” American Economic Review Vol.93:551-72).

Transaction cost theory attracted the attention of sociologists because it in part arose from, and therefore spoke to, concerns articulated in organization theory. Sociologists tended to accept the first of transaction cost theory’s basic behavioral assumptions, that economic agents are “rational but boundedly so” (Simon 1962). The second assumption, that agents may be opportunistic, was treated more skeptically. Granovetter (1985), for example, expressed concern that this latter assumption suggested an “undersocialized” view of economic life. Granovetter argues that, because individuals are usually embedded in networks of social relationships, even risky transactions can often occur through market exchange. Granovetter’s article set the stage for a debate that continues to be important in the strategy literature, particularly the literature on contracting and alliances.

The secondary readings fall into four areas that build upon the foundational research. Research on internal organization in organizational economics has focused on the limits for firm size and scope. A major theme in this work is that firms cannot perfectly replicate the high-powered incentives available in market, and that this accounts for the limits to firm size and scope. Examples of this include Holmstrom and Milgrom (1994) (theory) and Argyres (1995) (empirical). An important counterpoint to these papers is Fligstein (1985), who challenged the assumption in organizational economics that internal organization is driven by efficiency, offering evidence that power and other considerations are more important.

Transaction cost theory’s was originally developed to explain vertical integration, and much research on that phenomenon has followed. The readings in this section include some of the early empirical research that aimed to develop appropriate measures of transaction cost theory’s main determinant of vertical integration, asset specificity. A notable example of high quality research here and one of the first empirical studies directly linking organizational form to performance is Masten, Meehan, Snyder (1991). This paper was the first to examine transaction cost hypotheses using methods that account for the self selection problem; that is, firms choose organizational form based on the perceived performance attributes of such forms. This problem makes it difficult to draw clear inferences from simple qualitative choice models of make-or-buy choices, a pervasive problem in any study seeking to link organizational form with performance. Masten et al. (1991) go beyond these simple models to estimate the costs of different organizational forms for a particular transaction, including the costs of internal organization. As such, the paper represents a substantial advance over prior work both conceptually and empirically, moving beyond a discussion of the costs of the market to a more thorough analysis of integration. Other important studies in the area include Poppo and Zenger (1998), who offer a test of transaction cost theory against other theories in strategy aiming to explain make-or-buy choices, and Nickerson and Silverman (2003), who show that sociological considerations can affect the speed at which economization on transaction costs occurs.

Alliances have been a very popular topic of research in the strategy field, and contributions from scholars using transaction cost theory have been very influential. These scholars have extended the transaction cost framework to operationalize relevant inter-firm drivers of transaction costs, such as concerns over intellectual property leakage. Two of the listed studies emphasize the importance of intellectual property protection considerations in determining the structure of alliance relationships (Pisano 1989; Oxley 1997), while Sampson (2004) shows that alliance governance choices that are misaligned from a transaction cost standpoint are costly. Sociologists have studied other factors that may substitute for formal governance; Gulati (1995) finds evidence both for the effects of transaction costs as well as prior ties between partners on alliance structure. He argues that this finding reflects the development of trust, a factor that many sociologists have emphasized.

This theme of trust versus safeguarding through formal contractual provisions also appears in the literature on the details of contract structures. Macaulay (1963) emphasized that non-contractual relations are important in business to the extent that formal contracts may not only be unnecessary, but limiting, and many sociology-oriented scholars argue that trust substitutes for formal governance in contractual relationships. However, some recent empirical work on contract structure suggests that trust and formal governance may in some circumstances act as complements rather than substitutes. This may in part be due to the communication and learning benefits that formal contracting can provide to the parties (Poppo and Zenger 2002; Mayer and Argyres 2004), or because contracts clarify when renegotiation is required as well as when partners may legitimately either impose informal sanctions and/or terminate the relationship (Ryall and Sampson 2009). The fifth reading in the section, Joskow (1987), was among the first to find support for a transaction cost explanation of contract duration.

There are a number of important empirical challenges that researchers face when studying the determinants of organizational form and its relationship with firm performance. For example, the theoretical mechanisms described in these theories operate at a very micro level, so that tests require detailed data that is often difficult to collect. In addition, it is often difficult to fully account for differences in transaction characteristics. Researchers must be aware that unobserved factors may drive a firm’s decision to select a particular organizational form. Without a thorough understanding of what underlies the choice of organizational form for a particular firm, inferences from data on form and performance will be limited and possibly biased. The empirical studies highlighted here represent a sampling of those papers that have made inroads at thinking deeply about the organizational form problem and/or taken novel empirical steps to correct for unobserved drivers of organizational form.

It is important to note that the economic and sociological theories that are featured on this reading list have been applied to a much wider range of phenomena than are represented here. Other areas of application, some of which relate to those on lists already in (or to be added to) this reader, include diversification, corporate governance, the multinational enterprise, and political strategy.