Reaffirming relationship-specific investments.

• Author: Masten, Scott E.
• Position: Response to article by Yoshiro Mirwa and J. Mark Ramseyer in this issue, p. 2636

Comments on Miwa and Ramseyer's `Rethinking Relationship-Specific Investments'

I, too, have a work-related anecdote from my youth to relate. During one summer break from college, I had a job on the night shift in the canning plant of a Coca-Cola bottling franchisee in my hometown in New Hampshire. The process of canning tonic (known as "soda" outside of New England) consisted of three stages, beginning with the fabrication of cans in a room at one end of the building and concluding with the filling and sealing operations in a room at the other end. In between, workers in a third room inspected cans as they arrived by conveyor from the fabrication facility, loaded empty cans onto pallets for storage, and unloaded cans of the appropriate type (Coke, Tab, and Shasta) back onto the conveyor to assure a continuous supply of containers to the filling room "downstream."

The year was 1976 and Oliver Williamson's Markets and Hierarchies, published just the year before, had not yet found its way into the undergraduate curriculum. Nevertheless, I recall finding it curious at the time to learn that the can fabrication operation in the adjoining room was a separate company from the bottling franchise that stored, filled, and, ultimately, distributed the cans.

If youthful experience colors one's perceptions, it would seem that I, rather than Professors Miwa and Ramseyer, should be the one expressing skepticism about the importance of relationship-specific investments in organization decisions. Whereas the separate ownership of can fabrication and filling operations, despite the location specificity of the facilities, at least appears to conflict with the relationship-specific investment hypothesis, the circuit factory in which one of them once worked exhibited neither specific investments nor special governance arrangements -- exactly as the theory predicts! What's more, as I will argue below, the same can be said of virtually all of their evidence.

To be fair, Miwa and Ramseyer neither intend nor claim to refute the relationship-specific investment hypothesis. On the contrary, they say the theory makes sense.(1) And the evidence -- their "empirical vacuum" derogation notwithstanding -- does, by their own account, provide "substantial evidence of the relation between RSIs and governance."(2) The problem, as Miwa and Ramseyer see it, is that settings in which investments are large and specific enough to affect the choice of governance arrangements are so few and anomalous as to render the theory irrelevant for understanding the organization of everyday industries. Modem manufacturing simply entails far fewer relationship-specific investments than transaction-cost economists would have us believe. And where such investments do arise, reputation and standard market contracting adequately deal with the associated problems. As a logician might put it, the theory, though valid, lacks existential import.

Miwa and Ramseyer are not the first, nor even the most prominent, scholars to express reservations about the theory's domain. Indeed, no less an authority on the transaction-cost determinants of organizational form than Ronald Coase has questioned the importance of specific investments on repeated occasions -- most recently, in his extensive critique of Klein, Crawford, and Alchian's analysis of General Motors' 1926 acquisition of Fisher Body.(3) From his observations of the U.S. automotive industry in the 1930s, Coase saw that "suppliers were often unwilling to sell too great a proportion of their output to one customer"(4) for fear that that customer might take advantage of its position "to drive down the price to a level which yields no return on such investments."(5) Further inquiry, however, led him "to doubt not the reality of this risk, but its importance"(6): "Even though the costs of contracting [may] increase more than the costs of vertical integration as assets become more specific and quasi rents increase, vertical integration will not displace the long-term contract unless the costs of contracting become greater than the costs of vertical integration ...."(7) And that, Coase adds incisively, "might never happen for any value of quasi rents actually found."(8) The reason is that "the propensity for opportunistic behavior is usually effectively checked by the need to take account of the effect of the firm's actions on future business" and by "contractual arrangements."(9)

The importance of relationship-specific investments to organization is ultimately an empirical question and, as such, Miwa and Ramseyer's claims require an empirical response.(10) Good empirical research is always rooted in theory, however. So before turning to the evidence, some consideration of the underpinnings of the relationship-specific investment hypothesis is in order.

1. THE THEORY

   Miwa and Ramseyer summarize the relationship-specific investment hypothesis as follows: "According to this intuition, the scope and size of RSIs can directly affect the governance arrangements firms choose. Whether business partners negotiate long-term contracts, spot contracts, equity investments, franchise arrangements, or even mergers can depend vitally on the RSIs at stake."(11)

   Though perhaps intentionally hyperbolic, this characterization nevertheless overstates the role that specific investments play in the theory. To be sure, relationship-specific investments have been prominent, arguably even central, in the "operationalization" of transaction-cost reasoning. But the effect of specific investments on organizational form in the theory is neither direct, exclusive, nor decisive. First, contrary to Miwa and Ramseyer's portrayal of the relationship-specific investment hypothesis as predicting the adoption of "extra-contractual governance arrangements," transaction-cost economists have always regarded contracts and other organizational arrangements as alternative responses to the appropriability hazards engendered by specific investments.(12) Klein, Crawford, and Alchian, for example, expressly identify vertical integration and contracting as substitute devices for curbing opportunism: "The primary alternative to vertical integration as a solution to the general problem of opportunistic behavior is some form of economically...