|
Article Excerpt
Introduction
Utterback and Abernathy's theory of the product cycle has been highly influential among scholars of technology management and strategy (Utterback and Abernathy 1975, Abernathy and Utterback 1978). According to this theory, firms compete primarily on alternative product designs and positioning early in a product's history, whereas cost becomes the most important basis of competition once a dominant design emerges. Extending these ideas to the industry level, Porter (1980) discussed the implications for the ways in which firms' strategies must change over the industry life cycle. Transaction-cost theory has also been influential in technology management and strategy, but has argued that cost reduction--economizing on transaction costs in particular--is a superior strategy in general, without distinguishing between stages of the industry life cycle (Williamson 1993). In this paper, we seek to reconcile transaction-cost and industry life-cycle theories by investigating whether aligning transactions with governance modes in accordance with transaction-cost prescriptions has a uniform impact on firm survival across stages of the industry life cycle, or whether its impact varies across these stages in ways consistent with life-cycle theories.
We measure transaction alignment and misalignment by observing whether or not firms in the early U.S. auto industry were following transaction-cost prescriptions when deciding whether to make or buy a key piece of an automobile's technology: its engine. Our study thus aims to improve our understanding of whether, and how, a firm's technology strategy must adjust as the industry in which it participates evolves, and therefore to contribute to the literature on the contingency effects of technological evolution on firm strategy. This study is also of interest in light of recent literature on the performance effects of transaction misalignment. That literature finds that firms in the trucking industry that were not organized in accordance with transaction-cost principles--i.e., that were "misaligned"--tended to suffer greater organizational mortality and ultimately lower profits (Silverman et al. 1997, Nickerson and Silverman 2003). Similarly, firms in the overnight delivery industry that were misaligned according to transaction-cost theory tended to display slower delivery times (Nickerson et al. 2001). In this paper, we examine the relationship between the performance effects of transaction misalignment on the one hand, and the industry life cycle on the other--a relationship that has yet to be addressed.
Our approach is to study the determinants of organizational mortality in the U.S. auto industry during the critical transition years of 1917-1933. This period is of particular interest because its first six years were marked by high rates of entry and exit, after which time a dramatic "shakeout" began to occur. This shakeout initiated a long secular decline in the number of firms that continued until the post-WWII period. The 1917-1933 period is thus a useful laboratory for studying the effects of transaction-cost economizing in the early, fragmentation stage of the auto industry life cycle and the later shakeout stage.
We measure the extent to which firms were misaligned over the life cycle by observing the degree to which they were following transaction-cost prescriptions regarding the procurement of engines over time. Engines are, of course, a critical component of automobiles because they have such a large effect on both total vehicle cost and performance. (1) Therefore, transaction-cost theory might well predict that a firm with an inefficient procurement policy toward engines would be putting its survival at risk. We used detailed information on the characteristics of automobile engines and procurement policies to develop a measure of "misalignment." We measure a firm as "misaligned" if it tends to produce standard engine types itself, and/or to source unique engine types from outside suppliers. By contrast, firms are "aligned" if they tend to produce unique engines themselves, and/or to source standardized engines from outside suppliers. Engine uniqueness is thus our proxy for asset specificity.
The main finding of our study is that while our measure of engine procurement misalignment did not make a significant impact on firm mortality over the 1917-1933 period as a whole, it had a significantly greater impact during the later shakeout period 1923-1933. We find that a one-unit increase in a firm's misalignment during the shakeout period--i.e., organizing one more engine transaction inefficiently--was more than twice as likely to lead the firm to exit than if the same increase occurred during the earlier period. The penalty associated with engine procurement misalignment was thus much greater in the shakeout period than in the earlier period.
Transaction-cost theory does not specifically address the possibility of such a differential impact of transaction misalignment on firm performance across stages of the industry life cycle. We argue, however, that the finding is consistent with theories of the industry life cycle, which have emphasized the ways in which selection forces wax and wane over the life cycle. Indeed, we suggest that the findings are consistent with models of the industry life cycle such as Klepper (1996), as well as with the arguments of Abernathy and Utterback (1978) concerning the role of dominant designs in life-cycle transitions. We conclude that combining transaction-cost theory with theories of industry life cycles produces a deeper, more nuanced understanding of the dynamics of firm strategy than either approach alone.
We begin by discussing the assumptions about firm survival, selection, and efficiency in transaction-cost theory. We then discuss evolutionary theories, focusing on theories of industry life cycles. We then present our data, methods, and results. We conclude with suggestions for future research.
Transaction-Cost Theory, Efficiency, and Firm Survival
The essence of transaction-cost theory is that firms economize on transaction costs by making "discriminating alignments," in which they match the governance structure for a transaction with the underlying characteristics of the transaction (Williamson 1985). Transactions featuring greater bilateral dependence between the parties (generated most prominently by high-asset specificity) will tend to be organized internally to avoid the hazards associated with organizing them through arm's length contractual relationships. These hazards stem from potential opportunism, especially hold up, by a contractual partner. On the other hand, transactions featuring less bilateral dependence will be organized through contractual relationships that incorporate more or less elaborate safeguards against such hazards, depending on the level of bilateral dependence (Williamson 1991). Mediation by the market is more efficient for these more generic transactions because bureaucratic costs associated with internal production are avoided, and because generic transactions are more likely to benefit from economies of scale that may be available to suppliers from aggregating the demands of multiple buyers.
As a positive theory, transaction-cost theory thus assumes that efficient discriminating alignments will be observed much more frequently than misalignments because misalignments "invite their own demise" (Williamson 1996). This assumption stems from the idea, articulated most prominently by Alchian (1950) and Friedman (1953), that market pressures act to "select out" inefficient firms in a manner similar to that of natural selection. Williamson (1985, pp. 129-130) writes:
Natural selection forces do not always operate quickly.... Firms that are buffered against product market rivalry ... and against capital market discipline ... can postpone the reckoning. But these would appear to be the exception rather than the rule. Where incumbent managements are not pressed to adopt the new procedures by economic events, successor managements, often in conjunction with the appointment of a new chief executive, commonly will.
Transaction-cost theory thus presumes that in general, a firm cannot rely on market power to make up for its inefficiencies. As Williamson (1993) argues, "... most firms lack market power of the kind that is routinely assumed by the strategizing literature" (p. 80). Any advantages that a firm may possess that provide it market power are assumed to be temporary because rivals can in most instances imitate them in a process that Schumpeter (1947, p. 155) termed "handing on." Hence, economizing, rather than strategizing, becomes transaction-cost theory's prescription for superior financial performance. This presumption in transaction-cost theory that market forces tend to select firms that economize on transaction costs and achieve efficient alignment is made without regard to the industry life cycle. It therefore suggests the following hypothesis:
HYPOTHESIS 1 (H1). Firms whose key transactions are misaligned in transaction-cost terms will display greater firm mortality than those whose transactions are aligned, regardless of the stage of the industry life cycle.
An important question that remains, however, regards the time required before the market forces that promote efficiency have their effects. How long before a truly inefficient policy will be eliminated, either through the exit of the firm promulgating it, or by adjustment of the policy by the firm? Williamson (1985, p. 23) briefly suggests that efficient transaction-cost economizing might occur over 5-10 years, although the time scale required to achieve efficient organization is rarely addressed in empirical studies. One exception is Nickerson and Silverman (2003), who found that institutional constraints on firms in the U.S. trucking industry significantly slowed their efforts to economize on transaction costs after deregulation.
Evolutionary theories of economic organization, on the other hand, have paid explicit attention to selection processes and their implications for efficiency. For example, Nelson and Winter (1982) show that if firms tend to search locally for solutions to problems, rather than continuously optimizing, then firms pursuing nonoptimal solutions (e.g., nonoptimal capacity utilization rules) can survive in equilibrium. Moreover, because selection occurs at the level of the firm, rather than at the level of the transaction, evolutionary theory suggests that firms may carry inefficiently organized transactions along with them for some time. Winter (1988,...
|
