Revisiting a Classic on Legitimation.  Is it Legitimate?

Rao, Hayagreeva. "The social construction of reputation: Certification contests, legitimation, and the survival of organizations in the American automobile industry: 1895–1912." Strategic Management Journal 15.S1 (1994): 29-44.

The central claim of the paper is that winning third party certification contests leads to greater legitimacy / reputation which then leads to survival. Rao, 1994 has been cited over 1,000 on Google Scholar, and enjoys over 300 Social Citation Index cites. The paper is generally cited as evidence of one or both of the following theoretical links: 1. winning third party certification contests leads to legitimacy/reputation 2. legitimacy/reputation leads to survival or 3. both links.

The paper’s results have been used to build further theory and may have also been used to advise entrepreneurs regarding worthwhile activities in the early stages of their ventures.

Empirical Context: The paper measures survival of US automobile manufacturers at the start of the industry in 1896 through 1912. Data on automobile races and competitive trials are collected from an early trade magazine. Other control variables, primarily at the industry level are also collected.

What is good: The paper invokes an historical setting in its early stages of the industry. Learning from history should be commended. The paper also invokes an interesting set of hypotheses surrounding the role of achieving mind-share of the public and stakeholders in the survival of new enterprise.

What is less good: The paper has several problems that call into question its empirical contribution.

●      An empirical link is shown between winning an automobile race and firm survival. Legitimacy is never measured or observed, but rather assumed to accrue to winners of the races. The lack of evidence in support of the causal mechanism calls raises the specter of alternative mechanisms besides the accrual of legitimacy/reputation causing the survival.

            Endogeneity:

The strong critique: Both winning and survival may stem from an unrelated third factor, such as managerial competence: Better companies’ cars may have been more likely to enter races. Because these companies are better, they would have been more likely to survive. Thus, both winning and survival are symptoms of better management or technology. In this scenario both legitimacy and winning races play no role.

The weaker critique: The decision to enter a race is assumed to be that of the automobile entrepreneur. However, the racing industry is an emerging industry itself. Many if not most contests awarded prizes to winners, and trade press coverage devoted significant ink to the drivers. Thus, the decision to enter a race was not always that of the entrepreneurs, but of entrepreneurial enterprises designed around races. Winning may not have been important, but rather participation. Since one must participate to win, winning may be proxying for participation. Since very inept companies may not have been able to enter a race, the results may be simply picking up the quality differences between those firms that were able to enter races and those that were not.  But in this case, participation would not likely have resulted in the same legitimacy and reputation enhancing benefits and winning, thereby undermining the theoretical contributions of the paper.

It is unclear what the marginal contribution of invoking the concepts of legitimacy over the straightforward winning as a Spencian signal.

Conclusion: There is a disconnect between the empirical exercise and the theory. While the results are consistent with the theory, the mechanisms in the theory are assumed, not demonstrated empirically. Moreover, alternative mechanisms are as plausible as those promoted in the paper. As such, citing articles are incorrect when citing the paper as evidence of the the role of legitimacy and reputation in the survival of firms.