Abstract: Two of the most prominent Japanese corporate governance scholars, Professors Miwa and Ramseyer ("M&R"), have recently published numerous articles and a book setting out their contrarian free-market theory of Japanese corporate governance. According to their theory, contemporary Japanese corporate governance is, and always has been, driven by free-market forces and not government incentives. M&R's theory is enchanting in its simplicity and universality, as it uses standard economic theory to provide a single, and seemingly logical, solution to a myriad of complex legal, institutional, historical and cultural conundrums that have challenged observers of Japanese corporate governance for decades. Unfortunately, M&R's theory is also incorrect. This article demonstrates why and by doing so provides evidence against the broader convergence theory that looms large in the comparative corporate governance literature.
M&R's theory fails to explain the systematic lending of trillions of yen by Japanese banks to "loser firms," at below-market interest rates, to rescue them from bankruptcy, throughout the lost decade (1990 - 2003). According to M&R's free-market theory, lending to loser firms at below-market rates is not a rational, optimal, or credible governance strategy. Therefore, to claim that such behavior systematically occurred in Japan's banking system for over a decade would be to create a myth.
A myth it is not. Empirical and case study evidence demonstrates that Japanese banks did in fact systematically lend trillions of yen to loser firms at below-market interest rates to rescue them from bankruptcy. This paper reveals the matrix of institutional incentives that made it a rational strategy for Japanese bank managers to engage in such seemingly irrational behavior. The result is that unique institutional incentives, and not universal free-market forces, drove Japanese corporate governance-which is weighty evidence against the broader corporate governance convergence theory.
Academic theories are attractive-especially to academics. When the theory is simple, contrarian, and championed by eminent Tokyo University and Harvard professors, it is almost irresistible. However, when it fails to make sense of reality, it quickly loses its appeal. So, despite the temptation to embrace Professor Miwa and Professor Ramseyer's ("M&R") new theory of Japanese corporate governance,1 that finds decades of research to have constructed "a myth," I resist.2
The story painted by M&R is enchanting in its simplicity and universality. In their world:
Whether in the United States or in Japan, firms raise funds in competitive capital markets, and buy and sell in competitive labor, service, and product markets. Whether here or there, in order to survive, they will need good governance schemes.... The scheme they pick will vary from firm to firm. The fact that they will pick the optimal scheme or die will not.3
It all sounds logical, because it is-unless unique and perverse institutional incentives, and not free-market forces, drive corporate governance.4 In which case, the incentives for "bad governance" and "suboptimal schemes" may be greater than those for "good governance" and "optimal schemes." Perverse it is, but mythical it is not.
The simple fact that taints M&R's conclusions is that Japan-indeed, every country-is unique and that uniqueness matters in corporate governance. Japan's-again, every country's-unique institutional framework provides incentives that drive the decisions of corporate executives.5 The market in Japan-again, every country's market-is uniquely imperfect in the face of exogenous institutional forces. In theory, economic efficiency guides the "invisible hand" to wash away differences between disparate corporate governance systems. In practice, however, institutional forces combine with imperfect markets to create a plethora of unique incentives-some of which lead firms to take suboptimal inefficient actions that M&R's theory, which is based on the supremacy of free-market forces (as opposed to institutional incentives), would not predict. …