The Ownership of Japanese Corporations in the 20th Century

The Japanese insider ownership system began to fall apart approximately twenty years after it came into operation at the beginning of the 1970s. In our paper, The Ownership of Japanese Corporations in the 20th Century, which was recently made publicly available on SSRN, we suggest that the insider system emerged in the first place because the alternative institutions for promoting outside ownership failed. The problem was not with the legal framework, which was relatively strong in Japan. Instead, the failure was due to the absence of institutional reputational capital in equity markets equivalent to that embedded in the business coordinators and zaibatsu earlier in the century. The first point that this brings out is that the destruction of institutions, such as zaibatsu, can be serious in terms of economic performance. The second point is that the creation of new institutions of trust to replace previous institutions is complex and not readily achieved by design.

What does this imply for Japan and other countries in the 21st century? Having experienced a decade of deleveraging and restructuring, Japan is now beginning to emerge with what looks like outside ownership. Some of the past hostility to the emergence of a market for corporate control appears to remain, but there is one important difference from the experience of the 1960s and that is the acceptance of foreign ownership. Much of the shareholding in Japan today comes via foreign financial institutions alongside the emergence of some indigenous institutions. In our paper, we show that outsider ownership has risen from 37.4% in 1990 to 55.9% in 2009, and foreign investors’ share has risen from 6.1% in 1990 to 19.5% in 2009. Foreign institutions have the advantage over domestic ones in that they are not so readily subject to domestic capture and influence but might not be as committed to Japanese investment and growth as the domestic institutions of the past. We show that, despite the growth of foreign investors, they have not acquired the largest shareholdings: the most significant shareholders remain domestic corporations, insurance companies, investment trusts, and pension funds.

Japan, therefore, has more outside investors but neither behaves like an outsider system in the Anglo-American sense nor has the institutions of trust that characterized Japan in the first half of the 20th century. The recent case of Olympus is an illustration of the conflicts that this halfway house can create between the two parties and the potential vulnerability of the system to the problems that eroded outside ownership in Japan in the 1960s and 1970s.

The breakdown of trust in post-WW2 Japan has wider implications for economies outside of Japan and for the development of institutions of trust. As described in Mayer (2013), commitment and trust have fundamental implications for patterns of corporate finance and investment around the world. In the absence of adequate contractual protection, the terms on which different parties can trade with each other and the costs of employment and finance are dependent on the ability of parties to commit to each other. This article has referred at several points to how countries develop a variety of institutional mechanisms for coping with this and how these arrangements vary over time and by location.

The main lesson to be learned from Japan is one of caution in seeking to import institutional structures or regulatory practices from elsewhere. Institutions of trust take time to establish and embed in local arrangements. They are highly country- and context-specific, and laws and rules that function in one country may be inadequate or inappropriate in another. This is particularly relevant to emerging markets that are currently seeking to reform their corporate governance arrangements, most notably China and India. The Japanese experience should be a reminder to us of how little we know about institutional and legal design and how cautious we should be in making policy recommendations about it.

The full paper is available for download here.