Ownership and trust
In unique research, Julian Franks, Professor of Finance at London Business School, has looked back at over 100 years of data on corporate ownership. The research (carried out with Colin Mayer and Stefano Rossi) found that relationships of trust were a crucial historical ingredient in the ownership of firms
Listen now | View all podcasts
Owners often knew each other and lived relatively locally. But does this have any relevance in the era of international corporations with hugely dispersed shareholders? Professor Franks argues that trust is not an historical concept. Indeed, organisations such as eBay are built very openly on notions of trust - as, too, is the London Stock Exchange whose motto is "my word is my bond". Relationships of trust are as important as ever.
One of the best known facts about corporate ownership is that ownership of large listed companies is dispersed in the U.K. and U.S. and concentrated in most other countries. In more than 50 per cent of European companies there is a single voting block of shareholders that commands a majority of shares. In contrast, in the U.K. and U.S. it is less than 3 per cent.
Two theories have been proposed to explain these differences. The first is that U.S. legislators responded to a populist agenda in the 1930s by limiting the power of large financial conglomerates. The second argues that concentrated ownership is a response to inadequate investor protection. In the absence of protection, investors seek to protect their investments through large share blocks.
In both these theories dispersed ownership is associated with strong investor protection. The difference in ownership concentrations can be attributed to weak investor protection in Continental Europe and strong investor protection in the U.K. and U.S.
Differences in legal structures are deep rooted with a long history. One would expect differences in investor protection also to have a long history, particularly in the U.K. where common law originated. But this is not the case. At the beginning of the twentieth century, the U.K. was actually devoid of anti-director rights provisions and protection of small investors.
Our research examined how the ownership of a panel of firms evolved over a hundred years and the extent to which law contributed to that evolution. The research was made possible by the existence of an unusually rich source of data in the U.K. For more than a century, Parliament has required companies to deposit information, including accounts and a register of shareholders, at a central depository open to the public.
From this, we selected three samples - one from companies incorporated around 1900 that have been in existence since then, a second from firms incorporated at the same time but which are no longer in existence and a third from companies incorporated around 1960 and still in existence.
We found that ownership of the sample of firms incorporated around 1900 was rapidly dispersed with the shareholdings of inside directors more than halving over the 40 years to 1940. Interestingly, when investor protection was finally strengthened in the second half of the century, it had little effect on either levels or rates of ownership dispersion. Ownership of well-established companies was already dispersed and rates of dispersion of newly incorporated firms, for example of the sample of firms incorporated around 1960, were similar to those of firms incorporated at the start of the century.
An obvious question this raises is how ownership dispersion could have occurred in the absence of investor protection? Our conclusion is that informal relations of trust, rather than formal systems of regulation, promoted the development of capital markets and dispersion of ownership.
In trust we trust
In economics, trust is associated with reputation and commitment between players engaged in repeated games with each other and with the emergence of particular institutional arrangements. In law, a distinction is drawn between contracts where there are reciprocal arrangements between parties and trust law where there are unidirectional agreements between beneficiaries and trustees. Where there is no reciprocity in trusts on the part of the beneficiary there is a greater fiduciary responsibility on the other party, namely the trustee. Trust arrangements carried over to company law because of the analogous relationship between directors and their dispersed shareholders with that of trustees and their beneficiaries.
Our concept of trust in a corporation draws on both economics and law and refers to actions by the director of a firm that are dictated neither by contract nor by regulation. By trust we mean conformity with accepted norms of behaviour in the absence of explicit incentives or penalties to do so. Trust can derive from repeated interactions, from moral and ethical codes, or from social conventions and networks.
Our research found informal relationships of trust existed in a number of ways. For example, we found that the local nature of stock exchanges played an important role in the development of trust between directors and investors. Ordinary shareholders lived close to the company's city of incorporation and its board of directors and relations of trust flourished as a result.
In addition, relations of trust created the conditions in which interactions between firms and investors were repeated and where directors had incentives to sustain their reputations among local communities. For example, looking at the way in which firms made offers to shareholders in takeovers and mergers, we found the same price was offered to all shareholders even in the absence of specific regulatory or contractual requirements.
Among the most interesting aspects of the research was the comprehensive data gathered on the shareholder records of twenty-six companies in 1910. We recorded the names and addresses of shareholders, and calculated the distance between the shareholder's address and the city of incorporation. In the 1910 sample, the average number of shareholders was 320; the mean distance between ordinary shareholders' addresses and firms' cities of incorporation was 52.2 miles and the median 15.4 miles. The proportion living within six miles of the city of incorporation was 56 per cent.
Geographical concentration was remarkably high even where ownership was dispersed. For one company, GKN, we found that geographic dispersion sharply increased over the period 1910 to 1950. The average distance of the shareholders to the head office in Birmingham rose from 69.5 to more than 150 miles.
Looking back, it was clear that shareholders had little recourse in courts but much influence in the communities and local markets of which they and their firms were a part. Even as it became dispersed, ownership remained geographically concentrated and directors were concerned to maintain their reputations among local investors. All shareholders, including insiders, sold their shares to acquiring firms at the same price and directors of target firms were frequently promoted to the boards of merged firms. Eventually, as local relations of trust became harder to sustain then formal investor protection emerged to substitute for them.
Was the U.K. an isolated case? Our previous research found a surprisingly high level of ownership dispersion in Japan at the start of the twentieth century, with a large number of listed companies and a very large number of shareholders. Again ownership dispersion cannot be attributed to investor protection. It was almost as weak in Japan as it was in the U.K. But there were differences. Takeover waves were less in evidence in Japan and the purpose to which new equity was put was primarily to finance internal rather than external expansion.
Again ownership dispersion appeared to have more to do with informal relations of trust than investor protection. However, the institutional arrangements that fostered trust were not the same. Local stock markets were not prevalent in Japan. Individual promoters of shares took on a more active role in the oversight and management of firms in Japan than in the U.K. As a result, different mechanisms evolved to establish the basis of trust on which ownership could become dispersed.
Trust now
How is all of this relevant to today's turbulent and truly global trading environment? Well, the motto of the London Stock exchange remains "My word is my bond", as it has been for centuries since brokers traded with each other literally across the floor of the exchange. The globalization of the capital markets has changed all that. Trust is usually no longer so intimate an experience. But, we still need to create or generate new mechanisms of trust.
One trust generating mechanism using modern technology is eBay. Here anonymous buyers and sellers are able to access the views of past purchasers and sellers, when selling or buying an item on the site. If a purchaser has previously sold ten items, the levels of satisfaction of past purchasers are stored and made available to the next buyer. Trust rules.
The relationships of trust discovered by our research are timeless. This means there is still a home bias for investors, a preference for buying investments in one's own country and even in one's own state or region. This home bias reflects a view that common cultures, common laws, and geographic proximity are more likely to produce honest and fair outcomes between investors.
Julian Franks (jfranks@london.edu) is a Professor of Finance at London Business School.