Amongst the events that predictably lead to demands for government action are business failures and corporate scandals. Demands for government action to improve corporate governance are, however, based on a dual mistake. They wrongly presuppose that the problems have been caused by a lack of sufficient regulation, and they erroneously assume that government regulation can make things better.

Experience has shown, however, that all such pronouncements need to be considered sceptically. Best practice rather than regulation may indeed be the government’s publicly preferred starting point. But judging from their past performance, it is unlikely to be the preferred – or the actual – outcome.

Today there are suggestions that ‘best practice’ guidelines should be enforced by the Financial Services Authority, a government agency with sweeping powers (though less accountability). The danger of ‘regulatory creep’, in which guidelines are transformed over time into mandatory requirements, is all too real.

The likelihood of regulation is exacerbated because many of the respondents who participate in corporate-governance reviews regularly advocate regulatory strictures. Some seek government regulation to make directors serve a social or political agenda other than the direct interests of shareholders. Others, including many who ostensibly represent business interests, would enlist government support to prescribe some favoured model of best practice, regardless of shareholders’ own preferences.

Both such approaches are mistaken: by its very nature, corporate governance should be the responsibility of the shareholders themselves. There is much that could and should be done to increase the effectiveness of non-executive directors, but it should be up to the shareholders of each corporation to determine the rights and responsibilities of their directors, both executive and non-executive.

A Genuine Market for Corporate Control

Most of the changes that are needed to improve corporate governance do not involve any government action. They can and should be provided by the marketplace itself. In the US, many firms have already voluntarily chosen to show the cost of granting options as an expense in their profit and loss accounts, and investors are voting down far more option plans than they did five years ago. The best way to bring about beneficial changes would be to encourage maximum experimentation in the marketplace, and to allow different forms of corporate governance to compete for the support of investors.

Contrary to popular belief, the free-rider problem is not an insuperable barrier to such non-regulatory solutions. An investor acting independently would unilaterally incur the costs of action from which all shareholders would benefit. But if the costs of the corporate governance action were for the account of the corporation, they would be automatically shared by all shareholders, in proportion to their shareholdings.

The kinds and amounts of shareholder support paid for by the corporation could be one of the ways in which companies competed in a genuine ‘market for corporate control’. The ‘market for corporate control’ conventionally refers to the use of takeovers to transfer corporate ownership. But it can be used more broadly, to refer to the market in which companies compete for shareholders, and investment managers compete for funds, in part on the degree and kinds of accountability they afford to owners and investors.

Subjects of such competition might include, for example, the nature of the constitutional corporate objectives, the extent to which strategic and operational matters required shareholder approval, company election proced-ures, the independence and quality of directors, the extent and quality of performancerelated remuneration, and the types of disclosure and audits.

Companies could compete for shareholders on the different ways in which their directors were selected and elected. Or experiment with the levels of disclosure they required from directors (including their reasons for resignation), with the personal or professional or other qualifications they required directorial candidates to have, and with the number of times that directors could be re-elected. Or they might try varying the responsibilities of the lead non-executive director, and having differing percentages of non-executive directors.

Corporations might even vary in the extent to which they allowed their directors to be executives of other companies. Shareholders might seek to employ the services of ‘professional directors’ – directors who would not be the executives of any firm, but who would be chosen specifically for their ability to safeguard shareholder interests. Even if such directors acted in a non-executive capacity for more than one firm, they might be less prone to the damaging conflicts of interest which now typically arise between executive directors and owners.

Companies could also compete for shareholders on the basis of the different sorts of financial and structural support offered to directors. Companies might reimburse some or all of the expenses that directors incurred in investigating company matters, and in taking specialist advice. Companies could also vary the extent to which directors’ (and advisors’) liabilities were indemnified, contractually limited, or covered by errors and omissions insurance at company expense.

But these are just a few of the very many different ways in which companies might compete in respect of corporate governance. One of the many advantages of free markets, is that they elicit innovative solutions to problems as they arise in all their real-life variety and complexity. Markets also effectively test those solutions and efficiently disseminate best practice.

The best way to ensure good corporate governance is to allow shareholders the greatest possible freedom to control their own corporations. The value of doing so is clear. According to a recent analysis of 1,500 stocks by the (US) National Bureau of Economic Research, companies with the most restricted shareholder rights had annual earnings and valuations between 1990 and 1999 that were almost 9% lower than companies with the fewest restrictions. Shareholder freedom is associated with both good corporate governance and superior corporate performance.