What is a company? What is its purpose? Who owns it? What should be its goal? These are four closely related questions. The answers society gives will determine the future of capitalism. Those answers are also hard to find, because the limited liability company, though an extraordinarily successful institution, is a hybrid institution: it is in the market economy, but not of it.
So what is a company? As the late Ronald Coase, a Nobel laureate in economics, taught, if one wants to organise production and sales of complex products and services, a (semi)-permanent institution will outperform an array of small businesses forced to deal with one another through market contracts. Companies exist because hierarchies -- 'command and control' -- beat markets. The genius of the company is to import the hierarchical structure of older institutional forms -- bureaucracies and armies -- into the market economy.
The advantage companies possess is the mirror image of the costs of creating and monitoring a vast number of detailed contracts under uncertainty about future requirements. Organising economic processes successfully often requires the scale of an army and the longevity of a tortoise.
A life insurer is of little use if it will not meet its obligations 80 years hence. A maker of jet engines is of little use if it will be unable to service and replace its engines over their lifetime. A car manufacturer is of little use if it is unable to use what it has learned from today's models to build tomorrow's. A company is built upon a set of relational, or implicit, contracts.
These say something like the following: 'we will purchase your services for a more or less indefinite period; we will look after you; and, in return, you will do what we tell you'.
The corporate form was a brilliant innovation. But, like many innovations, it has created new challenges. In order to work, such entities need vast amounts of capital. In the beginning, these funds are mainly provided by shareholders. Thereafter, they mostly come from retained earnings and further borrowing.
In return for providing risk capital, shareholders are entitled to the stream of corporate profits, whether paid out as dividends or share buy-backs, or retained within the company. If things go well, shareholders have a profitable investment. Under limited liability, if things go badly, they will lose their investment, but no more than their investment.
What is the goal of a company? It is an institutional mechanism for adding economic value. This is the social function of any and all companies, subject to an important proviso: the company should not add value by inflicting negative externalities, such as environmental degradation. Society has given the corporate form important privileges. In return, society has a right to expect obedience to the law and a measure of decency: even if it is not illegal, dumping toxic waste or rigging one's affairs so as to pay minimal taxes to the jurisdictions that provide the environment within which the company can generate its profits is indecent. It is freeloading. A company adds value by organising its array of assets -- skills, knowledge, values, traditions and loyalties -- into an effective and flexible whole. It is succeeding if it prospers in a competitive market. It fails if it does not. It is almost always possible for a company to make higher shortterm profits by sacrificing investments that allow it to prosper in the long run. But companies were created in order to endure. Ownership can change hands, be fragmented, or re-unified, without significant impact on the company itself. Caring about its future is part of a company's raison d'être. Who owns and controls a company? The formal answer of economists has been that shareholders own and control a company. This is simplistic. Shareholders do not own companies in the normal sense of ownership. They cannot walk into the property they 'own' and demand that 'their' employees do certain things, as they could in a shop or a farm that they owned. Indeed, all they can do -- and, for the most part, only with difficulty -- is to help change management by voting or selling their shares. This is a highly qualified form of ownership. A closely related question is whether shareholders should control companies. In the English-speaking countries, the answer is immediate and unambiguous: yes. Almost anywhere else, it is, to a greater or lesser extent, no. The difference is more intellectual and cultural than legal.
In Germany or Japan, for example, shareholders have a right to consideration, but not to control. Control rests, instead, with management, whose responsibilities are to the company as an enduring entity. In such countries, the answer to the question of who owns a limited liability company is that nobody really does, any more than someone can own, say, a university. Is it clear who is right? No, because, as noted above, the limited liability company is a hybrid entity. The arguments in favour of shareholders' control rights are two: first, shareholders are the residual claimants and so bear the residual risks; second, somebody has to have the capacity to evict management. If it is not the shareholders, who should it be? The advantage of an Anglo-Saxon shareholder-governed company is that fundamental shifts in management and direction are relatively simple to make.
The argument against this view is that in a shareholder-controlled company some valuable implicit contracts -- for example, willingness to undertake unpleasant and inconspicuous tasks in the hope of reward in subsequent employment, or willingness to invest heavily in one's ability to work within a specific team -- will not be made. The explanation for this difficulty is that contrary to the normal view, shareholders do not bear the largest proportion of the uninsurable risk in a company. Shareholders can diversify their portfolios and usually do, thus insuring themselves against company-specific risk. The people who cannot do this are those who make large investments in company-specific skills, knowledge and relationships, protected, they hope, by the implicit contracts that justify the corporate form. This category may include both long-serving workers and dedicated suppliers. Without a measure of control, such implicit contracts are unlikely to prove worth the paper they are not written on. If those who are called upon to invest in implicit contracts know they have limited, if any, influence on those who control the company, they must then also know that opportunistic default is possible and, in the right circumstances, certain. They will rationally not make those investments. In such cases, shareholder control actually undermines corporate success. Finally, what should the operational goal of a company be? Should it be to maximise shareholder value, defined, as it should be, as 'maximising the present value of free cash flows from now until infinity, discounted at a rate that reflects the risks of these cash flows'?
The answer is yes if and only if two conditions hold. The first is that the prices of the goods and services (including labour services) that a company buys and sells reflect their true social costs and benefits. The second is that the goal can be made operational in a beneficial rather than a perverse way. Neither is plausible. The biggest difficulty with the first of these conditions lies in the labour market. The social costs of lay-offs are not internalised by the company. But that can be dealt with by imposing a regulation or tax. A far greater difficulty is the second. Shareholders do not know what policies will maximise the present value of a company's free cash flow to infinity. In fact, they have virtually no idea. Most of them also have no incentive to invest in the relevant knowledge either. Unless they own a large share of the company, they will suffer from the collective action problem described by the late Mancur Olsen: the costs of investment in knowledge is borne by each of them, but the benefits are shared amongst all. As a public good, knowledge is always undersupplied in the market. This is, note, not the same as saying stock markets are inefficient in their ability to evaluate the prospects of one company vis-à-vis another (though they are certainly inefficient at the aggregate level). That is because shareholder ignorance is likely to shape what the company does. It is a self-fulfilling prophecy.
Martin Wolf is Associate Editor and Chief Economics Commentator, Financial Times