The ownership of firms

Who owns a firm and why? Investor-owned firms are the most common form of ownership for large-scale enterprises in most of the world, but in a market economy there are a number of other forms of ownership which are commonly utilised. We see, in many sectors of the economy, producer cooperative, customer cooperatives, nonprofit firms and mutual enterprises. Why? What determines who owns a firm?

This is a question addressed by Hansmann (1988, 1996, 2013). To understand Hansmann's theory of firm ownership we first need two definitions. Hansmann defines a firm's owners to be those persons who have two formal rights: the right to control the firm and the right to appropriate the firm's profits. That is, owners have control and income rights over the firm(1). Compare this to the property rights approach to the firm which see a firm's owners as those who have just control rights over the firm. Note also that this definition means some firms do not have owners. In nonprofit firms the control rights are separated from the income rights, no one can receive the residual earnings, meaning such firms are ownerless. A second useful definition is that of a firm's "patrons". Patrons are those persons who transact with the firm either as purchases of the firm's outputs or as suppliers of factors of production to the firm.

Significantly nearly all large firms are owned by persons who are also patrons.

"In principle, a firm could be owned by someone who is not a patron. Such a firm's capital needs would be met entirely by borrowing; its other factors of production would likewise be purchased on the market, and its products would be sold on the market. The owner(s) would simply have the right to control the firm and to appropriate its (positive or negative) residual earnings. Such firms are rare, however" (Hansmann 1988: 272).

But

"[i]f a firm were entirely owned by persons who were not among the firm's patrons, then all the firm's transactions involving inputs and outputs would take the form of market contracting. Although feasible in principle, in practice this is likely to be quite inefficient. Market contracting can be costly, especially in the presence of one or more of those conditions loosely termed ``market failure"-for example, where there is an absence of effective competition, or where one of the parties is at a substantial informational disadvantage. [\dots] For the present we need simply note that, where these costs are high, they can often be reduced by having the purchaser own the seller or vice versa. When both the purchaser and the seller are under common ownership, the incentive for one party to exploit the other by taking advantage of market imperfections is reduced or eliminated. Assigning ownership of a firm to one or another class of the firm's patrons can thus often reduce the costs of transacting with those patrons-costs that would otherwise be borne by the firm or its patrons. To assign ownership to someone who is not among the firm's patrons would waste the opportunity to use ownership to reduce these costs" (Hansmann 1996: 20).

There are two basic types of relationship between firms and patrons, a "market contract" or "ownership". Under a market contract the patron interacts with the firm only through a contract and is not an owner. Under ownership the patron is also an owner of the firm.

This raises an obvious question of what are the costs of contracting and of ownership? We will briefly survey the most significant types of costs that arise with market contracting and ownership.

The costs of contracting include market power, "lock-in", asymmetric information and the risks of long-term contracting.

In some situations a firm can have market power with respect to some of its patrons. The affected group of patrons have an incentive to own the firm to avoid price exploitation. For example, if the firm has monopoly power with respect to its customers, the customers could takeover and reorganise the firm as a consumer cooperative. Or if the firm has monopsony power with regard to its suppliers, the suppliers could create a producer cooperative as a way to avoid receiving the monopsony price (Hansmann 1996: 24-5).

Monopolistic exploitation can occur after a person has been dealing with the firm for sometime, even if at the beginning of the relationship there was competition for which firm the patron would contract with. Such ``lock-in" can occur if the patron has to make transaction-specific investments, which have little or no value outside of the relationship, and the contract between the firm and the patron is incomplete so that not all important aspects of future transactions can be specified in an ex-ante contract. This leaves the patron open to ex-post exploitation by the firm. Assigning ownership to the effected patrons mitigates the effects of such lock-in. This can be, for example, one reason for worker ownership of firms where firm-specific investment is required by the workers (Hansmann 1996: 25-6).

Asymmetric information can increase contracting costs via both moral hazard and adverse selection. If, for example, a firm knows more about the quality of its products than consumers know, then the firm may offer lower quality products than it says it will. This is the "lemons" problem (adverse selection) made famous by Akerlof (1970). In such situations consumer ownership has the advantage that it reduces the incentive for the firm to exploit its informational advantage. Consumers will not exploit themselves. Another example would be where workers know more about their effort levels than the firm knows. This gives rise to moral hazard issues. Worker ownership could act to reduce the incentive for workers to act opportunistically, since as owners the workers suffer more if effort levels are low. They also are likely to be better at monitoring each others level of effort. Asymmetric information can also give rise to strategic bargaining. A firm's management has private information about the firm and the firm's patrons have private information about their preferences and opportunities. If the patrons do not own the firm then each group has little incentive to reveal their information. Each group can use their private information strategically when bargaining with the other and thus increase the costs of negotiation. Patrons ownership reduces strategic behaviour since it decreases the incentive to hide information or to take advantage of information the other party lacks (Hansmann 1996: 27-9).

When patrons can not credibly communicate their preferences the costs associated with contracting increase. Not knowing patrons true preferences the firm's management may not be able to determine the least-cost combination of contractual terms that satisfies the firm's patrons. In addition patrons have an incentive to misrepresent their preferences to their strategic advantage. Ownership by patrons reduces the conflict of interest between patrons and owners and thus makes communication easier (Hansmann 1996: 30).

It is often the case that firms have to treat all patrons in a given class the same even when individual patrons in that class have different preferences. In such a case market contracting may result in an inefficient compromise being made among the patrons' differing preferences. This is because market contracting satisfies the preferences of the marginal patron whereas efficiency generally requires the choosing of conditions that satisfy the preferences of the average patron(2), and the marginal and average preferences can be very different. When patrons own the firms and a voting rule - in particular the majority rule - is used to make decisions then the results tends to favour the median patron rather than the marginal. The median will not in general be the average, but it will likely be closer than the marginal. Thus patron ownership has an advantage in decision making when patrons preferences differ (Hansmann 1996: 30-1).

Long-term contracting comes with its own problems. One such issue is that transaction-specific investment can leave one of the parties exposed to opportunistic behaviour by the other. Another is how can the parties allocate specific risks between them. Yet another issue is how can the parties mitigate the problems of adverse selection that occur in insurance and related markets (Hansmann 1996: 27).

Next Hansmann considers the costs of ownership, which involve risk bearing, principal agent problems and collective decision making.

One group of patrons may be able to better bear the risks associated with owning the firm than others. They may, for example, be better able to diversify their investments. Assigning ownership to this group can bring important economies. This is a common argument for having investors as the owners of a firm. However, Hansmann argues that investors are not the only low-cost risk bearers and that the evidence suggests that the importance of risk bearing as a reason for investor-ownership is overstated (Hansmann 1996: 44-5).

Hansmann (1996: 35-8) divides agency costs onto two board groups, costs of monitoring the firm's managers and the cost of opportunism by the managers given they can not be perfectly monitored. For effective monitoring of the managers, patrons must: 1. inform themselves about the operations of the firm and 2. communicate among themselves to exchange information and made decisions and enforce those decisions upon the managers. Each of these activities is costly but the costs will vary among the different groups of patrons. Patrons who transact with the firm frequency, for important transactions, over a long period of time will, most likely, be better informed about the firm. The ease of organising patrons for collective action will also affect the costs of monitoring. Thus patrons who are physically closer to one another and the firm will have lower costs. The size of the group of patrons will also affect monitoring costs. Given the public good nature of monitoring the larger is the group, the smaller is any individual's share of potential gains from monitoring, and thus the smaller is the incentive to monitor. Therefore when the group of patrons is large it can be prohibitively expensive for the owners to undertake effective monitoring.

"In itself, this [costly monitoring] argues for smallest group of owners possible--preferably a single owner. The fact that, despite this, a large firm often has a very large class of owners therefore suggests that either or both of two things must be true. First, the costs of market contracting would be much higher under any alternative assignment of ownership. Second, the costs of managerial opportunism are modest even though the firm's owners cannot actively supervise the managers" (Hansmann 1996: 36-7).

Managerial opportunism can be thought of as being divided into two groups: self-dealing ("putting their hand in the till") and deliberate managerial inefficiency. There are a number of sanction on self-dealing that do not require collective action by the firm's owners. There are moral, contractual, tort and criminal actions that can be taken against a transgressing manager which, for the most part, limit self-dealing activities. While self-dealing may be limited, this does not mean that managers always work hard and make the best commercial decisions. Managers may not always be acting in the best interests of the owners. But Hansmann argues that for the types of people who win the winner-takes-all management type contest and make it to the top of the management hierarchy pride and moral suasion provides important motivation. Also the need for the firm to prosper so that a manager's career can be enhanced provides additional incentive for effort. But Hansmann, more controversially, argues that the potential gains from better monitoring are often exaggerated. There is one area in which Hansmann does think managers may be able to act against the interests of owners, excessive retention of earnings. Retentions provide managers with a buffer against adverse times, enhance the survival of the firm and increase the size of the firm the manager controls. But retentions are costly to owners if the rate of return on retentions is lower than that on outside investments or if retentions can not be recovered by the current owners. This is seen in some mutuals and cooperatives.

Hansmann explains that in most firms collective decision making is enacted via some form of voting scheme. But when the interests of owners are diverse such schemes involve costs. Such costs fall into two general groups: costs from inefficient decisions and costs of the decision making process, see Hansmann (1996: 39-44).

As has already been noted majority voting leads to the outcome which is preferred by the median voter while efficiency requires the outcome preferred by the average voter. Where these two preferences differ it can lead to inefficient decisions. Also the decision making process can become dominated by the majority even when the costs of the agreed outcome are greater than the benefits. Hansmann gives the example of a broken lift in a hypothetical four-story cooperative apartment building. If residents on the first two floor don't use the lift and outnumber those on the top two floors then a vote would result in the cheapest method of repair being chosen rather than the quickest, despite the fact that any money saved could be less than the pecuniary and nonpecuniary costs borne by the top floor residents. Alternatively the decision process could be dominated by a motivated, better informed, if unrepresentative minority resulting, again, in an inefficient outcome. All that is needed for a costly outcome is for the dominant decision makers' self interest to be given greater weight than the interests of others.

The process of collective choice can be costly in and of itself. Time and resources must be invested to obtain information about the firm and other patron's preferences as well as in attending meetings and carrying out other activities required to reach and enforce decisions. The possibility of voting cycles increase as differences in preferences among patrons increase. This can be costly if it results in continued alterations to a firm's policies. Such cycles also give power to whomever gets to set the voting agenda. Setting the agenda strategically can help the setter get the outcome they want. Strategic behaviour by owners will also increase costs via the costs of hiding or finding information or making and breaking coalitions.

But the costs of collective decision making need not be large even when the owners of the firm have heterogeneous interests as long as there is a simple criteria for harmonising those interests. If there is not then reaching agreement can be long and costly. Think of a worker owned firm where the workers differ in the work they do. Reaching agreement on the division of earnings may be straightforward as long as it is easy to account for the net benefits to the firm of each owner's efforts. If, on the other hand, it is difficult to measure every owner's net benefits then agreement may be difficult to reach.

The process of collective decision making may bring benefits to participants as well as costs. Hansmann identifies three such benefits. First, individuals may enjoy the experience of participating in decision making as a social activity in and of itself. Second, it can be argued with regard to worker ownership that workers gain psychological satisfaction from being in control, that is, from participating directly in decision making. Third, again with regard to worker ownership, participation in the firm's decision making process is useful training for participation in the democratic processes of society in general. Hansmann does note however that while such benefits are real the evidence suggests that they do not outweigh the costs of collective decision making.

The takeaway message from Hansmann is that

"[t]he preceding survey points to several reasons for the dominance of investor-owned firms in market economies. One is that contracting costs for capital are often relatively high as compared with contracting costs for other inputs--including labor--and for most products. A second reason is that, however poorly situated investors may be to exercise effective control, there is seldom any other group of patrons who are in a better position to assert control. Where either of these conditions fails, other forms of ownership arise. Thus, when there are serious imperfections in the firm's product or factor markets, the firm is often organized as a consumer or producer cooperative or as a nonprofit. Similarly, when some group of patrons other than suppliers of capital is in a good position to exercise collective control, consumer or producer cooperatives often arise even when the patrons in question are faced with only modest problems of market failure. This suggests either that the effectiveness of the oversight exercised by shareholders--even with the assistance of the market for corporate control--is distinctly limited or that other factors may be more important in constraining managerial opportunism.

In determining whether the costs of ownership are manageable for a given class of patrons, homogeneity of interest appears to be an especially important consideration. In particular, it is evidently a significant factor in the widespread success of the modern investor-owned business corporation, and it may be among the best explanations for the relative paucity of worker­ owned firms, which otherwise have some significant efficiency advantages" (Hansmann 1988: 301-2).

Footnotes:

1. While having both sets of right controlled by the same people is not necessary it is the most common form of allocation. For efficiency reason you would expect that both sets of rights would be held together. Put simply control rights give you the ability to do things while income rights give you the incentive to do things. Having them controlled by the same people lowers the cost of achieving an efficient outcome.
2. In a footnote Hansmann makes reference to Spence (1975) as an explanation for this result. See Krouse (1990: section 6.1) for a simple exposition of Spence's result. There is a result from public choice theory that says for a public good the Pareto efficient quantity of the good is the same as the medians' voters demand only if the median voter is also the average voter. So when the median and average differ the result is inefficient.

References:

• Akerlof, George A. (1970). 'The Market for 'Lemons': Quality Uncertainty and the Market Mechanism', Quarterly Journal of Economics 84(3) August: 488-500.
• Hansmann, Henry (1988). 'Ownership of the Firm', Journal of Law, Economics, & Organization, 4(2) Autumn: 267-304.
• Hansmann, Henry (1996). The Ownership of Enterprise, Cambridge, Mass: Harvard University Press.
• Hansmann, Henry (2013) 'Ownership and Organizational Form'. In Robert Gibbons and John Roberts (eds.), The Handbook of Organizational Economics (pp. 891-917), Princeton: Princeton University Press.