Corporate Governance: A Contractual Assessment

The study of corporate contracting is complicated by interdependencies within and between contracts; changes in one set of terms commonly require realign- ments in others. It will nevertheless be more helpful to examine the contracts of corporate constituencies in a sequential rather than a fully interactive way. That completed, I shall then examine interaction effects.

1. Framework 

Recall the two-technology schema in Chapter 1. One is the general purpose technology—technology that is useful over a broad range of transactions and therefore involves no exposure of transaction-specific assets. Such resources can be redeployed easily should either party terminate the contract. Special purpose technology, by contrast, incorporates transaction-specific assets. They cannot be redeployed easily or costlessly if the contract is prematurely terminated or if continuity of the exchange relation is otherwise upset. Using k as a measure of transaction-specific assets, transactions that use the general purpose technology are ones for which k = 0. When transactions use special purpose technology, k > 0. Such trades experience a fundamental transformation, hence take on the attributes of bilateral dependency.

Although classical market contracting suffices for transactions when k = 0, unassisted market governance poses hazards whenever transaction-specific assets are placed at risk, because parties then have a special incentive to safeguard investments. Let s denote the magnitude of any such safeguards. A situation in which j = 0 is one in which no safeguards are provided. A condition of complete safeguard will obtain if s = k. A refusal to provide a contractual safeguard will, of course, show up in the price. If p is the price at which the firm procures a good or service when s = 0, and if p is the price for the same good or service when s > 0, then p‾ > pˆ, ceteris paribus.

What is referred to in Figure 1-2 as a node A outcome obtains if k = 0. Node B is the k > 0, s = 0 result. And node C corresponds to k > 0, s > 0.

The above all relates to the governance branch of transaction cost eco- nomics. The measurement branch, however, is often relevant (and, as discussed below, is here). Thus, despite trading safeguards of a node C kind, there may be special circumstances where additional benefits would accrue if information pertinent to the exchange were more fully disclosed. Sometimes the disclosure will enable the recipient more successfully to anticipate future developments and plan accordingly. Sometimes such disclosure will reduce informational asymmetries which, if unrelieved, will cause the less informed party to disbelieve the representations of the more informed and lead to a costly contractual impasse.

It bears repeating, however, that such disclosures are not needed if assets are nonspecific. Investment plans do not turn on bilateral trading in such circumstances. And where neither party values continuity in its relationship with- the other, a costly effort to reduce informational asymmetries serves no useful veracity purposes either.

2. Applications 

Two classes of membership on the board of directors will be considered: voting membership and participation only to secure information. Voting membership invites a constituency to participate in what Eugene Fama and Michael Jensen (1983) refer to as the ratification of corporate decisions and the follow-on monitoring of corporate performance. Informational participation allows a constituency to observe strategic planning and to be apprised of the information on which decisions are based, but allows no vote on investments or management. Those responsibilities are reserved for the voting subset of the board.

2.1. LABOR

Supporters of codetermination regard participation for informational pur- poses as inadequate. They maintain that codetermination should extend the influence of workers to include “general issues of investments, market planning, decisions about output, and so forth” (Schauer, 1973, p. 215).4

That argument is clearly mistaken as applied to workers with general purpose skills and knowledge (node A). Such workers can quit and be replaced without productive losses to either worker or firm.5 Consider, there-fore workers who make firm-specific investments and are located at nodes B or C Ordinarily, it can be presumed that workers and firms will recognize the benefits of creating specialized structure of governance to safeguard firm- specific assets. Failure to provide such safeguards will cause demdhds for higher wages. Also, as discussed in Chapter 7, inefficient utilization decisions will result from node B outcomes. Accordingly/efficiency purposes will be served if labor of the k > 0 kind is located at node C by aligning incentives and crafting specialized bilateral governance structures that are responsive to the needs of firm and labor at this contracting nexus.

Consider therefore the relation of node C labor to the board of directors in informational respects. A chronic difficulty with long-term labor agreements is that misallocation will result if wages are set first and employment levels are unilaterally determined by management later. The inefficiency was first noted by Wassily Leontief (1946) and has since been elaborated upon by Robert Hall and David Lilien (1979) and by Masahiko Aoki (1984). Even if wages and employment are both established at the outset, the agreement may drift out of alignment during the contract’s execution to the disadvantage of the less informed member of the contracting pair. Such a result might be avoided by imparting more information to labor. Labor membership on the board of directors for informational purposes is one means of achieving that result. Indeed, Aoki contends that the “true value of co-determination is to be found in its being an instrument through which important and accurate information is shared” (1984, p. 167).

Labor membership on boards of directors can be especially important during periods of actual or alleged adversity, especially when firms are asking workers for give-backs. Labor’s board membership might mitigate worker’s skepticism by promoting the exchange of credible information.6 Douglas Fraser’s inclusion on the Chrysler board during the company’s recovery is an illustration.

The practice does not, however, enjoy widespread support. Some oppo- nents fear that it will be difficult to resist the transformation of informational  roles into decision-making participation. It is also possible, however, that the informational benefits of labor membership dre not adequately appreciated.


The term “owners” is usually reserved for stockholders, but debt-holders sometimes assume this status. However described, suppliers of finance bear a unique relation to the firm: The whole of their investment in the firm is potentially placed at hazard. By contrast, the productive assets (plant and equipment; human capital) of suppliers of raw material, labor, intermediate product, electric power, and the like normally remains in the suppliers’ pos- session. If located at node A, therefore, these suppliers can costlessly redeploy their assets to productive advantage. Suppliers of finance must secure repayment or otherwise repossess their investments to effect redeployment.112 Accordingly, suppliers of finance are, in effect, always located on the k > 0 branch. The only question is whether their investments are protected well (node C) or poorly (node B).

  1. Equity. Although a well-developed market in shares permits individual stockholders to terminate ownership easily by selling their shares, it does hot follow that stockholders as a group have a limited stake in the firm. What is available to individual stockholders may be unavailable to stockholders in the aggregate. Although some students of governance see only an attenuated relation between stockholders and the corporation, that view is based on a fallacy of composition. Stockholders as a group bear a unique relation to the firm. They are the only voluntary constituency whose relation with the corporation does not come up for periodic renewal. (The public may be regarded as an involuntary constituency whose relation to the corporation is indefinite.) Labor, suppliers in the intermediate product market, debtholders, and consumers all have opportunities to renegotiate terms when contracts are renewed. Stockholders, by contrast, invest for the life of the firm, and their claims are located at the end of the queue should liquidation occur.

Stockholders are also unique in that their investments are not associated with particular assets. The diffuse character of their investments puts share- holders at an enormous disadvantage in crafting the kind of bilateral safeguards normally associated with node C. Given the enormous variety, the usual strictures on the feasibility of comprehensive ex ante contracting apply here in superlative degree. Inasmuch, moreover, as unanticipated events cannot be addressed and folded into the contract at contract renewal intervals, because the equity contract runs for the life of the firm, the parties appear to be at a contracting impasse. Absent the creation of some form of protection, stockholders are unavoidably located at node B.

Recall that suppliers located at node B demand a premium because of the hazard of expropriation that such contracts pose. That premium can be regarded as a penalty imposed on the firm for its failure to craft node C safeguards. The incentive of the firm to secure relief from the penalty is clear (Jensen and Meckling, 1976, p. 305). What to do?

One possibility would be for entrepreneurs to supply all of their equity financing directly—from their own funds or from friends and family who know and trust them and can apply sanctions that are unavailable to outsiders who are not members of friend or family networks. This would place a severe limit, however, on the amount of equity funding available. It is no solution, moreover, to increase the amount of debt financing to compensate for those restraints.10

A second possibility is to invent a governance structure that holders of equity recognize as a safeguard against expropriation and egregious mis- management. Suppose that a board of directors is created that (1) is elected by the pro-rata votes of those who hold tradable shares, (2) has the power to replace the management, (3) has access to internal performance measures on a timely basis, (4) can authorize audits in depth for special follow-up purposes, (5) is apprised of important investment and operating proposals before they are implemented, and (6) in other respects bears a decision review and monitoring relation to the firm’s management. Such a governance structure arguably moves the node B relation that would otherwise obtain toward a node C result, with the attendant benefits that are associated therewith.

The board of directors thus arises endogenously, as a means by which to safeguard the investments of those who face a diffuse but significant risk of expropriation because the assets in question are numerous and ill-defined arid cannot be protected in a well-focused, transaction-specific way. Thus regarded, the board of directors should be seen as a governance instrument of the stockholders. Whether other constituencies also qualify depends on their contracting relation with the firm.

Such protection for stockholders can be and often is supplemented by other measures. Corporate charter restrictions and informational disclosure requirements are examples. Firms recognize stockholders’ needs for controls, and many attempt responsibly to provide them.113 Some managements, however, play “end games” (undisclosed strategic decisions to cut and run before corrective measures can be taken), and individual managers commonly disclose information selectively or distort the data. Additional checks against such concealment and distortion can be devised to give shareholders greater confidence. Arguably, an audit committee composed of outside directors and a certification of financial reports by an accredited accounting firm promote those purposes. Another possibility is the required disclosure of financial reports to a public agency with powers of investigation. The efficacy of those devices is difficult to gauge.

2. Lenders. In certain atypical circumstances, lenders may also deserve board representation. Unlike stockholders, lenders commonly make short-term loans for general business purposes or longer-term loans against earmarked assets. Proof that the firm is currently financially sound, coupled with short maturity, affords protection for short-term lenders. Such lenders do not need additional representation. Lenders who make longer-term loans commonly place preemptive claims against durable assets. If the assets cannot be easily redeployed, lenders usually require partial financing through equity collateral. Thus, long-term lenders usually carefully align incentives and protect themselves with safeguards of the sort associated with node C (Smith and Warner, 1979).

As Mervyn King observes, however, firms in countries where the stock market is poorly developed are forced to rely more extensively on debt (1977, p’.’ l56). Adequate safeguards are more difficult to provide in such circumstances. As the exposure to risk increases, these debt-holders become more concerned with the details of the firm’s operating decisions and strategic plans: With high debt-equity ratios the creditors become more like shareholders, and greater consultation between the management and its principal creditors results. A banking presence in a voting capacity on the board of directors may be warranted in those circumstances. More generally, a banking presence may be appropriate for firms experiencing adversity, but that should chahge as evidence of recovery progresses.

  1. A Digression on Optimal Finance. The above discussion suggests that the manner in which an investment is financed will vary systematically with the attributes of the assets. For the reasons given in Chapter 2, the usual fixed cost- variable cost distinction will not do. Rather, the crucial matter is one of redeployability. Equity financing, according to the approach taken here, will vary directly with the degree to which assets are nonredeploy able. Theories of finance that do not make the asset specificity distinction predict, by contrast, that there will be no such association. The Modigliani-Miller theorem (1958), which maintains that the cost of capital is independent of the capital structure in the firm, is thus at variance with the asset specifici- ty/govemancc structure approach.


Whether or not suppliers of raw material and intermediate product have a stake in a firm depends on whether they have made substantial investments in durable assets that cannot be Tedeployed without sacrificing productive value if the relationship with the firm were to be terminated prematurely. The mere fact that one firm does a considerable amount of business with another, however, does not establish that specific assets have thereby been exposed. At worst, suppliers located at node A experience modest transitional expenses if the relation is terminated. Neither specialized bilateral governance nor membership on the board of directors is needed to safeguard their interests. The protection afforded by the market suffices.

Suppliers who make substantial firm-specific investments in support of an exchange will demand either a price premium (as at node B, where the projected breakeven price is p) or special governance safeguards (as at node C). Progress payments and the use of hostages to support exchange are illustrations of node C safeguards. An agreement to settle disputes through arbitration, rather than through litigation, is also in the spirit of node C governance. (The issues here are those developed in earlier chapters, where the governance relations between suppliers and buyers are examined.)

Considering the variety of widely applicable governance devices to which firms and their suppliers have access, there is no general basis to accord suppliers additional protection through membership on the board of directors. There could be exceptions, of course, where a large volume of business is at stake and a common information base is needed to coordinate investment planning.115 Ordinarily, however, the governance structure that firm and supplier devise at the time of contract (and help to support through a web of interfirm relationships) will afford adequate protection. Membership on the board, if it occurs at all, should be restricted to informational participation.


The main protection for customers located at node A is generally the option to take their trade elsewhere. Products that have delayed health effects are an exception, and consumer durables can also pose special problems. Membership on the board of directors is not, however, clearly indicated for either reason.

Health hazards pose problems if consumers are poorly organized in relation to the firm and lack the relevant information. If consumers can organize only with difficulty, because they are unknown to one another or because of the ease of free-riding, then a bilateral governance structure between firm and consumers may fail to materialize. Protection by third parties may be warranted instead. A regulatory agency equipped to receive complaints and screen products for health hazards could serve to infuse confidence in such markets.

Whether consumer membership on the board would afford additional protection is problematic. Who are representative consumers? How do they communicate with their constituency? Token representation may create only unwarranted confidence.

Similar problems of consumer organization and ignorance arise in con-junction with consumer durables. That is true whether the consumer durable requires no follow-on service or a great deal of such service.16 Among the available types of consumer protection are brand names, warranties, and arbitration panels. Shoppers who choose node B are presumably looking for bargains. They will spurn the additional protections in favor of a lower price. Such customers implicitly accept a higher risk and should accept occasional disappointments. There are other consumers however, who value protections at node C. Some are prepared to pay a premium for a brand name item. Brand names effectively extend a firm’s planning horizon and create incentives for the firm to behave “more responsibly.’’17 (To be sure, customers must be wary against firms that build up a reputation, thereafter to expend it by taking advantage of lagging consumer perceptions.I8) Warranties are explicit forms of follow-on protection, and many are available on optional terms. The recently introduced consumer arbitration panels are likewise responsive to concerns over consumer protection. Consumers concerned about fair play during the service period will presumably concentrate their purchases on brands for which arbitration is available.

Further innovations to offer consumer protection on a discriminating basis may be needed. With the possible exception of large customers with special informational needs, however, a general case for inclusion of consumers on the board of directors is not compelling.


Community interest in the corporation is a very large subject. I consider two concerns here: externalities and the hazards of appropriation.

Externalities commonly arise where the parties in question do not bear a contracting relation to each other. Pollution is one example. Corrections can be interpreted as an effort by the community to impose a contract where none existed. For example, the community may place a pollution tax (price) on the firm, or it may stipulate that pollution abatement regulations must be satisfied as a condition for doing business.

A chronic problem in this area is to secure the knowledge on which to base an informed pollution control policy. Firms are often in possession of the necessary knowledge and may disclose it only in a selective or distorted manner. Public membership on the board of directors could conceivably reduce misinformation. But the remedy would come at a high cost if the corporation were thereby politicized or deflected from its chief purpose of serving as an economizing instrument. Penalties against misinformation coupled with moral suasion may be more effective. It is an area in which there may simply be no unambiguously good choices.

The hazards of expropriation are even less of a justification for public membership on boards. Communities often construct durable infrastructures to support a new plant or renewal investments by old firms. Expropriation is possible if the firm is able to capitalize these public investments and realize a gain upon selling off the facility. Such concerns are much greater if the firm makes general purpose rather than special investments. Communities that make investments in support of a firm should therefore scrutinize the character of the investments that the firm itself makes.

As elsewhere, expropriation hazards will be mitigated if the parties can locate themselves at node C. Insistence that the firm make specialized invest- ments is akin to the use of hostages to support exchange. In general, specially crafted node C protection, rather than public membership on the board of directors, has much to commend it as the main basis for safeguarding commu- nity investments.


There is one constituency that curiously goes unmentioned in most discussions of corporate governance: the management. Perhaps analysts assume that management is appropriately assigned a mediation role between contesting constituencies.19 And some critics maintain that management is already overrepresented in the affairs of the firm: Management participation on the board of directors is the problem, not the solution. The issues here are developed in section 3, below.

3. Contracting in Its Entirety

Suppose, arguendo, that voting membership for node B constituencies is granted. Suppose further that constituencies located at node C ask-for voting participation. Two arguments might be advanced in support of the proposal: A spirit of generosity warrants node C inclusion, and democratic purposes would be served by broadening the board in that way. What are the costs?

One obvious cost is that of supplying information. Huge educational needs arise if specialized constituencies are to be informed participants on the board. Representatives of each specialized constituency would need to learn a great deal about the overall character and agenda of the corporation. Such participation also risks deflecting strategic decision-makers from their main purposes by forcing them to redress operating-level complaints. That squanders a valuable resource. More serious, however, is the prospect that the inclusion of partisan constituencies on the board invites opportunism. A constituency that had reached a bilateral bargain with the corporation would, if it participated in board level decisions, gain leverage to extract additional concessions from the corporation during the execution of the contract. Opportunism is especially likely where many partisan constituencies are represented on the board and logrolling is feasible. Also, and related, corporate assets may be dissipated in the support of “worthy causes” with which specialized constituencies sympathize.

Unwarranted” participation in the decisions of the board of directors by such poorly suited constituencies will, moreover, cause subsequent adaptation by other parties who deal with the firm. For one thing those who are asked to provide general purpose corporate funding will adversely adjust the terms under which corporate finance will be made available. Moreover, the bilateral contracts affected by the deflection, distortion, and dissipation of corporate assets will be realigned. Not only will the original terms (price) differ in anticipation of later efforts by a constituency to strike “better” deals, but also bilateral safeguards are apt to be reduced. Node C governance will thus move toward node B. In extreme cases special purpose technologies and involvements will give way to general purpose ones, and node A governance will result. Since membership on the board of directors by constituencies located at node A lacks economic purpose, it is naive to believe that the board of directors’ franchise can be extended without cost.

Broadening the franchise is not, therefore, a simple matter of effecting a redistribution of wealth away from those who had the franchise previously in favor of those to whom it is newly awarded. Absent the prospect that a contractual defect will be corrected by awarding a place on the board of directors to a previously unrepresented constituency, broadening the franchise will have two adverse effects: Future terms of finance will be adversely adjusted, ànd the terms of the bilateral bargain between the firm and the affected constituency are apt to deteriorate. Here as elsewhere, contracting must be examined not at a point in time but in its entirety.

Informational participation does not appear, however, to pose equally serious concerns. In the degree, therefore, that informational participation promotes contracting confidence and deters possible abuses (of the kind dis- cussed in 3.4, below), such participation has much to commend it.

Source: Williamson Oliver E. (1998), The Economic Institutions of Capitalism, Free Press; Illustrated edition.

Leave a Reply

Your email address will not be published. Required fields are marked *