Why They Give at the Office: Shareholder Welfare and Corporate Philanthropy in the Contractual Theory of the Corporation

Volume 84 Issue 5July 1999

Article 2

Why They Give at the Office: Shareholder Welfare and Corporate Philanthropy in the Contractual Theory of the Corporation

Henry N. Bulter

Fred S. McChesney

Part of theLaw Commons

WHY THEY GIVE AT THE OFFICE: SHAREHOLDER WELFARE AND CORPORATE PHILANTHROPY IN THE CONTRACTUAL THEORY OF THE CORPORATION

For centuries legal, political, social, and economic commentators have debated corporate social responsibility ad nauseam.2 An important component of that debate concerns the legitimacy of corporate philanthropy. Courts largely have resolved the legal issue in favor of allowing corporate management wide latitude in making philanthropic contributions.3 Yet there remains considerable controversy over whether corporate managers should have the authority to engage in these acts. Indeed, one reads more and more about the "problem of corporate philanthropy."4

It is ironic that many now view management's philanthropic acts as a problem. The modem debate over corporations' social responsibility began with claims that corporations gave too little, not too much. Challenging the prevailing views of the time in an important and then-controversial article, Milton Friedman argued that acts of pure corporate philanthropy (i.e., philanthropy inconsistent with maximizing a firm's profits) were a mere waste of shareholders' assets and thus contrary to management's function.5 Friedman must be pleasantly surprised to find contemporary mainstream commentators worrying about the same waste of assets that concerned him. Nevertheless, current newspaper stories about corporate philanthropy are more likely to praise the "giant corporation" for demonstrating "social responsibility" through liberal philanthropy.6

This paper explores the "problem of corporate philanthropy" from the standpoint of the contractual theory of the firm. That is, it considers corporate philanthropy within a framework of the types of contracts that shareholders generally would find desirable to govern relations among themselves and between shareholders and managers. Part I explains this general framework, noting in particular that shareholders desire the firm to assume only those functions that the firm can perform better (or more cheaply) than individuals themselves could undertake them.

This desire explains why investors want their firms to make some sorts of charitable donations but not others, as Part II discusses. But to achieve the benefits of corporate philanthropy, investors who leave philanthropic decisions to firm management must countenance the

possibility that management's own objectives, not those of shareholders, may motivate some philanthropy. This problem is no different, however, from any other agency cost that shareholders willingly confront when they choose to invest in firms that have a split between ownership and control. Thus, the reasons for a separate literature on philanthropy (often calling for separate legal rules to regulate the "problem" of philanthropy) remain obscure.

Part III explores these concepts of philanthropy against the backdrop of the recent suggestion that corporate philanthropy is different from the other agency costs that shareholders must tolerate in exchange for others running their firm.7 Invoking the team-production explanation of firms, some suggest that managers ought to be able to support charity in ways that do not maximize firm profits and that the law recognizes the desirability of unprofitable philanthropy. Part III argues against both the claims that corporate philanthropic giving unrelated to firm profitability is desirable and that the law validates it.

I AGENCY COSTS, MARKET FORCES, AND THE (LImED) ROLE

The logical starting point for both the economic and legal analysis of corporate philanthropy is the raison d'etre of the corporate firm itself. The existence of firms reflects two distinct, if related, phenomena. First, firms exist because production is optimally undertaken by more than one individual, typically because of gains from specialization.8 These gains from specialization lead to team production among the various actors in the firm.9 In law firms, for instance, lawyers, paralegals, secretaries, and other workers each specialize in various sub-tasks that constitute the team production of legal services for clients.

Second, firms exist because the transactions costs of organizing production through market transactions often make it desirable to take some activities from the market and perform them in the firm. This observation in fact complements the team-production theory: gains from team production are a necessary but not sufficient condition for the existence of firms. Team production can always be accomplished by hiring other team members in the market for specific

tasks without combining them into a firm. For example, the lawyer can hire temporary paralegals and secretaries in the "spot" labor market as needed without creating a law firm. But the transactions costs of doing so, Ronald Coase has explained, are often so high that it is preferable to create a firm in which labor is typically available as production needs arise, avoiding frequent and thus costly trips into the spot labor market.10 The nature of the firm, then, is the use of ongoing internal direction by the firm's managers to control labor and other resources, rather than negotiating a series of external contracts in the marketplace as needs arise. When the costs of internal control rise relative to the use of spot contract markets, entrepreneurs will substitute external market (i.e., contract-based) direction of resources in place of internal direction by managers. Thus, there is a limit to the scope of firm production.

There is an important corollary to Coase's insight that firms exist as lower-transaction-cost alternatives to contracting in spot markets: firm owners will only commit their resources internally to firms to the extent that the firm can make better (i.e., more profitable) use of those resources than individuals themselves could. This is simply a restatement of Franco Modigliani's and Merton Miller's fundamental point that firms in well-functioning markets can earn no greater returns by combining capital (i.e., debt and equity) within the firm than individuals can by investing in debt and equity markets." Unless firms can achieve optimal portfolios more cheaply than can dispersed individuals, investors have no reason to entrust their funds to firms. So, for example, it would be unusual to find individuals investing in firms whose assets consist solely of treasury bills and which pay the interest as dividends to shareholders. Individuals who want treasury bills can just as easily buy them and collect the interest themselves.

The points above apply to all firms, notjust corporations. Among firm types, corporations are distinct because those directing the use of resources internally (managers) may not be the ones who own the firm (shareholders).12 Thus arises the traditional concern, which Adolph Berle and Gardiner Means popularized, about the separation of ownership and control in the large publicly traded corporation.'3 Non-owner managers may be tempted to maximize their own welfare rather than the profits of the firm that employs them, preferring

themselves over the shareholders who own the firm. The literature on this topic is voluminous, the more recent literature referring to the separation of ownership and control as creating a "principal-agent" problem.14 Managers should act as agents of the firm, but they have some incentive to maximize their own utility at the expense of firm profits (and thus the welfare of the firm's owners).

This well-founded concern about manager-agents disregarding their shareholders' interests has resulted in disagreement among legal commentators about how to confront the problem. Berle and Means, of course, believed that changes in corporation law and securities regulations were necessary for the development of corporate democracy and the protection of shareholders.15 Many other commentators have followed in the Berle-Means tradition of addressing the principalagent problem through legal solutions.16

An alternative (and less legalistic) concept of the corporation is the contractual, or market, theory of the corporation.17 The contractual theory of the corporation also starts with the recognition of the principal-agent problem. This theory, as its name suggests, stresses that private contracts and anonymous market forces act as the primary restrictions on managerial discretion and thus on agency costs that reduce shareholder welfare. Legal restrictions on managerial discretion necessarily take a back seat because they are often unnecessary to resolve any problem.

In this view, the corporation is based on mutually beneficial exchange (that is, contract) among various suppliers of inputs to the firm. Although some of these exchanges are explicit, legally enforceable contracts, many are informal or implicit contracts that market mechanisms, such as repeat dealing and reputation, enforce.' The gist of the contractual theory of the corporation is that market forces-not the threat of legal sanctions-give corporate management the incentives to act as if it has the shareholders' best interests at

heart. The contractual theory of the corporation is rigorous and supported by numerous empirical studies.19

Under the contractual theory of the corporation, the primary external force holding the publicly traded corporation together is the market for corporate control.20 This market forces managers to behave by threatening them with the loss of their jobs through hostile takeovers, proxy battles, board revolts, or mergers. Faithless manageragents must fool not just the shareholders who vote them into management positions but all other possible owners of the firm as well. Outside investors will see in sub-par management a profit opportunity that is available through takeover of the firm. This threat is external to the corporation and-when legal restrictions on changes in corporate control do not impede it-acts as a powerful constraint on managerial discretion. Additional market forces put pressure on managers to act in their shareholders' best interests. Capital markets, internal and external markets for managerial talent, product markets, and service markets give managers the incentive to maximize share value.21 Corporation law does play an undeniable role in the contractual theory of the corporation.22 It provides a standard-form governance structure, including a set of legal remedies that are available to shareholders when managers get too far out of line. The law governing boards of directors illustrates this standard-form characterization of corporation law. Increasingly, shareholders are free under state corporation statutes to define the duties for which a director will be liable if they do not like the "off-the-shelf' definition the statutes provideor in some altogether.23

to dispense with a board of directors

instances,

The flexibility under the law for shareholders to shape directors' duties reflects the law's recognition that boards do matter and that shareholders who elect boards of directors expect them to function as

agent-monitors over corporate affairs.24 Although some mistakenly view it as a formalistic, legally imposed requirement of a publicly traded corporation, the board plays an important economic role in the contractual theory of the corporation.25 The board of directors supplements the market for corporate control by providing an internal source of monitoring to force managers to act in their shareholders' best interests.

In the contractual theory of the corporation, the primary role of the board of directors is to assure shareholders that agency conflicts are under control. Eugene Fama and Michael Jensen have explained that when decision-making agents do not bear the wealth effects of their decisions (and thus do not necessarily have an incentive to act in their shareholders' best interests) the decision-making process will split between agents who manage and agents who control.26 Thus, in the publicly traded corporation characterized by the separation of ownership and control, decision management (initiation and implementation) is the responsibility of senior management, and decision control (ratification and monitoring) is the responsibility of the board of directors.

The contractual theory of the corporation explains why boards of directors have the legal authority to define the perimeters of managerial decision making. In this view, the board of directors is a marketinduced mechanism for monitoring management on behalf of shareholders. The relevant question then becomes whether directorseither individually or collectively-have the incentive to do their job. Corporate law provides standard-form fiduciary duties of care and loyalty, but it is widely recognized that directors easily satisfy and rarely violate the standard of care.27 On the other hand, market forces do provide directors with an incentive to monitor the performance of se-

nior managers.28 Specifically, competitive markets for outside directors' services reward directors if the firms they monitor perform well. Also, directors know that they could lose their positions if the firm is taken over in a control transaction due to poor performance.

The contractual theory explicitly recognizes that in many instances market constraints have a greater effect on managerial discretion than legal constraints. Corporate philanthropy provides a classic illustration of this point.29 For legal commentators steeped in the Berle-Means tradition of skepticism regarding managerial motives and behavior, calls for increased corporate social responsibility represent nothing more than yet another excuse for managers to abuse shareholders. By law, managers have a great deal of discretion, including the ability to give away corporate assets to various groups. Moreover, it is certainly easy to come up with profit-maximizing or value-maximizing rationales that deflect legal challenges to corporate giving. Thus, it is clear that corporations could give away more money than they currently give or traditionally have given without fear of shareholder actions.

A. Good Philanthropy

The contractual nature of the firm provides a useful framework for analyzing the "problem" of corporate philanthropy. First, it is clear that shareholders as individuals may well have a desire to commit philanthropic acts. Not only is this understood as a matter of formal economics,30 but it comports with common sense and everyday observation. Individuals give money to churches, schools, hospitals,

and countless other institutions-not to mention family members and friends.

However, the issue is why individuals might prefer to make charitable contributions through the corporations of which they are shareholder-owners. The answer typically given is that corporate giving often benefits the firm by increasing its "goodwill"-the extent of future patronage by those who become familiar with the firm's name. Individuals, by hypothesis, are not looking to increase their personal goodwill, or at least cannot increase its value to the extent that the firm can increase its own.

Often, the use of corporate resources for conduct that appears to be nonmaximizing [of profits] can be justified on a straight maximizing basis, because it is simply a special form of ordinary business expense. For example, General Motors subsidizes Ken Burns in making his documentaries for public television.... [I]t gets its name associated with a classy product and it gets to put a fifteensecond commercial before and after the documentary.3' But the fact that giving by corporations is valuable, though undoubtedly true, does not explain why shareholders give through corporations. After all, General Motors (GM) shareholders could just give their own money, designating it a contribution of Smith, GM shareholder. The Ken Burns special, which General Motors sponsored, still might say as a condition of the shareholders' donations that these donations came "from the shareholders of GM," indicating that GM (via its shareholders) is the sponsor of the show.

In other words, the mere fact that charitable donations benefit the firm does not answer the Modigliani-Miller question of why shareholders would give at the office rather than at home. As Milton Friedman noted, "The stockholders or the customers or the employes [sic] could separately spend their own money on the particular action if they wished to do so."32 But it is not hard to see why in fact shareholders would prefer to give at the office, assuming that there is corporate goodwill to be captured by the philanthropy. By hypothesis, the firm already has the earnings (current or past) necessary for the philanthropy. Distributing the earnings as dividends which Smith can contribute individually simply imposes an additional transaction cost.

Second, and perhaps more importantly, dispersed share ownership creates a potentially important free-rider problem among shareholders. Assuming that there is value to be had in contributions propagating the firm's name, that value accrues to all shareholders in

proportion to their share holdings. Shareholders would prefer to give at the office precisely because giving through the firm forces all others who will also benefit from giving at the office too.

The point is straightforward, but Figure 1 will help clarify the dis-

cussion that follows.

The vertical axis represents the dollar amounts ($) of benefits and costs from giving. The horizontal axis measures the amount of dollars (Q) that can be donated, the marginal cost (MC) of every dollar donated naturally equaling $1.33 As long as the benefits from giving that accrue to the firm exceed the costs to the firm, the firm will continue to give. Assume that the benefits of corporate philanthropy to the firm decline with giving, producing the marginal benefit curve MB: good managers will choose the highest return donations first and then lower yielding ones. Eventually, when the benefits of giving (MB) fall short of the costs (MC = $1), the firm will stop giving. In

Figure 1, that point is realized when corporate giving reaches the level Q'. The total gain from giving is the summed difference between marginal benefits and costs at each level of giving, as the triangle ABC represents.

Alternatively, shareholders could give individually. But that giving can only be done at higher cost. Each shareholder must send in his own check; write a letter explaining that the gift is made in the firm's name, not that of the individual shareholder; and require mention of the firm as a condition of the donation. In addition, each shareholder must contend with free riding by other shareholders. Each of these complications causes the costs of giving to rise above the dollar amount of the gift-from MC to some MC'.34

The total gains from giving shrink from ABC to a smaller triangle, ADE. The shaded area EDBC in Figure 1 represents a loss in wealth that shareholders will suffer if individual giving occurs; to put it another way, EDBC measures the gains from giving at the office. That gain explains why shareholders leave corporate giving in the hands of the same managers that direct the rest of the firm's affairs.

Maximization of the gains from corporate philanthropy explains why shareholders ordinarily will prefer to make some (but hardly all) of their charitable donations through the firm. The fir can make donations that are uniquely beneficial to the firm. Furthermore, although shareholders could make these donations themselves, the firm can make them more cheaply, avoiding both transaction costs and free-rider problems. For these reasons, the law wisely accords wide deference to managers' choices of both the amounts and the targets of corporate giving.35 Doing so recognizes that shareholders want the firm, rather than themselves, to take the lead in choosing the objects and amounts of corporate charity.

B. Bad Philanthropy

This recognition, of course, does not mean that managers never will abuse shareholders' trust in making charitable donations. Managers are human, and personal utility maximization, rather than shareholder wealth maximization, doubtless exerts some influence over some managerial action sometimes. It does not follow, however, that shareholders are worse off in a system in which managers, sometimes venal and corruptible, make firms' philanthropic decisions.

Consider Figure 2: in addition to any benefits the firm may reap from a corporate contribution (MB), individual managers get a personal benefit of a in some form (e.g., prestige, free tickets) from the contribution.36 The value to the firm plus the additional a of personal benefit yield a total benefit curve of MB'. The firm's total contributions expand from Q' to Q", with the manager herself realizing a personal gain of a Q" (area AXYZ in Figure 2).37 To the firm, there is a loss equivalent to the area BYZ, the excess of costs over the benefits to the firm of the contribution.

This, then, is the "problem" of corporate philanthropy. But one should put the problem in perspective. Note, for instance, that the loss to the firm, and thus to shareholders, is considerably less than the gain to the utility-motivated manager (BYZ < AXYZ). Furthermore, there seems to be no reason to think that even the BYZ-type losses to firms are very great; in fact, most commentators believe that any problem with philanthropy is relatively slight.38 This belief does not mean that the problem is nonexistent, of course. Nor does it mean that courts should ignore the possibility that certain "philanthropic" acts are more utility maximizing to the manager than they are profit maximizing to the firm.39

But consider again the contractual nature of the firm. What would shareholders want courts to do concerning a possibly tainted contribution-one that violates the manager's contractual obligation to advance the firm's welfare? It would depend upon courts' ability to distinguish good philanthropy from bad. Stricter legal controls on bo-

FIGuRE 2

A

$1 C I

N

MB

DOuARS DONAD

gus philanthropy could impose substantial costs on shareholders by deterring the value-increasing, profit-maximizing philanthropy that, it seems agreed, ordinarily motivates charitable giving. In a world in which (1) corporate philanthropy usually makes good business sense, (2) transaction and information costs are positive, (3) judges have difficulty distinguishing profit-maximizing from utility-maximizing philanthropy, and (4) market forces constrain managerial discretion, fully informed shareholders ordinarily would choose legal rules that give managers a great deal of discretion.

If some courts would easily confuse profit-maximizing philanthropy with utility-maximizing philanthropy, managers might decline to make certain value-enhancing decisions for fear of shareholder challenges.40 To shareholders as a group, the relevant question concerns the cost of holding managers to a higher legal standard. Recoupment of the loss (BYZ) from improper giving is beneficial in itself. But if that benefit comes at the cost of deterring truly profit-

enhancing donations, it may not be worth the cost. The issue, then, is whether the transaction-cost gains from giving at the office outweigh the costs of giving at an office that sometimes utility-motivated managers oversee. Which is greater: area EDBC in Figure 1 (the benefits of giving at the office) or area BYZ in Figure 2 (the costs that inevitably arise when managers include their own welfare in philanthropy considerations)?

The problem is the familiar statistical one of Type I versus Type II error.41 Judges too inclined to find venal motives where in fact there are none (Type I error) create a disincentive for managers to contribute in the first place. True, judicial forbearance may mean that some truly inappropriate acts of so-called philanthropy go unpunished (Type II error). But Type II error in this context is essentially selfcorrecting. If corporations changed their practices and began distributing cash in ways that resembled utility maximization more closely than profit maximization, then one would expect to see changes in economic and legal constraints. For example, the changes could be a catalyst for institutional investors to take a more active role in monitoring charitable contributions.42 Other possible actions to constrain managers' utility-maximizing donations might include shareholder resolutions, charter amendments, and changes in corporate law default rules.43

Given the current array of market constraints, however, there is no reason to expect that either shareholders or interest groups want the law to change. Ultimately, the issue is an empirical one, but purely gratuitous corporate transfers without an expected benefit to the granting corporation seem rare.44 The economic explanation for this rarity turns on market constraints on management; legal constraints seem largely irrelevant to managerial decision making. Be-

cause legal rules can have unintended adverse consequences, notably in the form of Type I error, shareholders would not necessarily seek stricter legal prohibitions on managerial discretion to engage in corporate philanthropy.

Moreover, in addition to Type I error problems, legal punishment of utility-motivated philanthropy presents rather unique problems of remedy. In ordinary agency and other business situations, one who uses the principal's (i.e., the firm's) property for her own purposes is liable for any resulting gains. Thus, for example, an agent who "borrows" $100 from the firm till, wagers it at the racetrack, and wins $1,000 may not return the $100 and keep the $1,000. Rather, she must reimburse the firm for the gains she obtained using the firm's property in addition to the money she originally took.45 But how could one possibly measure the gain to a corporate manager who improperly gives $10 million to her alma mater, Cornell University, hoping for front-row ice hockey tickets in return? The personal gain almost certainly would be considerably less than the $10 million Cornell received. If the manager nonetheless must account to the firm for the full $10 million, the problem of deterring legitimate, profit-enhancing charity reappears.

Perhaps, then, the basis for damages should be the loss to the firm. Because the firm could have invested the improper donation at some reasonably determinable rate, the loss of interest on that investment could constitute a measure of firm damages. But consider Figure 2 once again. Although the manager has invested more than is appropriate (Q" rather than Q'), one cannot measure the firm's loss by the use to which the firm could have put that entire sum. There was still some benefit to the firm, as the area under the MB curve between Q' and Q", area Q'BZQ", represents. As discussed, the true loss to the firm is the much smaller area BYZ.

But, one might argue, in at least some instances it will be clear that the firm should have made no investment, so that the appropriate choice for the manager was not Q' but zero. Consider Figure 3: the value to the firm (again represented by MB) of different levels of investment never equals the $1 cost of investing.

Only with the additional personal value (a) that the manager attaches to a particular donation (as the vertical distance between MB

S$1

MC

and MB' represents) will some amount of philanthropy, Q", result. It might seem that because no charity at all was warranted, reasonable interest on the Q" that the firm did give adequately would approximate the firm's loss.

But again, as in Figure 2, measuring damages by the full value donated would overcompensate the firm and thus would create the risk of Type I error that deters profit-maximizing philanthropy. Overcompensation arises because the firm obtains some benefit from the donation, as the area OLMQ" under the MB curve measures. That is, it is the rare act of philanthropy that results in zero benefit to the firm, even if the benefits fall short of the costs.46

C. Much Ado About Little?

Finally, it is worth considering why scholars spill so much ink on the subject of corporate philanthropy in the first place. After all, the points in this section apply to any conduct in which managers' personal interest rather than the firm's profitability is responsible for expenditures using firm funds. Figure 2 describes the "problem" of office sizes just as well as it does the philanthropy "problem." An office of a certain size (Q") is necessary for a manager to. perform his profit-maximizing tasks; many managers doubtless have bigger offices (Q:"). Figure 2 likewise portrays the "problem" of company cars. Those with company cars doubtless use them on personal drives as well as company business, and so they drive more miles (Q") than the amount (Q') that purely business reasons justify. One could add any number of "problems" to the list: managers playing more golf than they should, having thicker carpets than optimal, ordering too many office supplies and taking some home for personal use, and so on.

Supra-optimal office sizes, excessive use of company cars, and all these other "problems" are just particular manifestations of the generic agency costs that afflict any team production enterprise. Profits await those who can solve these "problems" at lower cost than the available benefits. There is no specific literature on the problem of office sizes or company cars. Nor do scholars think that particular legal rules are necessary to govern managers' non-maximizing behavior concerning offices or cars. These matters are essentially for shareholders to figure out and resolve for themselves. The value of a separate literature and the need for separate legal rules concerning philanthropy are both far from evident.47

The previous Part would seem to provide a complete justification for both corporate philanthropy and the state of the law concerning that philanthropy. In many instances, shareholders find giving through the firm to be value-maximizing. For those instances in which managers' personal utility maximization rather than profit maximization motivates corporate giving, shareholders have a variety of mechanisms at their disposal to correct any "problem." Information and other relevant transaction costs necessary for distinguishing good from bad philanthropy make courts a poor substitute for sharehold-

ers' own self-help remedies. But there is nothing unusual about corporate charity in this respect. Whenever ownership and control are divided, managers face incentives to benefit themselves at shareholder expense, whether the form of the benefit is charity, office size, or abuse of company cars. Ordinarily the law leaves shareholders to their own devices rather than fashioning special rules to cover specific margins from which agents may stray in favor of their own interests and in derogation of those of their shareholders.

In the spirit of Ockham's Razor, therefore, it would seem appropriate for the corporate law world to declare victory on the issue of corporate philanthropy and move on to other matters.48 Nevertheless, new academic approaches to corporate philanthropy appear regularly. These new approaches, even if they ultimately conclude that the corporate philanthropy "problem" is not especially worrisome, are nonetheless significant. Every new approach trying to invent around the more familiar justifications for corporate philanthropy furnishes new ground for new disagreements that, potentially, would undermine the approach toward philanthropy already in place.

A. Team-Production vs. Principal-Agent Models of the Firm

The most recent addition to the literature justifying corporate philanthropy is that of Dr. Margaret Blair,49 apparently based on her work with Professor Lynn Stout.50 Dr. Blair and Professor Stout begin the development of their team-production theory of corporation law by rejecting the principal-agent problem, which Parts I and II discuss above, as the primary institutional challenge to the success of the modern publicly traded corporation.

Their rejection of the contractual theory of the corporate firm appears to rest on a misunderstanding of that model. Dr. Blair objects to the absence of a formal, legally enforceable system of agency contracts binding various corporate actors:

[T] he idea that managers and directors are "agents" of "shareholders" has always been at odds with the way corporation law actually works. Under American corporate law, directors and officers of a corporation are not agents of shareholders. Managers are agents of the corporation itself, and directors are sui generis. Directors are not subject to direct control by shareholders, and they owe no duty of obedience to shareholders. Confronted by these uncomfortable facts, contractarian legal scholars generally retreat to the "metaphor" argument: It doesn't matter that there is no explicit legally enforceable agency contract between shareholders and directors.51

This misunderstanding is merely semantic. Admittedly, certain nonattorneys writing in the contractarian tradition may have been careless about nomenclature, referring to "contracts" when no legally enforceable arrangement existed. But to economists, the notion of "contract" extends beyond legal agreements enforceable in court, as Part I discusses above. Many principal-agent relationships are implicit arrangements that market constraints rather than legal remedies enforce-"self-enforcing agreements" is a frequent description.52 Criticizing the contractarian model of the firm for lack of legally enforceable agreements, as Dr. Blair does, misses the central point of the contractarian model, in which private ordering matters more than the law.

Disposing of the contractarian notion of the corporation with its focus on agency costs, Dr. Blair views the "intractable contracting problems" of team production as the primary explanation for the structure of corporation law.53 Team production truly is an important aspect of the theory of the firm,54 in particular because of the problems for the firm that team production entails. Team production refers to production processes in which the interaction of team members increases the value of the output, as Coase mentioned.55 But

team production carries with it a problem that Coase left on the table-the difficulty in determining the marginal product that each input adds, an essential point that Armen Alchian and Harold Demsetz analyzed a generation later.56 Because it is difficult to monitor the productivity of individual team members, members have the incentive to shirk their productive activities. In light of these incentives, some individuals prefer market exchange transactions to participation in a firm-again, as Coase noted.57 The ease or difficulty of solving the contracting problems of team production explains why some activities occur within firms instead of through discrete arm's-length transactions.

Entrepreneurs recognize the potential gains from solving the team-production contracting problems. For example, profit incentives lead the team members to hire a monitor to oversee the overall performance of the team and to distribute rewards according to the monitor's perception of the productivity of team members. In the Alchian-Demsetz firm, the monitor is the residual claimant of the net cash flows that the team generates after all team members have received payment for their contribution to the firm. In this sense, the monitor is the owner of the firm, and there is no principal-agent (or separation-of-ownership-and-control) problem.

In many instances, however, the owner-monitor will not have (or will refuse to provide) all the financial resources necessary for the firm to operate optimally. Additional capital must come from other sources. Financing can come from numerous sources including banks, general partners, limited partners, and bondholders, or by selling stock to many dispersed shareholders. The Berle-Means dispersed-owner corporation is at one end of the spectrum of possible financial structures. In the dispersed-owner corporation, specialization causes shareholders to become the residual claimants and the monitors of team production to become the managers. In this regard, the managers generally act as both the directors and senior officers. In terms of the principal-agent literature, the managers are the shareholders' agents. A variety of legal and market mechanisms evaluates how well the managers perform their function as organizers and monitors of team production. In particular, the external market pressures monitor the monitors of team production. In this way, team production plays an important role in the contractual theory of the corporation. Team production not only explains the existence of firms, but it also explains particular aspects of internal corporate-gov-

ernance structures. Team production, in this sense, is internal production.

Numerous other economists have extended and applied the team-production theory.58 But the essence of the economic discussion has remained contractarian. For example, Dr. Blair summarizes the work of Sanford Grossman, Oliver Hart, and John Moore:

Rather than thinking of the inputs of the team members as actions that are difficult to monitor, they think of the inputs as investments that must be made-for example, in firm-specific skills-that are difficult to specify and verify contractually. The team will be most productive if all of the team members invest, but the ex ante incentive that each team member has to invest depends on how much of the rents she thinks she can capture, which, in turn, depends on who has what bargaining power over the rents ex post.59

Dr. Blair also summarizes her criticisms of the solutions that this economic literature on team production offers:

Basically, the handful of scholars who have worked on this problem have focused on one possible solution, that of simply assigning property rights to one member of the team. Property rights means that one person gets to make all the decisions. The problem with that is that if you really need complex and difficult-to-monitor inputs from all of the other participants, the parties who don't have property rights can lose their incentives to make the necessary effort or investment.60

Thus, according to Dr. Blair, there is need for an understanding of the solutions to team-production contracting problems.

Although one could quibble with Dr. Blair's characterization of the problem,61 there is nothing remarkable in her claim that solving team-production problems poses a major challenge to firms. Indeed, an endless progression of business bestsellers appears to address pre-

cisely that goal.62 What is remarkable is Dr. Blair's claim that corporation law is the solution to a major problem facing businesses. Dr. Blair claims that "[t] he central contractual problem to be solved by the formation of a corporation, and by the structure of corporate law, is not a principal-agent problem (which could be solved through contract law). It is a 'team production' problem."63

As Dr. Blair explains, corporation law has provided the framework to solve team-production problems because the law provides managers with the discretion to reward difficult-to-measure contributions to team production. Of course, the manner in which a particular corporation deals with internal business matters, such as whether to invest in firm-specific human capital, avoids the second-guessing of the traditional business judgment rule. But it is a huge step to argue that because the law shields decisions about a particular type of production process, legal protection explains the entire structure of corporation law. In addition, it is difficult to see why the world needs a new theory that explains the already widely recognized need for managerial discretion over the internal activities of the corporation.

to return

Clearly, Dr. Blair and Professor Stout have developed some creative arguments related to team production and the resulting contracting problems. But to a central point of the contractarian view of the corporation, each Blair-Stout problem that this discussion has identified represents a profit opportunity for entrepreneurs who develop private solutions without the aid of corporation law. Their treatment of the contracting problems associated with firm-specific human-capital investments illustrates the fundamental problem with their approach.

That employees pay for investments in their general human capital and that employers pay for investments in firm-specific human capital are widely accepted in economics-and are indeed the stuff of the Nobel Prize.64 The logic of this analysis is straightforward:

It is useful to conceptually distinguish between two types of training: General training increases an individual's productivity to many employers equally, and specific training increases an individual's productivity only to the firm in which he or she is currently employed. General training might include teaching an applicant

tual problem" that Blair and Stout identify,69 both because it is unnecessary given market correctives and because almost nothing in corporate law constrains the terms of the contracts among shareholders, boards of directors, officers, and employees. To repeat, the business judgment rule, which generally protects management's disinterested decisions, shields practically all of management's decisions regarding the monitoring of team production.

This protection is hardly surprising. Corporate fiduciary duties are particularly ill-suited for correcting complicated, idiosyncratic contracting problems with an input to the firm. Corporate law is a stan(and their contracting dard-form contract, available for firms components) to use "off the rack" when seeking standard language and interpretations. By definition, then, standard-form default language and rules are not intended to deal with the particularized teamproduction situations that individually negotiated contracts solve better.

B. The Team-Production Defense of Corporate Philanthropy

Dr. Blair claims that the Blair-Stout team-production theory provides an economic basis for "socially responsible" corporate philanthropy. It is not clear whether Dr. Blair regards "socially responsible" corporate philanthropy as giving that is only profit maximizing to firms or whether this idea also includes donations that are utility maximizing to managers.70 If she is arguing that their theory justifies managerial discretion to make "socially responsible" contributions consistent with maximizing profits, then no new theory of corporate law is required. As explained above, "socially responsible" corporate philanthropy that increases firm value is consistent not only with

Milton Friedman's famous position on philanthropy, but also with the contractual theory of the corporation in general.

On the other hand, if Dr. Blair is asserting that the Blair-Stout team-production theory justifies and even requires managers to make profit-reducing charitable contributions, then the logic of her analysis needs closer examination. Because Dr. Blair makes strong statements about the superior explanatory power of the Blair-Stout team-production theory,71 the discussion here assumes that Dr. Blair's extension of the team-production theory implies that managers should have the power to engage in true non-profit-maximizing philanthropy.72

Dr. Blair claims that the Blair-Stout theory provides ajustification for corporate philanthropy contrary to Milton Friedman's famous maxim that the social responsibility of business is to maximize profits.73 Normatively, Dr. Blair seems to argue in favor of non-profit-maximizing corporate philanthropy. But she also argues positively that the Blair-Stout theory is superior because it explains why corporate law allows non-profit-maximizing corporate philanthropy.74 Thus, she must mean that corporate law allows socially-responsible (but profitreducing) corporate philanthropy. In essence, utility-maximizing philanthropy requires the diversion of corporate assets to people or institutions who are not members of the team in any meaningful sense.

As a normative matter, Dr. Blair's application of the Blair-Stout team-production theory to philanthropy that is unrelated to advancing firm profits appears to go beyond the meaningful boundaries of the firm. At some point, the firm must end and the market must begin. It is admittedly difficult to identify the precise location of the

boundary line between the two. But the claim that "team production" implies that directors must have discretion to distribute corporate assets to anyone they please has no logical limits. Surely, the limits of directorial discretion must relate to either the "team" aspect or the "production" aspect of "team production." Unless the "team" includes everyone in society, then, by definition, philanthropy that does not further profits is not productive to members of the team in any meaningful sense.75

As a positive matter, it is not at all clear that corporate law doctrine is as permissive as Dr. Blair suggests.76 In particular, the doctrine of waste is inconsistent with her application of the Blair-Stout team-production theory of the corporation. The doctrine of waste is a constraint on corporate managers. If they give away corporate assets without a reasonable expectation of benefit for the firm, courts will respond either by voiding the gift (to the extent that it involves corruption) or by holding the real donors financially responsible for it.77 Rules that require the unanimous consent of shareholders to ratify gifts are the clearest evidence that non-profit-maximizing philanthropy is not part of the legal theory of the firm.78

In the end, one should be clear about what is and is not different about Dr. Blair's notions of corporate philanthropy, as there are superficial similarities between her analysis and that which Part H presents above. Dr. Blair's analysis does lead her to the same ultimate point as that in Part II: the law should not concern itself with philanthropy, even that which is not profit maximizing to the firm. But her method of analysis is quite different from that of Part IE.

She views non-maximizing giving as a first-best outcome that serves important economic purposes. But in the ordinary agency model of the firm, permitting philanthropy that is not profit maximizing is not a first-best outcome; it is an inevitable cost that attends philanthropy that truly is profit maximizing. Non-maximizing giving, therefore, is necessary in a second-best world in which judges find it impossible to distinguish between good and bad philanthropy. It is certainly correct that the granting of discretion to corporate managers serves an important economic purpose. As explained above, however, it is also clear that market forces and contracts constrain managerial discretion (of all kinds) much more than the default rules of corporation law.

Dr. Blair finds support for her team-production theory of nonprofit-maximizing corporate philanthropy in the structure of modern corporation law.79 However, the long-term survival of default rules that provide for a wide range of contributions, including philanthropy that may be unrelated to advancing the firm's profitability, does not prove that shareholders or other team members want managers to engage in such philanthropy. The same default rule allows managers to choose the size of their offices, but that choice does not imply that shareholders applaud offices that are larger than necessary. A default rule prohibiting philanthropy driven by managers' personal tastes, not firm profitability, could have the deleterious effect of deterring profit-

maximizing philanthropy. The Type I and Type II error problems would dictate that all philanthropy be approved, even when some charity is admittedly not motivated by profit maximization.

This circumstance is particularly true when there is no evidence that charity driven by managers' tastes rather than firm profits is a large-scale problem. It is even truer when one takes into account the force of market correctives. The evidence indicates that market forces constrain managerial discretion to advance their own utility, even when legal rules permit greater discretion. Thus, Dr. Blair's reliance on the structure of corporation law to support her extension of the Blair-Stout team-production theory is misguided, because it is based on a purely legal model of the corporation-a model that ignores the economic power of markets to constrain managerial discretion.

But in another important regard, Dr. Blair is not concerned about managerial discretion or giving managers the legal incentives to act in shareholders' interests.80 Traditional corporate law concerns about the "separation of ownership and control" and "principal-agent problems" are not important in Dr. Blair's approach to corporate philanthropy because, in her view, shareholders are not the "owners" of the corporation. Dr. Blair states:

An essential feature of that [corporate] governance structure is that all of the participants in the firm agree to give up property rights over key inputs used in the joint enterprise, as well as any direct claims to the outputs. Inputs contributed by various stakeholders (especially those contributed by shareholders and employees) become the property of the corporation itself, and decisions about their use and allocation are governed by an internal hierarchy. At the top of that hierarchy is the board of directors, which has extremely wide discretion under the law to make decisions about the use of the assets, and about the allocation of any economic surplus (rents) generated by the enterprise. This type of governance structure (a hierarchy headed by a board of directors) is virtually unique to the publicly-traded corporate form. Careful contractarian scholars, then, understand that calling shareholders the "owners" of corporations, or referring the assets of corporations as "shareholders' property," or the profits as "shareholders' money," is a rhetorical trick that, while powerful, is misleading, since it is neither an accurate description of the legal role that shareholders

to

Regardless of the accuracy of this statement, or even the charge of employing rhetorical tricks,82 there remains the fundamental question of whose property managers are giving away when they engage in non-profit-maximizing philanthropy.

Again, the contractual theory provides the answer. The corporation is a nexus of contracts that determines the rights (and priority of rights) to the cash flows that the combination of inputs generates. The shareholders are the residual claimants who receive what remains after the firm has met all explicit and implicit contractual claims. Under this nexus-of-contracts approach, the nature of corporate philanthropy determines whether or not the managers have diverted shareholders' cash flows against their contractual expectations. By definition, profit-motivated philanthropy does not harm shareholders. If the managers expect a charitable contribution to increase the value of the firm, then they have given nothing away. The philanthropy is an investment.

On the other hand, if philanthropy of Dr. Blair's "socially responsible" type is not expected to increase firm value, and in fact causes a reduction in firm value, then it is clear that the managers have distributed the shareholders' wealth. It is important to recognize that shareholders are the only stakeholders in the firm that have no guarantee (or payment in advance) of a fixed price.88 Shareholders are the residual claimants on the "bundle of cash flows," and the waste of nonprofit-related charity reduces the amount of the residual. Either the money goes to the shareholders (who own the residual claim), or the firm gives it away. When managers engage in philanthropy that advances their own utility but not firm profits, they are giving away the shareholders' money. This conclusion holds under any characterization of the corporation-resources diverted to corporate charity are the shareholders' wealth regardless of whether the corporation is an entity, a nexus of contracts, a bundle of assets, or a team. Dr. Blair's

theoretical contribution gives managers the ability to separate shareholders from their wealth.

Yet Dr. Blair claims that we need not be concerned about the "separation of ownership and control"-which, of course, is the traditional concern of corporation law-because under the Blair-Stout theory the "team members in a corporation will work out among themselves... who zigs when, who zags when, who is responsible for what, and who gets what."84 This view reflects some very strong assumptions about the loyalty of managers, once they have a license to engage in philanthropy that is not profit-related, to act not only in the residual claimants' best interests, but also in the best interests of society.

In this regard, Dr. Blair rejects the Friedman view of the social responsibility of corporations in favor of making socially responsible contributions in the name of the corporation's responsibility to the public.85 An implicit assumption in this approach is that managers and directors have an absolute advantage over shareholders in deciding what is "good for society." However, shareholders may have very different views on what is good for society. Even if they do not, there is no reason to channel non-profit-maximizing charity through the firm. The firm has no advantage-in greater benefits or lower costsin making donations that profit-maximization does notjustify. Shareholders have no need to give at the office when giving at home produces the same results at the same costs.

It is highly unlikely, of course, that management will choose the same charities as would shareholders.86 By definition, non-profit-maximizing philanthropy derives from management's own tastes. Instead

of allowing corporate managers the discretion to impose "taxes" on shareholders, it might be best and more democratic for governments to address pressing social problems. Corporate managers have enough trouble meeting the challenges of maximizing shareholder value without diverting their attention to saving the world.

Under the standard portrayal, corporate philanthropy can be either good (profit maximizing) for shareholders or good (utility maximizing) for management, which advances its own interests with shareholders' assets. But as illustrated here, that assertion is too Manichean a view of corporate philanthropy. In a (real) world in which shareholders gain by having management make decisions about corporate philanthropy, even personally-motivated management decisions still can work to shareholders' advantage overall, given the apparent advantages of giving at the office. When personally motivated decisions begin to outweigh the advantages of giving at the office, competitive markets will tend to penalize managers and corporations that engage in corporate philanthropy that hurts share value overall. Market forces act as the primary constraint on corporate management. Legal constraints on corporate philanthropy are largely irrelevant.

Of course, an omniscient judiciary would make the world a better place by separating good from bad philanthropy, punishing only the latter. Butjudges are not omniscient, and separating good from bad philanthropy would be virtually impossible to accomplish from outward appearances. Liability for philanthropy would often be meted out when in fact the motivation was firm profitability (Type I error). Avoiding that error by allowing managers wide discretion over philanthropy of course means that some bad philanthropy will go unpunished (Type II error). But given that there is no evidence that bad philanthropy is widespread and that shareholders have their own remedies for any defalcations, the law's relatively permissive attitude concerning philanthropy makes perfect sense.

The economics and law of the philanthropy "problem" see philanthropy as just one of any number of situations in which manageragents can advance their own welfare at the expense of shareholders. Philanthropy is just one manifestation of a generic agency problem, but one that is easily analyzed in the contractarian model of the firm. Attempts to invent new explanations are unnecessary and, in the case of Dr. Blair's advocacy of unbridled corporate giving, pernicious.

It is difficult to accept the logic of a theory of corporation law that distinguishes itself on the ground that it provides a rationale for managers to harm shareholders. The principal-agent model of the

firm recognizes that managers are not altruists, and expects that shareholders will tolerate a certain amount of excess charity motivated by managers' personal utility. The law's general refusal to interfere with philanthropic decisions within the firm is tolerable as well, given the overall benefits of philanthropy to the firm, the general sense that profit maximization motivates most philanthropy, and the difficulty of distinguishing profit maximization from utility maximization. But tolerance of inevitable costs is not the same as applause for philanthropy not motivated by firm profitability.

--- FOOTNOTES ---

Henry N. Butler4 & Fred S. McChesneylt

You've got to give a little Take a little...

-TnE FiV KEYsI1

INTRODUCTION

t Fred and Mary Koch Distinguished Professor of Law and Economics, University of

Kansas. t"

James B. Haddad/Class of 1967 Professor, Northwestern Law School and Professor, Kellogg Graduate School of Management, Northwestern University. The authors thank Dick Biondi, Bobby Comstock, Brian Goodman for research assistance, Alan Palmiter and Larry Ribstein for helpful discussions, and especially William Carney for very useful comments.

1 Billy Hill, composer, The Glory of Love (Aladdin Records, 1951) (ASCAP). 2

See generallyJAmEs WILLARD HuRsT, THE LEGITIMACY OF THE BUsiNESS CORPORATION IN THE LAWv OF THE UNITED STATES, 1780-1970 (1970) (discussing the historical changes in public policy regarding corporations).

See, e.g., Theodora Holding Corp. v. Henderson, 257 A.2d 398, 405 (Del. Ch. 1969) (noting that a loss of shareholders' profits because of a corporate gift "is far out-weighed by the overall benefits flowing from the placing of such gift in channels where it serves to benefit those in need of philanthropic or educational support").

4 E.g., Melvin Aron Eisenberg, Corporate Conduct That Does Not Maximize Shareholder Gain: Legal Conduct, Ethical Conduct, the Penumbra Effect, Reciprocity, the Prisoner's Dilemma, Sheep's Clothing, Social Conduct, and Disclosure 28 STETSON L. REv. 1, 1 (1998) ("This paper is part of a symposium on the problem of corporate philanthropy."); Faith Stevelman Kahn, Pandora's Box: ManagerialDiscretion and the Problem of Corporate Philanthropy, 44 UCLA L. REV. 579 (1997); see also Marianne Jennings & Craig Cantoni, An Uncharitable Look at Corporate Philanthropy, WALL ST.J., Dec. 22, 1998, at A18 (criticizing Boeing Corporation for its phi-

lanthropy of $51.3 million in 1997, despite losing $178 million that year and eliminating 48,000 jobs in 1998).

5 Friedman stated:

What does it mean to say that the corporate executive has a "social responsibility" in his capacity as businessman? If this statement is not pure rhetoric, it must mean that he is to act in some way that is not in the interest of his employers. For example,. ..that he is to make expenditures on reducing pollution beyond the amount that is in the best interests of the corporation or that is required by law in order to contribute to the social objective of improving the environment. ...

[The problem in this case is that] the corporate executive would be

spending someone else's money for a general social interest.

Milton Friedman, A Friedman Doctrine-The Social Responsibility of Business Is To Increase Its Profits, N.Y. TIMEs, Sept. 13, 1970 (Magazine), at 32; see also MILTON FRIEDMAN, CAPrrALISM AND FREEDOM 133 (1962) (stating that "there is one and only one social responsibility of business-to use its resources and engage in activities designed to increase its profits").

6 For example, Exxon and Mobil have become so famous for their corporate charity that commentators have expressed concerns that the announced merger of these firms will cause their combined levels of charitable giving to fall. SeeJ.C. Conklin, Cultural Groups Worry Mobil-Exxon Deal May Portend Drop in Charitable Giving, WALL ST. J., Dec. 2, 1998, at A8; Sara E. Mel6ndez, Letter to the Editor, Exxon, Mobi" Philanthropic Champs, WAIL ST.J., Dec. 23, 1998, at A15; Irvin Molotsky, Corporate Medici Lost to Mergers, Arts Groups Fear, N.Y. TIMEs, Jan. 5, 1999, at El.

OF CORPORATE LAw IN CONTROLLING

MANAGERIAL DISCRETION

See Margaret M. Blair, A Contractarian Defense of Corporate Philanthropy, 28 STETSON L.

REv. 27, 33 (1998).

See ADAM SMrTH, AN INQUIRY INTO THE NATURE AND CAUSES OF THE WEALTH OF NA-

TIONS 3-21 (Edwin Cannan ed., The Modem Library 1937) (1776).

See Armen A. Alchian & Harold Demsetz, Production, Information Costs, and Economic

Organization, 62 AM. EcoN. REv. 777, 779-81 (1972).

See R-H. Coase, The Nature of the Firm, 4 ECONOMICA 386, 390-92 (1937).

10 11 See Merton H. Miller, The Modigliani-Miller Propositions After Thirty Years, J. ECON. PERsP., Fall 1988, at 99; Franco Modigliani & Merton H. Miller, The Cost of Capitai Corporation Finance and the Theoy of Investmen; 48 AM. ECON. R v. 261 (1958).

12 To the extent that managers do own the firm-that a firm is "closely held"-the

problems that this Article addresses do not arise for the most part.

See ADOLF A. BERLE, JR. & GARDiNER C. MEANS, THE MODERN CORPORATI1ON AND

PRIVATE PROPERTY (1932).

14 See generally Michael C. Jensen & William H. Meckling, Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure, 3J. FIN. EcoN. 305 (1976), for a development of the principal-agent concept. The principal-agent problem within firms has its analogue in spot labor markets in the ability of one contracting party to shirk its performance for the other.

15 See BERLE & MEANs, supra note 13, at 233-46 (discussing the inability of the law at

the time to control corporate managers).

See, e.g., MELvIN A. EISENBERG, THE STRuCruRE OF THE CORPORATION: A LECAL ANAL-

-iSIS (1976).

17 For a synthesis of the literature, see Henry N. Butler, The Contractual Theory of the

Corporation, GEO. MASON U. L. REv., Summer 1989, at 99.

18 For a recent demonstration and citations to the relevant literature, see Werner Gfith et al., An Experimental Study of a Dynamic Principal-Agent Relationship, 19 MANAGERIAL & DECISION ECON. 327, 339 (1998) ("While principal-agent theory is exclusively built on incentives, actual behavior is sometimes better explained by trust on the side of principals and reciprocity on the side of agents.").

19 For summaries of the empirical research, see Frank H. Easterbrook, Managers'Dis-

cretion and Investors' Welfare: Theories and Evidence, 9 DEL. J. CORP. L. 540 (1984); Gregg A. Jarrell et al., The Market for Corporate Control: The Empirical Evidence Since 1980, 2 J. ECON. PERsp. 49 (1988); Michael C. Jensen & Richard S. Ruback, The Market for Corporate Control: The Scientific Evidence, 11 J. FIN. ECON. 5 (1983).

20 The seminal article is Henry G. Manne, Mergers and the Market for Corporate Control,

73J. POL. ECON. 110 (1965).

21 Because competition in all of these markets has increased with economic globaliza-

tion and the implementation of new information technologies, managerial discretion is more constrained today than at any time since the publicly traded corporation emerged as the primary vehicle for large business firms in the late nineteenth century.

22 SeeBarry D. Baysinger &Henry N. Butler, The Role of Corporate Law in the Theory of the

Firm, 28J.L. & ECON. 179, 180-81 (1985).

See, e.g., DEL. GEN. LAw tit. 8, § 102 (b) (1) (1991); MODEL Bus. CORP. Aar

§ 2.02(b) (2) (iii) (1998).

24 For an excellent discussion of the basic allocation of directors' powers and duties, see ROBERT CHARL.S CLAR, COR'ORATE LAW § 3.2 (1986). In buying a share of stock, shareholders in effect are paying for the services of directors. Directors who fail to function as boards violate their contract with shareholders. Shareholders cannot lose the promised services of directors because of a mere majority vote of shareholders to remove direction of the firm from the board. See, e.g., McQuade v. Stoneham, 189 N.E. 234, 236 (N.Y. 1934) ("Directors may not by agreements entered into as stockholders abrogate their independentjudgment."). Should they unanimously waive the services of their board after purchasing their shares, however, shareholders may dispense with the board's services both statutorily and by common law. See, e.g., Galler v. Galler, 203 N.E.2d 577, 585 (Il1. 1964) (stating director role in the context of a closely held corporation); RmSED MODEL Bus. CORP. AcT § 7.32(b) (1).

25 For a summary of the literature on this subject, see Henry N. Butier, Boards of Direc-

tors, in 1 THE NEv PALGRAVE DICTIONARY OF ECONOMICS AND THE LAxW 165 (Peter Newman ed., 1998).

See Eugene F. Fama & Michael C.Jensen, Agency Problems and Residual Claims, 26J.L.

& EcoN. 327, 331 (1983).

See William J. Carney, The ALl's Corporate Governance Project: The Death of Property

Rightsi, 61 CEO. WASH. L. REv. 898, 922-30 (1993).

II CORPORATE PHILANTHROPy IN =H CoNTRAcTuAL THEORY OF THE CORPORATION

See Eugene F. Fama, Agency Problems and the Theory of the Firm, 88J. POL. ECON. 288, 294 (1980). According to the contractual theory of the corporation, the market constraints on managerial behavior give managers the incentive to strive to improve the operations of their firms. Managers have great leeway in this regard: they can write incentive contracts; organize teams; create new divisions or subsidiaries; focus on total quality management, value-based management, market-based management, or some other trend in management philosophy; launch massive advertising campaigns; invest billions in longterm research-and-development programs with low probabilities of success; merge, diversify, and spin off; file lawsuits against competitors; lobby the government to file lawsuits against their competitors; lobby regulators and legislators to change laws that help their business; and so on.

29 As do the factors mentioned in the preceding discussion of the incentives that di-

rectors have to monitor managerial performance.

SeeJames M. Buchanan, The Samaritan's Dilemma, in ALTRUIsM, MoRA.U=, AND EcO-

NOMIc THEORY 71 (Edmund S. Phelps ed., 1975); Harold M. Hochman &James D. Rodgers, Pareto Optimal Redistribution, 59 Am. EcON. REv. 542 (1969); see also William M. Landes & Richard A. Posner, Salvors, Finders, Good Samaritans, and Other Rescuers: An Economic Study of Law and Altruism, 7J. LEGAL STuD. 83 (1978) (discussing "rescue" from an economic

perspective); Richard A. Posner, Gratuitous Promises in Economics and Law, 6J. LEGAL STuD. 411 (1977) (discussing "gratuitous promises" from an economic perspective).

31 Eisenberg, supra note 4, at 14 (footnote omitted). 32 Friedman, supra note 5, at 33.

FiGuRE 1

DOLLARS DONALD

33 This assertion assumes that there are no transaction costs in giving, a point this

Article will consider shortly. See infra text accompanying note 34.

34 Of course, corporations also incur administrative costs associated with their contributions. The point here is that the corporation can make decisions and solve collective action problems more efficiently than individual shareholders.

(i.e., at lower transaction costs)

See, e.g., Theodora Holding Corp. v. Henderson, 257 A.2d 398, 405 (Del. Ch. 1969).

36 Margaret Blair recites an example of the sort that Figure 2 models. See Blair, supra note 7, at 46. Shareholders of Occidental Petroleum filed suit against the firm's then-CEO, Armand Hammer, who had committed substantial corporate money for the capital and operating funds needed to create the "Armand Hammer Museum of Art and Cultural Center," to be located adjacent to Occidental's headquarters in Los Angeles. See id. Shareholders eventually agreed to allow the firm to commit a much smaller sum.

37 Of course, one can view the personal gain to the manager as part of her compensation package. Indeed, some commentators suggest that giving managers discretion over the direction of corporate contributions is an effective compensation strategy. See, e.g., Transcript of Proceedings-Corporate Charity: Societal Boon or Shareholder Bust', 28 STETSON L. Rxv. 52, 70-71 (1998) [hereinafter Corporate Charity Transcript] (statement of Margaret M. Blair). But it would take an unusual set of circumstances for compensation to take such a form. A manager who prefers personal philanthropy to additional salary can always give away extra salary. The employee almost always will prefer to receive extra compensation in the form of salary instead merely of capturing some of the value by having the ability to determine how managers make donations. Perhaps there are some constraints on how much firms can pay employees, and thus the granting of this discretion is an important and less costly way of compensating a valuable employee. At a minimum, however, the rule that their contribution decisions not harm the corporation still would constrain the employee.

See, e.g., Eisenberg, supra note 4, at 18-19. In fact, the opinions in cases like Theodora Holding demonstrate that courts are

attuned to that possibility.

B

MC

!

Q,

I

40 Consider in particular that, if successful, challenges to corporate philanthropy

likely will result in personal liability to the donating manager.

41 For a description of Type I versus Type II errors thatjudges may make, see CHRLys

J. GOETZ & FRED S. MOCHESNEY, ANTITRUST LAW: INTERPRETATION AND IMPLEMENTATION 6769 (1998).

42 For example, the dispute between shareholders and management at Occidental Petroleum that eventually settled, see supra note 36, caused an outcry from institutional investors, who felt that shareholders settling the suit (and their lawyers) had been too lenient with management. Indeed, institutional investors challenged the settlement in court, and they publicized the fact that, even with the settlement, management, in their view, was wasting shareholders' money. See Corporate Charity Transcript supra note 37, at 6265 (statement of Nell Minow); Adam Bryant, The Corporate Critic: Nell Minow Uses Her Zealfor Films to Investors'Advantage, N.Y. TimEs, Jan. 19, 1999, at G2.

Interest groups have played important roles in the evolution of corporation law. See, e.g., Henry N. Butler, Nineteenth-CenturyJurisdictional Competition in the Granting of Corporate Privileges, 14J. LEGAL STUD. 129 (1985); Jonathan R. Macey & Geoffrey P. Miller, Toward an Interest-Group Theory of Delaware Corporate Law, 65 TEX. L. REv. 469 (1987).

See Blair, supra note 7, at 47 ("[T] hese contributions exhibit a pattern that strongly suggests that the donations are being made to institutions and causes that are linked to the business goals of the companies.").

See HAROLD GILL REUSCHLEIN & WILLIAM A. GREGORY, HANDBOOK ON THE LAW OF AGENGYAND PARTNERSHIP 128 (1979) ("As one might well suppose, if an agent has received money which he was under a duty to pay to his principal or if he has benefited as the result of breach of duty, he is liable for the value of the benefit received."). Under the Uniform Partnership Act of 1994, a partner must "account to the partnership and hold as trustee for it any property, profit, or benefit derived by the partner in the conduct and winding up of the partnership business or derived from a use by the partner of partnership property." UNIF. PARTNERSHIP AcT § 404(b) (1) (1994).

FImuiE 3

DOLLARs DONATE

46 To analogize, suppose I would like a new Jaguar, but find the cost of buying one more than I am willing to pay. My spouse continually nags me to get the car, to the point that I ultimately buy the Jag, which we both drive. My full loss is not the cost of the car because I derive considerable pleasure from driving it-but just not enough to make it worth buying in the absence of the nagging.

Ti-m TEAm-PRODUrIMON MODEL OF CORPORATE LAW AND PmLANTHRopy

47 Thus, we heartily second the points of Professor Eisenberg, who locates the "problem" of corporate philanthropy within the larger set of agents' non-maximizing behavior. See Eisenberg, supra note 4, at 1.

48 The philosophic principle of Ockham's Razor essentially holds that multiple explanations for a given phenomenon are not needed and that the explanation with the widest application is preferable to others. See 3 THE NEW PALGcRAVE: A DICrIONARY OF ECONOMICS AND THE LAW 691 (John Eatwell et al. eds., 1987). "[P]liurality is not to be assumed without necessity.... [W] hat can be done with less is done in vain with more." Id. Ockham's Razor is also known as the "principle of parsimony." THE OXFORD COMPANION TO PHILOSOPHY Ockham's Razor, or the Principle of Panrimony 633 (Ted Honderich ed., 1995).

See Blair, supra note 7.

49 50 Dr. Blair states that her article on team production derives from an argument "developed in detail" in Margaret M. Blair & Lynn A. Stout, A Team Production Theory of Corporate Law, 85 VA. L. Rzv. (forthcoming 1999). Blair, supra note 7, at 30 n.8. Requests to its authors for a copy of this article failed to yield one. Thus, the present discussion of the joint Blair-Stout approach to team production derives from Blair's solo description of it. Accordingly, this Article is not an attempt to rebut the Blair-Stout team-production theory of corporation law. The Article's purpose is limited to challenging the logic of Dr. Blair's application of the team-production theory, as she describes it, to corporate philanthropy. In keeping with Ockham's Razor, one cannot establish the superiority of the team-production theory of corporation law by asserting that it is consistent with the current law of corporate philanthropy, because the existing contract model of corporate law already fully describes the law.

51 Blair, supra note 7, at 34 (footnote omitted). 52 E.g., L.G. Telser, A Theory of Self-Enforcing Agreements, 53 J. Bus. 27 (1980); see also Benjamin Klein & Keith B. Leffler, The Role of Market Forces in Assuring Contractual Performance, 89J. POL. ECON. 615 (1981) (arguing that market forces serve to enforce contracts). 53 Blair, supra note 7, at 36-38 ("Several solutions to the intractable contracting problems involved in team production have been proposed .... Blair and Stout offer a third solution. They argue that structuring the firm as a publicly-traded corporation helps to solve the team production problem ....

" (footnote omitted .

Indeed, the value of team production, which the gains from specialization animate, is the starting point for the contractarian analysis of the firm. See supra text accompanying notes 9-10.

See Coase, supra note 10.

See Alchian & Demsetz, supra note 9, at 779-81.

See Coase, supra note 10, at 894 ("Naturally, a point must be reached where the costs of organizing an extra transaction within the firm are equal to the costs involved in carrying out the transaction in the open market ....

.").

58 For an application of the Alchian-Demsetz model to the law firm, see, for example, Fred S. McChesney, Team Production, Monitoring, and Profit Sharing in Law Firms: An Alternative Hypothesis, 11 J. LEcAL STUD. 379 (1982).

59 Blair, supra note 7, at 36. 60 Corporate Charity Transcrip4 supra note 37, at 69 (statement of Margaret M. Blair). 61 For example, her statement that "[property rights means that one person gets to make all the decisions," id., really misses the significance of property rights in the theory of the firm. The owner of property rights faces the challenges of allocating decision-making rights within the firm, devising methods of rewarding individuals, and designing a structure of systems to evaluate the performance of both individuals and business units. These are important decisions, and if the decisions do not turn out well, the owner of property rights will lose those property rights. For a thorough and accessible treatment of managerial challenges and solutions from the perspective of property rights and principal-agent contracting, see JAmEs A. BRiucia ARcHrraruRE (1997).

ET AL., MANAGERIAL EcoNoamcs AND ORGAzA-noAL

See, e.g., MICHAEL HAMMER & JAMES CHAMPY, REENGINEERING THE CORPORATION (1993); PETER M. SENGE, THE FIFTH DISCIPLINE: THE ART AND PRACTICE OF THE LEARNING ORGANIZATION (1990).

63 Blair, supra note 7, at 35. 64 The seminal contribution to this literature is GARYv S. BECKER, HUMAN CAPrrAL (2d ed. 1975). Becker's work builds on concepts of human-capital investment developed earlier by Theodore Schultz. Schultz and Becker won the Nobel Prize in economics in 1979 and 1992, respectively. See THE FORTUNE ENCYCLOPEDIA OF ECONOMICS 777-78, 832-33 (David R. Henderson ed., 1993).

basic reading skills, or teaching an aspiring paralegal how to conduct legal research. Specific training might include teaching a secretary how to use a law firm's unique filing system. Employees, in the absence of some type of contractual restraint, will tend to bear the costs of general training because the employees are generally free to take it with them to another employer. Often times, employees bear these general training costs in the form of lower wages during a training period or an apprenticeship. Lower salaries for judicial clerks, and entry level associates in law firms can be explained from this perspective. On the other hand, the employer is expected to pay for specific training, since the employer will receive higher productivity from the employee from that training, yet the employee will be unable to receive any benefits in the form of higher wages from that training by moving to another job.65

The Blair-Stout team-production theory of corporation law focuses on firm-specific human-capital investments for which, according to basic theory, the employer should pay. It argues that team production improves when firm-specific investments in employees' human capital increases the productivity of the team. Yet-and this is central to their approach-there are well-known problems with contracting for firmspecific investment in employees' human capital. Employees are concerned that if they make the firm-specific investment, then the firm could expropriate the nonsalvageable investment.66 The firm is concerned that if it pays for the firm-specific investment in an employee's human capital, the employee can threaten to leave the firm unless it pays the employee a premium not to destroy or remove the investment.

In the Blair-Stout world, sub-optimal investments in firm-specific human capital would occur in the absence of corporation law. By focusing on corporation law, they neglect other contractual solutions. In the real world, parties make firm-specific investments because they have found ways to contract around the problems.68 In general, market reactions, reputation effects, and ex post settling up constrain opportunistic behavior by both employees and employers. If so, corporation law has nothing to do with solving the "central contrac-

65 HENRY N. BuTLER, ECONOMIC ANALYsIs FOR LAWYERS 502 (1998) (emphasis omitted). Economists have built on these basic insights to explain some widely observed exceptions, such as why firms sometimes make general investments in their employees and why employees sometimes make firm-specific investments in their employer.

See generally Benjamin Klein et al., Vertical Integration, Appropriable Rents, and the Com-

petitive Contracting Process, 21 J.L. & ECON. 297 (1978).

See id. at 298 ("After a specific investment is made[,] ... the possibility of opportu-

nistic behavior is very real.").

68 For a convincing argument that the contracting problems facing employees are not as difficult as some corporate law commentators suggest, see William J. Carney, Does Defining Constituencies Matter?, 59 U. CIN. L. REv. 385 (1990).

69 Blair, supra note 7, at 35. 70 For example, the following statement is consistent with both profit-maximizing and

utility-maximizing corporate philanthropy:

Although the team production theory of corporate law was derived from economic reasoning, it suggests that corporations are fundamentally political and social institutions. Since team members can be expected to use the political tools available to them, in addition to economic and legal tools, to try to capture as much of the rents from the joint enterprise as they can, corporate managers and directors must also be allowed to use such tools. They must be free to encourage the participation and cooperation of employees, for example, with a variety of incentives, including moral suasion, social pressures, and gift exchanges, in addition to contractual reward and punishment incentives. And, they may also find it necessary or useful from time to time to utilize those tools in their relationships with customers, or lenders, or suppliers, or even community leaders, voters, and members of the media.

Blair, supra note 7, at 48-49 (footnotes omitted). If these activities seek to increase the value of the business, then they are consistent with profit-maximizing philanthropy and the contractual theory of the corporation. Under the contractual theory, it is also important for managers to use corporate resources to manage their legal environment.

71 Empirically, Dr. Blair focuses on two cases that she claims are inconsistent with the contractual theory of the corporation. See id. at 45-46. But surely, the identification of two cases out of millions of corporate "charitable" transactions cannot represent an empirical validation of either the Blair-Stout theory or Dr. Blair's application of it to corporate charitable contributions.

72 This seems a reasonable interpretation of Dr. Blair's position, which she emphasizes is different from that of Milton Friedman: "I argue that Friedman's conclusions apply only to a very narrow and special case ....

" Id. at 29. See id. ("This paper provides an alternative defense of managerial discretion with respect to corporate philanthropy that embraces the contractarian reasoning of Friedman and his proteges, but follows that reasoning to a different conclusion.").

74 Dr. Blair states:

Under a theory of corporate law based on a "bundle of assets that belongs to shareholders" model, these decisions would be troubling. But under a theory of corporate law based on a model of corporations as solutions to team production problems, these decisions seem much more understandable.

... Professor Stout and I claim that this view of the nature and purpose of corporations goes a long way toward explaining aspects of corporation law that do not make sense when viewed from the perspective of principalagent theory.

Id. at 48.

Indeed, the title of Dr. Blair's article, A Contractarian Defense of Corporate Philan-

thropy, assumes that members of the team must contract to benefit society. Dr. Blair does not specify why team members would choose the corporation as the device to achieve this goal. Alternatively, the title could represent a new social contract theory-we're all on one big team. If Dr. Blair's leap from internal team production, as well as the resulting contracting problems, provides the basis for a social contract, then there is no logical limit to who is on the team. See id

In Dodge v. Ford Motor Co., 170 N.W. 668 (Mich. 1919), plaintiffs claimed that Henry Ford operated the firm "as a semi-eleemosynary institution," id. at 683, rather than maximizing shareholder profits. The court stated:

There should be no confusion (of which there is evidence) of the duties which Mr. Ford conceives that he and the stockholders owe to the general public and the duties which in law he and his codirectors owe to protesting, minority stockholders. A business corporation is organized and carried on primarily for the profit of the stockholders. The powers of the directors are to be employed for that end.

170 N.W. at 684. A recent survey found Dodge v. Ford Motor Company at the top of a list of the 10 most important judicial decisions concerning corporate law. See Charles M. Elson, Courts and Boards: The Top 10 Cases, DiRCroRs & BoARDs, Fall 1997 ("The court did not rule, however, that a corporation was prohibited from making charitable donations, only that one must have some valid long-term corporate motive for doing so-the ultimate maximization of corporate profitability and consequent shareholder wealth.").

In the important case of Revlon, Inc. v. MacAndrews & Forbes Holdings, Inc., 506 A.2d 173 (Del. 1986), the Delaware Supreme Court rejected management's claims that it was authorized to help bondholders at the expense of stockholders. See id. at 175-76. The decision specifically states that management can only provide this help when it also benefits (or at least does not hurt) stockholders. See id. at 176. This decision, of course, is consistent with profit-maximizing philanthropy.

78 As discussed supra note 70, Dr. Blair argues that corporations must be free to encourage the participation of employees with "gift exchanges, in addition to contractual reward and punishment incentives." Blair, supra note 7, at 49. But the doctrine of waste always has put a constraint on gifts: unless there is some anticipated benefit to the "corpo-

ration," however indirect, gifts constitute waste, as in the case of a bonus to a departing employee. If the "gifts" seek to encourage participation and cooperation of employees, then they are consistent with the contractual theory of the corporation, and Dr. Blair is saying nothing new about the theory of the firm.

79 Dr. Blair argues:

If participants in the joint enterprise being undertaken by the firm are uncomfortable with giving corporate executives and directors such wide discretion, they could, presumably, structure their relationship differently. They could organize the firm as a limited partnership or a close corporation rather than a public corporation; or they could lend the firm money rather than investing in equity; or they could work as subcontractors rather than as employees. The fact that so much economic activity is organized within the private governance structures created under the law of public corporations suggests that participants in these enterprises find it in their interests to operate this way. This, in turn, suggests that the discretion granted to corporate managers and directors is serving an important economic purpose.

Blair, supra note 7, at 49.

80 Dr. Blair seems to assume that managers will just do the right thing. See, e.g., id. at 47 (stating that "while abuses do occur, the available evidence suggests that corporate managers and directors have not, in general, grossly abused the discretion given them by the legislatures and courts"). Perhaps this lack of concern about principal-agent problems reflects the implicit understanding that markets generally are constraining managerial discretion. But if that is her view, then there is no reason to place so much emphasis on the particular legal rules governing corporate philanthropy.

play in corporations nor a particularly informative statement about their economic role.81

Id. at 31.

82 To say that "stakeholders" "contribute[ I" inputs that are "governed by an internal

hierarchy" ignores the form of and motivation for the contributions. Id. Virtually all employees sell their labor, either at will or by contract. They do not contribute it as that term is ordinarily understood; they expect some contingent return, and they get a specified price for their services. See Carney, supra note 68, at 405; see alsoJonathan R. Macey, Externalities, Firm-Specific Capital Investments, and the Legal Treatment of Fundamental Corporate Changes, 1989 Durra L.J. 173, 179-80 (arguing that participants in a corporation trade rights and obligations to those "who value them the most"). Rhetorical tricks that blur the line between firm and market transactions are a first step toward developing a theory of corporate philanthropy without logical limits. See Macey, supra note 82, at 180.

84 Blair, supra note 7, at 40. 85 Dr. Blair claims that corporations always have had a "public purpose," id. at 42, and she uses this claim to justify her expansion of managerial discretion to encourage socially responsible charitable contributions. Although it is true that special corporate charters and general incorporation statutes usually state that a corporation has a public purpose, this boilerplate fails to distinguish those statutes from any other legislation. Most legislation is cloaked in some form of public-regarding language. See generallyJonathan R. Macey, Promoting Public-Regarding Legislation Through Statutory Interpretation: An Interest Group Mode 86 COLUM. L. REV. 223 (1986) (arguing that statutory interpretation can ensure publicregarding legislation). Moreover, even though many early special corporate charters created corporations for the specific purposes of providing public infrastructures such as bridges and toll roads, it is clear that most of those acts were special-interest legislation conferring rents on politically influential rent seekers. See generally Butler, supra note 43 (offering several examples of this process and hypothesizing that the rapid growth of interstate commerce helped destroy state legislators' intrastate monopolies). Finally, to the extent that some statutes still do charge corporations with promoting the "public purpose," that language is just as consistent with the views of Adam Smith and Milton Friedman as with the interpretation of Dr. Blair.

86 See Carney, supra note 68, at 417-24, for a detailed description of the difficulties that arise when a board with multiple responsibilities must confront interest groups with different preferences.

CONCLUSION