Daniel J.H. Greenwood

Home | Previous Page

Faith Based Investing: Why Shareholders Receive Dividends

Daniel J.H. Greenwood*
Forthcoming as:
The Dividend Puzzle: Are Shares Entitled to the Residual
in Dana Gold (ed.), Law and Economics: Towards Social Justice, Vol. 24 of Research in Law and Economics (Amsterdam: Elsevier 2008).

Table of Contents

I. Introduction
II. The Economics of Dividends

III.  Beyond Market Rationality IV. Conclusion

I. Introduction

Everyone knows that shares receive dividends because they are entitled to the residual returns of a public corporation. Everyone is wrong. 

Basic principles of market competition, applied to the actual law of corporations, sharply contradict the conventional wisdom.  First, shareholders are not residual claimants in any legal or economic sense so long as the corporation remains in existence.  Second, standard economic theory implies that shareholders should receive dividends only under non-competitive conditions, and not often even then. 

Standard accounts elide the problem by a number of mutually contradictory rhetorical tropes, each seeking to portray payments to shareholders as philosophic entitlements or market necessity.  None stands up to serious questioning.  Instead, payments that corporations make to shareholders result from the then-current economic and political power of the stock market and, especially since the taming of the hostile takeover market, from ideological victories in a multilateral struggle over corporate surplus.  Shareholder returns depend on shareholder power—both the soft power of ideology and the hard power of law and the legally structured market—at least as much as on underlying business success. 

Moreover, because the struggle is largely over rents, if the current results are unattractive, we can change the rules, modify the power of the various players, or seek to persuade them to accept different views of their ethical rights and obligations.  The distribution of rents matters: if corporate returns go to shares, we may generate more investment but we definitely end up with far more inequality, short-term outlooks, and tolerance of anti-social externalization; if returns go to top managers, we reward entrepreneurialism but invite corruption; if returns go to rank and file employees, they are shared most widely but there may be a cost in lost flexibility; if returns remain inside the corporation, they may contribute to growth or simply subsidize slack.  But nothing in the nature of the corporate form requires that the current distribution be viewed as sacred entitlements or necessary consequences of our way of life.

II. The Economics of Dividends

On the standard economic view, shareholders contract to receive corporate profits, which are the same as the residual, that is, what is left over after all contracting parties have been paid their contract amounts.[1]  This account is quite problematic.  

Public shareholders are perfectly fungible providers of the most perfectly fungible of commodities in our most competitive of markets.  In a competitive market, the price of a commodity should be its marginal cost.  Accordingly, shareholders should expect to be paid no more than the marginal cost to the firm of retaining them.  But corporate law provides that shareholders have no right to corporate distributions at any given time—shareholder contributions are permanent capital which the firm may retain without further payment for as long as it deems appropriate.  The marginal cost of existing shareholders is therefore zero, and that should be the price they can command in a competitive capital market.  Similarly, in competitive product markets, any firm that paid shareholders a current return would be uncompetitive:  its costs, and therefore its prices, higher than those of competitors that defer shareholder payouts.  Moreover, even if the firm has disequilibrium or monopoly profits, it need not give them to shareholders at any given time.  In a system of rational maximizers, dividends would always be paid tomorrow, and tomorrow would never arrive.  Accordingly, under standard economic pricing theory shareholders should expect no return.

Shareholders in public firms have no contractual or other legally enforceable right to any payment at all.  Thus, dividends bear a certain resemblance to bonuses or tips–an after-the-fact payment enforceable only by custom, not law.  Traditional tips, however, occur in personal service markets where repeating personal relationships and reputational concerns can enforce after-the-fact payments without recourse to contract; bonuses, similarly, occur in multi-period markets where bonus payers have incentives not to renege because disappointed recipients will quit (and employees have sufficient market power to be willing to accept the risk that they will be fired without full payment).  In contrast, shareholders are both fungible and anonymous.  Standard explanations for the stability of tips or bonuses will not work in this context.  Instead, the usual economic models predict that dividends as know them should not exist.

Shareholders of closely held corporations, unlike publicly held ones, actually control the firm and, like real owners, can use that power to simply take any corporate surplus (disequilibrium profits) that may exist.  Therefore, closely held firms do not present the full dividend puzzle.  In the era of hostile takeovers, public shareholder returns were based primarily on the threat that shareholders would sell the company to a private owner.  When public shareholders had the power to force managers to accept takeovers, managers had to treat public shareholders much as if they had the powers of a dominant owner because otherwise they would sell to someone who would have those powers.[2]  Today, however, takeovers are quite difficult without the consent of incumbent boards.  To explain the continued high returns to public shareholders, we must look to ideology.  Shareholders receive returns, in large part, because corporate decision makers think that they ought to.

The following sections explain these claims in detail.

II.A.  Defining Profit

The conventional claim is that shareholders are entitled to the corporation’s profit, but profit is an ambiguous term.  For current purposes, the accounting definition is circular; the claim must rely, instead, on the economic understanding. 

By accounting convention, profits are the difference between a corporation’s income and expenses, but accountants label dividends and similar payments to shareholders (e.g., stock buybacks) as distributions of profit rather than corporate expenses.[3]  In this trivial sense, of course, shareholders necessarily receive the residual (or at least part of it) by definition: accountants classify any corporate proceeds that go to shareholders as residual, but call sums committed to any other participant expenses.   On this definition, the standard view is just a claim that shareholders are entitled to receive whatever shareholders receive.

The standard economic understanding of profit is more useful here.  Accounting profits are a measure of what remains in the firm or is distributed to shareholders in any given period.  In contrast, economic profits measure the surplus the firm has created, regardless of whether it retains it or to whom it distributes it.  The concept should be familiar to all lawyers who deal with closely-held businesses: economic profit is what we have mind when we talk about closely-held firms subject to the corporate income tax lowering their taxes by paying out profits to their shareholder/owner in the form of rent, interest, or wages rather than dividends. 

Economic profit is usually defined as the difference between the lowest expenses the corporation would have had, if it paid the lowest (quality adjusted) market price for its inputs and the income it would have had, if had charged the highest (quality adjusted) market price for its output.  Defined in the subjunctive in this way, economic profit is the profit the firm could make, given its skills, technology and competitors.  It represents the pool of funds that can be given to any corporate participant without threatening the viability of the firm—not only shareholders, but other investors, suppliers, employees, taxing authorities or consumers.   

Thus, the economic concept clarifies the standard view.  Rather than an empty claim that whatever shareholders receive should be labeled profit, it is a falsifiable claim that corporations do, or should, distribute the difference between the price they could charge and the costs they must pay–their economic profit—to shareholders rather than to any other corporate participant.  That is, the standard view is a claim that employees, suppliers and debt holders should be paid as little as possible, and customers charged as much as possible, in order to maximize the amount available for distribution for shareholders. 

On the economic understanding, it should be clear that, contrary to accounting convention, only some payments to shareholders should be called profit.  Conversely, some payments to other corporate roles or factors of production should be considered distributions of profit rather than costs.  In free markets, factors of production will participate in the firm only if they are paid to do so.  Accordingly, payments to any factor must be classified as costs to the extent that they are necessary to purchase the asset or service needed.  For these purposes, capital—whether contributed to the firm in the form of equity or debt—is no different than labor or raw materials.  The firm must pay if it is to attract the financing it needs—whether it pays in the form of interest, dividends, promises of future capital gains, or (as sometimes happens in closely-held corporations) more imaginative ways. 

One part of what accountants call profit, then, must be reclassified as a cost, not part of the residual at all.  Normal returns to capital—the market cost of capital adjusted for the risk associated with the particular investment—are a cost of doing business just like wages or raw materials.  Any payment or distribution the firm must make to attract and retain the capital the firm needs is a cost, not profit.   

Conversely, whenever any factor of production is paid more than necessary—whether that payment is called wages, bonus, interest, dividends, stock buy-back, loan, rent, price, featherbedding, sloth or even green cheese—the unnecessary payment should be considered a distribution of economic profit.  Indeed, even if the firm lowers its prices below the market clearing price, that forgone profit should be viewed as a distribution of economic profit to consumers. 

Once it is clear that the claim that “shareholders are entitled to the residual” is a claim that shareholders are entitled to economic profits, we can begin to see the full difficulty of the conventional wisdom.  As detailed below, from an economic perspective, it is hard to see how such a deal could be made in a competitive market.  From a legal perspective, it is hard to find any evidence that such a deal has been made.  Normatively, the claim that such a deal should be made defies ordinary understandings of the morals of the marketplace.  The only reasonable conclusion is that if in fact shareholders do receive the residual, or part of it, it must be by a combination of firm monopoly power in the consumer market with shareholder ideological power inside the firm.  In the next sections, I expand these perhaps surprising claims seriatim.

II.B. In Competitive Markets, the Firm Cannot Charge for Shareholders’ Contribution

In competitive markets, competition reduces prices to marginal cost.[4]  That means that at equilibrium the residual—the surplus created by the corporation—goes to consumers, not shareholders.  Even if competition does not preclude the corporation from seizing the residual, economic profits are disequilibrium rents.  No one has a pre-legal claim on rents in a capitalist market.  Thus, any claim that shareholders are “entitled” to the economic profits must be based on some identifiable aspect of their legal rights, not somehow inherent in the nature of the corporation. 

II.B.1. At equilibrium, economic profit goes to consumers   

Standard economic theory predicts that normally rents should be competed away as prices adjust to the marginal cost of efficient producers. In other words, in competitive markets at equilibrium, the residual goes to consumers. 

Corporations exist when they are able to organize employees, capital and other resources to produce a product or service for less than the highest price consumers are willing to pay.  Economic profit–the difference between the firm’s marginal cost and the marginal consumer’s reservation price—is the social value the firm creates. 

Ordinarily, no firm participant has a special claim to being the source of the corporate surplus.  On the one hand, each participant needs the firm:  if it did not, it would be able to produce without incurring the firm’s organizational, decision-making and bureaucratic costs, and drive it out of business. [5]  On the other hand, each of the firm’s participants is likely to be a but-for cause of the surplus: if the firm did not need them to produce its product, why would it retain them?  Thus, the very existence of the firm is evidence that any residual it creates must be credited to the entity itself rather than any participant or factor of production taken separately.  Capital without labor would be as barren as the reverse.

The law recognizes this fundamental unity of the firm in two important ways. First, it assigns ownership of the economic profit to the firm itself in the first instance.  Thus, contract, property and tort law each treat the corporation itself as a legal person, meaning that the corporation, not any of the humans or roles associated with it, is liable for its obligations and entitled to the proceeds of its sales.  Then, the law defers to internal corporate decisions regarding the surplus’s allocation: the authorized corporate decision-making body–the board–is entitled to decide, under the broad protection of the business judgment rule, what contracts to enter into and what uses to make of corporate resources.  Courts rarely intervene in the firm’s own decision-making to protect or promote particular visions of corporate purpose or interests of particular firm participants.  The upshot is that, as a matter of law, the firm—not its shareholders, employees or other participants—owns and controls the firm’s surplus until the firm decides to dedicate it to some particular purpose. 

However, although the firm is the creator and the owner of the surplus, it will be unable to retain it in a competitive market.  Instead, competition should drive prices down to no higher than the marginal cost of the next-to-lowest cost producer.  Under competitive conditions, the surplus created by the corporation goes entirely to customers, not shareholders.

Of course, this does not mean that shareholders should earn nothing.  Shareholders, like any firm input, will refuse to contribute unless they expect to be paid as much as they can command elsewhere, namely, the normal return to capital.  Thus, at first glance, we should expect shareholders to receive at least a normal return.

In competitive markets, the normal return should also be the most shareholders can charge.  If they try to charge any more, corporations seeking capital will go elsewhere.  Moreover, any corporation that agreed to pay capital–a fully fungible commodity–more than a normal return would have above-market expenses and be forced to charge above market prices.  Its competitors in the consumer market, paying only the market cost for capital, would quickly drive it out of business.

In short, shareholders should expect to earn neither more nor less than a normal return, which, in turn, should be just the marginal cost of capital.   That is not economic profit or residual at all.  It is just the part of accounting profits that the economic conception views as a mislabeled cost.

II.B.2.  Shareholder contributions to the firm are a sunk cost that should not be reflected in competitive prices

For firms operating in even moderately competitive markets, there is an additional barrier to being able to distribute surplus to shareholders.  Standard price theory holds that in a competitive market, the firm should be able to command a price equal to the marginal cost of production.  But normal returns to capital are an average cost, not a marginal cost.[6]  As explained below, the marginal cost of shareholder capital contributions is zero. As a result, firms in competitive markets should not be able to include even normal returns to capital, let alone economic profits, in their prices.  Shareholders in competitive product markets should expect no returns at all (and, foreseeing this, should decline to become shareholders in the first place).[7]

The marginal cost of shareholder capital is zero because of the legal rights public shareholders have or, rather, do not have.  Other capital suppliers have clear rights to require the firm to pay them.  Sole proprietors can put money into or take money out of the business largely at will, limited only by fraud and fraudulent conveyance rules. Controlling shareholders in closely-held corporations have almost the same rights as sole proprietors, providing they follow formalities designed to make clear the boundaries between firm and personal funds.  Partners have the inalienable right to end the enterprise and withdraw their capital, even in violation of agreements they have made.[8]  Lenders and bondholders have contractual rights to interest payments and to return of their capital on a date certain.  In contrast, public shareholders have no right to interim payments, no date on which their capital is to be returned, no right to withdraw their capital, and no right to force the firm to wind up, liquidate or sell at any fixed time.[9]  Dividends are after-the-fact payments made in the discretion of the corporation.  As a matter of law, only the board, acting for the corporation as a whole and restrained by its fiduciary duty to the corporation, has the right to transfer corporate funds to shareholders, whether by dividend or otherwise, and it has no obligation to do so at any time.[10]  Similarly, directors must initiate any decision to liquidate or sell the firm.[11]

Individual shareholders may “exit” by selling their shares, but the seller is paid by its buyer, not the firm:  the original capital contribution remains with the firm.  Instead of the legal rights of creditors or the control rights of owners, public shareholders have only a collective political right to elect the board (but not to restrict its discretion to act in the best interests of the firm) that is virtually meaningless so long as the firm remains public.

A corporation has no legal obligation to pay a dividend or equivalent at any given time and shareholders have no legal right to force it to do so.[12]  The firm can always defer dividends to a later date and the final distribution, like the messianic age, can remain always tantalizingly close without ever actually arriving.  

To be sure, popular ideology and occasional courts hold that the firm is supposed to earn a return for shareholders at some point.  But the right to delay means that in any given time period, the firm always has the right to continue to use capital contributed by shareholders without paying for it.  Thus, the marginal cost to the corporation of shareholder capital is zero.   

Under standard price theory, however, prices match marginal—not average—costs at equilibrium in competitive markets.  Any firm that attempts to raise its prices to generate funds to pay shareholders will be underbid by competitors that defer paying shareholders.  At equilibrium, all firms will be forced to reduce their prices to their marginal costs, which do not include paying shareholders in the current period.  Since marginal costs are lower than average costs, all firms in competitive markets can expect to run an economic loss at equilibrium (which accounting conventions will record as a less frightening break-even), and shareholder returns will have to wait until a future that will never arrive. 

In short, in competitive product markets, corporations will not be able to charge anything at all for their use of shareholder capital.  They should, therefore, have great difficulty paying shareholders a normal return, let alone residual or economic profits.  Of course, to the extent that shareholders can predict this in advance, they will simply decline to invest in the first place.  Standard price theory ineluctably predicts that public stock markets should rarely coexist with competitive product markets.

II.B.3.  Residual retained by the firm or distributed to firm participants, including shareholders, must be disequilibrium returns to innovation or monopoly

Looked at another way, if the firm does have a residual (in the sense of economic profits) to give to shareholders, then it must be charging more than its marginal cost.  In turn, this means that it is not be selling a commodity in a fully competitive market at equilibrium.  This could be because of the innovativeness or efficiency of the firm:  if the corporation is able to produce at a lower cost or higher quality than its competitors, it can earn more than a normal return until they catch up.  Alternatively, it could be the result of longer-lasting sources of pricing power:  cascades or network gains in which consumers find benefits in using the same product as other consumers;[13] increasing returns to scale that make the largest producer also the most efficient;[14] legal monopolies such as patent, copyright or trademark protection; or even simple anti-trust violations.[15]

If, on the other hand, competition prevails, the entire surplus created by the firm will go to consumers and the firm’s inputs, including shareholders, will earn no more than their cost at the margin.  The airline industry may be an example: constantly hovering at the edge of bankruptcy, as a group the companies in this highly competitive industry probably have earned no returns for shareholders since deregulation.    

II.C.  Rational Maximizers Will Not Give Corporate Surplus to Shareholders

Consumer markets often are not fully competitive, many products are not commodities, and firms often are able to charge more than marginal cost, thus covering their average costs and even earning disequilibrium economic profits.  Not coincidentally, many publicly traded companies fit into these categories.  Some, such as Proctor & Gamble and Walmarts, are mass producers with increasing returns to scale that allow the largest producers to earn economic profits because their size gives them lower marginal costs than their competitors. Others benefit from network effects that make their product more useful than those of competitors, like Microsoft or Disney, or have legal monopolies of non-fungible trade secrets, trademarks, patents or copyrights, like Coca-cola, Nike and most pharmaceutical and chemical companies.

These economic profits are rents: disequilibrium profits from monopoly or innovation that can be distributed to any participant in the system without efficiency implications. 

Shareholders, however, are perfectly fungible providers of the most perfectly fungible of commodities in our most competitive of markets.  A firm paying shareholders for providing capital has no reason to pay more than a competitive price for their product, even if the firm makes super-competitive returns.[16]  If shareholders attempted to charge more, corporations would simply look elsewhere for their capital (by finding new shareholders, borrowing, or using internally generated funds). 

Moreover, as corporate finance theorists normally assume, the returns capital commands should be a function of the non-diversifiable risk it assumes—which is another way of saying, again, that the price of capital is set in an anonymous market, independent of the (fully diversifiable) risk of success or failure of a particular firm.  In other words, the price of capital—the normal return that firms must pay to attract the capital they need—should be equal to the marginal cost of producing it, not to its productivity in a particular context. 

Accordingly, even if the firm has disequilibrium or monopoly profits, shareholders should not have market power to charge the firm any part of them.  Conversely, it is hard to see why corporations—our epitome of rationality—would voluntarily pay more than the market price.  Rational maximizers do not give free gifts. 

The second puzzle, then, is why the shareholders in a publicly held firm should expect to receive the corporation’s returns to innovation and monopoly rents even when there is a residual.  Real owners (including the shareholder of a close corporation) can simply take the residual; creditors may be able to contract to receive it.  Public shareholders can do neither.  The next sections explain why public shareholders should be seen, like other investors, as fungible sellers of fungible capital that should receive no more than the market price for their product. 

II.C.1.  Fungible capital should command only commodity prices

A firm can obtain capital from three basic sources:  retained earnings, borrowing (from banks or by issuing debt) or equity.  Companies with investment opportunities can finance them internally by using economic profits—that is, by charging consumers more than they must and paying suppliers, bondholders, shareholders, executives, other employees or the public fisc less than they could.  Or they can borrow from banks or by selling bonds, in effect renting capital.  Or they can issue equity, buying permanent capital by granting the rights of stock ownership. Modern finance theory, at least since Miller & Modigliani’s indifference theorem, teaches that that the three are entirely comparable and largely fungible, differentiated largely by price. 

If the financial markets were fully competitive and fully informed, the firm would pay the same risk-adjusted equilibrium cost for capital regardless of which source it tapped. In the real world, of course, financial markets are rarely at equilibrium, and even if they were to hit that magic point, no one would know.  Accordingly, the theoretical insight that the different sources of capital are the fundamentally the same has the paradoxical real world consequence that investors and corporations alike focus their energies on finding the differences that remain.  From the firm’s perspective, there is only one: price. [17]  Corporate finance teaches corporate executives that their job is to maximize corporate returns by obtaining capital for the firm at the lowest possible cost–so firms will switch from debt to equity, for example, depending largely on which market is offering money for a lower price. 

If the sources of capital are interchangeable, however, the position of shareholders is quite strange.  On the corporate finance view, all they are is sellers of cash and, perhaps, some risk-bearing services.[18]  But capital is a highly fungible commodity.  If one supplier seeks to charge more than the going rate, corporations can switch to another.  Equity capital competes directly with debt.  If shareholders were to attempt to charge more than lenders, firms presumably would shift from equity to debt by issuing more debt and using the proceeds to buy back stock, as is done in a leveraged buy out.  Conversely, if equity capital appears to be charging less than debt capital, we would expect firms to increase their purchases of equity, for example by going public or issuing additional stock.

Thus, shareholders should be able to charge no more than the cost of the commodity they have to sell.  In the case of stock sales directly by the corporation, this would be the general cost of money—at equilibrium, the same amount the firm would pay bond holders or bank lenders.  Lenders receive a fee (interest) that is closely related to the general cost of producing money (i.e., general interest rates), adjusted to reflect the expected risk of the particular firm. They do not expect to participate in extraordinary firm earnings, except perhaps to the extent that such earnings reduce risk for which the creditors have already been compensated.  In a competitive capital market, firms should refuse to sell stock if shareholders demand any more than the normal return lenders would receive.

Once stock is issued, however, the shareholder’s position worsens.  Current shares represent a past service–capital transferred to the firm when the share was originally purchased from the firm.  Since shareholders have no right to reclaim the capital represented by their shares, they are contributing no ongoing service.  As we saw in the previous section, past services should be insufficient to even claim a normal return, let alone any share of the economic profit.

In the end, firm participants can obtain more than the marginal cost of their contribution only if they themselves have some sort of disequilibrium or monopoly power.  Public shareholders—because they are fully fungible providers of a fully fungible commodity in a highly competitive market—are the least likely firm participants to have that kind of power.  Thus, even when firms have the market power to charge more than marginal cost so that they can earn economic profits, it is hard to see how shareholders would have the market power to seize them. 

II.C.2.  What have you done for me lately?

Firms are best thought of as a system of ongoing, continuously re-negotiated relationships.  In any given period, contracting parties have a contractual right to receive the payments they negotiated for in the prior period.  Over time, however, the important determinant of what each corporate participant receives is not past contract but current market strength.

Thus, for example, most employees have at-will contracts.  An at-will contract can be terminated at any time for any reason.  This right to terminate means that contractual terms are always subject to renegotiation—at any given time, the contractual period has always just ended.  Friction aside, then, contractual terms in the current period should always be a function of market conditions in the current period.  Other firm participants are in similar positions.  If a supplier has a contractual right to a payment that is no longer justified by current market prices, the corporation will quickly find an alternative supplier, and past contractual rights will, at most, govern past relationships.  If interest rates drop, the firm will replace its past loans with newer ones reflecting current interest rates. 

Even if a particular corporate participant has obtained contractual protection against such renegotiation–for example, employees who have won long term contractual protection for their pensions or against layoffs or bondholders with long term guarantees of interest rates—the protection is likely to be meaningful only to the degree that the firm can escape market competition.  If the firm fails to renegotiate long term contracts to reflect current market conditions, its competitors will.  In even moderately competitive markets, customers will find an alternative supplier providing similar goods at better prices or better goods at similar prices.  Either the legacy contracts will be breached, or the legacy firm will fail; in either case, long-term contracts collapse into spot markets.

Consider the history of the steel, automobile or airline industries.  In each case, the leading companies made extensive promises to employees to pay benefits in the future–pensions, medical insurance, employment through recessions, increasing pay with seniority.  Presumably, the firms received current benefits:  they paid employees less than they would otherwise have had to because of these promises.  But when the time comes for the company to pay for these past benefits, it has a problem—any new competitor automatically has lower costs, even if it is using the same technology, the same knowledge, the same capital, the same employees, and even the same promises (so long as they have not yet matured).

The old company must run faster just to stay in the same place.  Sometimes, organizational stability, smarter design, or locked-in customers will allow it to do that.  But if the newcomers also have disruptive technology (as in the steel industry), or if technical innovation means that work forces are shrinking so that promises to past employees become a larger part of current costs (as in most US factory-based production), or if the new company is successfully able to imitate the older one (as regularly happens to airlines), or worse yet able to provide a better product through better design sense, technical proficiency or managerial skill (as in the car industry), the old contracts are not going to be worth much.  If the firm is to remain competitive, it must find a way to renege on its past promises.  If the contractual beneficiaries are able to insist on their legal rights, the older firm will have higher costs, and in a reasonably competitive market, the effect will be drive it out of business, or at least into bankruptcy, where courts will allow it to reform contracts to reflect current market conditions.  And as the new firms mature, they too will be forced to renege on their promises by newer firms making newer promises for more distant futures.

In short, any time a firm participant forces the firm to pay above (current) market prices for (current) services—or any price at all for past services, no matter how legitimate the legal claim—the economic effect is to give the firm a cost that competitors do not have.[19]  At this point, a new company—even with no additional skill or technical improvements, and, if it can find someone willing to accept its word, even making the same long term commitments—can out-compete the old one.  Mobile factors of production such as capital, executives and the most productive employees will shift to the new one, while the old one is left with the least mobile participants and they, in turn, are left with claims against a failed entity.

Shareholders present the same issue as legacy cost employees, but more so.  Like employees with future expectations, shareholders make their contribution in one period and expect to be paid at a later time; firms that actually pay, then, will automatically be at a cost disadvantage relative to younger firms that do not yet have to pay or older ones that find a way to renege on past promises. 

The problem is worse for shareholders, however, because unlike employees, shareholders are entirely fungible with no continuity advantages to offer.  Experienced employees may learn the idiosyncrasies of the company’s technology and bureaucracy, the special needs of customers or how to get the most out of suppliers.  They may be better able to make the incremental improvements that characterize progress in many fields or better able to maintain old machines or long-standing relationships.  These bodies of hard and soft knowledge can be hard for an upstart to recreate.  Where this is so, an older company may be able to maintain competitive costs with a newer one despite legacy promises.  Shareholders, in contrast, represent nothing more than past cash investment.  If a new firm can obtain capital more cheaply than the old one, there is no possibility of an offsetting legacy advantage stemming from continuity.

The consequence should be clear.  Like retirees and unionized workers in older manufacturing industries, shareholder claims are a cost without a current benefit.  Managers in even a semi-competitive environment will feel compelled to reduce these costs to the maximum extent possible.  Trained to view equity capital as largely interchangeable with retained earnings and borrowed money, corporate managers should look askance as shareholder claims to the residual:  in a capitalist economy, no one has a right to rents, and commodity providers are never entitled to more than the cost of replacing them.  If shareholders seek current payments for past services, the rational manager’s first response ought to be to just say no.

II.C.3.  Waiting for Messiah

Managers, however, have a better option.  Instead of rudely telling shareholders to get lost, they can politely tell them to wait their turn.  The result will be the same.

One of the core doctrines of corporate law is that the managers have no obligation to maximize shareholder returns in any particular time frame.[20]  The rational managerial response to shares, then, is to promise great returns in the future.  Later, we should see managers promising, they will pay huge dividends; now, they need to use shareholder contributions in order to grow the company.[21]  Later, however, as every procrastinator knows, need never arrive.[22]  As in Red Queen’s contract, it’s “jam tomorrow and jam yesterday, but never jam to-day.”[23]

More generally, shares represent a promise that maximizing investors should view with extreme skepticism: “purchase our equity when we are starting up and have many investment opportunities, and we will pay you dividends later, at a time we determine in our sole discretion, when we lack sufficient opportunities.”  If investors accept this promise, firms should respond by selling as much equity as they possibly can and then constantly deferring the date on which the promised dividend payout actually occurs.  Rational investors, however, should see that a promise to pay dividends tomorrow, with no guarantee that tomorrow will ever arrive, is a promise that is unlikely to generate much value.  Therefore, they should decline to invest based on it.  In turn, rational corporations should see that rational investors are not going to trust them and should see no point in even making the offer. 

Were the stock market and corporate issuers to behave according to rational maximizing principles, no dividends would be paid to public shareholders even when less than fully competitive markets allowed the firm to earn a surplus.  Anticipating this, potential public shareholders would decline to invest in initial offerings.  The IPO market ought to fail in a classic market for lemons. 

II.D.  Alternative Explanations For Dividends Within The Economic Model

IPOs, however, do exist.  Something, then, is wrong with the theoretical picture.  Either investors and managers are systematically irrational, investing based on faith that the normal rules of capitalist behavior will be suspended for their benefit, or the economic theory is not capturing the true dynamics.  In the next subsections, I explore some attempts to rescue the maximizing economic theory, before concluding that the theoretical problem requires moving to a different paradigm altogether. 

II.D.1.  Dividends as reputational investments

If shares have neither the legal right to current dividends nor the market power to demand them, perhaps firms simply give them out as appreciation for past services rendered.  Dividends, in other words, perhaps should be seen as a form of corporate charity or a type of tip, analogous to the common law understanding of pre-ERISA pensions.

Rational profit maximizers, like difficult relatives, do not give free gifts.  Seeming gifts always have strings attached. As we have seen, corporate managers are trained to profit maximize and, at least when they are operating within the corporate finance framework, to view equity as simply one cost like all others.   If equity is just another cost, there is no reason to give shareholders any more than they are legally entitled to (for past services) or have the market power to demand (for present and future services).  As we’ve seen, shareholders have no legal right to payment for past or present services.  Thus, within the rational maximizer model, payments to shareholders must be made in order to induce future services.   

Perhaps, then, corporations pay dividends in order to entice investors to purchase stock in future offerings, much as maximizers are said to tip in order to enhance their reputations with dates or waiters. [24]  But end game problems should make this reputation rationale collapse in the corporate context even if it works in the restaurant business.

First, managers should recognize that paying dividends to ease future stock offerings is rational only if the firm intends to make such an offering in the future.  Indeed, only if the firm plans to sell stock in the relatively near future:  given the short job tenure of participants on both sides of this game—both top management positions and Wall Street have notoriously high turnover—managers should assume that any reputation they purchase by paying dividends will have a quite brief half-life.  Accordingly, if the firm is not planning to issue equity in the relatively near future, it has an incentive to cash in on its past reputation by ceasing to pay dividends. 

Second, recognizing that firms may act in this manner, investors should heavily discount any promises to pay dividends after the point at which the firm is not in dire need of new equity capital.   

The explanation of dividends as tips, then, predicts that firms ought to issue dividends when they need capital at the beginning of their life cycle, and then reduce them as they mature.   Moreover, investors, understanding that dividends are paid in order to attract future capital rather than to compensate past purchasers, should be willing to invest only if the company looks likely to need to attract future equity investors, and should price shares based on the assumption that all returns will come early on.   In almost the exact opposite of conventional “life-cycle” theory, investors would expect that once firms can internally generate adequate investment capital, they will stop paying dividends.    

These predictions bear a certain resemblance to some parts of the private equity start-up market, where investors expect to cash out quickly, but none whatsoever to the public stock market.  If investors viewed public stocks in this fashion, prices would be low indeed.  The theory that dividends are tips or reputational investments to induce future stock purchases fails.  

II.D.2.  Going private

Rational maximizer models, as we have seen, suggest that publicly traded corporations should always defer dividends to a future period that never arrives, and that shareholders, anticipating this behavior, should be unwilling to invest in publicly traded equities.  Instead, firms should raise capital by borrowing, because lenders, unlike equity holders, have legally enforceable rights to returns.

However, public corporations need not remain public.  The possibility of going private may add some otherwise unobtainable value for the public shareholders.  That is, shareholders may expect not dividends but a final payment when the company goes private.  If this expectation is rational, it would explain why investors are willing to become shareholders (although it still leaves dividends prior to the going private payment somewhat mysterious).  

II.D.2.a. Closely held and family dominated firms

The shareholder of a closely-held corporation has a dramatically different bundle of rights than shareholders in public firms.  In public firms, elected directors make major firm decisions in accordance with their views of the interests of the corporation and its shareholders.[25]  In contrast, in closely held firms, the will of the sole shareholder controls – not someone else’s views of its or the firm’s interests.  Interests are replaced by will in two ways.  First, the board serves at the sole shareholder’s pleasure and thus must respond to its will: a sole shareholder can call, without board approval, a shareholder meeting at which it is the sole voter and at which it may appoint or dismiss directors without cause or reference to elected terms of office.[26]  Second, a sole shareholder is not constrained by the directors’ fiduciary duties, except in the most extraordinary circumstances: generally, only shareholders can enforce the directors’ fiduciary duties, so no one is has standing to force the board to ignore the sole shareholder’s wishes in favor of judicially-sanctioned views of corporate interests.[27]  The only meaningful restraint on the sole shareholder using the firm as it wishes is alter-ego doctrine—it must respect the formal separateness of the corporation if it wishes to retain protection against the firm’s creditors in the event of insolvency. 

A shareholder with this degree of control is fairly called an “owner” of the firm.  Like other owners (but unlike shareholders of a publicly traded corporation), it has the power to do with the corporation what it will, including operating it in the way it wishes, determining business plans without regard to the views of others, and—most importantly for present purposes—simply taking any surplus the corporation may generate.  This shareholder receives a dividend because it chooses to, not because it can bargain for it (assuming the firm has the market power necessary to generate a surplus in the first place).

Similarly, even when there is more than one shareholder, a dominant shareholder with enough votes to elect its delegates to the board of directors (even by a plurality) has a degree of control that the shareholders of a public firm lack.  With control come attributes of ownership.  Like a real owner, it need not contend that it is morally entitled to appropriate the surplus, beg for tips, or assure the firm of its future usefulness.  If the firm has the market power to generate a surplus, a controlling shareholder can use its control to take it.

To be sure, unlike a sole shareholder, controlling shareholders are be limited by shareholder and board fiduciary duties.[28]  But those duties are silent with regard to struggles over firm surplus between shareholders as a group and all other firm participants and roles—a controlling shareholder is free to allocate all or any fraction of the firm surplus to shareholders as a group.  The law is somewhat more protective of minority shareholders.  Most importantly, shares of the same class must be treated equally[29]—if the controlling shareholder declares a dividend on its shares, it must pay the same dividend on shares of the same class held by others. 

The upshot is that passive shareholders may actually be better off in a firm with a dominant shareholder than in a classic Berle and Means firm, particularly if control is in the hands of a family and outsiders can purchase shares with the same dividend rights as those held by the family.  American conventional wisdom is skeptical about family-controlled publicly traded corporations, fearing that the insiders will take too much of the firm value in roles that do not require them to share with outsiders–for example, by hiring themselves on above-market terms or using company resources for personal consumption.  In contrast, in many other countries most publicly traded firms are family controlled.  The reason may be that even if the family takes a disproportionate part of the returns to shareholding, it also has the power to ensure that there are returns to shareholding in the first place.  Passive shareholders, then, simply tag along, assuming that dividend seeking family members will have influence beyond that granted by the statutes and cases.  Still, if investors in Google expect that Sergey Brin and Larry Page (or their heirs) will find it in their personal economic interest to declare dividends, they must be taking a long view indeed.  For the foreseeable future, Google’s controlling shareholders should be able to consume all they want without sharing the company’s largess with the public shareholders.

II.D.2.b. Leveraged buyouts and private equity

Family owned firms combine the ancient problems of aristocracy and capital accumulation.  On the one hand, inheritance is an ineffective system for finding talented and skilled leaders and on the other, few families have enough assets to finance a large modern organization.  Modern closely held firms usually take a completely conventional approach to the first problem: they are managed by professionals chosen and supervised in more or less the same way as classic Berle and Means firms.  As to the second problem, they have been reluctant to abandon the real advantages of diversification and cheap capital that public investors can provide.  Accordingly, in the modern era, leveraged buyouts and private equity firms have sought to solve the dividend puzzle without returning to the feudal era, and specifically without depending on individual or family owners. 

In a leveraged buy-out, a firm converts most of its equity into debt.  Often the limited amount of equity left is transferred in large part to the professional managers.  The upshot is that the company is closely held, but the investors remain public, re-labeled debt holders rather than shareholders.   Private equity firms function in much the same way:  they raise their own capital in the debt markets, from more or less the same pools of funds that finance the public equity market and the debt market that underpins LBOs, and use those funds to take companies private, often causing the company to issue still more debt in the process. 

For current purposes, the common elements are more important than the differences.  In each case, the firm remains under institutional, not family, control, generally with the same professional managers that it had prior to the transaction.  In each case the professionals often receive a significant part of the remaining equity, which (depending on the details of the deal) may offer the possibility of vastly higher returns than professionals have traditionally been able to command.  The new executive compensation schemes appear to have the possibility of higher volatility as well, but in practice that possibility seems to materialize only rarely. 

Significantly, the vehicles change the firm’s relationship to the public capital markets in similar, and similarly limited, ways.  In each case, widely held equity is replaced by almost as widely held debt.  The debt market, to be sure, is less liquid and more institutionalized than the equity market.  But in each case, the investors are largely institutions with expertise in investing rather than operating—both the former stockholders and the new debtholders are from Wall Street rather than Main Street, specialists in gathering financial assets from the public and deploying them in the capital markets rather than in actually running businesses.  And in each case, the investors remain passive, with no effective means of control over the corporation.  Even when buyout firms or private investment firms exercise control over the managers of the underlying corporations, nothing fundamental has changed.  Even Berle & Means corporations have division heads whose autonomy is limited by the presence of superior officers; the LBO firms and private equity firms, like the head office of the conglomerates they are descended from, merely add another layer of professionals policing professionals.  In short, even after the LBO or private equity buyout, the investors remain basically passive pools of retirement and endowment funds, and the company remains basically run by professional managers. 

Passive debt investors are more or less the same as passive equity investors, and the capital they provide to the firm is more or less interchangeable, with one major distinction.  Unlike equity, debt has legally enforceable rights to specified returns at specified times and therefore has a positive marginal cost.  

At first glance, this might seem to solve the dividend puzzle.  Firms that pay interest, unlike firms that pay dividends, are legally bound to pay, and so public investors can invest based on reason and not faith alone.  One might imagine, then, that the public equity market was an oddity, a speculatively driven moment between debt dominated public markets (the nineteenth century railroads, the post-World War II boom), and a return to debt dominated public markets through LBOs and private equity.  The Japanese and German systems, which have often seemed anomalous to American observers in their dependence on bank financing, would instead appear as paradigms of dependence on debt.  It would become clear that at least over the long run, we differ not because of our deeper stock markets but because of our deeper, and less bank dominated, debt markets.  

But the world is not that simple.  So long as the equity market does not collapse, a debt financed firm in a competitive market will face competitors that remain equity financed.  They should have lower marginal costs (since in any given period they need not pay for their equity capital) and therefore able to out-compete debt-financed firms.  Thus, it is not immediately obvious how debt financed firms survive absent a collapse of the equity market.   

Most explanations of the LBO phenomenon have assumed that LBOs compete by being more efficient.  Proponents have pointed to the strong incentives on managers, particularly if they hold some of the highly leveraged stub stock.  As a stick, they must generate enough cash flow to pay the debt or face bankruptcy and the possibility of disgrace or displacement.  As a carrot, if they generate cash flow beyond the claims of the debt, it rapidly increases the value of the stock. 

Unfortunately for this explanation, basic problem remains that LBO firms must pay for their capital, while the dividend puzzle is precisely that equity markets appear willing to provide capital for free.  It is hard to see why more expensive capital and closer supervision by the capital markets would increase innovation.  Similarly, neither the carrot nor the stick seem potent enough at the margin to transform managers.  Even CEOs without equity stakes are a well-paid and hard working crowd.  The marginal carrot here seems unlikely to strongly motivate dedicated professionals to work harder still—but it does offer those CEOs who are most motivated by their own personal wealth a multitude of new opportunities to extract existing corporate assets instead of working harder to grow them for others.  The stick seems equally inadequate: modern bankruptcy reorganization is closer to a business planning tool than a disgrace for managers or an existential crisis for the firm.

Instead, the power of debt lies in redistribution, not efficiency.  Debt financed firms may be able to deliver more of the firm’s surplus to investors.  First, as is well-known, debt financing allows the corporation, in effect, to repeal the corporate income tax, shifting to investors corporate resources that otherwise would have gone to the public.[30]

The tax story can be generalized.  High leverage increases the value of the firm to investors (shareholders plus debt holders) primarily because it is a useful tool for convincing other firm participants to reduce their claims on the corporate pie—redistributing rather than creating wealth.  In equity-financed firms, if investors seek a larger share of corporate returns, they must ask employees to take less:  “you need to work harder for less pay so that the company will make higher profits so it can voluntarily pay higher dividends.”  This is a tough position to justify—the cause of upward redistribution is not commonly one that inspires idealistic sacrifice.  Employees are likely to resist, if not by strikes then at least by turnover, slowdown and morale drops. 

On the other hand, in a debt-financed firm, the company can go to the same employees with a very different claim:  “if we don’t make our interest payment, our corporate survival is at stake.  It is time for all members of the team to pitch in.” Pulling together for the good of the team is much more inspiring than reverse Robin Hood.  Firms struggling to stay out of bankruptcy may well be able to convince employees to work harder for less—or, if they do end up in reorganization, be able to enlist the court’s assistance.  But the substance of the two programs is the same.  Investors are taking a larger part of the corporate pie, leaving less for everyone else.

II.D.2.c. The power of the credible threat

If dominant shareholders or high interest debt holding investors can simply take the surplus, then the value of company stock (even when passively held) potentially includes the possibility that the company will be sold to a dominant shareholder or levered up.  A decade or two ago, most theorists looked to this market for corporate control to solve the problems of passive shareholders. 

The intuition was that if managers do not act to maximize surplus and distribute that surplus to shareholders, the company will be worth more held privately than held publicly.  Arbitrageurs should then take companies private or lever them up unless managers are willing to operate public firms as if they were private.  At equilibrium, the public stock market should converge on private market values and managers of public firms should behave as if they were under the supervision of dominant shareholders.  If they failed to, the public stockholders would find it attractive to sell control to an investor that would value the company more because of its ability to force managers to give it any surplus. In short, the threat that public shareholders might sell control to a dominant shareholder that would have effective ownership rights paradoxically can force managers to act as if the public shareholders were in fact owners without the sale ever taking place.  

In the heyday of the market for corporate control, in the late 1980s, this picture bore a strong resemblance to reality.[31]   The market for corporate control forced managers to act in the interests of the stock market as mediated through the price of the shares.  The threat of a junk bond-financed buy-out offer ensured that even less leveraged companies shifted company resources to capital from other participants.  The actual shareholders, to be sure, had no more authority, power or legitimacy than in the initial descriptions above, but the stock market, by its power to set the price of the company’s stock lower than its value in the private or leveraged market, could coerce managers into giving the shares the surplus.   

After the Poison Pill and the just-say-no defense,[32] however, the market for corporate control no longer seems powerful enough to solve the dividend puzzle.  Shareholders have never had the power to sell the company; the board must first negotiate and approve any potential deal.[33]  Nor does fiduciary duty give shareholders significant additional rights—Delaware reviews board decisions regarding when or whether to put a company up for sale with the usual deference given to all other board decisions under the business judgment rule.  Boards, not courts, shareholders or contracts, determine the time frame within which corporations operate.[34]  The pill and its statutory equivalents mean that sales of control blocks of shares are governed by basically the same rules as sales of the company—as a practical matter, both require advance approval of the board prior to shareholder action.  Accordingly, just as the board may decide to defer dividends to a future that need never arrive, so too it may defer changes in control indefinitely, largely independent of shareholder or judicial views.

Under this legal regime, the market for corporate control replicates the dividend puzzle rather than solving it.  If managers choose not to privatize or lever, shareholders have no way of making them do so.  If managers decide that privatization is in their own interest, they can count on shareholder support at a low price (since shareholders, as a group, are trapped in the company and, as we have seen, their rational expectations as public shareholders are quite low, they should be happy to sell their limited birthrights for relatively cool bowls of porridge).  Consequently, the primary beneficiaries of the threat of privatization or levering should be those in control of the transaction:  top managers and directors. 

In short, if shareholders had the right to force companies to go private or buy them out, we would expect them to use those rights to appropriate the bulk of the corporate surplus.  Corporate law has not given shareholders those rights at any time in the modern era.  For a brief moment in the 1980s, however, outside arbitrageurs had a relatively unfettered right to buy companies without the consent of the incumbent management.  This meant, in effect, that the stock market as a whole—not the corporation’s own shareholders—had the right to sell public corporations, because the stock market as a whole determines stock price and thus made it economically worthwhile, or not, for the arbitrageurs to act.  This active (hostile) market for corporate control could coerce corporations into acting in the interests of the stock market.  With the demise of the hostile takeover, however, arbitrage to move a company into the private or highly leveraged markets is possible only with consent of the incumbent managers, and thus one would expect those managers, who in any event are less fungible than shareholders, to obtain the bulk of the proceeds.  Even the possibility of going private is not enough to make it likely that public shareholders will earn a return. 

III.  Beyond Market Rationality

We have seen that the legal rights of public stockholders make it quite unlikely that rational public shareholders would expect to earn even a normal return, let alone a share of the corporation’s surplus.  In markets characterized by economic rationality, shareholders should expect to earn a return only in the presence of an active market for hostile takeovers, or in family controlled companies where not all family members are employed, family ties prevent the insiders from rationally maximizing their own interests without regard for the interests of their relatives, and public shareholders hold stock with rights tied to the rights of the non-insider family members.  The latter may be a caricature of the public markets in Canada, Italy and Korea; the former is a simplified picture of the United States in the boom of the late 1980s.  In the vast American investments of the nineteenth century the public participated, as predicted by these models, largely as debt-holders, not through the stock market.  But the modern era American stock market simply defies these models.

The stock market is perhaps the economic sector that is closest to the economic model’s assumptions of anonymity, absence of outside commitments or values, and maximizing behavior.  If it cannot be explained by the rational maximization model, then perhaps it is time to turn to more complex models.

One possibility is that shareholders receive dividends because it is right and just that they do so.  The rational actors of economic models are not moved by ideological or moral claims, but real people are.  If shareholders have a strong pre-legal moral claim to the residual, even rational actors might expect to receive the residual because of the power of morality, even without help from market, law or raw power.  Perhaps, then, the stock market functions because of its fundamental justice—or perceptions of justice. 

In my view, the normative claim of shareholders to the residual is weak as a rational matter, but powerful as a matter of sociological ideology.  Thus, while it is implausible that shareholders receive (or expect to receive) the residual because it is just that they do so, it seems likely that the system functions at least in part because many people believe that it is. This section, then, explores faith-based economics—the possibility that the market depends on a non-rational belief that dividends will be paid that is strong enough to override rational maximizing behavior, game-theoretical strategies predicated on assumptions that others will behave in a rationally maximizing way, and serious attempts to think about what capitalist (or more general) norms actually require.  

III.A.  Moral Rights

Moral claims to economic product can stem from a number of sources:  creation, need, equity, settled expectations or ownership, contract or agreement, or simple power are perhaps the most likely.  None work for passive public shareholders, who did not create the corporate surplus, are poor proxies for the neediest, have not been robbed or defrauded, have neither agreement nor established property rights granting them the surplus, and need ideological justifications precisely because they lack actual power. 

If corporate surplus is returns to innovation, surely the party with the strongest claim is the innovators.  Often, it will be difficult to tell which corporate participants are the sources of innovations, particularly when the innovations themselves are the product of the collective enterprise rather than individual initiative.  Still, passive shareholders, as the purely fungible providers of a purely fungible commodity, are the least likely candidate.  Capital is an important contribution to every enterprise, but it is rarely innovative.  If the corporate surplus is instead the result of situational or other monopoly, the moral claim of shareholders is even weaker.  No one has a pre-legal entitlement to economic rents in a capitalist system.[35] 

Moral claims based in need, equity or settled expectations fare no better than claims based in creation.  Shareholders are poor candidates for paternalistic intervention to change the results a free market would generate.  First, roughly one quarter of shares are held by extremely wealthy individuals and families and another half are owned by institutions.  These shareholders are quite capable of defending their own interests.  Second, even smaller shareholders, who may have difficulty bargaining on their own behalf, benefit from the ample protections of a highly sophisticated market, powerful and sophisticated intermediaries and advisors, and the extensive consumer protection and truth in labeling provisions of the federal securities regime.  Finally, investment capital is fully mobile and therefore should be able to bargain quite effectively against other corporate participants, most of whom (especially the human employees) suffer from varying degrees of lock-in.  Leaving a job is always somewhat traumatic, involving at a minimum changing communities and patterns of life even if not uprooting spouse and children.  Finding an alternative IPO or other passive investment is hardly in the same category.  

Shareholders, then, should have few claims beyond their actual contractual rights: the right to receive whatever distributions makes to them whenever it decides to make them; the right to vote on a basis of one vote one share for the ultimate corporate decision-makers; and the right to transfer stock, with the associated collective ability to drive the stock price down to the point where the directors may see advantages to going private.  As we have seen, standard economic rational maximizer models cannot explain how those rights can generate shareholder returns.   

III.B. The Power of Words

Moral claims need not be correct to be powerful.  They need only persuade, and the persuasion need not be logically defensible.  If directors are persuaded that shareholders ought to be treated as if they were owners, trust beneficiaries, principals or contractual insurers with an inexplicably favorable deal, they may conclude that shareholders’ claim to the surplus is just and right.

The metaphors of ownership and property rights, agency, trust and contract are pervasive in corporate law discourse and management training alike.[36] These metaphors fail as serious attempts to analyze either law or market. But their failure as analysis is, in the end, the best indication of their power. They are not descriptions of the workings of the market but moralistic calls to action; their importance lies in their uniformly pre-Copernican placement of shareholders at the center of the corporate universe.

Ironically, the central pillar of the shareholder value maximization principle is its demand that directors refuse to act as the self-interested, amoral, rational maximizers of economic theory. Instead, it teaches them, they should be a peculiar sort of altruist, working only to maximize the “well-being” of a self-interested, amoral, uncommitted but not particularly rational stock market.

More ironically still, this reverse-Rawlsian altruistic demand that managers promote the interests of the best-off also turns out to be the best available justification for managers helping themselves to an ever larger share of the corporate pie. Shareholder returns are the ideological price managers pay in order to defend the otherwise indefensible: fiduciaries using their trust to enrich themselves.

But this game will not last forever. The more managers are able to pay themselves, the less they look like fiduciaries and the more they look like entrepreneurs or owners. If they are ever able to convince themselves or us to see them in that light, I suspect all the metaphors of shareholder ownership and managerial altruism will fade away, leaving the moral claims of shareholders as weak as their legal ones.

Let us take these counter-intuitive contentions more slowly.

III.B.1. The problem of paying managers as fiduciaries

From one perspective, corporate managers are capitalist free agents who contract with the firm in an arms length transaction governed by the “workday morality of the marketplace.”[37] In this role, they are entitled, indeed expected, to sell their work for the highest price they can get.

But once hired, corporate managers, including CEOs, are agents of the firm, required by the fiduciary norms of agency law to set aside their own interests and instead work for the interests of the firm. Directors are not agents—on the contrary, board decisions are acts of the principal itself—but they too have a fiduciary duty to act in the interests of the firm that closely resembles the duty of an agent. Taken seriously, this duty of loyalty requires “not honesty alone, but highest punctilio of an honor the most sensitive” pursuant to which “thought of self was to be renounced, however hard the abnegation.”[38]

The conflict between these two norms—one praising the self-centered maximizing behavior postulated in simple economic models and ordinarily seen only in sociopaths, and the other requiring extreme altruism of a degree that would qualify the self-abnegating practitioner for sainthood were the goal less mercenary—pervades corporate law.[39] It is most pronounced, however, in the question of CEO pay. The duty of loyalty means that a faint odor of impropriety necessarily emanates from every CEO pay package, demanding explanation and justification.

For many years, corporate leaders resolved the tension by emphasizing that they were public servants, stewards of the enterprises that provide America’s jobs and drive our economy, and thus, like all public servants, entitled to a reasonable professional income. This justification, however, lost its usefulness when CEO perquisites broke ranks with the professions. Dictators, not democratic leaders, use their positions to enrich themselves.

Moreover, most companies require a degree of employee loyalty to survive and prosper. Employee loyalty, in turn, is best encouraged by appealing to team spirit—convincing the employees that the company’s interest is their interest. But the rapidly increasing income and status gaps between CEOs, as team captains, and their employees, as team players, is obviously destructive of esprit de corps. A certain degree of egalitarianism is essential to create the sense that “we are all in this together.” As a result, any credible organizational theory will have difficulty justifying vast pay or prestige gaps between leaders and led within an organization.

Relatedly, our publicly traded corporations are bureaucracies, and any bureaucracy is limited by its ability to process information up and down the hierarchy. Vast social gaps between the decision-makers at the top and the actors at the bottom predictably lead to bad decisions and poor implementation. No doubt, there are leaders who are so smart and so charismatic that they can successfully lead without communication from those below them—but they are exceptions. This is why we expect egalitarian capitalist economies to outperform dictatorial ones and why we expect generals who lead from the midst of democratic armies to defeat emperors who rule from distant palaces. Astronomical CEO salaries, prima facie, should look like dereliction of duty if the core of the CEOs duty is to rally the troops or listen to them (and the relevant troops are the employees).

III.B.2. Appropriating surplus and the duty of loyalty

Share centered ideologies do not solve these real sociological problems. They do, however, conceal them.

If the corporation is understood as its shareholders, then managers can reasonably compare their share of the corporate pie to shareholder returns rather than employee pay. A different team means a different comparison group. As leaders of an institution, they were implicitly entitled to be paid as sophisticated professionals, more than but in the same general range as the rest of the bureaucracy. But if the corporation is its shareholders and its purpose is returns to shareholding, then managers, as leaders of the pursuit of profit, can reasonably claim a percentage of profits—a far larger sum.

Moreover, if the corporation is its shares, then it is not its employees. We think differently about motivating outsiders than insiders, looking less to loyalty and more to carrots and sticks. If the employees are outsiders, then it will seem more reasonable to motivate them by fear of firing, on the one hand, and on the other, a competitive tournament in which individual executives are motivated to compete against each other in order to win the grand prize. On a view of the firm as fundamentally a team, high CEO pay is likely to look like a sign that the leaders are betraying the team, corruptly lining their own pockets, and thus as an invitation to similar behavior throughout the organization. But if individual executives are seen, instead, as competitors to be motivated by a prize if they win the game, then higher pay for the incumbent CEO looks less like misappropriation of corporate assets and more like a stronger motivation to his subordinates.

If the central problem of corporate leadership is not overcoming the bureaucratic problems of communication but rather assuring that CEOs have common interests with shareholders, then high CEO pay can appear as a solution instead of a central part of the problem. High pay, after all, will make CEOs more like capitalists than employees, and high pay tied to stock returns even more so. Pay them enough, and perhaps they will begin to feel the difference between themselves and their subordinates, creating exactly the break in empathy that is necessary for a regime that motivates by threats of layoffs and competitions in which only a few can win. Make them rich enough, and they may begin to think of the company in the risk-neutral manner of diversified portfolios.

For each of these reasons, the share-centered view of the corporation justifies high CEO perquisites. CEOs, thus, have a tremendous incentive to accept and promote the share-centered metaphors: in the share-centered corporation, they are doing their duty by making themselves into the new American aristocracy.[40]

The share centered ideologies are useful even beyond justifying high CEO pay. For directors and managers, they solve the biggest conundrum of their position. Fiduciary duties are owed to the firm itself. But firms are not preexisting natural entities endowed by the Creator with rights and interests. Rather, the very fiduciaries who are charged with working in its interests also may set and reset the firm’s boundaries, members, goals and purposes, defining what it is and what its interests might be, even as they aim to serve them. Directors and managers who attempt to think seriously about their responsibilities will discover that far from being “firm,” the object of their fiduciary duty is much like the poet’s description of Oakland: there is no there there.[41] The share-centered ideologies replace this dizzying indeterminacy with easily understandable and calculable directions: maximize the returns of an imaginary, undiversified stock investor with no commitments to firm or society beyond the stock holding itself.

Finally, so long as investors believe that they are entitled to the corporate surplus, they are likely to expect that they will receive it, and so long as they expect to receive it, they will value stocks as if they were going to receive it. In turn, as Miller and Modigliani correctly pointed out, if shareholders will ultimately receive the surplus, they should be indifferent as to when they receive it. Even if dividends or other distributions are deferred, some other investor should be willing to purchase the stock based on expected future distributions—and the stock market, collectively, can go up based on new investor savings and expected profits rather than actual distributions. The perpetual motion machine is in everyone’s interest.

III.B.3. Where shareholder returns come from

The short answer to the puzzle of shareholder returns, then, is this: shareholders receive returns because investors, directors and CEOs believe they are entitled to them.

The underlying faith in the rightness of operating the corporation for the shares is not based on any compelling economic, moral or legal argument. However, CEOs who wish to justify seizing a significant part of the corporate surplus for themselves find the share-centered metaphors of the corporation more comfortable than the alternatives.

CEOs, unlike shareholders, have actual economic and political power in the corporation. They use that power to pay dividends and otherwise act to keep stock prices relatively high because the share-centered ideology tells them they should. They subscribe to the share-centered ideology because it offers the best available justification for their own perquisites. Meanwhile, investors, believing that they will get the corporate surplus in the end, meanwhile act in a way that makes it largely irrelevant that they are only partly receiving it in the meantime.

Outside the chief executive’s office, the share-centered metaphors also have some appeal: they offer an explanation of a world that would otherwise seem unfair, exploitative, anti-democratic and dysfunctional. If corporations exist for the sake of their shareholders, then perhaps their employees, customers and neighbors have no legitimate complaint when they are treated as no more than means to an end not their own. And if you are being treated that way, it may be more appealing to have an explanation than simply a simmering sense of injustice.

Cynically, simplistic, unprofessional, and uninteresting as this explanation is, it must be a major part of the truth. The great secret of the great manager/shareholder conflict at the heart of corporate law is that, conflicts notwithstanding, the shareholder centered view unites managers and shareholders to shift the corporate surplus from all the other corporate participants. Only then do the arguments over the spoil begin.

In the end, both the political and the market power of public shareholders are quite limited. Most politics is by informal persuasion, in the shadow of the power of the electorate, donors, masses, violence, organization, guns or wealth. In public corporations, shares have votes, but anonymous, transient, unorganized and fungible shareholders have little connection with the actual sociological firm.

The basic protection shareholders have today is the principle that all shares must be treated equally when a dividend is declared and the assumption that stockholding managers will, in the end, want to issue dividends, both to extract larger sums from the corporation than they feel they can award themselves in salary and to maintain the value of their own stock or options. That said, when managers determine that their interests are no longer aligned with the other shareholders, there will not be much the outside shareholders can do about it.

And in the end, managers are likely to find that the public stock market has outlived its usefulness. The share-centered ideologies taught CEOs to treat the actual human beings with whom they work as mere means to an end, tools to be exploited in the cause of profit maximization. But the only justification the ideology offers for why managers who have trained themselves to betray the people with whom they work should then altruistically turn their gains over to market abstractions, is a mélange of metaphorical claims that shareholders own the company.

Top managers largely control the corporation. They have now established that unlike other employees they are critical to the corporation’s success and that corporate profits should be viewed as their marginal product, entitling them to be paid (in the interest of shares) like entrepreneurs. In short, they have the control that characterizes owners, they perform the basic ownership function of making critical decisions, and they take a large part of the residual. For at least two decades, top managers have regularly regularized their positions as owners entitled to appropriate the surplus by using their control of the corporation to structure a leveraged buyout or private equity financed going private transaction. Not surprisingly, the premia offered to the public shareholders to approve these transactions have decreased over time. The remaining issue is how much longer it will take to find a new language in which to discuss still-public corporations, allowing top managers to take the surplus without paying to take the company private.

IV. Conclusion

Shareholders are not entitled to the residual by law, justice, nature or even economic bargains and incentives. Accordingly, shareholder expectations of returns are more fragile than generally assumed, the economic consequences of changes in shareholder rights probably would be less than generally assumed, and there is no reason to think that the current distribution of corporate surplus is more or less efficient than alternative ones.

Instead, the struggle for the corporate surplus—the residual—is a political struggle over economic rents to which no party has any special claim precisely because the surplus would not exist but for the contribution of all the claimants. From an economic perspective, the shareholders, as fully fungible providers of a fully fungible commodity, have a weaker claim than most. From a political perspective, the power of the shareholder claim depends on the power of the shareholder role, which, as we have seen, is more limited than a casual student of the corporate governance literature might imagine.

The key weapons the shares have are two. First, an ideological claim to entitlement, founded in the metaphors of ownership, contract, and agency.[42] This claim contains its only critique—it is only because public shareholders are not owners or principals and have no legal rights to corporate residual that their defenders feel the need to so strenuously insist that they are. In the end, the claim that shares ought to be treated as if they had more rights then they do is simply a demand for upward redistribution.

Second, in recent decades top management has found the rhetoric, and sometimes the reality, of share-centeredness useful for its own purposes. Managers, nearly everyone agrees, are fiduciaries for the corporation with no independent claim to ownership or entitlement. So long as this remains the law and the social understanding, CEOs must justify their actions as on behalf of the corporation, not themselves. The share-centered views of the corporation conveniently rationalize high CEO pay and perquisites—if CEOs are major shareholders, they are more likely to think like shareholders than like employees. Moreover, by viewing the stock market rather than the bureaucracy as central to corporate life, the share-centered metaphors distract attention from the ineffectiveness of our modern CEO aristocrats and their failure to create decent jobs at decent wages.

The theory of economic determinism failed in the communist regimes. It fares no better in the capitalist ones. In the end, we must decide through politics, not economics, how much power we wish to give to the stock market, how much to democratic processes, and how much to leave to unguided professionals in semi-autonomous quasi-private organizations. The distribution of the surplus is part of the issue, but the more fundamental issues are how we want our public corporations to operate, towards which goals, and with what relationship to the people who make them up. So far, the defense of the share-centered corporation has proceeded mainly by invoking rhetorical claims to a shareholder supremacy that simply does not exist. But metaphors of ownership or agency are no substitute for social analysis—the question is not whether shareholders “are” owners of the corporation, but whether pretending they have so much power in the corporate-governance structure is good for society as a whole.

The alliance of convenience between CEOs and the stock-market has proven only partially productive, creating cheap goods but failing to produce enough good jobs outside the upper echelons. The metaphors of share ownership have obscured the real world power structures. The academic task is, first, to fully elucidate the false metaphors, and second, to shift the debate’s imagery to metaphors that illuminate instead of concealing. The political task is to find a model which produces wealth for all, not just a small elite; which promotes decent work places and productive careers, not just consumers’ playgrounds; which respects our ecosphere, not just a narrow vision of the econosphere.

* Daniel J.H. Greenwood, Professor of Law, Hofstra University School of Law; J.D. Yale; A.B. Harvard. http://law.hofstra.edu/greenwood. Special thanks to Dean Kellye Teste, Dana Gold and the Seattle University Law School for organizing the conference where this chapter was first presented, and to Bill Bratton, Tim Glynn, Kent Greenfield and my fellow participants for their helpful comments. This chapter is a revised version of The Dividend Puzzle: Are Shares Entitled to the Residual?, 32 Journal of Corporation Law 103-159 (2006).

[1] Richard A. Brealey & Stewart C. Myers, Principles of Corporate Finance 364 (5th Ed. 1996).

[2] Daniel J.H. Greenwood, Fictional Shareholders: ‘For Whom is the Corporation Managed,’ Revisited, 69 Southern California L. Rev. 1021 (1996).

[3] Legal “surplus” is closer to accounting profit than economic surplus or residual. See, Del. Code Ann. tit 8 § 154.

[4] R.H. Coase, The Problem of Social Costs, 3 J. L. & Econ. 1 (1960).

[5] R.H. Coase, The Nature of the Firm, 4 Economica 386 (1937).

[6] Ronald Coase, The Marginal Cost Controversy, 13 Economica 169 (New Series, 1946) (discussing the relationship between marginal and average costs).

[7] Compare Warren Buffet’s famous investment advice to invest only in companies with a “franchise” – by which he means some ability to charge more than the commodity price.

[8] U.P.A. §§ 31, 38.

[9] As Blair and Stout point out, this is one of the key benefits of the corporate form. Lynn Stout & Margaret Blair, Specific Investment, 31 J. Corp. L. 719 (2006).

[10] Del. Code Ann. tit. 8 § 170. Cf. Blair & Stout, Director Accountability, 79 Washington U. L. Q. 403, 406 (2001);

[11] Del. Code Ann. tit. 8 §§ 141, 203; Smith v. Van Gorkum, 488 A.2d 858 (Del. 1985) (holding that directors may not delegate determination of whether to sell company to shareholders); Paramount v. Time, 571 A.2d 1140 (Del. 1989) (court may not “substitut[e] its judgment as to what is a ‘better’ deal for that of a corporation’s board”).

[12] See, e.g., Paramount v. Time, 571 A.2d 1140 (Del. 1989) (“The fiduciary duty to manage a corporate enterprise includes the selection of a time frame for achievement of corporate goals. That duty may not be delegated to the stockholders.”)

[13] Brian Arthur, Positive Feedbacks in the Economy, Sci. Am., Feb. 1990, at 92. Even where there is no obvious advantage to standardization, it still often remains more important to go to the same movies, listen to the same music, wear the same clothes, or join the same club as our peers than it is to find the best of those products. See, e.g., Henry Hansmann, The Role of Nonprofit Enterprise, 89 Yale L.J. 835, 892 (1980) (describing this phenomenon in the context of country clubs). When a cascade occurs, the producer of the favored product may be able to charge monopoly prices despite the existence of competitors. Even if the competing product has similar technical specifications, without the customer base, it cannot provide true substitutability.

[14] Alfred D. Chandler, Jr., Scale & Scope: The Dynamics of Industrial Capitalism 56-58 (1990).

[15] John Micklethwait & Adrian Wooldridge, The Company: A Short History of a Revolutionary Idea 62 (2003) (describing J. P. Morgan’s attempts to raise prices above marginal cost by suppressing “ruinous competition”).

[16] Daniel J.H. Greenwood, Enronitis: Why Good Corporations Go Bad, 2004 Columbia Business L. Rev. 773, 818 (2004).

[17] The discussion may seem slightly strange to readers used to thinking of investors as buyers of securities rather than sellers of capital. From that perspective, the product—the securities—is not quite as fungible. Securities vary not only in price, but in expected return, risk, legal and market enforceability, time frames, and so on. But the inversion of perspectives changes nothing fundamental. For securities buyers, firms appear mainly as bundles of risk and return, and portfolio theory teaches that bonds and stocks are just different ways of allocated proceeds from the same pool. The firm’s ability to generate returns to capital is fundamental; its division of (economic) profit between different securities is merely contingent.

[18] Theorists have contended that shareholders provide risk-bearing services to the firm, absorbing short term business shifts that are more expensive for other corporate participants to bear. I find this argument somewhat implausible—dividends generally seem to be more stable than employment, not less as ought to be the case if the stock market were in the insurance business. But for current purposes, we can accept the claim and note that no where else in the economy do insurers expect to share in extraordinary profits earned by the insured. Risk bearing services, even if rendered, do not change the analysis in the text.

[19] The problem is a variant of the well-known market for lemons issue. See, George A. Akerlof, The Market for "Lemons": Quality Uncertainty and the Market Mechanism, 84 Q. J. Econ. 488 (1970).

[20] Supra, n. 11.

[21] See, e.g., Louis Lowenstein, Sense and Nonsense in Corporate Finance 125 (1991) (describing instances in which managers did not pass on economic profits to shareholders but instead used them to expand, diversity, etc.)

[22] See, Manuel Utset, A Theory of Self-Control Problems and Incomplete Contracting: The Case of Shareholder Contracts, 2003 Utah L. Rev. 1329.

[23] Lewis Carroll, Through the Looking Glass and What Alice Found There 53 (1871)

[24] [Find an article comparing dividends to tips]; Freakonomics [or other source] (giving maximizing explanation of tips).

[25] Aaronson v. Lewis, 473 A.2d 805, 811 (Del. 1984) (“A cardinal precept of the General Corporation Law of the State of Delaware is that directors, rather than shareholders, manage the business and affairs of the corporation”).

[26] See, e.g, Del. Code Ann. tit. 8 § 228 (right of shareholders to act without a meeting by unanimous consent); § 141 (k) (removal of directors without cause); § 109 (right of shareholders to amend bylaws); § 141 (b) (right of corporation to modify the size of its board by bylaws).

[27] In contrast, in public corporations, courts may impose their own view of the “real” interests of shareholders or the corporation instead of the expressed will of actual shareholders, individually or collectively. Dodge v. Ford Motor Co., 170 N.W. 668 (Mich. 1919) (overruling majority shareholder’s view of dividend policy), State ex rel. Pillsbury v. Honeywell, 191 N.W.2d 406 (Minn. 1971) (declaring actual views of actual shareholder irrelevant); Paramount v. Time, 571 A. 2d 1140 (Del. 1989) (ignoring expressed views of majority of shares); Revlon v. MacAndrews & Forbes, 506 A.2d 173 (Del. 1986) (holding that corporation must act in interests of “shareholders” imagined to have no interests beyond their shareholdings); Unocal Corp. v. Mesa Petroleum Co., 493 A.2d 946 (Del. 1985) (holding that some shareholders are less equal than others). See generally, Fictional Shareholders, supra n. 2.

[28] So long as there are minority shareholders with standing to bring complaints of breach of fiduciary duty, there will be judicially imposed limits on the ownership rights of the dominant shareholder. The board of directors has a fiduciary obligation to act in the interests of the corporation, and dominant shareholders may have fiduciary obligations to minority shareholders. However, these fiduciary duties are limited by the business judgment rule; courts remain quite deferential to the internal corporate decision-making processes so long as dominant shareholders and their boards respect corporate formalities.

[29] But see, Unocal Corp. v. Mesa Petroleum Co., 493 A.2d 946 (Del. 1985) (allowing discrimination against particular shares based on identity of their owner as “greenmailer”).

[30] The corporate income tax is levied on corporate income, defined more or less along the lines of accounting profit. Dividends are deemed to be paid out of income; interest, in contrast, is deemed to be an expense that reduces income. Thus, if the firm is generating a surplus and retains it or pays it to investors in the form of dividends, it must pay income tax on it. If it pays the same surplus to investors in the form of interest, the payments reduce its net income and avoid corporate income tax. Taxation at the investor level may create countervailing pressures. However, may of our largest investors are tax insensitive (retirement plans, endowments, mutual funds). Money that is not paid in income taxes is available for other firm purposes or participants.

[31] See, e.g., Fictional Shareholders, supra n. 2.

[32] Del. Code Ann. tit. 8 § 203 (statutory pill); Moran v. Household Int’l, 500A.2d 1346 (Del. 1985) (upholding pill); Paramount Communications, Inc. v. Time Incorporated, 571 A.2d 1140 (Del. 1989) (upholding just say no).

[33] Del. Code Ann. tit. 8 § 141 (powers of board); Smith v. Van Gorkum, 488 A.2d 858 (Del. 1985) (holding that board must exercise its fiduciary duty to determine whether sale of the company is in its best interests prior to allowing shareholders to vote), ratified and modified by Del. Code Ann. tit. 8 § 146 (permitting board to agree to allow shareholders to vote with a recommendation that they reject should later developments cause it to reverse its initial approval).

[34] Paramount Communications, Inc. v. Time Incorporated, 571 A.2d 1140 (Del. 1989) (upholding right of board to determine time frame in which company operates). I oversimplify, of course. Once the company is up for sale or the board takes targeted defensive measures, judicial scrutiny increases, although it remains largely procedural. See, e.g., Omnicare, Inc. v. NCS Healthcare, Inc. (Del. 2003) at fn36-40 (emphasizing that even a “enhanced scrutiny” of the “reasonableness of the substantive merits of the board’s actions” is largely procedural—devoted to establishing the “adequacy of the decisionmaking process” and evaluating the merits only to determine if they were “within a range of reasonableness” “in the context in which action was taken.” )

[35] On the concept of “rents” as used in the public choice and law and economics literature, see, for example, Mark Kelman, On Democracy-Bashing: A Skeptical Look at the Theoretical and “Empirical” Practice of the Public Choice Movement, 74 Va. L. Rev. 199, 227 (1988) (describing rent-seeking, when it is worthy of condemnation, and ambiguities in concept).

[36]. For preliminary discussions of the metaphors of corporate law, see Daniel J.H. Greenwood, Markets & Democracy: The Illegitimacy of Corporate Law, 74 UMKC L. Rev. 41 (2005); Enronitis, supra n. 16 at Parts IV.A, IV.B.3; Daniel J.H. Greenwood, Introduction to the Metaphors of Corporate Law, 4 Seattle J. Soc. Just. 273 (2005).

[37] Salmon v. Meinhard, 249 N.Y. 458 (1928).

[38] Salmon v. Meinhard, 249 N.Y. 458 (1928). The duty of loyalty clearly bars corporate fiduciaries from promoting their own interests at the expense of the corporation’s. Cardozo notwithstanding, however, corporate law’s duty of loyalty is quite a bit weaker than the equivalent duties in trust law or governmental civil service norms, let alone the highest punctilio of an honor the most sensitive. Nepotism, conflicts of interest, and even what Boss Tweed unsuccessfully defended as “clean graft” typically are permitted in the corporate sector, at least with adequate disclosure and approval by disinterested parties. Bayer v. Beran, 49 N.Y.S.2d 2, 10 (N.Y. Sup. Ct. 1944) (allowing corporation to hire CEO’s spouse so long as the corporation was no worse off than if it had hired a third party); Del. Code Ann. tit. 8 § 144 (2006) (regulating ratification of interested transactions). This rule is often an appropriate response to information problems: insiders may be willing to provide better terms than could be had in arms length transactions and apparently corrupt transactions may have benefits that are not obvious to outsiders. However, the information problems are identical in the public sector and, indeed, seem more likely to be correlated to scale than business form, so the explanation does not justify the difference between public sector and corporate norms. More likely, the acceptance of nepotism in the corporate world is a reflection of the survival of the personalized norms of feudal loyalty (epitomized in modern law in the property right to be idiosyncratic and the employer’s right to fire at will), while the public sector has more completely adopted the market and bureaucratic norms of anonymity reflected in notions of single price, careers open to talent, equality before the law and anti-discrimination law.

[39] Enronitis, supra n. 16, at 825.

[40]. See supra note --.

[41] Gertrude Stein, Everybody’s Autobiography, Ch. 4 (1937).

[42]. See supra note --.