The Moral Economy: Why Good Incentives Are No Substitute for Good Citizens - Samuel Bowles (2016)
II. A Constitution for Knaves
Having noticed a suspicious bunching of sick call-ins on Mondays and Fridays, the Boston fire commissioner on December 1, 2001, ended the fire department’s policy of unlimited paid sick days. In its stead, he imposed a limit of fifteen sick days; firemen exceeding that limit would have their pay docked. Here is how the firemen responded: the number calling in sick on Christmas and New Year’s Day increased tenfold over the previous year.
The fire commissioner retaliated by canceling the firemen’s holiday bonus checks.1 The firemen were unimpressed: during the next year, they claimed 13,431 sick days, up from 6,432 a year earlier.2 Many firemen, apparently insulted by the new system, abused it or abandoned their previous ethic of serving the public even when injured or not feeling well.
I admit to having some sympathy for the commissioner. I once offered my teenage kids a price list for household chores as a way of topping up their modest weekly allowance. In response, they simply stopped doing the housework that they had once more or less happily done without incentives.
The commissioner’s difficulties and my failed experiment in home economics are far from exceptional. As we have already seen, imposing explicit economic incentives and constraints to induce people to act in socially responsible ways is sometimes ineffective or even, as Boston’s fire commissioner discovered, counter productive. Is this a problem? My hunch is that a larger penalty would have worked. The firemen’s massive sick call-ins on Christmas and New Year’s Day do not mean they had lost interest in money. Had the fire commissioner imposed a heavier penalty, the firemen would surely have shaped up, even if their anger and distrust would, as a result, have eclipsed their sense of duty. Economic interest would have substituted for pride in serving the public.
But these constraints and incentives may have limits. Heavy fines or more draconian punishments might have deterred phony call-ins, but would they have motivated the subtler and more immeasurable aspects of a fireman’s professionalism and bravery? Even if extreme penalties could do the job, a liberal society might find them repugnant. Instead of letting penalties substitute for the firemen’s sense of duty, the commissioner might have looked for policies that would affirm and enhance their civic pride.
Whether you think the firemen’s response to the commissioner’s incentives is a problem lines you up on one or the other side of some venerable and unresolved questions in the philosophy of governance. These are, roughly, whether a constitution for knaves could possibly work, and if so, would it be a good idea to govern according to one. I will begin addressing these questions by telling the remarkable story of how the constitution-for-knaves idea came about and the radical turn it took in the hands of the economists who transformed the market into a morality-free zone, beyond the reach of the ethical judgments we routinely make in the family, the polity, and the neighborhood.3
Political philosophers from Aristotle to Thomas Aquinas, Jean-Jacques Rousseau, and Edmund Burke recognized the cultivation of civic virtue not only as an indicator of good government but also as its essential foundation. “Legislators make the citizens good by inculcating habits in them,” Aristotle wrote in the Ethics. “It is in this that a good constitution differs from a bad one.”4 A century earlier, Confucius had provided advice about how this might be done, and about the pitfalls to be avoided: “Guide them with government orders, regulate them with penalties, and the people will seek to evade the law and be without shame. Guide them with virtue, regulate them with ritual, and they will have a sense of shame and become upright.”5
But through a twenty-first-century lens, reference to virtue and shame as the basis of a well-ordered society appears quaint, and to some even pernicious. Friedrich Hayek celebrated markets as “a social system which does not depend for its functioning on … all men becoming better than they now are, but which makes use of men in all their given variety and complexity, sometimes good and sometimes bad.”6 In the aftermath of the stock market crash of 1987, the New York Times headlined an editorial “Ban Greed? No: Harness It,” which continued: “Perhaps the most important idea here is the need to distinguish between motive and consequence. Derivative securities attract the greedy the way raw meat attracts piranhas. But so what? Private greed can lead to public good. The sensible goal for securities regulation is to channel selfish behavior, not thwart it.”7 An economics Nobel laureate, James Buchanan, illustrated how this might work, describing a visit to a farm stand near his home in Blacksburg, Virginia: “I do not know the fruit salesman personally, and I have no particular interest in his well-being. He reciprocates this attitude. I do not know, and have no need to know, whether he is in the direst poverty, extremely wealthy, or somewhere in between … Yet the two of us are able to … transact exchanges efficiently because both parties agree on the property rights relevant to them.”8
Jurists paralleled economists in this way of thinking. “If you want to know the law and nothing else,” Oliver Wendell Holmes Jr. told students in 1897 (and every entering law school class since has been instructed in the same belief), “you must look at it as a bad man, who cares only for the material consequences which such knowledge enables him to predict, not as a good one who finds his reasons for conduct, whether inside the law or outside it in the vaguer sanctions of conscience … The duty to keep a contract at common law means a prediction that you must pay damages if you do not keep it—and nothing else.”9 Hayek attributed a similar but more nuanced view to Adam Smith: “There can be little doubt … that Smith’s chief concern was not so much with what man might occasionally achieve when he was at his best, but that he should have as little opportunity as possible to do harm when he was at his worst.”10
The long road from Aristotle’s Legislator inculcating good habits in citizens to a system of economic governance and law for “bad men” began in the sixteenth century with Niccolò Machiavelli. Like Aristotle, he was concerned about social customs that would ward off what he termed “corruption,” but he gave rather different advice, in a passage anticipating Hume’s maxim on knaves (in the epigraph to this book) by more than two centuries: “Anyone who would found a republic and order its laws must assume that all men are wicked [and] … never act well except through necessity … It is said that hunger and poverty make them industrious, laws make them good.”11 Machiavelli’s “laws make them good” might sound a bit like Aristotle’s Legislator “inculcating habits” in the public. But here, as with his “all men are wicked,” Machiavelli uses “good” (buoni) and “wicked” (rei) to describe actions, not aspects of character.
The political philosopher Leo Strauss traced the genesis of this thinking among twentieth-century economists and others to the sixteenth-century Florentine: “Economism is Machiavellianism come of age.”12 But while the origins of what Strauss termed “economism” can indeed be found in his writings, Machiavelli, like Aristotle but unlike many modern economists, did not imagine that good governance was possible in a self-interested (“corrupt”) citizenry: “Neither laws nor orders can be found that are enough to check a universal corruption. For as good customs have the need for laws to maintain themselves, so do laws have the need for good customs so as to be observed.”13
For Machiavelli, laws have two functions: providing incentives and constraints to harness self-interest to public ends, and at the same time maintaining the good customs on which the effectiveness of the laws depends: “Good examples [of virtù] are born of good education [which is] born of good laws.”14 Machiavelli thus endorsed exactly the synergistic policy paradigm that I advance in the closing chapters, in which good laws and good customs are complements rather than substitutes.
Nonetheless for Machiavelli, government’s task was preeminently to induce citizens motivated by the “natural and ordinary humors” to act as if they were good. Machiavelli makes clear, especially in The Discourses on Livy, that it is not the morality of its citizens that ensures that a republic will be well governed, but rather the capacity of a statesman to “order its laws.”15 Compared to Italy, he wrote, Spain and France were well governed; but this was a distinction deriving “not much from the goodness of the people, which is for the most part lacking, … but from the way that these kingdom are ordered.” “France,” he continued, was “a kingdom moderated more by laws than any that has been known in our times.”16
The message was unmistakable: citizens of ordinary predispositions and desires could nonetheless be well governed if their behavior was “moderated … by laws.” The new idea here was that the quality of the governance of a society was not any simple aggregate of the quality of its citizens. Good governance was less a matter of a society being composed of good citizens than of how social institutions ordered interactions among citizens.
Modern day physical scientists might rephrase Machiavelli to say that the quality of the governance of a society is an emergent property of the polity, that is, a property of the whole that cannot be directly inferred from the characteristics of the citizens making it up. To Machiavelli, good government, then, was an emergent property of a well-ordered society.
Two centuries later, a radical version of this idea was the key message of Bernard Mandeville’s scandalous The Fable of the Bees. In it, the eccentric Dutch doctor turned Londoner held that for the maintaining of social order, virtue was dispensable, even pernicious. Mandeville’s hive thrived on licentious greed and invidious competition. But when the bees turned virtuous, collapse and disorder ensued. (Mandeville could not have known that members of the genus Apis, among the most cooperative of all species, are genetically programmed not to compete.) Mandeville’s insight that by dampening the demand for goods, the virtue of frugality might be a source of economic collapse is thought by some to be a precursor to the paradox of thrift, which is the foundation of Keynesian economics. The 1714 edition of his Fable announced in its subtitle that the work contained “several discourses to demonstrate that human frailties … may be turn’d to the advantage of civil society, and made to supply the place of moral virtues,” with the result, wrote Mandeville, that “the worst of all the multitude did something for the common good.”17
In case the reader might fail to decipher the lesson of the Fable, Mandeville explained in a prose commentary that: “Hunger, Thirst and Nakedness are the first Tyrants that force us to stir; afterwards our Pride, Sloth, Sensuality and Fickleness are the great Patrons that promote all Arts and Sciences, Trades, Handicrafts and Callings; while the great Taskmasters Necessity, Avarice, Envy and Ambition … keep the Members of the Society to their labour, and make them submit, most of them cheerfully, to the Drudgery of their Station; Kings and Princes not excepted.”18 To Mandeville, the benign consequences of what Machiavelli called the “ordinary humors” is not a natural fact about human society. Just as Machiavelli saw the foundation of good government in the human capacity to order the laws, for Mandeville it was “the dextrous Management of a skilfull Politician” that allowed the “Private Vices” to be “turned into Publick Benefits.”19
In contrast to the Aristotelian view that good laws make good citizens, Mandeville’s Fable suggested that the right institutions might harness shabby motives to elevated ends. It was left to Adam Smith to explain how this improbable alchemy might be accomplished, and it comes in his famous description of the motivation of the businessman, the consumer, the farmer: “He intends only his own gain, and he is in this, as in many other cases, led by an invisible hand to promote an end which was no part of his intention. Nor is it always worse for the society that it was no part of it. By pursuing his own interest he frequently promotes that of the society more effectually than when he really intends to promote it.”20 Competitive markets and secure, well-defined property rights, Smith explained, would order a society so that the invisible hand could do its magic: “It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own interest.”21
Thus, under the right institutions, elevated consequences may follow from ordinary motives.
A Constitution for Knaves
Attention turned to the design of institutions that might accomplish this. In his Essays: Moral, Political, and Literary (1742), David Hume had recommended the following “maxim”: “In contriving any system of government … every man ought to be supposed to be a knave and to have no other end, in all his actions, than private interest. By this interest we must govern him, and, by means of it, make him, notwithstanding his insatiable avarice and ambition, cooperate to public good.”22 In a similar spirit, for the design of public policy Jeremy Bentham offered his “Duty and Interest junction principle: Make it each man’s interest to observe … that conduct which it is his duty to observe.”23 In his Introduction to the Principles of Morals and Legislation, the first text of what we now call public economics, Bentham laid out the public policy implications of Hume’s maxim.
But while harnessing knaves was their leitmotif, these and other classical economists did not believe that economic actors and citizens were indeed amoral. Quite the contrary.
Hume pioneered the study of the evolution of social norms; and in the sentence immediately following the passage about knaves quoted above, he mused that it is “strange that a maxim should be true in politics which is false in fact,” Smith, in his Theory of Moral Sentiments, held that “How selfish soever man may be supposed, there are evidently some principles in his nature that interest him in the fortunes of others, and render their happiness necessary to him, though he derives nothing from it except the pleasure of seeing it.”24 In practice, the policies advocated by the classical writers did not overlook appeals to ethical and other-regarding motives. Bentham, as we will see, believed that punishments should be “moral lessons.”
The same tension between the assumption of unmitigated self-interest and the empirical reality of more complex and elevated human motivations did not seem to trouble the twentieth-century legal thinkers who took up the Homo economicus paradigm. Just a few lines before directing law students’ attention to the “bad man,” Holmes insisted that “the law is the witness and external deposit of our moral life.”25 Legal practice today, like the classical writers’ policies, recognizes a broad range of social dispositions rather than simply the presumed bad man’s self-interest. Market regulation, for example, combines fines for violations with requirements for public disclosure of the wrongs done, in order to shame the blameworthy.
Even Machiavelli introduced the idea of corrupt citizens, quoting a widespread expression of the time—“It is said that all men are wicked”—as a prudent assumption, not as evidence of a malign human nature. In his Discourses, Machiavelli rejected this assumption on empirical grounds: “Our reasonings are about those peoples where corruption is not very widespread and there is more of the good than of the rotten,” adding, “Very rarely do men know how to be completely bad or completely good.”26 Aristotle had much worse things to say in this regard. “Most men are rather bad than good and the slaves of gain … as a rule men do wrong whenever they can.”27
Thus the appeal of the constitution for knaves was not that citizens were in fact knaves. Rather, it was, first, that the pursuit of self-interest had come to be seen as a benign or at least harmless activity compared with other, more disruptive “passions,” such as religious fervor or the pursuit of power; and second, that as an empirical matter the virtues alone provided an insufficient basis for good government on the scale of the national state.
In the Middle Ages, avarice was considered to be among the most mortal of the seven deadly sins, a view that became more widespread with the expansion of commercial activity after the twelfth century.28 So it is surprising that self-interest would eventually be accepted as a respectable motive, and even more surprising that this change owed little to the rise of economics, at least at first. The year before Adam Smith wrote about how the self-interest of the butcher, the brewer, and the baker would put dinner on our table, James Boswell’s Dr. Johnson gave Homo economicus a different endorsement: “There are few ways in which a man can be more innocently employed than in getting money.”29
Smith’s passage above is widely cited as one of his few references to the invisible hand. But it should be remembered too for advancing the new idea that motives other than self-interest could be pernicious: “By pursuing his own interest he frequently promotes that of the society more effectually than when he really intends to promote it.”
It was the shadow of war and disorder that made self-interest an acceptable basis of good government. During the seventeenth century, wars accounted for a larger share of European mortality than in any century for which we have records, including what Raymond Aron called “the century of total war,” recently ended. Writing after a decade of warfare between English parliamentarians and royalists, Hobbes (in 1651) sought to determine “the Passions that encline men to Peace” and found them in “Feare of Death; Desire of such things as are necessary to commodious living; and a Hope by their Industry to obtain them.”30 Knaves might be preferable to saints.
The second appeal of the constitution-for-knaves approach was more directly related to Machiavelli’s work and to the practical turn in political theory that he advocated. “It seems more appropriate to go the reality of things rather than how we might imagine them,” he wrote. “There are many imagined republics and principalities that have never been seen nor known to exist in reality.”31 A century and a half later, Baruch Spinoza opened his Tractatus Politicus with: “No men are … less fit to govern … than theorists or philosophers … [who] sing the praises of a human nature nowhere to be found, [who] … rail at the sort which actually exist [and] conceive of men not as they are but as they would like them to be.”32 A generation after Spinoza, Mandeville introduced his Fable with virtually identical language.
But it was more than realism about human motives that recommended the retreat from virtue as the sine qua non of good government. If the “others” with whom we would seek to govern were our kin, neighbors, or friends, then our concern for their well-being and our desire to avoid social sanction or retaliation for violating social norms might induce us to act in ways that would take account of their interests and contribute to good governance. But with the growth of cities and the consolidation of the nation-state, the metaphor of the polity as a family or even a lineage became untenable. The scope of governance had expanded too much. With the increasing size of nations and the reach of markets, individuals interacted not with a few dozen familiars but with hundreds of strangers, and indirectly with millions.
The new policy paradigm was a response to the concern that when large numbers of strangers interact, ethical and other-regarding motives would be an insufficient basis for good government, which therefore would need to adopt a system of constraints and incentives to supplement the civic virtues. It was the insufficiency of the civic virtues, not their absence or irrelevance, that worried Machiavelli. The classical economists who shaped the new policy paradigm knew that no economy or social system could function well in their absence. Even the scandalous Mandeville reassured his readers on this point: “I lay it down as a first Principle, that in all Societies, great or small it is the Duty of every Member of it to be good; that Virtue ought to be encouraged, Vice discountenanced, the Laws obey’d and the Transgressors punished.”33
Similarly, the “natural liberty” that Smith endorsed was constrained by morality. His famous line “Every man … is left perfectly free to pursue his own interest in his own way” was conditioned in that very sentence by the proviso “as long as he does not violate the laws of justice.” Justice, he explained, required “every member of the society [to be protected] from the injustice or oppression of every other member of it.”34 Mandeville expressed the very same idea memorably, comparing human proclivities to the unruly growth of an untended grapevine: “So Vice is beneficial found, when it’s by justice lopt and bound.”35
The classical economists were thus quite aware that what later came to be called Homo economicus was a simplification, one that differed from what they knew about human behavior, but one that would clarify, among other things, how policies that altered economic incentives would affect behavior. Here is John Stuart Mill, among the last of the classical economists, laying down boundaries and key assumptions of economics that remained with us until very recently: “[Political economy] does not treat of the whole of man’s nature … It is concerned with him solely as a being who desires to possess wealth … It predicts only such … phenomena … as take place in consequence of the pursuit of wealth. It makes entire abstraction of every other human passion or motive.”36 He termed this “an arbitrary definition of man.”
The advent of the neoclassical school of economics in the late nineteenth century did not change the status of self-interest as a handy but empirically false abstraction. F. Y. Edgeworth, a founder of the neoclassical paradigm, expressed this view in his Mathematical Psychics: “The first principle of economics is that every agent is actuated only by self-interest.”37 But in the same passage he recognized as a fact that “the happiness of others as compared by the agent with his own, neither counts for nothing, nor yet ‘counts for one.’”
But like Mill, Edgeworth held that political economy could study the effects of incentives that appeal to the wealth-maximizing side of individuals without reference to the other motives that they both fully recognized but took to be beyond the purview of the discipline.
Separability of the Moral Sentiments and Material Interests
What the classical economists (and most economists since) missed is the possibility that moral and other prosocial behavior would be affected—perhaps adversely—by incentive-based policies designed to harness self-interest. “Why should it be,” Kenneth Arrow asked in his review of Richard Titmuss’s The Gift Relationship: From Human Blood to Social Policy, “that the creation of a market for blood would decrease the altruism embodied in giving blood?”38Until recently, most economists have been sure enough of the answer that they have not bothered to reply. But to Arrow it was “really an empirical question, not a matter of first principles.”
To most economists, however, the unspoken first principle was that incentives and morals are additively separable, a term from mathematics meaning that the effects of variations in the one did not depend on the level of the other. When two things are additively separable, they are neither synergistic—each contributing positively to the effect of the other, like a duet being better than the separate parts—nor the opposite.
I return to this separability assumption in subsequent chapters. You have already seen where it can go wrong. The firemen’s sense of duty toward the citizens of Boston was not separable from their self-interested regard for their own pay: a policy addressed to the latter appears to have diminished the former. The whole in this case was less than the sum of the parts. It is exactly this possibility that the self-interest-based policy paradigm overlooked.
Separability is hardly an everyday word, so it is worth pausing to consider an example. Amy Wrzesniewski, Barry Schwartz, and their coauthors studied the motives that led young men and women to attend the U.S. Army’s academy at West Point.39 They used questionnaires administered by the institution to nine annual cohorts of incoming cadets to assess whether an individual had sought admission for instrumental motives (“to get a better job,” the “overall reputation of West Point” presumably being seen as a plus on one’s résumé) or intrinsic motives (“desire to be an Army officer,” “personal development”). They then followed the 11,320 cadets for a decade after graduation to see whether these motives for admission correlated with later success.
Figure 2.1. Instrumental and intrinsic motives as substitutes in the postgraduation performance of West Point cadets The measure of success on the vertical axis is the estimated probability that a cadet will become a commissioned officer. Thus, each point on a particular line refers to the expected success probability for a cadet with the instrumental motivation score given by the horizontal axis and the intrinsic motivation score given by the label for the line. The instrumental motive score measures self-interested reasons for attending West Point. The intrinsic motive score measures the intrinsic motives mentioned in the text. High scores on this factor refer to those at the ninety-fifth percentile; low scores refer to those at the fifth percentile. (Data from Wrzesniewski et al. 2014.)
Figure 2.1 presents reports of how instrumental motives were statistically associated with the probability of becoming a commissioned officer (one of the measures of success) for cadets with high intrinsic motives for attending West Point, median intrinsic motives, and low intrinsic motives. What do the data show?
First, notice that for those with a mean level of instrumental motives (the zero on the horizontal axis) having strong intrinsic motives is associated with a substantially increased likelihood of making commission (the line for high intrinsic motivation indicates greater success in becoming a commissioned officer). Second, for those with very low intrinsic motives (the bottom, upward-sloping line), stronger instrumental motives were also associated with an enhanced likelihood of making commission.
But the big news is the third observation: for cadets with median or high intrinsic motives (the two downward-sloping lines), having strong instrumental motives was associated with worse performance. Intrinsic and instrumental motives were not additively separable, they were substitutes: more of one diminished the positive effect of the other.
How might the military academy use this information to train officers? If it believed that its pool of potential recruits largely lacked intrinsic motives, it would take the bottom line in the figure as its guide. As a result, it would appeal to their instrumental motives, stressing the job opportunities in the armed forces and the value of West Point’s reputation in getting jobs outside the military. But if it knew (correctly) that many cadets are highly idealistic young men and women wishing to serve their nation, it would be guided by the two downward sloping lines, and downplay the appeal to instrumental motives.
Economists have routinely either made the first assumption—that intrinsic motives are absent—or, if they recognized motives other than self-interest, assumed (for the most part unwittingly) that the two sets of motives were separable. But if they were really separable, all three lines in the figure would be upward rising and parallel.
The implicit assumption of separability has led economists to ignore two important possibilities: first, the use of incentives to harness self-interest to the public good might attenuate civic virtue or its motivational salience, and second, there might be conditions under which ethical and other-regarding concerns can jointly flourish and synergistically promote societal outcomes.
Markets as Morality-Free Zones
The broad (though for the most part implicit) acceptance of separability meant that assumptions about economic behavior did not have to be squared with facts and observations about human psychology. Since the late eighteenth century, economists, political theorists, and constitutional thinkers have embraced Hume’s maxim and have taken Homo economicus as their working assumption about behavior. Partly for this reason, competitive markets, well-defined property rights, and efficient and (since the twentieth century) democratically accountable states are seen as the critical ingredients of governance. Good institutions displaced good citizens as the sine qua non of good government. In the economy, prices would do the work of morals.
From there, it was a short step to thinking that while ethical reasoning and concern for others should inform one’s actions as a family member or citizen, the same did not go for shopping or making a living. Lewis Carroll’s Alice had taken the economists’ message to heart. When the Duchess exclaimed, “Oh, ’tis love, ’tis love makes the world go round,” Alice whispered, “Somebody said that it’s done by everyone minding their own business.”40
How could people minding their own business take the place of love? This was the classical constitutional challenge posed by Bentham, Hume, Smith, and others, and it still constitutes the holy grail motivating policy design. The quest is to find laws and other public policies that will at once facilitate peoples’ private pursuit of their own ends, yet also induce them to take adequate account of the effects of their actions on others.
In economic language, this can be done if each actor internalizes the entire costs and benefits of an action, including the effects on others, rather than taking account merely of one’s private benefits and costs, that is, those that affect one’s own profit or loss, pleasure or pain.
Deliberately taking account of the effects of one’s actions on others is what it means to have other-regarding preferences. Variations on the golden rule and other ethical precepts are one way to pursue this objective (or as the Duchess put it, “love”).
The other approach—“everyone minding their own business,” as epitomized by Buchanan’s indifference to his neighborhood fruit seller—relies on prices doing the work of morals. This could, in principle work if two conditions are met.
First, everything that matters to people when they decide what to do must have a price. This “everything has a price” requirement applies not only to goods (identical goods must have the same price), but also to any other aspects of a transaction, such as the noise nuisance effects of the goods’ production on neighbors or its carbon emissions.
Second, taxes, subsidies, and other policies must affect prices in such a way that the price a buyer pays to acquire a good includes all the costs incurred by anyone as a result of its production and use, and the price a seller receives likewise includes all the benefits (to the buyer and to others) of making the good available. Prices must measure all the social costs and benefits of the good’s production and distribution, rather than just the private costs to the buyer and seller. Call the second condition “the prices are right.”
Under these conditions, “everyone minding their own business” means that simply by paying attention to prices, self-regarding people will implement (albeit unwittingly) the holy grail: they will take full account of the effects of their actions on others. This is exactly what Mandeville had in mind when he startled his readers with the claim that “the dextrous Management of a skilfull Politician” would allow the “Private Vices” to be “turned into Publick Benefits.”
Smith went further than Mandeville. The remarkable idea behind his invisible hand is that under the right institutions, the “skilfull Politician” might not even be necessary: prices determined by market competition could do the job by themselves, entirely unaided by subsidies, taxes, or other government policies. Smith stressed that competition among buyers and among sellers was critical to this result, and he warned about the ways in which monopolies and cartels could thwart the invisible hand: “People in the same trade seldom meet together, even for merriment and diversion, but the conversation ends in a conspiracy against the public; or in some contrivance to raise prices.”41 Smith was also well aware of the many areas of policy, such as the provision of public education, that lay beyond the reach of the invisible hand.
Economists since then have stressed that the institutional conditions necessary for the invisible hand to work go beyond competition. For every good to have the right price, all economic interactions must be governed by what economists call complete contracts. This means that every aspect of an exchange—anything valued by either the exchanging parties or anyone else—has a price that is included in a contract governing the exchange. Complete contracts assign claims and liabilities in such a way that each actor “owns” all the benefits and costs resulting from his or her actions, including those conferred or imposed on others.
If contracts were complete, the equilibrium result of competition among self-interested individuals would ensure both that “everything has a price” and that “the prices are right.” As a result, competitive markets would implement outcomes that are termed Pareto-efficient, meaning that there would not be another technically feasible outcome in which at least one individual would be made better off without anyone being made worse off.
Kenneth Arrow and Gerard Debreu demonstrated the above reasoning in what I term the “invisible hand theorem,” and won a Nobel Prize for it. The axioms of this first fundamental theorem of welfare economics, as it is known to economists—notably, the assumption that contracts are complete—clarified the nature of an idealized world in which governmental intervention was not needed to address market failures, namely, situations in which uncoordinated exchanges or other economic activities resulted in Pareto-inefficient outcomes.
Less heralded, but more important for our purpose, is that in this world—as Hayek and Buchanan hinted—good governance does not seem to require virtue. The first fundamental theorem is true regardless of people’s preferences, including entirely amoral and self-regarding ones.
Markets thus achieved a kind of moral extraterritoriality akin to the suspension of a host nation’s sovereignty and laws accorded to foreign embassies. One could frown on or even chastise a neglectful parent or a citizen who transgressed laws when it served his purpose. But a selfish shopper? Or for that matter, a selfish banker? As long as everything has the right price, one’s pursuit of self-interest in a market setting is constrained (by these right prices) to take appropriate account of the effect of one’s actions on others. This is why Buchanan felt no shame in expressing his indifference to the well-being of his fruit seller.
The voluntary nature of transactions and the efficiency of the results (under the assumptions of the theorem) made competitive exchange a special domain in which one could suspend the normative standards commonly applied to relationships among citizens or family members. Generalizing Buchanan’s indifference and designating the market a “moral free zone,” the philosopher David Gauthier held that “morality arises from market failure… . Morality has no application to market interactions under the conditions of perfect competition.”42
And so avarice, repackaged as self-interest, was tamed, transformed from a moral failing to just another kind of motive, like a taste for ice cream.
Economics for Knaves
But what if, unlike Buchanan and his fruit seller, two parties do not agree on the property rights relevant to an exchange? This happens when contracts are not complete and there are aspects of an exchange that are not priced: you breathe my secondhand smoke; farmer Jones’s bees pollinate farmer Brown’s apple trees. When Jones exchanges his honey for Brown’s apples, he typically cannot also charge for the free pollination services provided by his bees. Their assistance to farmer Brown and countless other orchard owners nearby is a spillover (termed an “external economy” or simply an “externality”), that is, a direct effect between economic actors that is not priced or covered by (is “external” to) the contractual terms of the exchange. The bees’ services are not included in the price Jones gets for his honey. As a result, the “everything has a price” and “the price is right” conditions do not hold, and Jones’s private revenues from the sale of honey fall short of the total social benefits his farm produces (the bees’ pollination services are not included). As a result, inefficiently little honey (and bee pollination services) will be produced.
Contracts are also incomplete (or nonexistent) in team production processes such as research and the provision of legal services, and in the voluntary provision of public goods such as neighborhood amenities or adherence to social norms. Public goods—such as radio broadcasts or ideas—are an extreme form of incomplete contracts because (by definition) they are both nonrival (my having more does not reduce the availability to you) and nonexcludable (the enjoyment of the public good cannot be restricted; if I have it, so can you).
In cases like farmer Jones’s pollination of farmer Brown’s bees, or the provision of public goods, unregulated market interactions among self-interested actors fail to implement efficient outcomes (too little pollination, too little of the public good). But this does not mean that self-interested competition in markets should be restricted; it means that public policy has to take on the task of getting the prices right.
In the early twentieth century, Alfred Marshall and Arthur Pigou spelled out the economic logic for nonetheless letting prices do the work of internalizing the effects of one’s actions on others, even when markets fail. Where contracts were incomplete, they advocated taxes on industry for the environmental damage (external diseconomies) it imposed on others, and subsidies for a firm’s worker-training activities, which benefit other firms when workers change jobs.
Farmer Jones would get a subsidy equal to the value of his bees’ pollination service to Farmer Brown so that Jones’s revenues (including the subsidy) would measure the full social benefits of his honey production. What came to be called optimal taxes and subsidies were those that recompensed an economic actor for the benefits his actions conferred on others and made him liable for the costs of his actions borne by others, when these benefits and costs would not otherwise be accounted for in the actor’s private revenues and costs.
Green taxes that “make the polluter pay” for environmental spillovers are an example. Where feasible, these optimal incentives would exactly implement Bentham’s “duty and interest junction” principle: altering the material incentives under which the individual acts, in order to align self-interest with public objectives. The optimal taxes and subsidies advocated by Marshall, Pigou, and economists since then are thus a substitute for complete contracts, an effort to extend the reach of the invisible hand to cases in which its assumption is violated. Ideally, such taxes and subsidies can put a price on everything that matters and also get those prices right.
The resulting guide for the policy maker clarifies what is required to induce citizens to act if they were good, namely, the kinds of incentives and constraints that motivate a self-regarding individual to act as if he valued the effect of his actions on others in the same manner that those who are affected would evaluate them. The job description of the wise policy maker in this case is no longer that of Aristotle’s Legislator, tasked with uplifting the population, but instead that of Machiavelli’s republican, tasked with ordering the right laws to induce citizens to act as if they were good.
Bucolic examples like farmer Jones’s bees and Brown’s apple blossoms were staples for economists teaching about incomplete contracts. The textbook example of a public good was the lighthouse, whose light can be seen by all if it can be seen by anyone. But the problem of incomplete contracts is not some curiosum on the periphery of the economy. We will see that it is a ubiquitous characteristic of markets for labor, credit, information, and the other central arenas of the capitalist economy.
The fact that incomplete contracts are the rule and not the exception sets in train a series of implications for the use and limits of incentives, and it is worth pausing to consider why it is true. Information about the amount and quality of the good or service provided in an exchange is very often either asymmetric, that is, not known to both parties, or nonverifiable, meaning that even if known to both, it cannot be used in the courts to enforce a contract. Where this is the case, some aspects of an exchange will not be in the contract: the contract is incomplete. As Emile Durkheim put it: “Not everything in the contract is contractual.”43 As a result, market failures are not confined to environmental spillovers, but occur in the everyday exchanges essential to a capitalist economy: labor markets and credit markets. It is impossible to write an enforceable employment contract that specifies that the employee will work hard and well. Credit contracts cannot be enforced if the borrower is broke.44
The labor- and credit-market examples share a common structure: a principal (the employer or the lender) wishes to induce an agent (the employee, the borrower) to act in a way beneficial to the principal but counter to the agent’s interests (work hard, use the money borrowed in a way that maximizes the expected repayment, not the expected returns to the borrower). But because information about work effort or the use of the money is either not known to the principal or not admissible in court, the conflict of interest between the two cannot be resolved by specifying the terms of a complete and enforceable contract.
When contracts are incomplete, the de facto terms of the exchange are determined in large part by the strategic interaction between the parties, not by the courts. The outcome of this interaction depends on the bargaining power of the two parties and their social norms. The same problem arises when a farmer pays a share of his crop to the landowner. In all three cases, the agent (including the sharecropper) does not own the results of his or her actions: the lender takes the loss if the borrower cannot repay; the employer enjoys most of the benefits of the employee’s hard work but likewise cannot recover back wages if the work has not been done.
Thus, in much of a modern capitalist economy, the complete-contracts assumption of the invisible hand theorem is violated. The great contribution of the mathematical representation of the workings of a market economy that allowed a proof of the “invisible hand theorem” was to clarify just what Adam Smith’s idea really required. Here is how Arrow later described the contribution of his theorem: “There is by now a long and … imposing line of economists from Adam Smith to the present who have sought to show that a decentralized economy motivated by self-interest and guided by price signals would be compatible with a coherent disposition of economic resources that could be regarded in a well-defined sense as superior to a large class of possible alternative dispositions … It is important to know not only whether it is true but whether it could be true” (original emphasis).45
Thanks to Arrow and others, the conditions under which it “could be true”—not only competition but also complete contracts in everything that matters—are now known to be highly restrictive, showing how unlikely it is that a policy of thoroughgoing laissez-faire would implement an efficient outcome. I know from experience that writers of introductory economics textbooks struggle to find empirical examples of even a single market that approximates the model on which the theorem is based.
From Machiavelli to Mechanism Design
But maybe the kinds of optimal subsidies and taxes proposed by the Marshall-Pigou tradition can provide surrogates for the complete contracts missing from the credit, labor, and other markets. If so, prices could yet do the work of morals; the domain in which “people minding their own business” would be a good policy could be vastly expanded.
Many ingenious systems of incentives have been proposed to this end. There is even an academic field called mechanism design that exists for this purpose. A mechanism is simply a set of property rights, incentives, constraints, or other rules governing how people interact. But as we will see (in chapter VI), the assumptions required to make these clever mechanisms work are about as remote from real economies as the axioms on which the first fundamental theorem of welfare is based. Mechanism design has not yet devised incentives that make ethical and other-regarding motives redundant, nor is it likely to. Getting entirely self-regarding people reliably to act as if they cared about the effects of their actions on others has so far eluded jurists and policy makers. I explain why in chapter VI.
So it is no surprise that except on the whiteboards of economics classrooms, people try to avoid dealing with Homo economicus. Employers prefer to hire workers with a strong work ethic; banks prefer to lend to people whom they trust to conduct their business as proposed rather than to adopt self-interested but riskier projects. Again except on the whiteboard, everyone knows that “the contract itself is not sufficient,” as Durkheim wrote a century ago; “regulation of the contract … is social in origin.”46 He was repeating the commonplace that handshakes matter; and where they do not, the economy underperforms.
In a paper explaining his invisible hand theorem, Arrow wrote: “In the absence of trust … opportunities for mutually beneficial cooperation would have to be forgone … Norms of social behavior, including ethical and moral codes [may be] … reactions of society to compensate for market failures.”47 In other words, because contracts are incomplete, morals must sometimes do the work of prices, rather than the other way around.
Arrow’s point is that social norms and moral codes can attenuate market failures when they have the effect of internalizing the benefits and costs that an individual’s actions confer or inflict on others. Despite the incompleteness of contracts, a modern economy’s major markets—for labor, credit, and knowledge—can sometimes function tolerably well because social norms and other-regarding motives foster a positive work ethic, an obligation to tell the truth about the qualities of a project or a piece of information, and a commitment to keep promises. “The moral economy” is not an oxymoron.
The importance of norms and other social motives extends far beyond what we usually term market failures. It encompasses many of the arenas of social life in which the effects of one’s actions on others is not governed by contract: the long-term climatic effects of one’s lifestyle choices, the creation of drug-resistant superbugs through the opportunistic use of antibiotics, and the traffic congestion resulting from one’s choice of a way to get around. The need for social norms to underwrite good governance is likely to increase as these and other problems pose ever-larger challenges to our well-being. The changing nature of work itself—from producing things to processing information and providing care, for example—further suggests that our economies will increasingly be characterized by contractual incompleteness.
The classical economists were right in thinking that ethical and other-regarding motives would be insufficient for the good governance of an economy in which many interactions are conducted between strangers. Nobody now doubts Smith’s once-remarkable claim that a citizen’s self-interest could be harnessed to “promote an end that was no part of his intention.” But Joseph Schumpeter, who pioneered the economics of innovation and technical progress, was likewise right when he wrote that “no social system can work … in which everyone is … guided by nothing except his own … utilitarian ends.”48 In this passage he was not describing families or the polity, domains where the importance of ethical and other-regarding motives is widely recognized, but the workings of a capitalist firm.
Machiavelli anticipated Rousseau’s injunction to take “men as they are” by more than two centuries (and mechanism design by more than four) in charging his republican lawmaker to devise a structure of governance in which people with “ordinary and natural humors” would nonetheless choose to act in ways that result in a well-governed republic. This was a major contribution to our understanding of how policies should be designed and laws ordered. But the same cannot be said of the radical extension of this good idea, first by Mandeville and subsequently by economists. Even today many in my discipline combine a professed indifference to the nature of individual preferences with excessive confidence in the ability of clever incentives to induce even an entirely amoral and self-interested citizenry to act in the public interest.
Ethical and other-regarding motives have always been essential to a well-governed society and are likely to be even more so in the future. Policies that ignore this fact and are indifferent to the preferences that motivate people’s actions may compromise these essential predispositions. This is why policy makers should be concerned about the firemen’s response to the commissioner’s punitive incentives, and about the parents who arrived even later at the day care centers after the fines were imposed.
In the pages that follow, we will put ourselves in the shoes of Aristotle’s Legislator, who shares these concerns and seeks to govern well, knowing that incentives and constraints, while essential to any social order, will never be sufficient and may have unintended adverse effects on ethical motivations. In the eyes of the Legislator, the policy tool kit based on the self-interest maxim, designed as it was for knaves and the wicked, may be part of the problem.