Everything Is Obvious

by Duncan J. Watts

  THE INDIVIDUAL AND SOCIETY

  Whether we can answer them or not, questions about how to differentiate luck from talent and individual contributions from collective performance can also inform our thinking about fairness and justice in society as a whole. This issue was raised in somewhat different language in a famous argument between the political philosophers Robert Nozick and John Rawls over what constitutes a just society. Nozick was a libertarian who believed that people, in essence, got what they had worked for, and therefore no one was entitled to take it from them, even if that meant putting up with large inequalities in society. Rawls, by contrast, asked what kind of society each of us would choose to live in if we didn’t know beforehand where in the socioeconomic hierarchy we would end up. Rawls reasoned that any rational person would prefer an egalitarian society—one in which the worst off were as well off as possible—over one in which a few people were very rich and many were very poor, because the odds of being one of the very rich was so small.24

  Nozick found Rawls’s argument deeply disturbing, in large part because it attributed at least part of what an individual accomplishes to society rather than to his or her own efforts. If an individual cannot keep the output of his talent and hard work, Nozick’s reasoning went, he is effectively being forced to work for someone else against his free will, and therefore does not fully “own” himself. Taxation, it follows, along with all other attempts to redistribute wealth, is the moral equivalent of slavery, and therefore unacceptable no matter what benefits it might confer on others. Nozick’s argument was appealing to many people, and not only because it provided a philosophical rationale for low taxes. By reasoning about what would be considered fair in a hypothetical “state of nature,” Nozick’s arguments also played well to commonsense notions of individual success and failure. In a state of nature, that is, if one man invests the time and effort to build, say, a canoe for fishing, no one else is entitled to take it from him, even if it means that the man lacking the canoe will suffer or perish. Individual outcomes, in other words, are solely the product of individual efforts and skill.

  And in a state of nature, Nozick might well be right. But the whole point of Rawls’s argument was that we do not live in such a world. Rather, we live in a highly developed society in which disproportionately large rewards can accrue to individuals who happen to possess particular attributes and who experience the right opportunities. In the United States, for example, two equally skilled and disciplined athletes—one a world-class gymnast and the other a world-class basketball player—are likely to enjoy wildly different degrees of fame and fortune through no fault or merit of their own. Likewise, two children with indistinguishable genetic endowments—one of whom is born into a wealthy, highly educated, socially prestigious family, and the other who is born into a poor, socially isolated family with no history of educational achievement—have dramatically different prospects for lifetime success.25 Finally, even random differences in opportunities that arise early in one’s career can accumulate, via the Matthew Effect, to generate large differences in outcomes over the course of a lifetime. Rawls’s claim was that because the mechanisms of inequality are essentially accidents—whether of birth, or of talent, or of opportunity—a just society is one in which the adverse effects of these accidents is minimized.

  Rawls’s claim is often misunderstood to mean that inequality of any kind is undesirable, but this is not at all what he was saying. Allowing for the possibility that through hard work and application of one’s talents one can do better than one’s peers is no doubt beneficial for society as a whole—just as libertarians believe. In a Rawlsian world, therefore, people are free to do whatever they want, and are perfectly within their rights to take whatever they can according to the rules of the game. And if the rules of the game have it that basketball players earn more than gymnasts, or investment bankers earn more than teachers, so be it. Rawls’s point was just that the rules of the game themselves should be chosen to satisfy social, not individual, ends. Bankers, in other words, are entitled to whatever they are able to negotiate with their employers, but they are not entitled to an economic system in which the financial industry is so much more profitable than any other.

  The counterintuitive consequence of this argument is that debates over individual compensation should not be conducted at the level of individuals. If it’s true that bankers are paid too much, in other words, the solution is not to get into the messy business of regulating individual pay—as indeed the financial industry itself has argued. Instead, it is to make banking less profitable overall, say by limiting how much banks and hedge funds can leverage their portfolios with borrowed money, or by forcing so-called over-the-counter derivatives to be traded on public exchanges. The financial industry could argue, of course, that leverage and customization offer benefits to their customers and to the broader economy as well as to themselves. And these claims, although self-serving, may have merit. But if the purported benefits are outweighed by the economic costs of increased risk to the economic system as a whole, then there is nothing inherently unjust about society changing the rules. We can argue about whether making banking a less profitable industry would, on balance, be a good or a bad thing for society, but that’s what the debate should be about—not about whether particular individuals deserve their $10-million bonuses. Libertarian arguments about what would or would not be fair in a state of nature are simply irrelevant, because in a state of nature nobody would be getting a $10-million bonus.

  For much the same reasons, arguments about the so-called redistribution of wealth are mistaken in assuming that the existing distribution is somehow the natural state of things, from which any deviation is unnatural, and hence morally undesirable. In reality, every distribution of wealth reflects a particular set of choices that a society has made: to value some skills over others; to tax or prohibit some activities while subsidizing or encouraging other activities; and to enforce some rules while allowing other rules to sit on the books, or to be violated in spirit. All these choices can have considerable ramifications for who gets rich and who doesn’t—as recent revelations about explicit and implicit government subsidies to student lenders and multinational oil companies exemplify.26 But there is nothing “natural” about any of these choices, which are every bit as much the product of historical accident, political expediency, and corporate lobbying as they are of economic rationality or social desirability. If some political actor, say the president or Congress, attempts to alter some of these choices, say by shifting the tax burden from the working class to the superrich, or by taxing consumption rather than income, or by eliminating subsidies to various industries, then it is certainly valid to argue about whether the proposed changes make sense on their merits. But it is not valid to oppose them simply on the grounds that altering the distribution of wealth itself is wrong in principle.

  PRIVATIZE THE GAINS, SOCIALIZE THE LOSSES

  Arguments about the claims that society can justly make on its members are also relevant to questions of accountability. For example, a great deal has been written recently about whether the banks and other financial firms that pose a serious systemic risk should be allowed to exist in the first place.27 Much of this discussion has revolved around whether it is the size or interconnectedness or some other attribute of a financial institution that determines how much risk its failure might create for the rest of the economy. Addressing these questions is important, if only to understand better how to measure systemic risk, and hopefully to limit it through thoughtful regulation. But it is also possible that there is no way to eliminate systemic risk in financial systems, or to guarantee the robustness and stability of any complex interconnected system. Power-transmission grids are generally able to withstand the failure of individual transmission lines and generators, but occasionally a seemingly innocuous failure can cascade throughout the entire system, knocking out hundreds of power stations and affecting millions of consumers—as has happened several times in recen
t years in the United States, Europe, and Brazil.28 Likewise, our most sophisticated engineering creations, such as nuclear reactors, commercial aircraft, and space shuttles, all of which are designed to maximize safety, occasionally suffer catastrophic failures. Even the Internet, which is extremely robust to physical failures of all kinds, turns out to be highly vulnerable to a whole range of nonphysical threats, including chronic spam, Internet worms, botnets, and denial-of-service attacks. It may be, in fact, that once a system has attained a certain level of complexity, there is no way to rule out the possibility of failure.29 If so, we need not only better tools for thinking about systemic risk, but also a better way of thinking about how to respond to systemic failures when they inevitably occur.

  To illustrate, consider the response of the banking community to the Obama administration’s proposal to tax certain trading profits as a way to recoup taxpayer bailout money. In the bankers’ view, they had already paid back their bailout money—with interest—and therefore nothing more could legitimately be asked of them. But imagine for a moment what the banking industry profits would have been in 2009 absent the several hundred billion dollars of government funds from which they benefited both directly and indirectly. We can never know for sure, of course, because we didn’t run that experiment, but we can make some educated guesses. AIG, for one, would probably not exist and its various counterparties, including Goldman Sachs, would be short several tens of billions of dollars that were funneled to them through AIG. Citigroup might well have collapsed, and Merrill Lynch, Bear Stearns, and Lehman Brothers might all have been dissolved rather than merged with other banks.

  All told, the banking industry might have lost tens of billions of dollars in 2009—quite the opposite of the tens of billions they actually made—and many thousands of bankers who in reality received bonuses would instead have been out of work. Now imagine that in the fall of 2008 the leaders of Goldman Sachs, J.P. Morgan, Citigroup, and the like were offered a choice between the “systemic support” world in which they were guaranteed government support and the “libertarian” world in which they would be left hung out to dry. Forget for a moment the devastation that would have been wreaked on the rest of the economy, and ask how much of their future compensation the banks would have been willing to concede in order to not be allowed to fail? Again, it’s a hypothetical question, but with their very survival in the balance it seems safe to assume that they would have agreed to commit more than the face value of their direct loans.

  That the banks and their allies have instead managed to portray themselves as victims of stifling government intervention and political populism run amok is therefore disingenuous, and not only for the obvious reason that the banks benefited from considerable government largess other than direct loans.30 The real reason is philosophical consistency. When times are good, banks wish to be perceived as independent risk-taking entities, entitled to the full fruits of their hard-won labors. But during a crisis they wish to be treated as critical elements in a larger system to which their failure would pose an existential threat. Whether this latter claim is true because they are too big or too interconnected, or for any other reason, is actually not so important. The real point is that either they are libertarians, who should bear the full weight of their own failures as well as their successes, or else they are Rawlsians, paying their dues to the system that takes care of them. They should not be able to switch philosophies at their convenience.

  BEARING EACH OTHER’S BURDENS

  In his recent book Justice, the philosopher Michael Sandel makes a similar point, arguing that all questions of fairness and justice have to be adjudicated in light of how dependent we are on each other—most obviously, on our networks of friends, family, colleagues, and classmates, but also on our communities, on our national and ethnic identities, and even on our distant ancestors. We are proud of what “our” people have accomplished, we protect them against outsiders, and we come to their aid when necessary. We feel that we owe them our loyalty for no reason other than that we are connected to them, and we expect them to reciprocate. It should come as no surprise, therefore, that social networks play a critical role in our lives, connecting us with resources, providing information and support, and facilitating transactions on the basis of mutual trust and assumed respect. So embedded are we in networks of social relations that it would be hard to imagine ourselves outside of them.31

  So far, this view seems uncontroversial. Even Margaret Thatcher, who claimed that “there is no such thing as society,” conceded that families mattered as well as individuals. But Sandel argues that the importance of social networks has a counterintuitive consequence for the notion of individual freedom. Whatever we might like to think, we are never entirely free, nor would we want to be. The very ties that give our lives meaning also constrain us, and it is precisely by constraining us that they give us meaning. From Sandel’s perspective, it makes no more sense to reason about fairness or justice exclusively from the perspective of individual freedom than it does to reason about what is fair exclusively by analogy with some imaginary state of nature. Neither is an accurate representation of the world in which we actually live. Like it or not, our notions of justice must deal with this tension between the individual and society as a whole. Yet this can be easier said than done. For example, Sandel argues that one ought not to feel proud of one’s heritage as an American without also feeling shame about the country’s history of slavery. Libertarians might argue that it was their ancestors, not them, who carried out such reprehensible deeds and therefore they have nothing to apologize for. But surely these same people are also proud of their ancestry, and would rather live in this country than in any other. In Sandel’s view, one cannot simply decide at one’s convenience when to identify with one’s ancestors and when to absolve oneself of them. Either you’re a part of that extended community, in which case you must share the costs as well as the benefits, or you’re not, in which case you get neither.

  Sandel’s argument that our individual actions are inextricably embedded in networks of social relations has consequences not only for arguments about fairness and justice, but also for morality and virtue. In fact, Sandel argues that one cannot determine what is fair without also evaluating the moral status of competing claims. And that in turn requires us to resolve the moral purpose of social institutions. We cannot decide whether gay marriage is right or just, for example, without first deciding what the point of marriage is. We cannot determine whether a particular university’s admission criteria are fair or unfair until we have first determined what the purpose of a university is. And we cannot decide if the way bankers are compensated is appropriate without first establishing what it is that banking should accomplish for society. In this respect, Sandel’s view harks back to the ancient philosophy of Aristotle, who also believed that questions of justice require reasoning about the purpose of things. Unlike Aristotle, however, Sandel does not espouse a view of purpose that is determined outside of the social system itself—say by divine decree. Rather, purpose is something that the members of a society must decide collectively. Sandel therefore concludes that a just society is not one that seeks to adjudicate disputes between individuals from a morally neutral perspective, but one that facilitates debate about what the appropriate moral perspective ought to be. As Sandel acknowledges, this is likely to be a messy affair and always a work in progress, but he does not see any way around it.

  What’s particularly interesting about Sandel’s arguments—at least to a sociologist—is how sociological they are. Sociologists, for example, have long believed that the meaning of individual action can only be properly understood in the context of interlocking networks of relationships—a concept that is called embeddedness.32 Even more so, Sandel’s claim that the values by which we judge fairness are necessarily the product of society reflects the idea, first advanced by sociologists in the 1960s, that social reality is a construction of society itself—not something that is handed to us by some ex
ternal world.33 An important implication of Sandel’s argument, therefore, is that the fundamental questions of political philosophy are sociological questions as well.

  How are we to answer these questions then? Certainly thinking about them in the way that Sandel does is one approach, and that has generally been the way that sociologists have approached them as well. But relying on unaided intuition to reason through these sorts of questions can also be limiting. It’s fine to argue, as many people have argued recently, that banks whose actions result in systemic risk ought to be held accountable for that risk, say by purchasing “systemic risk” insurance or being required to increase their capital reserves. But without a sound understanding of systemic risk, it is impossible to measure how much systemic risk a particular action creates, and therefore how much of a penalty ought to be imposed for taking it. Likewise, it is one thing to point out that we place too much emphasis on outcomes when evaluating processes, or attribute too much importance to “special people” in determining those outcomes. But it is quite another to come up with better measures of performance and a better sense of how complex social systems like companies, markets, and societies actually work. As important as it is to think through these issues, in other words, it is also important to do more than simply argue about them. And on this point it is worth asking what, if anything, social science might be able to offer.