Success and Luck: Good Fortune and the Myth of Meritocracy
Princeton University Press, 2017, 208 pp., $26.95
Robert Frank’s recent book, Success and Luck, is an engaging, partly autobiographical account of why and how most Americans underestimate the role of luck in economic success. Frank sees our tendency to overlook the role of luck as a cultural bias that helps explain why we are less likely than Europeans to favor high taxes on the rich and generous benefits for the poor.
Frank also thinks that compressing the distribution of income would make life better for the vast majority of Americans, including most of the rich, because it would strengthen our sense of community and make us treat one another better. He concedes, of course, that higher taxes would leave the rich with less money to spend on private luxuries, such as living rooms with a view of the Golden Gate Bridge or parties at New York’s most expensive restaurants. He argues, however, that the price of most such luxuries would fall if the rich had less money to spend. Meanwhile, the rich would still have more money for luxuries than anyone else, so they would still end up owning most of the homes with great views and eating most of the meals at expensive restaurants.
To assess these arguments we need to be clear about what Frank means by “luck.” The word is a shape-shifter, not only because it means different things to different people, but because it means different things in different contexts to the same person, including Frank. In everyday conversation, however, “good luck” usually refers to an unforeseen event that makes someone better off without requiring commensurate risk or effort. Finding a $20 bill in the street is a standard example. “Bad luck” is a little more complicated. Sometimes it is just the antonym of good luck: an unforeseen event that makes someone worse off, like losing a $20 bill. However, bad luck can also refer to the unpreventable consequences of widely anticipated events. The weather bureau can now predict a hurricane far enough in advance for almost everyone in its path to flee, for example, but homeowners still cannot prevent the hurricane from flattening their beachfront summerhouses. The common thread uniting “good” and “bad” luck is thus that they both involve events beyond our control, which means we do not deserve either credit or blame for them.
The kinds of events that both Frank and others describe as “lucky” are, however, seldom entirely beyond the control of the individuals they affect. In his preface to Success and Luck, for example, Frank quotes at length from an address that Michael Lewis delivered to Princeton’s graduating class in 2012. Lewis described his path to success this way:
One night I was invited to a dinner, where I sat next to the wife of a big shot at a giant Wall Street investment bank called Salomon Brothers. She more or less forced her husband to give me a job. I knew next to nothing about Salomon Brothers. But Salomon Brothers happened to be where Wall Street was being reinvented. . . . When I got there I was assigned, almost arbitrarily, to the very best job in which to observe the growing madness: they turned me into the house expert on derivatives. A year and a half later Salomon was handing me a check for hundreds of thousands of dollars to give advice about derivatives to professional investors.
Not long after Salomon handed Lewis that check he quit his job to write a book, about which he says:
The book I wrote was called Liar’s Poker. It sold a million copies. I was 28 years old. I had a career, a little fame, a small fortune, and a new life narrative. All of a sudden people were telling me that I was born to be a writer. This was absurd. Even I could see that there was another, truer narrative, with luck as its theme. What were the odds of being seated at that table next to that Salomon Brothers lady? Of landing inside the best Wall Street firm from which to write the story of an age? Of landing in the seat with the best view of the business?
Although Lewis ascribes this sequence of events to luck, that description turns out to be overly simple. First, he asks rhetorically what the odds were of his being seated next to the “Salomon Brothers lady.” At least according to Wikipedia, Lewis was seated there at his own request by his cousin, the Baroness Linda Monroe von Stauffenberg, who was on the planning committee for the dinner. Lewis needed a job and hoped that his cousin could help him land one at Salomon by seating him next to the wife of the man who ran Salomon’s London operations. Having a cousin who can seat you next to someone whose husband can give you a job at Salomon Brothers would qualify as “good luck” in the colloquial sense even if Lewis made no such request, but it seems he may have planned the whole thing in advance. The only part of this story that was really beyond his control was the fact that his cousin knew the “Salomon Lady” in the first place. That Lewis then made a favorable impression on both the lady and her husband was Lewis’s own doing.
Salomon’s decision to turn Lewis into an expert on derivatives may, in contrast, have been entirely outside his control. However, the fact that he got a check for “hundreds of thousands of dollars” a year and a half later was probably not luck but a byproduct of his being a quick study who used his skills to sell Salomon’s clients a lot of derivatives.
Lewis also implies, without explicitly saying so, that Liar’s Poker sold a million copies by luck alone. However, the book is both gripping and well written. Gripping, well-written books do not always sell a million copies, but their chances of doing so are a lot better than the chances of a boring, badly written book doing the same. Since some of his subsequent books were also best-sellers, the success of Liar’s Poker cannot have been pure luck.
Lewis’s account has at least two implications. First, he would not have written Liar’s Poker without some lucky accidents, of which the social connection between his cousin Linda and the wife of the big shot at Salomon was probably the most important. Beyond that, the role of luck is less certain. To echo Lewis, what are the odds that if another young man had been seated next to the Salomon Brothers lady she would have insisted that her husband hire him? Or that another young man, having been handed a check for hundreds of thousands of dollars after 18 months at Salomon, would have left to write a book rather than staying to see if subsequent checks got even bigger? Luck was probably a necessary condition for writing Liar’s Poker, but it was certainly not sufficient.
Both Frank and others often use “chance” as a synonym for “luck,” but that usage is often misleading. “Chance” typically refers to events that are fundamentally unpredictable, like winning the state lottery. The path of a hurricane is not pure chance, which is why forecasters can predict it with some accuracy. However, I might describe the fact that the hurricane destroyed your house but not your next-door neighbor’s house as chance, at least if I did not know that your next door neighbor’s house was designed to withstand hurricanes. That said, only a small subset of all “lucky” and “unlucky” events is attributable to chance alone.
The ancient Romans thought the goddess Fortuna had blessed individuals who were consistently lucky, while those who seemed consistently unlucky must have offended some other powerful god. Modern science discourages such thinking by assuming that impersonal causal chains are everywhere, even when we cannot see or understand them. We might therefore expect the spread of modern science to have reduced the frequency with which people attribute events to good or bad luck. We see this tendency reflected even in well-known hokum folk wisdom: “The harder I work the luckier I get.” However, in the early 1990s, when the British psychologist Richard Wiseman and his collaborators asked a large street sample of London shoppers whether they considered themselves lucky, unlucky, or neither, 50 percent said they were lucky, 14 percent said they were unlucky, and only 36 percent said they were neither.1
Despite the fact that Success and Luck is subtitled “Good Fortune and the Myth of Meritocracy,” Frank does not actually think meritocracy is a myth. On the contrary, he argues that big corporations have become steadily more meritocratic in recent decades. By this he means that they try to select their executives based on criteria that both they and their competitors think likely to maximize a firm’s profitability, such as how quickly candidates can master complicated subjects quickly whether they can persuade others that their analysis is correct, and how hard they are likely to work on their employer’s behalf. These criteria may not be very good at identifying the CEO most likely to maximize a firm’s future profits, but most corporations are trying to use the best predictors they can find.
One consequence of meritocracy’s spread has been the growing number of candidates considered for top corporate jobs. Until the 1960s most big American corporations looking for a new CEO considered only white American men who already worked at their firms. Today these firms search all over the globe and often consider candidates who are not American by birth or upbringing. They also consider (and sometimes even hire) women, non-whites, and managers who currently work at other firms. As a result, the number of potential candidates for most top jobs has grown substantially, as has the number of candidates for other high-level jobs.
If nothing else had changed, expanding the number of candidates should have driven down their compensation. However, many big firms have grown even bigger over the past fifty years, and when a firm’s revenue doubles, the value of managers who can add one percentage point to the firm’s profit margin also doubles. Global competition and technical innovation have also made corporate directors less complacent about firms’ ability to prosper simply by continuing to do what they have always done. Taken together, rising firm sizes and anxiety about the future have led big firms to pay whatever it takes to get a CEO who is widely viewed as one of the most competent and forward-looking in the business. As a result, CEO pay at big corporations has risen faster than that of any other occupational group, and the same is probably true for those who rank just below the CEO. That trend has encouraged a growing fraction of the world’s most ambitious university graduates to enter the race for these jobs.
Frank’s explanation for the growing importance of luck follows directly from the growing competition for top jobs:
Chance events are more likely to be decisive in any competition as the number of contestants increases. That’s because winning a competition with a large number of contestants requires that almost everything go right. And that, in turn, means that even when luck counts for only a small part of overall performance, there’s rarely a winner who wasn’t also lucky.
This observation is Frank’s most important contribution to understanding the role of luck in success. It does not imply that any single form of luck has become more important over time, only that as the number of highly qualified candidates grows, the number who meet all the traditional meritocratic requirements is also likely to grow. That will force selection committees to use other criteria as tie-breakers. Those non-meritocratic criteria will often be untested, sometimes unconscious, and sometimes counterproductive.
Frank hopes that by calling attention to the role of luck in success he will convince readers that top executives’ outsized pay packets, while usually traceable partly to their talent and hard work, are also traceable to lucky breaks for which they deserve no special credit. He hopes this realization will make more Americans support raising taxes on the rich and putting the receipts to more productive uses than buying highly rated bottles of wine.
That might happen, but most Americans already think the rich should pay more taxes. The decline in top Federal income tax rates from 91 percent in 1963 to less than 40 percent since 1987 did not derive from popular reluctance to tax the rich heavily. It probably derived mainly from legislators’ growing reliance on paid advertising rather than unpaid volunteers to get re-elected. Few legislators want to appear hostile to their biggest benefactors. Changing public opinion about lucky rich people won’t change that.
Because Frank cares about poverty as well as wealth, I was surprised that he devoted so few pages to the two forms of luck that many social scientists would say have the biggest impact on people’s economic success, namely the DNA they inherit from each of their parents and the physical and social environments in which they grow up. Children inherit half their genes from each of their biological parents, and economically successful parents can usually do more than unsuccessful parents to develop the skills and character traits that facilitate children’s eventual economic success. As a result, children’s earnings are positively correlated with their parents’ earnings at the same age, not just in the United States, but in every other nation for which we have data (although the strength of the correlation varies, of which more below).
The correlation between parents’ and children’s earnings is one measure (though certainly not the only one) of what I will call “familial luck.” By this I mean the luck that flows from having had one set of parents rather than another. Calling children’s social environment a matter of luck may seem misleading to adults who spend endless hours (and dollars) trying to maximize their children’s future opportunities and see others doing the same. To adults, the fact that parents who prosper tend to have children who prosper is a predictable byproduct of parental effort, despite the many exceptions. For a child, however, the fact that other parents can do more for their children than their own parents can do for them is a major discovery. Of course, no one has ever had a choice of parents, and we wouldn’t be us if we did. But from such a child’s viewpoint, at least, his parents’ inability to do as much as many other parents is bad luck.
Most modern democracies try to blunt the effects of familial luck by at least paying lip service, and often a good deal more than that, to the ideal of equal opportunity. This ideal has many possible interpretations, but in politics and the media it usually means that children’s chances of economic success or failure should not depend on their parents’ income, race, or other characteristics. Skeptics argue—correctly in my view—that completely eliminating the correlation between children’s and parents’ earnings is an unrealistic goal, since it would require both abolishing the nuclear family and standardizing children’s genes. However, if we redefine the goal of equalizing opportunity as simply reducing rather than eliminating the current correlation between parents’ income and their children’s income, the experience of other rich nations suggests that the United States could move considerably further in this direction without weakening the family or manipulating the gene pool.
Figure 1 estimates the average earnings gap in nine rich democracies between thirty-year-old sons whose fathers’ earnings differed by a factor of two (100 percent) at age thirty. In Great Britain fathers with earnings that differ by 100 percent have sons whose earnings differ by an average of 41 percent. In the United States the difference is 39 percent. Among Canadian fathers with earnings that differ by 100 percent, however, sons’ earnings differ by an average of only 14 percent. In most of Scandinavia the gap is 11 to 13 percent. France, Germany, and Sweden lie between these extremes. Figure 1 suggests, though it certainly does not prove, that the governments of nations may be able to exercise considerable control over the importance of familial luck, at least if their citizens have the political will to try.
Figure 1 also tells us something about where different kinds of English-speaking adults might want to live. If they expect to be (or already are) near the top of the economic ladder and want to pass along their advantages to their children, they should live in Great Britain or the United States. If they do not expect to be near the top of the economic ladder themselves but want their children to have a fair chance of moving up, they should live in Canada. (I could not find similar data for Australia or New Zealand.) Infants cannot make such arrangements for themselves, of course, so for them being born in, say, the United States rather than Canada is just a matter of luck.
Frank does not discuss familial luck in this book. His original—and stunning—contribution is his analysis of good luck’s role in reaching the top. Demonstrating that as competition for top jobs increases, the odds increase that luck will play some role in one’s chances of success could be a game changer. If Americans are lucky, this argument would reduce opposition to taxing the rich more heavily. Otherwise, it will continue to be true that, as E.B. White put it in a quip that Frank uses as the epigraph for this book, “Luck is not something you can mention in the presence of self-made men.”
1Richard Wiseman, The Luck Factor (Miramax Books, 1993), p. 10.