EIGHT

The Scarcity of Charm

Je suis soulfre et salpêtre, vous n'êtes que glace.

(I am sulfur and saltpeter, while you are ice.)

—Pierre de Ronsard, 1578

Scribes and Screevers

ONE OF THE SIDELINES of the Brahmin schoolteacher in the Indian village where I did fieldwork as a graduate student in the 1980s was to write letters on behalf of the villagers who could not write for themselves— nearly all of them, as it happened. They were usually bureaucratic letters—to government officials or personnel officers of large companies—but sometimes they were family letters to sons or cousins who had made good in a city somewhere. Sometimes they were letters seeking a marriage for a son or daughter, letters in which the appearance of literacy and prosperity might make all the difference to a young person's fortune and happiness. I never learned how much the schoolteacher charged for his services, but given the dependence of his profitable sideline on a continuing pool of illiterate fellow villagers, I did sometimes wonder about its effect on his motivation as a teacher.

I remembered the schoolteacher many years later when reading in the New York Times in 2007 the story of G. P. Sawant, who is a professional letter writer in Mumbai. Sawant also writes letters to all kinds of recipients: he even does pro bono work on behalf of young women who come to the city from the countryside to work as prostitutes. Many are illiterate, and for obvious reasons most do not want to tell their families exactly what work they are doing. So Sawant has not only to stand in as a writer but also to invent a life for them, a story about their struggles and successes in the city that they can feel comfortable transmitting back to their parents and siblings in their home village. It is a noble and poetic vocation in more ways than one, but his livelihood is now under threat, the Times reports, from the spread of cell phones: “Calling the village or sending a text message has all but supplanted the practice of dictating intimacies to someone else.”¹

It sounds poignant, this story of a modern technology driving out the humanity and charm of the old. Modern literature is full of touching metaphors of ways of life vanishing because of modern technology, or of old technologies turning slowly into folkloric rituals until their original point is forgotten. Jean Giraudoux writes of the rural France of his childhood: “A shepherdess who beat her two clogs against each other: twenty years ago it was an alarm against wolves, now it's against foxes, and in twenty years its only use will be against weasels.”² He clearly means his readers to mourn the richer timbre of life in former times.

But something about this particular story of the disappearance of the letter writers doesn't add up. A century and a half ago, Henry Mayhew devoted several pages of his massive study London Labour and the London Poor to “screevers, or writers of begging-letters and petitions,” whom he described as “a class of whom the public little imagine either the number or turpitude.” “Their histories vary as much as their abilities,” continued Mayhew: “Generally speaking they have been clerks, teachers, shopmen, reduced gentlemen, or the illegitimate sons of members of the aristocracy; while others, after having received a liberal education, have broken away from parental control, and commenced the ‘profession' in early life, and will probably pursue it to their graves.”³ It was not the moral disapproval of respectable folk that made this profession disappear, but it was evidently not the arrival of mobile telephony either. It was universal literacy.

Literacy has made slow progress in India, but there are proportionally many fewer illiterates than there used to be and many more literate people competing to help them. I'd like to think that the Brahmin schoolteacher in my village is now out of work, or at least out of that kind of work. Quite possibly he has now retired, and his children are selling a different skill to their fellow citizens: computer programming. India now has well over two million computer engineers, nearly as many as the United States, and they earn salaries that may be modest by American standards but are beyond what their parents' generation thought possible.⁴ Many more young people in schools and colleges up and down the country, seeing what a computer programmer can earn, are dreaming of the time when they too will have earned their qualifications and can make a good living. But many of their parents also dreamed that a university degree would bring guaranteed prosperity and were disappointed when (in a country that invested heavily in its universities and comparatively little in its primary schools) so many people had degrees that university credentials were no longer worth very much. The current enthusiasm for computing qualifications is likely to bring just as much disappointment in its wake. The fact is that skills acquired through hard work bring a sober and honorable return, but no more than that: the only way to make a fortune is to be skilled in ways that others cannot emulate and that are also in demand. One way to succeed is by acquiring a skill before it becomes fashionable, like the first generation of Indian computer engineers. Another is by acquiring a skill that is kept in artificially limited supply by the institutions that offer the training (like qualifications to practice law). And another is by having a skill that is too difficult for most people to copy, however hard they try.

There is justifiable excitement in India, as well as in the world's other poorest regions, such as sub-Saharan Africa, about the potential of information and communications technology to ease the lives of the very poor. As of late 2011, there were well over half a billion mobile phone subscriptions in Africa and more than three-quarters of a billion apiece in India and China.⁵ There are now around 6 billion mobile subscriptions in the world.⁶ Even allowing for multiple subscriptions, that means a large majority of the world's population now have mobile phones, gadgets that seemed the stuff of science fiction little more than a quarter of a century ago. Many others have the use of a mobile even if they don't have a subscription. Across Africa you can see people seated by the side of the road, selling mobile phone time by the second to those who don't have a phone of their own. And callers in poor countries are not just chatting, agreeable and important though that may be. They are using all kinds of services, like mobile banking and medical consultations, services that may save a livelihood or a life.

A camel driver using a cell phone in the Nubian desert in Egypt, 2010. © Jon Bower / Loop Images / Corbis.

The Internet has not yet reached as many people as mobile phones have. The International Telecommunication Union (ITU) estimates that China had 111 million Internet subscribers in 2009, India only 15 million. There are many more users, of course: the ITU estimates them at more than a billion in the Asia Pacific region, 105 million in Africa, and around 2.4 billion in the world—still a minority of the population.⁷ But it's an astonishing development in just two decades. Even if the Internet demands higher levels of literacy of its users than the mobile phone network, there's no doubt at all that it will change the lives even of the poor. Many people will gain access to a better standard of education than their geographical and economic situation would ever have permitted without it. (Already it seems likely that several million students have learned scientific subjects to a world-class standard thanks to the Khan Academy, whose free classes are available on YouTube.)⁸ Even one literate person in a family can improve the prospects for the whole family by knowing how to find contacts, markets, jobs, or help in an emergency. If mere literacy makes such a difference to the isolation of the poor, think what a fully Internet-connected education can do.

Yet Internet and communication technologies will improve the lives of the world's citizens only because they connect people to each other. Sometimes connections are valuable because they solve coordination problems, bringing together individuals who would otherwise find themselves in unconnected niches, as we saw in chapter 3. If one of the most tragic predicaments you can imagine is a child dying for lack of medical attention, how much worse is it for the child to die while a doctor sits in an empty surgery a few kilometers away because there is no communication between them? But there is a limit to the benefits connections can bring. More exactly, the benefits of matching technologies are limited by the scarcity of attention: not everyone to whom you want to be connected has the time or inclination to be connected to you. In the information economy, attention is the ultimate scarce resource.⁹ Many of the talents that natural selection has instilled in us, or that can be acquired by an expensive education if natural selection has not, are weapons for staking claims to the attention of others, weapons whose exercise means reducing the attention available to our rivals. By the logic of all arms races, the greater the number of people who acquire such talents, the less such talents are worth. More advanced communications technologies will reduce the amount of waste resulting from people's idle attention while increasing the amount of waste resulting from people's congested attention.

The history of writing is a long, epic story of the transformation of the magical into the ordinary. When the first scribes mastered cuneiform or hieroglyphic scripts, the knowledge they acquired was so mysterious to lay-people that it usually brought the writers priestly status as well as material rewards. The spread of education has slowly led to the democratization of writing, and the adoption of alphabetic scripts at different times in various cultures (most recently by fans of text messaging in China) has contributed to the process by making education less difficult and expensive to acquire. As a result, conventions about what you have to do to prove your talent have evolved: merely being literate impresses nobody any more. The rise of written poetry as an art form (as distinct from the kind that was spoken or sung to an audience) was the natural result of writing having spread widely enough for the mere art of putting meaning onto a page to have lost a good part of its original magic.

Charm in the Modern Workplace

We can see the same arms race at work in more modern skills as well. Little more than a century ago, the manufacture of moving pictures must have seemed as magical to its first, privileged spectators as the art of making writing had done five millennia earlier. When the Lumière Brothers held the first public screening of motion pictures at the Grand Café in Paris in 1895, the flickering images on the wall of the café caused a sensation and resulted in an invitation to tour in London, Bombay, New York, and Buenos Aires the following year. Now, far more technically impressive creations languish on YouTube, watched by a few indulgent friends of the creator at most. To the extent that technical sophistication has given us access to more satisfying creations, it represents progress for the benefit of all of us. But if it's easier to make technically accomplished movies today than it has ever been, it is harder than ever before to make a living doing so. Those who succeed have something extra, something that marks them out from their rivals who have the same technical skill. If you want a word for it, that word is charm. Charm sounds like a nice, comforting thing, but it's not: almost by definition, not everyone has it, or not when they need it, and for everyone whose charm is working, someone else is left seething and frustrated.

Charm plays a more important role in some professions than in others. One way of seeing it in action is to note how the earnings of people who do the same job can vary substantially according to how effectively they attract the attention of others: colleagues, customers, collaborators. Almost nobody ever does a job that is wholly independent of the contributions of others, and almost everyone depends on others in a fundamental and far-reaching way. This means that to do your job well, you need to be able to persuade others to work with you; and however conscientiously you do your job in other respects, if you can't persuade others to form a team with you, your rewards will be meager and your sense of fulfillment low.

We can see this in the variations over time of salaries in a number of professions where charm is important. Inequality in earnings has been increasing in recent years in the United States and in many other industrialized countries. You might have thought that was just because certain kinds of already well-paid job were being even better paid, relative to the others, than they were before: bankers' pay rose relative to that of bellhops, for instance. But for some occupations, inequality has been rising even among people doing the same job. And there are good reasons to think this is because the rewards of charm—that essential ingredient that not everyone has enough of—have been rising over time in those occupations as well.

Evidence from the US Bureau of Labor Statistics points in this direction. The BLS publishes salary-inequality figures by some very precisely defined occupational categories (more than eight hundred of them, in fact). We need evidence like this. It wouldn't tell us anything if we discovered that salaries for “car workers,” for instance, were becoming more unequal, because that category covers too many different types of skill, from the people who make tea in the General Motors canteen to people who design software in its research division. Widening disparities in earnings within that broad group could have been happening just because technology has been improving faster in computer-aided design than in tea making. So what do we learn from these finely graded occupational earnings statistics? In some occupations, there's a much bigger difference between the earnings of the best-paid and worst-paid practitioners than there is in others. For instance, the BLS reports a number of occupations that in 2008 had the same median earnings but very different distributions. Among signal and track switch repairers, those whose earnings were at the tenth percentile of the distribution took home a little more than $16.50 an hour, while those at the ninetieth percentile took home a little more than $30. Among film and video editors, by contrast, those at the tenth percentile took home just under $12 an hour, while those at the ninetieth percentile took home $54—around four and a half times as much.¹⁰ The point isn't that video editors are better paid than signal and track switch repairers—they're not, because at the median they're paid exactly the same (that's the whole point of this comparison). The point is that getting into the top 10 percent gives you a much bigger salary payoff in one occupation than in the other.

There are many possible explanations for this difference, but the role of charm seems one of the most likely. Any job requires a combination of method and flair, of things you can learn by repetition and things you can't, just as playing any musical instrument requires a mixture of perfected technique and a sense of musical interpretation. It seems likely that repairing signals and track switches needs more method and less flair than video editing does. (That's not to say it needs no flair at all: many stubborn repair problems need imagination as well as dogged persistence.) But the benefits to a flair for signals repair are probably occasional and serendipitous, whereas in video editing they're often the whole point. Every video made today has to compete for attention with a gazillion others. The editor who can make the whole creation sing may have just that edge that makes the difference. And a talented editor is not just someone who can make a random set of video material into a better and more attractive whole. It's someone who can attract, and be sought out by, talented scriptwriters, actors, and directors, people whose collaboration raise the odds of success in the first place. The most talented signal repairer is probably not going to be attracting collaborators to anything like the same degree, and the quality of the collaborators is going to matter a lot less anyway.

In any occupation, it's possible that many factors other than the return on charm may influence the distribution of earnings and may affect comparisons between occupations. So it's worth comparing how the distribution of earnings changed in various occupations between 2000 and 2009. (It's difficult to measure spreads by occupation over longer periods than that, because the Bureau of Labor Statistics regularly changes its definitions of occupations to keep up with the changing nature of the modern workforce.) Let's start by looking at two occupations where it's unlikely that charm plays much of a role in the determination of earnings: parking-lot attendants and fast-food chefs. Inequality is measured as the ratio of earnings at the ninetieth percentile to earnings at the tenth percentile. The spread in these two occupations is indeed as low as it gets, with individuals at the ninetieth percentile earning respectively 1.8 and 1.5 times as much as the individuals at the tenth percentile.¹¹ Nor did it change to any significant degree in the first decade of this century.

We can compare these to two other occupations where it seems unlikely that charm plays much more of a role than it previously did: accountants and paralegals. These are higher-paying occupations and also have somewhat higher earnings spread between the tenth and ninetieth percentiles: 2.8 times for accountants and 2.5 times for paralegals. That's probably due to the fact that more of these individuals work in teams, and individuals with charm are likely to end up working in higher-paid teams for higher-paying clients. The spreads have edged up very slightly over the decade (in 2000 they were 2.6 for accountants and 2.4 for paralegals), but the change is still fairly small.

The really interesting changes come, though, in various occupations in the motion-picture industry. It would be tempting to think that this is an industry in which earnings generally should have become more unequal, because of the way in which the industry has become polarized between very large studios and many small independent filmmakers, and the way in which the growing internationalization of movie distribution makes a few superstars able to command truly world-beating earnings. But this growing inequality characterizes only some of the occupations within the industry. Film editors, actors, and make-up artists have indeed seen substantial rises in earnings spreads over the first decade of the twenty-first century (spreads for editors rose from 3.8 to 4.5, for make-up artists from 4.2 to 5.7, and for actors, from an already very high of 7.4, to 9.3 in just nine years). But a number of other occupations in the same industry saw spreads fall—notably broadcast technicians (from 4.6 to 3.7), camera operators (from 4.5 to 4.0), and projectionists (from 2.7 to 2.1).

What was going on? In these latter, relatively technical occupations, it seems a reasonable guess that technology was slowly taking the charm out of the work. The same sophisticated and user-friendly technology that enables the likes of you and me to make a high-quality video clip and post it onto YouTube also makes it much harder for a really skilled technician to be creative enough to stand out from the crowd. In short, these skills seem to be offering less and less opportunity for the exercise of creativity as more and more people acquire them.

Statistics like these offer more opportunities for speculation, it has to be said, than for drawing hard conclusions. For instance, it's only to be expected that the spread between the tenth and the ninetieth percentiles in the earnings of fashion models has been rising (from 2.8 to 3.3), and a little surprising that the same measure for writers' earnings has been falling slightly (from 4.0 to 3.8)—until you reflect that the writers who earn really big sums are just a tiny fraction of the total, so the ninetieth percentile is hardly going to capture the earnings of those who write bestsellers. Writing is also a famously solitary occupation, at least compared to most of the occupations in a modern economy. The rising spread in the earnings of bailiffs (3.2 to 3.6) is a poignant commentary on how creative that profession had become by the end of the decade.

Finally, we can consider two statistics for occupations in the computer industry suggesting that the same kind of dual phenomenon may have been occurring there as in the movie industry. The spread in earnings of computer programmers rose slightly (from 2.7 to 2.8), while that of technicians working in computer support roles fell (from 3.0 to 2.7). While there is a lot of overlap between the two occupations, there are some broad differences too. Programmers have to think up creative solutions to novel problems, often in teams. Computer support technicians tend to work more on maintenance and debugging, tend regularly to revisit problems they have seen before, and tend to work on their own or in large call centers that allow them little individual autonomy. Like broadcast technicians, for whom the increasing sophistication of the technology takes some of the challenge out of the work, computer support technicians appear to be finding that the technology, while certainly not making them redundant, is increasing the proportion of routine in their work. And perhaps the increasing proportion of routine may be a sign that future innovations could threaten their livelihood even if it has not done so already. Certainly there is a correlation—small but significant—between the change in earnings spreads and both the change in employment and the change in median earnings over that decade.¹² Higher inequality, both across occupations and over time, seems to have been the price Americans paid for being in occupations with above-average prospects for employment and earnings growth.

Sexual Selection Again

It's important not to build too much on these numbers: they're suggestive, not conclusive. If the argument that computer programmers are not the answer to India's poverty problem (because their current high returns are an artifact of their relative scarcity, which will decline over time) doesn't persuade you, these statistics about earnings inequality won't—and shouldn't— make the difference. All kinds of things may be affecting earnings spreads, and the changing role of charm in these occupations is only one of them. But they may serve as a reminder that technology is not making purely routine the task of building the teams that manage production in a modern economy.

Today, producing goods and services is almost always a matter of teamwork. Sometimes the different members of the team can operate entirely at arm's length, doing their part of the job and passing it on to the next person through a market transaction. Even making so relatively simple an object as a shirt can require coordination of a team whose members are located on several continents, but a large part of that coordination can be brought about by the market. More commonly, though, teams of people work together to manage part of the production process, and before it's possible even to decide how to motivate and manage that team, its different members have to agree to work together.

You might have thought that putting together a production team is largely a matter of searching for the right kinds of people to take part in it. You might also expect that this process would be made vastly easier by the technologies so impressively spawned by the information economy, including, of course, Internet searching itself. But in addition to finding the right kinds of people, putting together teams is also a matter of persuading those people that they should cooperate. And the same search technologies that make the first task easier make the second task more difficult. The more easily Prince Charming can find his Cinderella, the more likely it is that instead of blushing winningly at him, she will reply: “I'll see if I can fit you into my schedule; maybe we should try and have lunch sometime.” The very technologies that help us to find our ideal partners, whether in our professional or our romantic lives, crowd the attention of those partners and raise the stakes in the contest to find ways to stand out from the crowd, requiring ever more elaborate and unpredictable ways of staking a claim to their attention. No piece of software, no algorithm, no formal training that anyone can copy will ever guarantee that those we want to work with will want to work with us.

This predicament—that more sophisticated matching technologies raise rather than lower the need for charm—may help to explain a puzzling feature of the networks discussed in chapter 7. Suppose it's indeed true that talented women pay an excessively high price for career choices such as taking career breaks and working less than the mind-numbingly long hours apparently standard for men in some kinds of job. If these choices really mean that such women receive lower rewards than similarly talented men, why aren't smart entrepreneurs sniffing out a profit opportunity by linking up with such women? I suggested the answer might be that entrepreneurs might know such women exist but have difficulty finding them, because women who take career breaks also drop out of the networks that entrepreneurs use to locate potential opportunities. Surely, you might think, this is where modern information technology can make a difference. Yet it's precisely because modern technology is so good at matching up people that it can make our selection problem more difficult rather than easier: on the Internet, everyone is Prince Charming, so only Prince Extra Charming will do. And the more efficiently the software works, the more likely is Cinderella to think that accepting Prince Extra Charming may be a less attractive option than waiting for Prince Utterly Charming to come along. And that's just the problem in the dating market. In the modern version of the story, once the matching problem is solved, Cinderella doesn't live happily ever after but takes a career break. Once she tries to return to work, hoping for something a little more elevated than the hearth sweeping with which she started her career, she learns, to her dismay, that everyone has glass slippers nowadays: they're no big deal.

Networks function because they help us navigate in a world crowded with information. The more technology enables us to rationalize and organize our networks, the more information we have to process, and the more, paradoxically, we may rely on our instinctive and intuitive networks to help us feel comfortable about other people we interact with. An employer looking for committed workers who has only three applicants to choose among may be willing to spend quite a lot of time interviewing them and finding out about their individual qualities. An employer who posts an online job ad and receives three thousand applications may resort to all kinds of unscientific rules of thumb to reduce the applicant pool to a manageable size. Employers scared of the consequences of receiving three thousand job applications may resort to various informal methods to avoid even getting to that point, such as asking the people they know to recommend candidates, and so on.

True, the law nowadays uses sophisticated methods to ensure that once women reach the official short list for a position, they do not suffer discrimination. But these methods may be powerless to ensure that women reach the short list that really matters, namely the unofficial one inside the head of the decision maker in today's informationally overloaded world. If women aspire to fully equal participation in the modern economy, they will have to find ways to level up their access to the most fundamental scarce resource of all in the twenty-first-century world—the bottleneck inside other people's brains.

This bottleneck affects men no less than women. Indeed, some have claimed that it affects men far more. Is that true, and does it mean that questions about male power are no longer relevant?

Is There a Crisis of Men?

As I noted in the introduction to part two, it has become fashionable in recent years to talk about a “crisis of men.” Discussions of this phenomenon conflate at least two observations. The first is that society's most marginalized groups—the homeless and the prison population in particular—contain far more men than women. This is certainly true and has always been true. It may be a more important phenomenon than any discussed in this book, but it is not evidence for the “myth of male power,” to use the title of one best-selling book published a few years ago.¹³ The fact that many men are powerless does not preclude the possibility that some other men may be very powerful indeed.

A second trend that is sometimes used as evidence for a “crisis of men” is that women's college enrollment rates in the United States have substantially overtaken those of men, and that as a result women in their twenties and thirties have higher education levels than those of men. This development is certainly not confined to the United States: it's true of many other industrialized countries (with the important exceptions of Germany, Japan, and Korea).¹⁴ But it's not the result of any change in the educational aspirations of men. According to the US Census Bureau, the proportion of US men in 2010 with at least a bachelor's degree is 30.7 percent for men aged between 65 and 69; for men aged 30 to 34 years it's 29.9 percent (a negligible difference, considering the fact that some people graduate after the age of 35). The trend for roughly three men out of ten to graduate from college has remained essentially unchanged for thirty-five years. What has changed dramatically is the level of women's aspirations. Only 23.3 percent of women aged 65 to 69 have a bachelor's degree or higher credential, while for women aged between 30 and 34, that proportion is 38.2 percent.¹⁵ In terms of educational achievement, women have pulled out as far ahead of men as they were formerly behind. That's a remarkable achievement, but in what sense is it a crisis of men?

A safer assessment of this trend is that it will be bad news for some men and good news for others, and will require a potentially stressful readjustment of priorities for everyone in the meantime. It's not unlike other economic relationships, in fact. When China enjoys faster economic growth than the United States, that's mostly good news for the United States, because it allows US workers to sell more goods to China and US consumers to buy more Chinese-produced goods. Although in some respects China is a rival to the United States (for example, in exporting to third countries), its role as an economic collaborator is far more important. It makes no sense to see economic relations as sporting contests, in which one country's success must mean another country's failure. But in the meantime, some people (those who made money dealing with China when it was poor) will have to raise their game, and the necessary adjustments may create stress on both sides.

Education, likewise, is not a sporting contest, in which a larger number of more highly educated women must portend the failure of men. As professional colleagues, as contributors to public life, and as personal friends, more-educated women are good news in principle for everyone. To their potential life partners, more-educated women are certainly good news too—so long as the women themselves don't have unrealistic expectations about those partners. The average American woman is a lot more educated than her mother, but if she thinks that entitles her to choose as her life partner a man who is a lot more educated than her father, she's riding (on average) for a big disappointment.

That disappointment is real, and it is in the marriage and dating stakes that it now seems to be felt most harshly. Marriage rates (among heterosexual couples) have been falling for many years, with fewer than 60 percent of US white adults married in 2010, compared to more than 70 percent in 1970. No doubt this drop partly reflects changing social conventions about the appropriateness of marriage for committed couples, but it probably also reflects a reevaluation of the extent of commitment many individuals feel to their current partner, if indeed they have one. Greater selectivity by women, many of whom feel their educational achievements entitle them to more educated partners than they can currently find, is likely to be part of the explanation. Greater selectivity by the minority of highly educated men, who now have more women to choose among, is probably involved as well. It's unlikely to be a coincidence that marriage rates have fallen fastest among African Americans, the group in which women's educational achievements have pulled furthest ahead of men's in recent years.¹⁶ However, although there is no shortage of corroborating anecdotes, reliable systematic evidence for these conjectures is not easy to find. The economists Betsey Stevenson and Justin Wolfers have claimed that women's happiness has declined in the United States over the last thirty-five years: if true, this decline may indeed reflect disappointment in relationships induced by such educational disparities.¹⁷ Stevenson and Wolfers find that the happiness of less-educated men fell over the period while that of more-educated men rose. But they also find that more-educated women suffered smaller happiness declines than less-educated women, the opposite of what would have been expected if relationship disappointments driven by educational disparities were the main factor behind these trends.¹⁸

If these speculations are accurate (and they remain speculations, given the state of current evidence), they certainly suggest a crisis in relationships for less-educated men—many of whom have lost their charm, in a word, for their female contemporaries. But if it's a crisis, it's one that affects women as well, at least as long as the expectations of women fail to adjust to the reality of the new educational imbalance. Education is good for improving your access to many valuable things in life, but access to a fixed stock of potential life partners is something it's hard for everyone to have more of at once.