In 1992, the 1,059 partners in ten of New York’s largest law firms averaged $957,000 as their shares of their firms’ profits. Some physicians affiliated with the country’s leading medical schools received incomes exceeding $1 million. Executives did even better. Among fifty companies recently examined by Fortune magazine, each compensated its leading executive in excess of $5 million, rising to $11 million at Philip Morris and $15 million at General Electric.1 Corporate salaries are commonly augmented by stock awards. The current chairman of Coca-Cola, for example, has already amassed a portfolio of more than $300 million.

Derek Bok, who recently retired after twenty-two years as president of Harvard, is concerned over “the acquisitive, self-centered goals reflected in our compensation practices.” He uses phrases like “undeserved income” and “unwarranted wealth” to characterize the stipends of all too many executives and professionals. While The Cost of Talent is a highly opinionated book, it also contains much valuable information, historical as well as current, about what can be earned in law and medicine, business, and public service, as well as in teaching in schools and colleges.

Whether we look at surgeons or casual laborers, the same question arises: What is an appropriate wage for the services they provide? Needless to say, we all have opinions about our own and others’ incomes. Most usually, these often tend to be reactive ones, stirred on hearing of payments we feel are too high or too low. But after that, it is by no means easy to set up universal standards for determining what a just wage would be. Dozens of factors intervene, from competitive considerations to notions of need and worth and equity. These ambiguities are illustrated throughout Bok’s book. For example, he uses the term “earnings” freely, even when describing payments he deplores. Yet when we speak of money as having been earned we often mean that the amount is merited. Even calling a figure “merited” can be equivocal, since it can express moral approbation or refer to a market valuation. It soon emerges that Bok tends to think incomes should be set by the market. So when he refers to some salaries as “undeserved,” he is commenting as an analyst, suggesting that society could get the same kind or quality of work for less than it is currently being billed for.


As I have noted, doctors can make more than $1 million. But such incomes are not earned in an open medical market, since few patients can pay the high current charges or fees from their own pockets. Bok shows how the rest of us have been subsidizing some physicians, owing to “the willingness of the government and private insurance plans to approve their fees and reimburse them accordingly.” Until recently, doctors have got virtually all they demanded from the repayment system, not least because no price was deemed too high for professionals who “save lives.” Yet this deference is waning, in part because fewer people have a close association with a family doctor. Moreover, if medical incomes do decline, this will be less the result of legislation than of decisions of employers, who are no longer willing to pay a large proportion of the bills. We are likely to see the emergence of a second tier of medical doctors, who work in managed-care clinics or store-front offices. Even so, some specialists have become adept at marketing themselves as having their own mystique, so they will continue to be very highly paid.

Bok has a harder time with well-paid lawyers. Last year, Sullivan & Cromwell billed its clients an average of $750,000 each for work done by the 463 partners and associates on its staff. No doubt equally expert legal service could be obtained elsewhere at lower cost. But companies seem willing to stay with the elite firms since trading down might cast doubts on their own solvency or status. Whether agreeing to pay more than you have to makes sense is an old puzzle. (Think of what we have paid for dinners or dresses only to impress someone who was not impressed.) In a similar vein, top firms give $90,000 to students fresh from law school, even though they could get most of them or people just as good for not much over half that amount. (Each year, some 36,000 men and women receive law degrees; among them there can hardly be a shortage of promising candidates.) But such symbolic sums are seen as needed, if clients are to be persuaded that the firm hires only “the best.”

The Cost of Talent never clarifies whether pocketing what clients freely pay can still be defined as “undeserved” income. The partners of a law firm are owners of a business, in an industry with no shortage of competitors. Since they can claim that clients consent to their fees, they see nothing odd about charging what the market will bear. As with dress designers and other entrepreneurs, profits can come from trading on one’s name as well as providing goods or services. Writers and publishers can prosper in much the same way, thanks to people’s willingness to pay $22.95 for a physical product that—once sales pass a certain point—may cost $1.20 to run off the press.


With executive compensation, even more ambiguities arise. Bok leads off by asserting that, in far too many cases, “pay for top executives has turned out to be a sham and an embarrassment.” He begins by invoking the model of a “perfectly competitive world,” in which “all chief executives would receive amounts approximating what they added to the net profits of their company.” Yet corporate boards often continue to reward managers of companies with mediocre records. Still, a second step must be taken after agreeing that chief executives’ paychecks should be related to performance. To cite one example, General Electric’s profits in 1992 came to $4.6 billion, a notable advance over the previous year’s figure of $2.6 billion. But a decision must then be made whether the chairman’s reward should be $200,000 or $2 million or $20 million. Nor is the last amount fantastical. A Year earlier, the Walt Disney directors gave Michael Eisner $37 million, in gratitude not only for spectacular profits, but—at least as important—for adding $5 billion to the value of the company’s stock. Even so caustic a critic of executive compensation as the economist Graef Crystal considered the figure Eisner received to be earned and appropriate.2 It would be interesting to know if Bok feels there could ever be circumstances that justify eight-digit compensations.

A few pages later, however, Bok suggests a rather different measure of executive worth. Here it is the salary it would take to keep a man or woman from leaving, assuming they are people the company wants to keep. Of course, the only real way to know is to wait until that person receives an outside offer, and then estimate by how much it must be topped. More usually, firms try to keep their key people contented by paying them well. But even here, Bok believes, much more money than is necessary is being passed around, including payments to executives whose market value has never been verified. He also reports that “more than 70 percent of all CEOs are hired from within and that most are not many years from retirement.” Bok seems to be suggesting the possibility that they might even take on the top job without much if any rise in pay. Since the United States cannot grant knighthoods, perhaps honorary degrees could be set aside for executives who would agree to more modest checks and more spartan perquisites.


If we overpay some professions, we are underpaying others. Bok is not alone in worrying that some critical occupations are no longer attracting applicants, or not enough of suitable quality. On the whole, it seems true that if people are given the opportunity to choose, they will select better-paying careers. We know that talented women once settled for nursing and teaching because they had few other choices. Today many of them avoid those occupations in favor of better paying professions. Young people also have higher expectations about the way they want to live. The salaries currently offered to parole officers and social workers cannot provide for the diversions which members of the current generation take for granted. Many will agree with Bok’s exhortation to increase the pay in such fields, but the questionable premises of his argument emerge when he writes:

Compensation has an important effect in determining which sectors of society claim the largest shares from the finite pool of intellectual talent.

In a chapter entitled “Summing Up,” he posits that every society contains only “a limited supply of highly talented, educated people,” and we will set our house in order only if we “begin thinking of talent as a finite resource.”

No doubt wage levels are central to the decisions people make about careers. But it is another matter to presume that the population of this or any other nation contains only a “finite” supply of capable people. So firmly does Bok hold this position that at no point in The Cost of Talent does he allow that the country’s pool of competent people might be enlarged.

There is, of course, another view. Throughout history, most human talents have remained dormant, because the systems under which people have lived lacked the capacity or interest to make use of their fullest potential. The chief reason for the great migrations to America has been the opportunities this country has offered for realizing the abilities people sensed that they had but could not develop in more constricted settings. The exigencies of wartime have also brought out all sorts of skills—including capacities for leadership—that people were never aware that they had. Or we may simply look at countries such as Germany which ask and expect more of factory workers, thus upgrading the aggregate intelligence of their labor force. But Bok seems unimpressed with these and similar lessons. Since he starts from a premise of a circumscribed stock of talent, he sees the problem as one of how to allocate very scarce skills. Perhaps such a position is not surprising in someone who presided over a college which takes pride in rejecting a very large majority of its applicants.


Equally arguable is Bok’s use of “intellectual” to describe the skills and aptitudes society most needs. He tells us that by current measures, the typical law student is at least “the intellectual equal” of doctoral candidates in the humanities and social sciences. This seems an arguable statement, since many graduate students show an affinity for ideas, something one encounters less frequently in law schools. However, we soon learn that Bok considers that grades on college transcripts and scores on multiple-choice tests are a reliable measure of “intellectual talent.” Good grades certainly attest to a willingness to do assignments to a teacher’s liking. And it may also be granted that a certain kind of agility is needed for supplying speedy answers to questions like the one reproduced on this page. Yet it often emerges that temperaments that flounder on such tests still have a strong capacity for scholarship in fields like philosophy, literature, and history.

Carrying his presumptions even further, Bok goes on to posit that high grades and scores attest to generic mental qualities, which can be deployed across a range of callings. The problems of primary and high school education, he believes, derive from the low qualifications of the teachers, many of whom ranked in the bottom of their college class. If we raise teachers’ pay, he believes, better students who are now choosing professions like law would redirect their talents toward inspiring third graders. Yet in a later chapter he comes much closer to the truth, when he allows that an intuitive flair for explaining matters to youngsters is what makes for effective teaching. Nor are those qualities revealed by test scores. Perhaps some law students could transfer their forensic talents to a classroom, but many would do better to keep to their torts and contracts.

Bok cites sources to support his assertion that high “grades and test scores appear to have some significant effect on success in one’s career.” His chief reference is a 1975 study, which found that college graduates who got mainly A’s ended up with salaries that were 18 percent higher than their classmates who received C’s. But an 18 percent edge hardly seems enough to differentiate the talented few from the mediocre many. Moreover, this study also contains the finding, not mentioned by Bok, that “non-academic attributes are as important as academic abilities in determining job performance.” Another paper that he cites concentrates less on students’ class ranking than on the costs and characteristics of the institutions they attended. And it concludes by saying that graduating from a state university with a major in engineering can turn out to be a better investment than an Ivy League degree.3


How then to define a “just wage”? Some people do not find this question a problem, since they feel they are already getting just that. Cardiovascular surgeons, who averaged $575,000 last year, doubtless believe that their fees are justified, and can offer explanations they find satisfying. Much the same certainty probably pervades those law firms whose billings confer on their partners upward of $1 million per year. But people with high incomes may not be the best persons to ask. (Indeed, how many people will actually admit they are overpaid?) Despite Bok’s allusions to “undeserved income,” in only a few egregious cases does he identify specific people that he believes are overpaid. Thus he cites the windfalls that come to lawyers who win huge settlements, and the $78 million that the late Steven Ross received from Time-Warner, plus the notorious $550 million that Drexel Burnham bestowed on Michael Milken.

As has been noted, Bok alters his criteria, sometimes on a single page, and as a result The Cost of Talent ends up without a coherent analysis. Thus while he repeatedly calls for making salaries “truly competitive,” when it comes to the academic world, he defends a different principle of “collegiality.” This makes it easier to justify the annual raises that tenured faculty members receive, even when their best work is behind them and they are no longer being sought by other institutions. Because of this solicitude for those who already have secure positions, the salaries that Harvard and other universities pay to their full professors can absorb three quarters of the total payroll. This leaves little money for bringing in younger people.

On other issues Bok raises important questions but does not acknowledge their implications. In addition to attacking high salaries, he seems to support a ceiling on the overall assets any person can amass. At least that appears to be his aim when he writes that “unwarranted wealth perpetuates injustice and frays the essential bonds that strengthen trust and hold societies together.” In America, contrasts of poverty and privilege are more marked than in any of the countries with which we compare ourselves. And we are paying for these disparities in violence and fear and social malaise. But it may be too reflexive to presume that the presence of rich people “frays” the social bond. Many ordinary citizens may resent the rich while reaching avidly for reports on their activities. And if the poor have to blame someone for their condition, it is not clear that their chief target is the extremely rich. Indeed, even Bok admits that “the munificent salaries of CEOs reflect the high regard Americans have for initiative and success.” This comes close to granting that all those huge checks have at least some popular consent.

Bok does not propose policies that might cut back on levels of wealth he deems to be “unwarranted,” nor does he define just what is the difference between wealth that is warranted and that which is not. James Madison, writing in the Federalist, noted that in all societies there will be some individuals who possess “faculties of acquiring property.” In most cases, they achieve their wealth by providing products or services that other people want. Twenty years ago, Robert Nozick offered Wilt Chamberlain as an example of someone with just such a faculty. Basketball fans were eager to see Chamberlain play and willingly paid to enjoy that experience. In so doing, they created a millionaire; nor is it easy to see how that eventuality could—or should—have been prevented.4 Moreover, it is not easy for government to appropriate accumulated wealth. According to Forbes magazine’s most recent compendium, America’s four hundred richest individuals or families have amassed fortunes of at least $300 million. (The list changes every year; but people lose their place because of bad luck or ineptitude, not governmental incursions.) In 1991, the most recent year for figures, the Internal Revenue Service received 52,019 returns declaring annual incomes in excess of $1 million. In 1978, only 8,964 returns declared incomes of $1 million in today’s dollars, so households at that level have increased sixfold in not much over a decade. At this bracket, salaries account for less than a quarter of all income, with the rest coming mainly from dividends, interest, and capital distribution and gains. Between 1978 and 1991, again using constant dollars, the median family income for the nations as a whole went from $35,594 to $36,841, an increment of barely 3.5 percent.5 It may be that this stagnation is at the root of Bok’s problem.


Since, in our kind of economy, people cannot be stopped from making money, Bok would make it harder for them to keep it. Hence his proposal for “a more steeply progressive income tax.” It is true that public revenues can be used to redistribute income. The most effective example has been raising Social Security pensions, which are now at levels well beyond what individuals put in as contributions. These subventions have had a marked effect in lowering poverty rates among the elderly. Indeed so much of this money now goes to middle-class retirees that increasing portions of the benefits are being subjected to tax. However, there is less inclination to transfer public funds to families headed by single parents. Stipends under the Aid to Families with Dependent Children program average only $4,680 per household, a sum which would have to be quadrupled to provide a decent living standard. Payments and food stamps for the poor now total about $56 million. On the other hand, medical expenditures on their behalf amount to some $77 billion. In other words, doctors and hospitals and drug providers get more public money for ministering to the poor than their patients receive as family budgets. (Nor can we be so sure that these patients, who get much of their treatment in emergency rooms, are that much healthier.)

Bok also finds it “grossly unfair” that executives receive “fifty, one hundred, even two hundred times the average pay of working people who toil away at much less interesting jobs. This raises the question of whether premiums should be accorded for work that is tedious or unpleasant. In the real world, this hardly ever happens. Coal miners secured higher pay not because what they did was debilitating and dangerous, but because they formed a strong union. In an open labor market, which Bok sometimes supports, it is left to employers to decide what they are willing to pay. They often offer even less for distasteful tasks, since they draw on pools of desperate people. If such wages are “unfair,” then we need some standards by which to assess what is an equitable wage for every occupation. As it happens, we have a minor industry of compensations experts, who advise employers on rates and scales. In a recent book, Steven Rhoads shows how by using “a predetermined system of factors and factor weights…each job is assigned a total number of points.” Thus an evaluation of positions on the Illinois payroll assigned a weight of 889 points for what accountants did, compared with 578 points for the state’s electricians.6 When weights are converted into dollars, accountants might get $41,783, while electricians would receive $27,116. Such systems try to quantify elements like the job’s difficulty, length of training, and responsibility; but thus far they have not given a weight to unpleasantness. Nor do they address what may happen if state agencies cannot attract competent electricians at $27,116.

It is true, as Bok observes, that “American corporations seem to employ a remarkable number of executives and managers compared with other advanced countries.” Germany and Sweden, for example, apparently produce more efficiently with far fewer people behind desks. Perhaps a possible way to reduce the income imbalance would be to reduce the number of administrative positions. While the United States has a higher managerial ratio, much of it results from a generous method of accounting. According to surveys conducted by the Bureau of Labor Statistics, the 1992 work force had more than ten million men and women in “executive, administrative, or managerial” positions. But a group of this size must reach fairly far down the supervisory ladder, since their median salary was a relatively modest $33,800, hardly what we would regard as corner-office pay.7 If Bok wishes to reduce their numbers, he has allies in the many companies which are currently slashing their middle-management staffs, bringing to a premature end many white-collar careers. One effect of this policy has been to convey to the coming generation that there will be fewer places for them of the sort their parents once occupied.

Trying another tack, Bok argues that the use of money to make people work harder remains unproven. In the business world and elsewhere, he proposes, people might continue to do their best work without the lure of higher pay. “There is no reason to suppose,” he writes, “that American executives would work less hard if they were paid several thousand dollars a year instead of several million.” I would like to think he is right. Many people who enter corporate careers come from fairly modest origins, and it is unlikely that they start out dreaming of big money. But their aspirations evidently change as they start moving up in management. Even if executives are too busy to live lavishly, they are still charmed by all those numbers on their checks. Nor is it so different in other fields. Why else, one may wonder, does Harvard make sure that its faculty salaries are the highest in the country? Perhaps its professors, like executives, take pleasure in knowing that they have reached the top, in material rewards as well as other measures. And, as the table on this page shows, it would seem that college presidents also need monetary inducements to take and do their jobs.



How much people should be paid is ultimately a moral issue. It must address not simply what each person may deserve, but the kind of society in which we want to live. Questions of incentives and allocation inevitably arise, as do judgments about competence and competition. According to a familiar egalitarian position, there can be no justification for anyone’s receiving more than a certain amount, regardless of skills, shortages, responsibilities, or the public’s willingness to pay. In this vein, Franklin Delano Roosevelt proposed a $25,000 ceiling during the Second World War, as a symbol of common sacrifice.8 That translates to $211,000 in today’s dollars, which is slightly over what we pay our current president. While Bok does not talk expressly about redistributing income, his call for much higher taxes on the highly paid would have that effect.

Higher taxes on income and inheritance might lower the living standards of at least some of those at the top. This happens in countries like Denmark, New Zealand, and the Netherlands, whose tax rates keep even chief executives in the middle class. But, as has been noted, efforts within the United States to improve the lot of poorer people have not shown much success. Nor is the question entirely one of economic extremes. As it happens, the wages currently received by middle Americans are not very auspicious. According to the most recent Census Bureau study, the 57.7 million men aged 25 to 64 who worked during 1992 had a median income of $22,467, while for the 49.3 million women, the median came to $16,227.9 It would take a substantial infusion of cash to raise these figures more than marginally. In families with a single wage earner the median income came to $26,292, whereas those with two or more workers achieved $45,779. On the whole, households with multiple earnings seem able to fend for themselves, although it should be noted that almost 30 percent of this group took in less than $35,000. If our objective is the creation of more better-paid jobs, these figures indicate how far we have to go.

How far might lowering the roof raise the floor? Of the 106 million federal tax returns received in 1992, some 763,000 declared incomes of over $200,000.10 Let us suppose—fancifully—that each of these households would be allowed to keep their first $200,000, and everything over that amount would be taxed away. At first reading, this would give the government a fund of $187 billion. But since this group already pays taxes at the top bracket, even taking everything over $200,000 would yield only $103 billion in new money. Were this sum shared equally among the rest of the nation’s households, each would receive a check in the amount of $1,081. Were the recipients to be confined to families having incomes of less than $35,000, each would get $2,173. In any event, save as a textbook case, no one seems prepared to propose an after-tax income ceiling of $200,000.

So other ways will have to be found to achieve greater equality in incomes and earnings. In the end, there is no substitute for increasing productivity, which would in turn increase economic growth. Of key importance is improving labor skills at every level, which will very largely depend on raising the quality of education from kindergarten through graduate school. Improving the schools and providing higher salaries for school teachers could be as important to a higher rate of productivity as anything else. While some of the rich get richer in periods of prosperity, so do a lot of other people. We might recall that during the 1960s, the bottom three fifths of the population took home a greater share of the nation’s income than they do today.

This Issue

March 3, 1994