In December 1983, Federal Judge Jack Tanner accorded the fullest legal recognition it has so far received to the novel economic doctrine of “comparable worth.” This doctrine holds that women in the work force are paid less than they are really “worth” in terms of the “value” of what they do. The recognition came in Judge Tanner’s decision, in American Federation of State, County, and Municipal Employees [AFSCME] v. Washington State, that Washington owes 15,500 female state employees over four years’ back pay and raises. Estimates of the cost of this judgment vary from $500 million to $1 billion. Whatever the eventual tally, Washington taxpayers will have to come up with it soon, as Judge Tanner has denied various motions by the state for time to accommodate its new burden. The case is currently under appeal to the Supreme Court, with the Reagan administration reluctant to side with Washington “against women” in an election year.

The history of AFSCME v. Washington reflects the development of the comparable-worth doctrine over the last decade, and illustrates as well the tactics on which comparable-worth advocates have come increasingly to rely. The idea itself has lately become the main item on the feminist agenda and has, moreover, been enshrined in the 1984 platform of the Democratic party. It is generally introduced with a statistic and a few examples intended to shock. One learns, first, that the average full-time working woman makes about sixty cents for every dollar earned by her full-time working male counterpart. To vivify the point, one may then be told that tree surgeons outearn librarians by several thousand dollars a year. As a sort of climax of unfairness, the advocacy literature of the National Organization of Office Workers (NOOW) stresses the case of a firm offering $745-1,090 per month for general clerks, who must analyze invoices and have “good telephone etiquette” and who are usually female, while offering $1,030-1,100 per month for shipping clerks, who need only write legibly and be able to “lift equipment in excess of 100 lbs.” but who are usually male. It is presumed to be self-evident that the ability to talk on the telephone entitles one to more money than a strong back.

The big question is why women earn so little if the “human capital” they embody is so valuable and the jobs they do so important. The answers given by comparable-worth advocates may be ranked by the degree of conscious malevolence they attribute to employers. At one extreme stands the accusation that (male) employers actually conspire against women in order to reduce labor costs (“to maximize profits” is the preferred formulation). Thus, in 1980, the program director of NOOW assured the Equal Employment Opportunity Commission that banks “colluded to hold down wages”; her evidence was that (a) bank presidents regularly met for lunch, and (b) men use “cultural stereotypes” when arguing with female union organizers. Such charges, however, are not taken very seriously even within comparable-worth circles, for wage-fixing cartels, quite apart from being illegal, would be vulnerable to raids by independent firms willing to pay more for talented females. Nor is it clear why employers willing and able to rig substandard wages for women would not also rig equally substandard wages for men.

The case for conspiracy being so flimsy, comparable-worth advocates prefer two other explanations of the wage gap. The first is the “herding” of women into a “pink-collar ghetto” of typically female jobs, which floods the market and depresses wages for these jobs. According to a study done for the Michigan Department of Labor:

The reason for sex-related pay differentials [is] the traditional sex-segregation of most jobs. Under this hypothesis, the result of “crowding” of females into the clerical and service occupations is an oversupply of labor in these occupations, with resulting lower wages.

But this “hypothesis” rests most uneasily with charges of sexual discrimination, for it concedes that at bottom female wages are determined by the market forces of supply and demand. The more would-be secretaries there are, the less an employer must offer to get secretarial help—just as, conversely, the recent scarcity of nurses has driven up nursing salaries. Secretarial wages may be considered “discriminatory” and the supply of secretaries an oversupply only if “crowding” is not a metaphor but a literal description of employers forcing women into some jobs and barring them from others. The supposition that such practices exist is entirely implausible in light of the Equal Employment Opportunity Act and Title VII of the Civil Rights Act, which explicitly forbid them; nor has any comparable-worth advocate produced a single instance of such activities occurring in the last decade. But as this citation from the Michigan Department of Labor suggests, one of the more ominous themes in the comparable-worth campaign is its implicit claim that the spontaneous social factors which affect women’s vocational choices are themselves forms of discrimination calling for government correction.

Since the notion of “ghettoization” is so dangerously ambiguous, comparable-worth advocates add a final factor to explain the wage gap: the systematic undervaluing of women’s work as part of society’s overall disregard for women. Whereas classical economics says that in a free market the wages of all workers reflect their “marginal productivity,” no matter what prejudices their employers may harbor, the comparable-worth case holds that market mechanisms “don’t work” for women. In the words of Women, Work, and Wages, a report prepared in 1981 by the National Academy of Sciences (NAS):

In many instances . . . jobs held mainly by women and minorities pay less at least in part because they are held mainly by women and minorities . . . the differentials in average pay for jobs held mainly by women and those held mainly by men persist when the characteristics of jobs thought to affect their value and the characteristics of workers thought to affect their productivity are held constant. [Emphasis added]

In a remarkable display of pseudo-rigor, the NAS even produced a mathematical “coefficient of discrimination” for calculating by how much the proportion of females in a job lowers its wages.

The tricky part, of course, is being sure one has identified all the “characteristics” that affect the wages for any particular job. In practice, this means hiring a team of “management experts,” like Hays Associates or Norman Willis or the Arthur Young Co. (authors of the Michigan report), to make a determination. As one might expect, this also in practice inexorably links the question of what factors do determine wages to the quite separate question of what factors, in the opinion of management experts, ought to determine wages.

The experts typically resolve any given job into a few standard factors—skill, effort, responsibility, and working conditions, or some variants thereof—and then assign a score to each factor. Jobs with equal point totals, when all dimensions are combined, are regarded as equally “valuable.” The heart of the procedure—the assignment of points—is entirely subjective and universally admitted to be so by the firms that specialize in the practice. Nevertheless, it was in terms of such evaluation surveys that the NAS found women to be victims of wage discrimination.

_____________

Thus we arrive at the position of the state of Washington in the early 1970’s when, heeding feminist imperatives, its governor commissioned the Willis firm to run a “point-factor analysis” of wage rates. Like most public employers, Washington at the time was compensating at the rates prevailing for the nearest corresponding jobs in the private sector. But when the first Willis survey found that predominantly female jobs (those done 70 percent or more by women) paid on average 20 percent less than jobs of the same “worth” done predominantly by men, the Washington Personnel Board resolved in 1976 to use the Willis analysis in setting wages. No steps, however, were taken in this direction, in part probably because of a memorandum cautioning that doing so would cost the state almost $65 million—in retrospect, a twentyfold underestimate. Eventually, it was not the results of the Willis survey per se, but Washington’s failure to abide by studies it had commissioned and resolved to respect, that Judge Tanner used to sustain his finding of discrimination. Washington was hoist on the petard of its own impeccably progressive intentions.

It is vital to emphasize that equal pay for the same work was not at issue in the AFSCME suit and has not been an issue since the passage of the 1963 Equal Pay Act. Moreover, there is not a single sentence in the U.S. Code or anywhere else to give comparable worth any statutory basis whatever. What enabled AFSCME to bring its suit in the first place was the Supreme Court’s 1981 ruling in Gunther v. Washington that litigants have a right to test the comparable-worth concept in federal court under the rubric of Title VII of the 1964 Civil Rights Act. (Gunther concerned female prison guards of female prisoners, who claimed that their work was as valuable as that done by better-paid male guards of male prisoners in higher security prisons.) AFSCME was simply the first droplet in the flood of litigation released by Gunther.

It is true that neither of these two key comparable-worth decisions directly endorses comparable worth, nor is either a bolt from the federal blue into the wage policies of localities or industry. At the same time, however, it would be naive not to expect AFSCME to be replayed in the near future. Over a dozen states have undertaken comparable-worth studies of their own work forces, and there is agitation for such studies in virtually every other state and in hundreds of municipalities. These studies are invariably accompanied by disclaimers of legislative intent—“Let’s simply see if our state is perpetuating discrimination”—but AFSCME is an obvious basis for legal action once these studies uncover wage discrimination, as they always seem to do.

To this commotion may be added the 1982 endorsement by the leadership of the Democratic party of “equal pay for work of comparable social value,”¹ a doctrine even more radical than “equal pay for work of comparable economic value.” Walter Mondale is on record as supporting “equal pay for comparable effort,” more radical still. Congress has held hearings on the issue. Comparable-worth advocates are predictably eager to impose it on the private sector; such is the stated position of the New York Times. An advocate of comparable worth has argued that women who occupy “traditionally segregated jobs” have a prima-facie case for wage discrimination, with the burden of disproof falling on the employer²—an impossible burden to discharge since “wage discrimination” was unheard of until recently, and so no employer could have acted affirmatively to prevent it. Clearly, comparable worth has followed some other notable ideas from the feminist agenda onto the national agenda.

_____________

II

Is it true that discrimination explains why women earn less than men? The National Academy of Sciences concluded from its survey of the statistical literature that “differences in education, labor-force experience, labor-force commitment, or other human-capital factors believed to contribute to productivity” could explain, at most, half the wage gap. It would seem evident that the failure to explain the wage gap by a given set of variables is consistent with the operation of undiscovered variables having nothing to do with discrimination. Yet comparable-worth advocates have managed to obscure this point by expeditious wordplay. In the jargon of the moment, they conceive discrimination as a “residual.” This means that certain variables are selected a priori as relevant to wages, and discrimination is then defined as what these variables cannot explain.

In the last two decades the traditional test for discrimination—“intent”—has yielded to the much more dubious test of “effects,” but the “residue” test goes beyond even “effects.” Stigmatizing a practice as “discriminatory in effect” requires at least that one identify some specific human activity and show that it has some definite effect on the sexes. The “residue” test is uncontrolled, conflating as discriminatory all situations in which gender differences are observed, possibly including those caused by subtle innate biological sex differences.

The NAS states:

The burden should rest on the designer of the job-evaluation system to identify and explicitly incorporate all factors regarded as legitimate components of pay differences between men and women, not merely to assert the possibility that including unspecified and unmeasured factors or improving the measurement of existing factors could reduce the “discrimination” coefficient.

In English, this means identifying discrimination with the wage gap itself, since in practice all “unmeasured factors” are assumed irrelevant. By fiat, then, the NAS is able to dismiss such factors as differences in drive and family sex roles as imaginary.

Thus, none of the studies surveyed by the NAS properly controlled for so obvious a variable as marital status: the wages of never-married women with full-time, continuous labor-force participation are virtually the same as those of the average married full-time working male. Women earn less than men, it seems, because they want jobs permitting easy exit from and reentry into the labor force, preferences which flow in turn from the average married woman’s perception of her family as being her primary responsibility, especially when her children are young. Men, on the other hand, see themselves as breadwinners, whose obligation to make money increases as children come along. Perhaps a further factor will prove to be the male’s amply-documented greater innate competitiveness, his tendency to do whatever is necessary to ascend hierarchies, including the hierarchies of modern business organizations.

Would it exonerate all concerned of the charge of wage discrimination if the wage gap turned out to be due to differences in family sex roles? Not at all; these basic differences between men and women simply become further injustices:

Criticizing women for “lower labor force participation rates than men, moving in and out of the labor force more frequently than men, and being more likely than men to be seeking part-time work only” fails to take certain fundamental facts into account. For instance, among Fortune 500 executives, who is taking care of the children? Whose responsibility is it to juggle career demands with child-bearing and child-caring? Did it ever occur to you that the reason women seem less committed to careers is that they don’t have the luxury of a wife to take care of home and children while they blaze their career paths?³

This rage at all difference in the basic social role of the sexes underlies the campaign for comparable worth.

_____________

III

Critics of comparable worth understandably concentrate on the biases inherent in any list of factors “deserving” compensation. For instance, all job-evaluation systems I have seen assume that a college degree enhances a worker’s “intrinsic value,” a bit of snobbery which shows only that the people who design evaluation systems have all been to college. Yet this is not what is fundamentally wrong with comparable worth. For while the range of objective judgments about “training” or “responsibility” is far too narrow to support sweeping doctrines like comparable worth, such judgments can nonetheless be made. Planning shuttle missions in space plainly carries greater responsibility in terms of lives and hardware than does teaching philosophy. And planning space missions is plainly harder than adding up columns of figures, if only because the latter skill is just one part of the former.

For the sake of argument, then, let us assume a skill which is objectively difficult and reflects much objectively measurable training on the part of its practitioner; but let us also suppose it is a skill no one is interested in. Consider a person who is adept at throwing arrows into the air and catching them with his teeth. This is extremely difficult to do, and takes endless practice. Basketball players earning six-figure salaries do nothing so demanding. Unhappily, nobody wants to hire our man to catch arrows. He must eke out a living as a street entertainer. Is he somehow being denied his intrinsic worth by passers-by who flip him quarters? Does a circus scout who offers him a pittance for his act undershoot what the knack entitles him to? The answers would seem to be no. Intrinsic moral and aesthetic merit aside, the skill is economically worthless—unable to command other goods and services—if no one will pay for it. Only someone willing to trade something for the service in question can confer economic worth on it.

Money itself is merely the conventional measure of the capacity of a thing to prompt people to exchange their own goods for it. A thing’s price summarizes the ebb and flow of its performance in exchange, and has no independent meaning. And here is the intellectual black hole at the center of comparable worth: there is no such thing as intrinsic economic value. It is a chimera. Conversely, the willingness to supplant the market price of labor or anything else means the willingness to override the liberty of exchange, association, and contract expressed by market prices. In each particular comparable-worth proposal, the question is only one of determining where freedom is to be suppressed.

This crucial point is easy enough to see in connection with material objects. It would be absurd to maintain that copper deserves to cost more than gold because it conducts electricity better; gold and copper, absent people’s actual desires for them, would just be stuff in the ground. The point is also reasonably clear where the “just price” of money—i.e., permissible interest rates—is concerned. But the market determination of wages meets much greater resistance, perhaps because it puts our value so squarely in the eyes of our beholders.

The example of the arrow-catcher may be dismissed as a contrivance. Contrived it is, which proves its point. The skill seems freakish, worthless, precisely because no one has ever valued it or is ever likely to do so. Our very perception of skillfulness, in other words, is determined by the market. The ability to draw, or set bones, or fix automobiles, seems “naturally” compensable because we are used to there being a market for such skills. But the once “naturally” compensable ability to make buggy whips does not seem so today, when people no longer want buggy whips.

_____________

A long line of thinkers, extending back through Marx to Ricardo, has attempted, and failed, to devise non-market criteria of economic value for labor; almost every proposal in this tradition can be found in the comparable-worth literature. The straight Marxist appeal to the “social value” of work founders on the need to specify “social value.” If something is “socially valuable” just because a great many people are willing to pay top dollar for it, we are back to the market; and there is no other way to tell that society wants something than its willingness to try hard to get it.

A more frequent proposal equates the value of a job with its contribution to the employer’s profit. We may overlook the obvious difficulty: that this would make wages fluctuate wildly with exogenous factors—think of proofreaders becoming superfluous when a publisher’s profits are due to a single best-seller. Let us, instead, consider wage rates. Classical economics holds that each employee tends to get his marginal return on his product as he bargains omnisciently with his employers: If Jones’s labor is worth $20 to Smith, Jones can return Smith’s wage offer of $15 with a counterdemand for $16, which is still attractive to Smith. But then Jones can hold out for $17, $19, $19.50 . . . until Smith must agree to pay the full $20. But in this case the market itself already bestows a contribution to profit. In reality, of course, the employer must retain some portion of his employee’s product if he is to have any reason for hiring him. But then, the contribution criterion means deciding what portion of a businessman’s profit he “deserves” to keep, an entirely arbitrary decision.

Now, if one assumes that males get the “right” return on their product while females don’t, it might seem fairer and simpler to give each worker in a firm the same return on his product, without having to decide on a proper rate of return. A secretary, say, would be entitled to 90 percent of her product if holders of benchmark jobs like engineers got 90 percent of theirs. But benchmark figures must be determined independently, which means by the market. A firm finds that to secure the labor of scarce engineers it must pay what amounts to 90 percent of what they earn for the firm. But if so, what fairness requires is not that the return for secretaries be proportionate to that for engineers, but that it be determined in the same way—which is to say, by the market. Making the availability of engineers a factor in compensating secretaries is neither equitable nor rational.

What lends an air of unreality to every version of the “contribution to profit” theory is the fact that each of several jobs may be necessary for a firm’s profitability. Since it is logically impossible to give all the profits to each of several job categories, what does and must happen in the real world, in which a firm would be paralyzed both without secretaries and without engineers, is recruitment of secretaries and engineers at market wages, the wages they all agree to accept. As the English economist Thomas Hodgskin wrote, there is no “principle or rule . . . for dividing the produce of joint labor among different individuals who concur in the producing, but the judgment of individuals themselves”

A cognate difficulty undermines the related test of basing a job’s wages on its contribution to the “organization’s objectives,” to quote the Michigan comparable-worth study. This is a measure often suggested for non-market organizations like universities, governments, and foundations. Yet the achievement of non-monetary goals also requires the cooperation of many people whose contributions cannot be isolated a priori. Furthermore, to compete with profit-seeking organizations in attracting a sufficiently talented work force, nonprofit organizations must in any case heed market decrees about wages. Moreover, determining some government post’s contribution to the “overall objectives” of government presupposes clarity about what the “overall objectives” of government are, a goal that has eluded political philosophers for some centuries. If government has accommodated comparable worth more readily than private organizations, this is not because governments have found comparable worth easier to construe; governments can finance their follies by taxation, with no worries about buyer loyalty.

Finally, comparable-worth criteria that stress job characteristics like working conditions face all the problems that beset the use of traits like skill. Collecting refuse may be “unpleasant,” but it is pleasant enough to attract applicants at the going rate. We do not need elaborate analysis to see that the market wage adequately compensates refuse collectors for the unpleasantness they endure. And, again, our very perception of this circumstance is a response to the market. Most American farm workers would find intolerable the backbreaking chores that constituted farming a millennium ago.

Skill, effort, and training do play a role in determining pay. Someone who has invested much time preparing for a hard job will drive a hard bargain. More importantly, the harder a job and the training for it, the fewer candidates there will be, driving up wages still further. This naturally high correlation among difficulty, training, and salary explains the feeling that demanding jobs “deserve” higher pay. In the end, however, these factors raise wages only by influencing the choices of bargainers.

_____________

IV

Appeal to the free market may appear slightly musty in these days of the minimum wage, banking regulation, the National Labor Relations Act, and cancer warnings on cigarettes. Yet these familiar measures are modifications of laissez faire, while the rationale and scope of comparable worth make it a frontal assault on any form of economic liberty.

To begin with, none of the extant labor laws is based on a non-market notion of value. The minimum wage was intended to insure everyone a “decent” wage, without pretense that every worker’s output is “really” worth $3.35 an hour. The National Labor Relations Act banned “anti-union animus” on the part of management not because collectively negotiated wages were thought to reflect value more accurately, but to secure “labor peace.” And, while the minimum wage truncates the range of possible bargains, and the National Labor Relations Act affects a wide range of negotiations, both measures still permit very extensive play to market forces. Unionized plumbers may not get strict market wages, but their wages approximate market value.

Overall, the many regulations that control banking, trucking, the airlines, mergers, work-place safety, and the like are (a) industry-specific or activity-specific, and (b) intended, rightly or wrongly, to preserve the free market against its own excesses. Wage and price controls, when imposed, are usually justified as “temporary measures” to curb inflation. Even socialism itself was originally conceived as a way of harnessing in a more efficient way the productive capacity generated by capitalism. Because “New Deal”-type regulations thus purport to secure specific results, they imply built-in tests to which they can be held. A ceiling on bank interest rates is supposed to help Savings and Loan Associations, so if S & L’s keep failing at the same rate after the ceiling is in place, the ceiling has not worked.

Comparable worth, by contrast, contains none of these self-limitations. Its scope includes every job in the work force; and since most jobs are “sex-segregated,” most pay scales would be open to challenge. Nor does comparable worth pretend to facilitate the best tendencies of the free market; rather, it is explicit about seeking to flout the market. It does this, moreover, without a clear specification of what positive goal is to be achieved beyond “justice”; under this mantle, the pursuit of comparable worth can discount any economic havoc it wreaks as irrelevant to its “success.” (A similar criterion now governs assessment of affirmative action, deemed “successful” if it increases the number of women and minorities in well-paying positions, no matter the cost in fairness or efficiency of putting them there.) Penn Kemble has called comparable worth “a feminist road to socialism”;⁴ if so, it is socialism without a plan.

It is instructive to reflect that comparable worth can never come about through voluntary agreement. If a firm can get secretaries at a market-clearing wage, so can its competitors. Were it to raise its secretarial wages for ethical reasons, its labor costs would rise without any gain in productivity, its products would cost more than those of its competitors, and it would slide into failure. Comparable worth presents what economists call a “coordination problem”: it must be introduced all at once, or not at all. Needless to say, the favored agency for solving coordination problems is the government.

_____________

The central point is this: whether or not the marketplace offers an appropriate ideal of justice, wages will in fact tend to be set at their market value so long as people retain anything like their accustomed economic liberty. The only alternative to the market is systematic state control. Comparable worth can be implemented only by endless government intervention, in the words of a pre-Gunther federal court, “pregnant with the possibility of disrupting the entire economic system of America.”

For if “wage discrimination” is the wage gap itself, whatever its causes, nothing less than the elimination of the gap will be allowed to count as ending discrimination. To grasp what this would mean, consider that in 1983 there were 49 million full-time working men whose median income was $20,683, and 31 million full-time working women, median income $12,172. To close the wage gap, employers would have had to pay each woman $8,500 more, or about $250 billion, or—in still other terms—about 15 percent of total wages paid in 1983. These costs would no doubt have been passed on to consumers as higher prices, and as higher taxes to finance government services.

One might think the gap could be closed more slowly by freezing or slowing the growth of men’s salaries until women’s salaries grew to parity. This is now the policy in San Jose, California, where the city government worked out a comparable-worth agreement with AFSCME. It seems unlikely, however, that men or their unions would tolerate this arrangement on a wide scale. A man who, thanks to comparable worth, gets a smaller raise than he would have otherwise gotten has, so far as he is concerned, suffered a pay cut. Comparable worth is advertised as a boon for working women struggling to help their families with a second income, but it does not help a family to hold down the husband’s wages so the wife’s can be artificially boosted. On balance, the likeliest short-term effect of comparable worth would be to boost everyone’s wages, thereby flooding the market with new money in the absence of new goods—the standard recipe for inflation. And this might be the occasion for the government to begin coordinating wage policy to assure the proper closure of the wage gap.

The longer-run consequence of inflating female salaries and holding down male salaries in defiance of the market would be a massive disincentive to work. Women would already be getting more without having to work harder, and men would not be permitted to get more even if they did work harder. Why, then, should anyone work harder? Nor would the work force become more “integrated,” since women would have no incentive to leave “women’s work” once it paid as much as less pleasant “masculine” work. If anything, men would try to invade the newly well-paid female sphere. And at the same time men were queuing up for the typing pool, there would be no reason for a man to undertake an unpleasant job like collecting refuse if high wages for it were no longer available as an inducement—so we would almost certainly see critical job shortages. Add to this the undoubted persistence of a quota system which would prevent management from firing women to make room for (possibly more desirable) men, and a crisis of extreme proportions would be upon us. If recent history teaches us anything, it is that governments meet crises of their own making with more of the coercive rules that created the crisis in the first place.

_____________

V

It is not surprising that a society which has absorbed busing, quotas, Miranda, and publicly funded abortion should look on comparable worth with numb bemusement. One can almost sympathize with a businessman happy to reach the end of a profitable quarter without some new horror from the Federal Register landing on him. It is disheartening nonetheless that the American Compensation Association should summarize the attitude of the business community toward comparable worth in these words: “Planning for the inevitable appears to be the appropriate response.”⁵ Like such otherwise perceptive commentators as John Bunzel (who sees comparable worth as part of the “revolution of rising entitlements”), the ACA has missed the distinctive element in the comparable-worth campaign.

Feminists realize at some level that men and women view work differently—but they believe that this should not be, and, as do all those who wish people were other than they are, feminists want the government to correct the situation. Inverting cause and effect, they ask the government to curb one manifestation of sex differentiation. The real enemy feminism has targeted, however, is not some isolable inequity in the wage system but the family and the rest of the malign “social conditioning” which make men and women act differently, in the economy and elsewhere.

That is why one must view with mixed feelings those critics of the comparable-worth idea who fall in with the “social-conditioning” theory. Lee Smith writes that “in large part, the [wage] disparity stems from traditions and prejudices that have channeled women into low-paying jobs.” Bunzel conjectures that “greater interests and responsibilities in the home” may explain the number of women in low-paying jobs, but that this may “reflect socialization toward traditional roles that, for many women, begins in childhood.” Cotton Mather Lindsay, the least compromising critic of comparable worth, cites as the most important factor limiting women’s access to high-paying occupations “the subtle socialization process beginning in childhood, which orients, women toward domestic careers.” Even John Warehman, who bluntly asserts that “women lack the necessary emotional strength to achieve executive success,” says “women’s ‘emotional deficiencies’ spring from patterns of dependency established during their childhood.” These writers treat the origin of this socialization itself as a mystery.

Economists may be forgiven for not knowing that research on sex differences shows these different social roles to be innate, or so close to innate as to be uneliminable features of every human society—but when they reject comparable worth only because it penalizes employers who “had nothing to do with producing . . . female inhibitions” (in the words of Carl Hoffman, who has conducted much research on male and female motivation in the business world), they abet the most irresponsible feminist tendencies. If an adolescent girl’s “socialization” really forced her to choose low-paying nursing over high-paying surgery, she may not have been cheated by her employer but she has certainly been deprived. Wouldn’t it be well to change custom and tradition so that little girls were raised to be prepared emotionally for becoming surgeons? And isn’t the state allowed to do a little of the altering for the general good? Thus is the state given a blank check for the sort of intrusion of which comparable worth is an extreme instance.

Prophesying the end of the free market is a sure way to be labeled a crank. On the other hand, I am not encouraged by confident assurances that the free market can survive anything. Howard Ruff, in one of his cheerfully apocalyptic guides to the future, compares the free market to Rasputin. Rasputin was poisoned, and he was shot, and still he survived. So, writes Ruff, the regulators have done everything to American capitalism, and still it survives. Unfortunately, Ruff overlooks a final detail. Eventually, Rasputin’s tormentors threw him in the river. That killed him.

_____________

¹ Rebuilding the Road to Opportunity, Report of the Democratic Caucus, U.S. House of Representatives, September 1982.

² Sheila Blumrosen, “Wage Discrimination, Job Segregation, and Title VII of the Civil Rights Act of 1964,” University of Michigan Journal of Law Reform, 1979.

³ Letter from Susan Sharp of Sharp and Co. to Fortune, September 6, 1982.

⁴ “A New Direction for the Democrats?,” COMMENTARY, October 1982.

⁵ “Comparable Worth . . . A Few Thoughts on the Issue,” ACA News, February 1984.