(Archived document, may contain errors)
337 March 2, 1984 COMPARABLE WORTH PART 2 THE HIG H COST OF BAD POLICY INTRODUCTION According to the Census Bureau, the average woman in full-time employment earns about 62 percent of the income of the average male in full-time emp1oyment.l To many feminists, this is proof of discrimination. In addition t o this general difference in earnings, it is claimed, there is a systematic gap in earnings between male and female work of llcomparable worth.Il The reason the earnings gap is as large as it is, maintain adherents to the comparable worth theory, is becau se of sex discrimination in the marketplace.
To remedy this perceived inequity, feminists are now demand ing that comparable worth be used to set "fair" wage levels.
This goes well beyond the notion of equal'pay for equal work It would require equal wages for work of supposedly comparable value to an.employer. Under this doctrine, pay would be based on the opinion of an government .board or similar body, whose decisions would derive from an estimation of the skill, effort and responsibility involved in on e job relative to another.
Comparable worth advocates assume that if such job evaluations were to replace the market system in setting wages, the pay of women would rise to a "fairer" level.
The concept is b ased on economic ignorance. There is no such thing as an objective scale of economic value. Market-level wa.ges are the product of the subjective evaluations of employers workers, and consumers. Yet the comparable worth notion already has been endorsed by many leading politicians. In addition Robert Pear Earnings Gap is Narrowing Slightly for Women," The New York Times, October 3, 1983, p. B15. 2 comparable worth advocates have been emboldened by a recent court ruling in Washington state that ordered highe r wages and back pay for many female state employees. The ruling could cost the state hundreds of millions of dollars if upheld, and it sets a precedent for the private sector. If similar rulings were applied throughout the country, the total cost to the e conomy could be as much as 320 billion a year.
Despite their efforts to make comparable worth the law of the land, supporters of the idea have yet to prove their allega tions of sex discrimination.2 And even if such discrimination existed, determining Ilfa irlI wages would be an impossible task given the subjective nature of such judgments. Moreover, the results of such a policy might be quite the opposite of the original intent. Indeed, women might simply find their employment opportunities vanishing rapid l y, as employers replaced them with men and machines P The comparable worth concept, put forward as a correction to an alleged failure of the marketplace, would have a devastating effect on the nation's economy. It would throw a monkey wrench into the bala n cing mechanism of the labor market, and lead to shortages and oversupply throughout that labor market. It would prevent unions from winning the best possible terms for their members, since wages no longer would be a product of collective bargaining. And i t would cause enormous resources to be devoted to influencing the Ilobjectivell decisions of boards of wage evalua tion. In short, comparable worth legislation would hamstring the marketIs.steady adjustment to the changing role of women and replace it with politicized central planning BACKGROUND Much of the current debate on comparable worth stems from a recent Washington State court decision. On December 14, ,1983 U.S. District Judge Jack E. Tanner ordered the state to award back pay and higher wages to mo re than 15,000 of its employees 90 percent of them women. His ruling was based on the alleged principle of "comparable pay" for llcomparable work.
The lawsuit before Judge Tanner accused the state of sex discrimination in its pay practices, based on the fa ct that women as a class were paid less than men. It did not claim that women received less pay for doing equal work, or that their oppor tunities were somehow limited by the state--only that women were not paid the same as men for work that was claimed t o be of comparable value. Tanner's ruling was based heavily on job evaluations conducted by !!experts who claimed that fields of See Peter Germanis Comparable Worth--Part 1: A Theory with No Facts,"
Heritage Foundation Backgrounder No. 336, March 2, 1984. 3 I8 work dominated by women consistently provided lower wages for work rated comparable to that in male-dominated fields.
The state of Washington had undertaken several job evaluations indicating disparities, but had not acted upon them (which was a reas on for the suit In 1973, for instance, Governor Dan Evans now a U.S. senator, commissioned the firm of Norman D. Willis and Associates to determine whether or not the state was paying women and men equal wages for work comparable in skill and responsibili ty.
The worth of the various jobs was evaluated by assigning points for four categories: knowledge and skills, mental demands accountability, and working conditions.
The study found that on this basis of worth, predominantly female jobs paid about 20 perc ent less than jobs held mainly by men.3 Later studies, using the same assumptions and methodology confirmed the original findings. Table 1 gives examples of jobs that earned equal points in the 1982 study and thus supposedly are comparable. It Table 1 Poi n ts 97 155 197 209 Male Dominated Jobs TOP Female Monthly Dominated Salary Jobs TOP Monthly Salary Truck Driver I $1,574 Laundry Worker $1,114 Electrician 1,918 Secretary I11 1,324 Equipment Mechanic 2,015 Attendant Counselor III*f: 1,392 Equipment Operato r I1 1,738 Attendant Counselor* 1,200 f: Provides care for retarded people.
Supervisor LA I. n Source: Jake Lamar, A Worthy but Knotty Question, Time February 6, 1984, p. 30 According to the survey, therefore, a truck driver is equivalent to a laundry work er and the two should earn the same salary-either by boosting the laundry worker or cutting the truck driver. In the Washington State judgment, the state was ordered to boost the salaries of those below the norms.
In addition to studies by Willis and Asso ciates, the plain tiffs introduced other evidence purporting to show sex discrimina tion, such as help wanted advertisements from 1959 to 1973 that separated state jobs into female wanted and male wanted See Geoffrey Cowley, Comparable Worth: Another Terr i ble Idea, The Washington Monthly, January 1984, p. 54 4 columns and certain state job specifications prior to 1972 that contained gender reference The state's case was severely hampered by Judge Tanner's refusal to allow the testimony of a number of state witnesses and the presentation of all of its exhibit The witnesses prevented from testifying included the state's wage setting process and affirmative action programs for women; June O'Neill, an economist with the Urban Institute who would have testified a bout the factors contributing to the wage gap and the absence of studies determining how much is due to discrimination; and an expert on job evaluation systems who would have criticized the methodology of the Willis firm. But Judge Tanner would hear only t estimony arguing that the state had economic reasons for justifying discrimination. He refused to hear evidence disputing the allegation of discrimination. Thus the state was forced to rest its case on the presumption that the comparable worth doctrine is a valid approach to wage determina tion a state official who would have described THE FAULTY LOGIC OF COMPARABLE WORTH Job' Evaluation Techniques Wage determination under comparable worth would work as follows: Jobs are identified that are segregated by s e x (e.g those where more than 70 percent of the jobs are held by one sex), and each is assigned points based on some supposedly objec tive value. The Willis job evaluation, on which Judge Tanner relied heavily in his decision, assessed each job classificat i on using the following four evaluation components (1 Knowledge and Skills (Job Knowledge, Interpersonal Communications Skills, Coordinating Skills 2) Mental Demands (Independent Judgment, Decision Making, Prob1e.m Solving Requirements 3) Accountability (F r eedom to Take Action, Nature of the Job's Impact, Size of the Job's Impact 4) Working Conditions (Physical Efforts Hazards, Discomfort, Environmental Conditions). The total value of these four components constituted the final point value of the class.6 Fo r more details, see the memo from Christine
0. Gregoire, Deputy Attorney General of Washington, to a number of elected officials in the state of Washington, dated October 4, 1983.
For a summary of the obstacles faced by the state, see ibid.
American Fed eration of State, County, and Municipal Employees v. State of Washington, No. C82-465T, Slip Opinion, cited in Phyllis Schaffly Equal Pay for Unequal Work script, p. 5 The Comparable Worth Concept unpublished manu5 To calculate the 'Ivaluel' of each of th e se components and subcomponents, a group of Ifexperts1l reviewed the various job descriptions, interviewed workers, and then assigned a final point score for each job In the Washington case, for example a laundry worker had 97 points, a truck driver 97, l i brarian 353 carpenter 197, registered nurse 573, and computer systems analyst 426 The results of this comparable worth study, in other words, suggest that a registered nurse is in some objective sense more valuable than a computer systems analyst and shou ld be paid more--even though the market actually rewards computer systems analysts with earnings that are 56 percent greater than those of nurses.
The principal problem with these job evaluations is that they completely ignore supply and demand. The Iffair wage deter mined by some expert is irrelevant if a position cannot be filled because potential applicants feel it is too low. Similarly, if the e x pert sets the wage above the market rate, there will be a flood of applicants, many of whom will have to be turned down because no positions are available, even though they would have been willing to compete by accepting a lower wage market wage system, p r ofitable and efficient companies can offer higher wages to fill vacancies, and workers can retain their jobs in difficult times by accepting wider differentials with workers in other industries Under a This process of adjustment would be ruled illegal by compar able worth legislation.
An iron rule would set wages, without regard for the reality of the labor market and without regard for the welfare of employees.
The llObjectivel' Value Fallacy Another problem with any job evaluation of this kind is that, although it is supposedly objective, it requires a subjec tive opinion.by an evaluator as to how much each job is worth.
Given the large number and diversity of jobs being rated, it is unlikely that any two people could come up with an identical valuation . This is not only because people disagree about 'Ithe facts," but because the whole notion of assigning an objective valuation on labor is absurd. Like beauty, value is in the eyes of the beholder. The value of a person's work depends not on how Ifsocial l y necessary" it is or how many lfpointsll it is awarded by a board, but on the willingness of consumers and thus employers to pay for it. In the final analysis, the comparable worth doctrine means job evaluators would simply replace the consensus of the m arket's subjective assessments with their own subjective assessment--and then the law would be used to enforce their views Schaffly, op. cit p. 6, and June O'Neill The 'Comparable Worth Trap The Wall Street Journal, January 20, 1984.
O'Neill, op. cit. 6 If Judge Tanner had accepted the evaluations of another respected consulting firm, Jeanneret and Associates in Houston instead of the Willis calculations, he would have been told that no disparity existed in pay among comparable male and female jobs on the W ashington State payroll. The Jeanneret system uses 45 factors, compared with four in the Willis system (broken down into 13 subcomponents) and a computerized factoring system instead of the Willis method of basing scores on committee consensus. This is no t to say that the Jeanneret system is any better than the Willis system, only that job evaluations of this type veer all across the board because they must be based on subjective opinions, rather than objective facts.
Under a comparable worth scheme, then, wages would no longer be based on'productivity and initiative, but on some bureaucrat's view of'the worth of the worker's occupation.
CONSEQUENCES OF COMPARABLE WORTH The Cost The ruling in Washington will cost the state between $500 and 800 million. The total cost to taxpayers throughout the country could be many times this figure, if other states adopted comparable worth legislation or similar court rulings are handed down. If such laws were applied nation wide and covered all employment, I the pay inc r eases required for female-dominated jobs could cost 320 billion, according to Dan Glassner of the consulting firm of Hay Associates.g I Robert Williams, a management representative in labor negoti ations, points out Unless we are prepared to alter radical l y our whole economic system, a solution that holds individual employers responsible for market conditions, or forces them to ignore the market in favor of purely internal value scales, simply cannot work.1 Employment Effects The pay increases in the publi c sector required by a compar able pay law would force governments to increase spending or cut services. If they opted for the first choice, the tax hikes to finance the additional spending would drain resources and jobs for both sexes) from the private se ctor. If governments chose the second option, many public sector workers would be laid off Jake Lamar A Worthy but Knotty Question Time, February 6, 1984 p. 30 lo Ibid.
Women's occupations would be especially hard hit since their labor costs would have risen most rapidly under such comparable pay laws.
The results would be even more disastrous if comparable worth were extended to the private sector. Initially there would be an enormous jump in the cost of doing business. These additional labor costs would bankrupt many firms with a high proportion of women in their workforce, thereby leaving these women without a job. Employers would also seek to replace many women with machines in cases where women's marginal productivity did not justify the increased rat e of pay. IlComparable worth,Il in fact, would hurt the least skilled women most, since their services would be first to be priced out of the market.
Ironically, by raising the pay of predominantly female work comparable worth legislation could slow or eve n reverse the employment gains women are making in such traditionally male dominated fields as medicine, law, and engineering It would do so by artificially distorting the wage structure, making the male-dominated professions relatively less attractive It could also encourage some men to enter the traditionally female occupa tions, generating greater competition for the jobs in these sectors.
Thus employment shortages and surpluses would develop because comparable worth rules would distort the price mechanism, which provides the signals necessary to induce people to enter fields where there are shortages and leave those that are oversupplied.
Generalized employment losses would also arise as the artificial and inefficient wage structure created by comparab le worth made U.S. industries less competitive. Forcing a change in the cost of one input in the production process--in this case female labor would cause employers to shift their resources around until they could once again minimize costs. This new mix o f resources however, would mean a rise in total costs (otherwise managers would have chosen it before and so push up the prices of domes tically produced goods and services. This, in turn, would make products from abroad more attractive to American consume rs leading to a decline in the demand for U.S. goods and increased unemployment in the affected industries.
Central Planning The proliferation of comparable worth laws, with wage boards determining the worth of every job, would be a significant and dangero us step toward central economic planning No individual or group of individuals, can possess the vast amount of knowledge necessary to set wages in a way that allocates resources through out the economy most efficiently. Only the free market price mechanis m can do this would attract too many people for that particular job, while wages set too low would create shortages. In a free market wages rise or fall to Lemove shortages or surpluses In a planned Any wage set too high by a wage board 8 economy, there is no self-correcting mechanism to deal with this problem A wage board cannot determine that a nurse and a tree trimmer are "worth1' the same. Only the valuation awarded by the market place, in the form of competitive wages, makes any economic sense balance d emand and supply. If women prefer nursing to tree trimming, resulting in an oversupply of nurses relative to tree trimmers (or any other occupation) and a relatively low wage, the result may not be to the liking of nurses, but that does not mean employers are responsible and should be penalized for the result And only wages emerging from this market process will Clearly wages set by a compar≤ worth evaluation board would render collective bargaining obsolete. If market factors were no longer to be the ba s is of wage setting, it would be irrelevant whether or not a union won concessions from a profitable firm--only the worker's ltobjective'l worth would matter. Unions would be reduced to toothless watchdogs, ensuring that management paid the wage rates dete rmined by the board of evaluation.
Rent-Seekinq Wages determined by wage boards would mean an end to pay increases reflecting productivity increases As such, workers would have less incentive to develop their skills if they felt that a point system would not reward them sufficiently other h and, workers would have strong incentives to lobby wage boards and expend considerable resources in an attempt to persuade the board to raise the wages of their particu.lar occupation by adjusting their llworthll score On the As some groups succeeded by s uch lobbying, other workers who might have been successful in a free market would be penalized.
This would be a particular problem of jobs subject to rapid tech nological change, where innovations could quickly alter the nature of a job, making it a candid ate for reevaluation on the comparable worth scale. Thus, resources would tend to be diverted from education and training, to petitioning wage boards--to the disadvantage of consumers, taxpayers, and labor.
CONCLUSION Women's advocacy groups often attribu te differing wage levels to persistent sex discrimination in the marketplace. Yet they overlook real differences in productivity, as well as basic supply and demand in the labor market. 'IComparable worth1' cannot end wage differentials. By trying to do s o , moreover, it would impose extraordinary costs and distortions on the economy. Among them: bureaucratic regulation; higher prices, higher taxes, increased unemployment, emasculation of labor union wage-bargaining power, reduced competitiveness in world m a rkets, lower productivity and perhaps worst of all, a signficant loss in freedom 9 Several steps could be taken to improve women's economic status without destroying the labor market basis of the U.S economy. Tax laws could be changed so that a household' s secondary worker (usually a woman) would no longer face high marginal tax rates on earnings, which now are simply added on top of the primary worker's earnings. Barriers restricting entry into certain occupations or raising the cost of employing certain g roups could be eliminated. Moreover, occupational licensing laws and restrictions on doing work in one's own home, which eliminate potential jobs for many women, could be repealed. In short, impediments that prevent the marketplace from adjusting to the c h anging role of women could be dismantled, allowing female wages to reflect their improving productivity.ll Comparable worth legislation would only slow down the progress being made. And it would probably increase discrimination rather than reduce it, beca u se the higher wages forced upon employers would lower their demand for traditionally female jobs while the number of applicants would rise, thanks to the mandated new pay scales. This would make it much easier for employers to discriminate. Comparable wor th legislation, in other words I would simply exacerbate the very problem it was designed to cure.
Peter Germani s Schultz Fellow See Catherine England and Robert Valero, "Working Women the Solution--Or the Problem?" Heritage Foundation Backgrounder No. 263 May 2, 1983 Is Uncle Sam