Comparable Worth: The Paradox of Technocratic Reform

by Sara M. Evans, Barbara J. Nelson
Comparable Worth: The Paradox of Technocratic Reform
Sara M. Evans, Barbara J. Nelson
Feminist Studies
Start Page: 
End Page: 
Select license: 
Select License


Comparable worth, the policy Eleanor Holmes Norton labeled "the issue of the 1980~~"~

is poised at the juncture of an inequitable past and a feminist vision of the future in a way that may clarify some of the problems and paradoxes we face as we struggle to move from one to the other. It evolved as a conscious strategy in the 1970s (although its roots were much older) as people recognized how deeply the history of discrimination is embedded in our economic institutions. Neither affirmative action nor equal pay for equal work had proved sufficient to address the inequities of a labor force segregated by gender, race, and ethnicity.

There is no need to belabor the structural and historic realities that prompted the development of comparable worth as a strategy for change. Work in all industrialized economies is segregated by sex, race, and ethnicity; women and minorities are concentrated in the lowest paid jobs with fewest opportunities for advancement.2 Comparable worth challenges not so much the distribution of workers among jobs as the valuation of the work itself. As the Na- tional Academy of Sciences recently found, "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."3

Comparable worth is a strategy intended to remove the negative weight of a history that devalued women, minorities, and the work they do. It holds the promise of a future in which wages are more equitably distributed, with all the side effects that could have for the independence and autonomy of women and people of color. But between the heaviness of the past and our vision of the future, we face the immediate realities of specific changes that bear unforeseen fruits. Although most discussions of comparable

Feminist Studies 15, no. 1 (Spring 1989). 1989 by Feminist Studies, Inc. 171

worth have focused either on analyses of labor force segregation and discrimination- establishing that there is indeed a problem re- quiring such a policy-on economists' predictions of long-range consequences for employment, or on the techniques of implemen- tationI4 we propose to discuss the concrete realities involved in the implementation of comparable worth in state and local jurisdic- tions. Our objective is to deepen the discussion by examining how the process of wimiig and implementing comparable worth affect its possibilities for social transformation.

Our analysis focuses on the experience of the state of Minnesota and twenty-two localities in that state (four counties, ten cities, and eight school districts).5 Minnesota's state employees are covered by the 1982 State Employees Pay Equity Act, which was fully implemented by 30 June 1987.'j Local employees are covered by the 1984 Local Government Pay Equity Act, a piece of com- parable worth legislation mandating that all local jurisdictions con- duct comparable worth studies; report their findings to the state Department of Employee Relations for transmission to the legis- lature; and develop plans to remedy any inequities found between male, female, and balanced job classes. To understand these laws and the policies they created we need to begin with some defini- tions.

Comparable worth is a wage policy designed to correct historically discriminatory wages of female- and minority-dominated jobs. Although there are many possible ways to identify and correct such discrimination, by the late 1970s most definitions of com- parable worth presumed the use of job evaluation technology. In this framework, comparable worth is a wage policy requiring equal pay within a jurisdiction or firm for male- and female-, majority- and minority-dominated job classifications that are valued equally in terms of skill, effort, responsibility, and working conditions.'

The single job evaluation system measures in detail the skill, ef- fort, responsibility, and working conditions of every job classifica- tion and combines the scores in each area to produce a single overall score for every classification. Job classifications with equal overall scores are considered to have equal value to the jurisdic- tion or firm.

A large number of studies have shown that if two job classifica- tions have the same value according to the job evaluation system, but one is held primarily (i.e., 80 percent or more) by men and the other held primarily (i.e., 70 percent or more) by women, the job held by men usually pays more.* Under a comparable worth wage policy, classifications with the same scores -regardless of whether they are dominated by women, minorities, or white men -should be paid equivalently. However, all individuals holding jobs within equal classifications would not be paid the same wages, because seniority, merit, and the quantity or quality of work done would continue to differentiate individuals' wages within equivalent classifications.

Theory and practice do not always coincide when comparable worth is implemented. In theory, at any one time, comparable worth compares male or white jobs and female or minority jobs within a workplace without reference to wage rates outside the firm. This is because market rates reflect the historic undervalua- tion of the work done by women and people of color. Essentially, the market is viewed as setting wages nondiscrirninatorily only for white men, and even then only under certain conditions.9 If wage rates for female- and minority-dominated jobs are compared with existing market rates, some of the undervaluation of female and minority jobs is reintroduced into wage setting.

The state of Minnesota closely conformed to comparable worth theory, comparing female-dominated positions to equally valued male-dominated positions but not to current wages for these posi- tions outside Minnesota. State implementation of comparable worth was "distributive," offering a remedy to historically low wages through an "add-on.""J In budgetary and legislative terms, distributive policies frequently have unorganized oppositions and are thought, by supporters at -least, to be desirable to society as a whole and not likely to be adopted without government interven- tion.''

Unlike the state, most local jurisdictions in Minnesota did not implement comparable worth in this manner. Like the settlement of the San Jose, California, comparable worth strike and the final results of the comparable worth dispute in the state of Washing- ton, most Minnesota local jurisdictions adopted some sort of pay- for-points policy, often comparing the salaries of equally valued jobs to the average salary of all jobs having that value and to market rates for those positions.12 In many Minnesota localities, like Hennepin County which employs 8,000 workers, comparable worth means that if two jobs have the same job evaluation score, the salary and benefits in the highly paid (usually male-dominated) position will grow more slowly than in the less well paid (usually female-dominated) position. Local implementation often was a legislatively and budgetarily redistributive policy, rearranging the wage bill more than adding to it. In general, politicians do not like redistributive policies, and they often try to make them more distributive. This was sometimes achieved in Minnesota local governments by making all "underpaid positions, regardless of what sex of workers predominated in them, eligible for wage in- creases. Some localities also made all "overpaid classes vulnerable to slower wage growth. These practices rely on definitions of fair- ness that do not take into account the historical undervaluation of female labor.

Examining the workings of a specific job evaluation system helps to demystify what is meant by "jobs of equal value." The state of Minnesota contracted with a management consultant firm,


Zookeeper  $967  Painter  $1,368  
Human Services     
Technician  $783  Social Worker  $1,103  

Poultry Improvement Specialist $1,227

Human Services Specialist $ 912

Police Training General Mainte- Director $1,841 nance Worker $1,027

Director of Administrative Nurses $1,583 Secretary $ 912

Fig. 1. Comparison of wages in equally valued jobs for 1981 (source: Min- nesota Department of Employee Relations).

Hay Associates, to conduct a job evaluation study that was completed and adopted in 1979.Subsequently,in 1981,the state Council on the Economic Statusof Women appointed a Pay Equity Task Force to reanalyze the Hay study by sorting out and comparing the wages of male, female, and balanced job classes. The Task Force study revealed systematic discrepancies between male-and femaledominatedjobs similar to those found in studies of other large employers and governmental jurisdictions.13 For example, according to the state of Minnesota'sjob evaluation system, a zookeeper's job and a human servicestechnician'sjob (e.g.,caring for the mentally retarded in a state hospital) received approximately the same number of points. The predominantly male zookeepers earned $967 per month, however, compared with $783 for human services technicians (see fig. 1). Similarly, equivalently rated jobs such as vocational education teacher (male)and registered nurse (female)differed dramaticallyin pay. The result of such disparities is a dual wage system, illustrated most effectively with a scattergrarn that plots male-and female-dominated jobs on a graph that has job evaluation points along the horizontal axis and wages along the vertical axis (see fig. 2).


2.4 -
a 2.1  
 2     -   
'-2 gi=2 xu < 2  1.9 -1.4 1.3   . * .* * . .* ' . FF*, *:FF'F:   I ..* .* F F  . . ;aF FF *  .  
 1.2 1.1 - F  FF  *#;I7FFF  FFF    
 1  F  i  :  l  ,  l  I  I  ,  i  I  ,  ,  I  ,  I  I  I  l  l  
 80  100  120  140  160  180   200  220  240  260  280  


Fig. 2. Job classes by Hay point and salary (source: Commission on the Economic Status of Women, 1985). * = male-dominated job classes (80 percent or more male incumbents]; F = female-dominated job classes (70percent or more female incumbents).

A scattergram is the most graphic, and the most common, method of illustrating the fact that most employers operate on a dual wage scale, a higher one for male-dominated jobs and a lower one for female-dominated jobs. The regression line drawn through the male jobs shows the mathematical average, or trend line, for those jobs, and a similar line could be drawn through the female jobs at a lower level. In the state of Minnesota, not a single female- dominated job could be found at or above the male wage trend line. This same method could be used to plot white- and minority- dominated jobs to discover racially based wage discrimination. However, because only 3.8 percent of Minnesota's state labor force are minorities (a figure slightly above the proportion of minorities in the population as a whole) an analysis by race and ethnicity is not possible from these data.

As discussed above, under a pure comparable worth policy, the wages of female- and minority-dominated jobs would be raised so that all jobs scatter around the male or white male wage line. In Minnesota, after four years during which members of female- dominated job classes have received special salary increments, the male- and female-dominated jobs now fall along a single line (see fig. 3).However, comparisons are made only between equivalent- ly valued positions. Therefore, comparable worth would not com- pare the salaries of, for example, secretaries and physicians, because a sound job evaluation system would not rate them as essentially equal in skill, effort, responsibility, and working condi- tions. Thus, while comparable worth works to achieve horizontal equity among equally valued jobs, it is not a strategy that ad- dresses hierarchical relationships in the workplace. On the con- trary, the technology of comparable worth -job evaluation -valorizes existing vertical status and pay arrangements.

Any close examination of the actual implementation of compar- able worth reveals that it is a highly technical reform. The concepts and values that underlie comparable worth are clear and easy to grasp. But the technique of job evaluation evolved in the twentieth century to rationalize the workplace and facilitate the work of managers, not to remedy historical discrimination and inequity. The result is a series of paradoxes that we would like to explore.



2.5 -F



2.2 -...r rr r. .


* F FF*.

3 2

z: .'.
.FFF F*'

F F' F

FF FL* .
1.7 xu

Tff .


1 1.6
F ZF.:

1 1111l111111~1~~~

80 100 120 140 160 180 200 220 240 260


Fig. 3. Job classes by Hay point and salary (source: Commission on the Economic Status of Women, 1985).* = male-dominated job classes (80 percent or more male incumbents); F = female-dominatedjob classes (70percent or more female incumbents).

We should begin by pointing out that comparable worth is one of many strategies for challenging inequity in a world organized by color and gender. Comparable worth is not in competition with but rather complements affirmativeaction and workplace organizing efforts. Like these other approaches, comparable worth is a long-term strategy for wage justice that requires constant political attention. But unlike other strategies, comparable worth requires long-term attention to the technocratic aspects of determiningjob worth and assigning wages to jobs with specific evaluation scores. Although there are many methods of raising wages for low-paid workers, currently there is no nontechnocratic method for implementing comparable worth.14 In this regard, comparable worth is like the Occupational Safety and Health Act (OSHA)or various types of water and air pollution prevention legislation. Because of these complexities, comparable worth is neither a single utopian answer to wage inequity nor is it dismissible because it does not do everything.

Problems arise from the fact that comparable worth, because it is technocratic in practice, intersects with and sometimes con- tradicts other democratic and egalitarian changes. Too often we forget that in solving one problem we may create others. Only by looking unflinchingly at possible conflicts may we think clearly about long-range consequences and future strategies. The three issues we will discuss are democracy, hierarchy, and jobs.

Democracy. In the first place, workplaces are not democratic, and comparable worth is not going to make them so. The exten- sive use of technical experts to implement comparable worth ex- acts a high cost both literally and figuratively. The price that con- cerns us here, however, is the erosion of an ideal of democratic and participatory decision making and the empowerment of the individual. Toward that end, many of us have worked to demysti- fy "experts" in many fields (healthcare, for example, with the self- help movement, or nuclear power). With comparable worth we find ourselves arguing for and relying on experts and also struggl- ing to demystify their expertise. Organizing for comparable worth can be both democratic and empowering as women and minorities rethink the value of their own work and challenge the hierarchy of values enshrined in the wage scale. Once employees win a com- mitment to comparable worth, however, the choice of technicians often remains in the hands of employers, and the process becomes management-dominated.

At the state level, Minnesota implemented comparable worth with amazing ease and little or no disruption of established bar- gaining relationships. This occurred for two reasons, the first of which was wholly accidental but proved crucial. The state had completed its job evaluation several years before it established its pay equity policy and distributed the first pay equity raises. Be- cause of this, pay equity was not tied to the often painful and dis- ruptive process of establishing job hierarchies. A second important condition was that all the important leaders in the implementation process understood in great detail the arguments supporters made about the structural origins of women's low wages. Most leaders in the American Federation of State, County, and Municipal Employ- ees (AFSCME) and the Department of Employee Relations whole- heartedly believed that jobs held primarily by women had low wages in part because women usually held those jobs. The com- mitment of these leaders to the underlying rationale of equal pay for jobs of equal value led to their decision to use the male pay practices line as the standard against which female-dominated jobs would be compared. When Commissioner of Employee Relations Nina Rothchild was asked why the state did not merely compute the pay practice line for all jobs and compare salaries in female- dominated jobs to it, she responded, "Iwas a math major at Smith College. I know enough not to include the salaries of women's jobs, which have been held down by discrimination, in my stan- dard for nondiscriminatory wages."15

Neither of these conditions obtained at the local level. Local jurisdictions usually had no job evaluation system, or even in many cases a job classification system, only a proliferation of job titles. In addition, because comparable worth was a state mandate, those most closely involved with implementation, both managers and employee representatives, were not strongly committed to the concept as had been the case with the state. Thus, issues of management prerogative and the process of job evaluation itself became central to local perceptions of comparable worth.

Advocates of comparable worth frequently recommend that job evaluation systems be jointly chosen and overseen by labor and management.16 Minnesota laws, however, clearly defined the choice of system as a management prerogative not subject to col- lective bargaining. In a few instances, such as in the cities of Min- neapolis and St. Paul, strong unions won some right to participate in this choice, but most managers chose simply to "meet and con- fer" with employee representatives, informing them of decisions already made.

The issue of employee participation becomes far more complex, however, once a job evaluation method has been selected. Many methods require a substantial amount of employee participa- tion- filling out questionnaires, serving on committees to evaluate jobs, or, as in the case of the University of Minnesota, even developing the factors and weights themselves. But the interpreta- tion of results is wholly managerial, and subsequent wage setting is, at best, negotiated. The final product of summed scores and re- designed job classes is often confusing and upsetting even to em- ployees active in the process.17 Regression lines, the definition of "male-," "female-," and "minority-" dominated classes with a fairly arbitrary mathematical cutoff, tend to leave people intimidated and confused. This places a special and continuing burden on pro- ponents of comparable worth, who must constantly explain both the policy and the method in order to maintain credibility, to res- pond to opponents, and to allow constituents to interact critically and knowledgeably with whatever system is in use. The fact that many local managers and politicians have chosen to use the "all jobs" pay line as their comparable worth standard signals both budgetary constraints and a lack of acceptance of the underlying premise of a comparable worth policy.

Hierarchy. Job evaluation, as a method, also exacts a price, and that price is hierarchy. Job evaluation systems evolved out of the movement to make management more rational and scientific in the early twentieth century. They presume hierarchy by providing a "scientific" method for ranking jobs from least to most valuable for the purpose of distributing pay. The factors used to rate jobs on skill, responsibility, effort, and working conditions are referred to in the literature as "compensable factors." Thus, on our way to equality, we find ourselves validating and even entrenching hier- archy as long as it is not biased by race, ethnicity, or sex. Compar- able worth policies should ensure that nurses and pharmacists are paid equivalently, but they do not raise the question of whether and how much more physicians should be paid than nurses.

Although any labor-management policy in effect accepts a hier- archical wage structure, the genuine pain experienced by many employees in the process of job evaluation and comparable worth analysis derives largely from the fact that such an analysis makes hierarchies explicit and rearranges them. Because the state of Min- nesota separated these two processes, employees experienced far less disruption. But, at the local level, comparable worth essential- ly "took the blame" for a variety of status concerns and anxieties raised by job evaluation. In part, these concerns clarify the fun- damental nature of gender in workplace hierarchies. In tiny Princeton, Minnesota, police at first refused to sign a contract when they learned about the substantial raises three clerical workers were slated to receive as a result of a comparable worth study. According to Police Association head Jules Zimmer, "we weren't aware of the large increases given to the female employ- ees."lE In St. Paul, firefighters joked bitterly that "it's very dangerous to be a librarian. Everyone knows a book could fall on your head at any time."lg

Two years later, however, St. Paul firefighters were far more upset that the comparable worth process had resulted in the loss of parity with the police. One informant told us, warning of the key status issues at stake, that "the police hate women and the fire- fighters hate the police." The bitterness of firefighters when police received a "comparable worth (really, pay-for-points) raise echoed that observation. When the city council refused to restore fire- fighters to parity with the police, the local firefighters' association president told them: "We will not go away. We'll be a wound that won't heal."20

Such issues arise in part because the technocratic nature of com- parable worth reform makes it easy to attach other, unrelated agendas to its implementation. Managers frequently see a demand for comparable worth as an opportunity to rationalize their job classification and wage system by doing an entire overhaul. They, of course, are as eager to find "overpaid as "underpaid classes and can, if they wish, use comparable worth to generate serious con- flict in an already fragmented labor force by arguing that com- parable worth really means "pay for points," a technocratic and meritocratic definition of fairness that avoids historical arguments or responsibilities.

The definition of comparable worth as pay for points not only fails to acknowledge the structural undervaluation of female and minority labor, but it also reflects the strength of opposition to any wholly or partially redistributive policy. Because redistributive policies transfervalued resources -establishing what is often viewed by participants as real, long-term winners and losers-politicians are pressured to reconfigure policies in a more distributive man- ner, where a greater number of people benefit, but often without targeting those most in need. The lack of commitment of many local managers to the goals of comparable worth, the necessity to invent the mechanisms to achieve the policy, and the dependence on outside technical experts allow these redistributive actions to take place.

Unions that encompass both female and male units can do much to educate their members about what is going on, as can managers who support the policy, but there is much room also for "divide-and-conquer" strategies that pit employees against one another. AFSCME, which represents 62 percent of Minnesota state employees, was in a position to control employee responses and keep comparable worth on track. Educational meetings em- phasized that the policy was fair, a long-overdue issue of pay equi- ty. But AFSCME conveyed this message about comparable worth far more vigorously to female-dominated units than to male- dominated units. Where male- and female-dominated jobs are represented by different unions, competition over scarce resources can easily undermine the original purpose of comparable worth, that is, the elimination of sex- and race-based wage discrim- ination. Some, such as the St. Paul police, have loudly and suc- cessfully demanded their "comparable worth based on job eval- uation studies, upsetting other male-dominated groups that could not do the same.

Jobs. The market may also exact a price in the loss and transfor- mation of jobs if we are not alert. Comparable worth is designed in part to modify wage setting on the assumption that current market wages incorporate into them past discrimination. Economist Heidi

I. Hartmann argues that comparable worth helps to correct a flaw in the market.21 Yet she and other economists also recognize that labor has always been rationed in industrial society, either through the market or through other mechanisms. Comparable worth joins a host of other policies designed to ration labor in ways the market would not. Immigration laws, for example, either en- courage or discourage the flow of workers into the country. In the latter case, the purpose is primarily to protect the jobs of workers who are already here. Similarly, debates over tariffs and trade pro- tection are, from labor's point of view, about retaining jobs in this country rather than allowing them to be exported. Plant closing laws, the minimum wage, and child labor laws all place con- straints on "free market" decisions, protecting some jobs at the ex- pense of others. In this sense, comparable worth is neither so new nor so radical as opponents would have us believe. But like these other policies it also will interact in a crowded arena with out- comes we can only partially anticipate.

Neoclassical economists predict that comparable worth will gen- erate significant unemployment among women as employers either substitute male for female workers or simply eliminate jobs. However, studies of Australia, where comparable worth was im- plemented for the entire economy in the space of a couple of years, have shown that the costs in terms of unemployment for women were quite minimal, in large part because the labor force remains gender-segregated, with few men competing for jobs traditionally held by women.22 When two economists simulated the impact of comparable worth policies on employment in U.S. state and local governments, they found the possible decline in female employment to be a "surprisingly small" 2 to 3 percent even though "our personal priors were that we would find larger esti- mates of potential job loss for females."23 (Indeed, even these un- employment estimates may be high because they are not pre- dicted on firm-level decision making about hiring and retention.)

At the state level in Minnesota, Commissioner Rothchild re- ported there has been no loss of jobs due to comparable ~0rth.24 But there may be pressures to reduce the size of work forces in some Minnesota localities, pressures that may be real or may be invented as threats in the bargaining process. Nationwide, the trends for part-time work and contracting out in the public sector have increased in the last decade. As David Lewin reported, 22 percent of state and local public employees were employed part- time in 1982, and only 7.9 percent of part-time public employees were unionized.25

Where might these pressures be felt? School boards have a num- ber of programs that are often required to be self-supporting, like latch-key programs or food preparation. Some school personnel, like a number in Stillwater, Minnesota, have indicated that they fear these programs might be eliminated or modified if the (most- ly) women who worked in them were paid higher wages. Similar- ly, city and county officials, like those in Olrnstead County, have remarked that home health workers might be vulnerable to per- sonnel cuts or contracting What is telling about these com- ments, whether they presage actual change or set the tone for labor negotiations, is that they refer to programs that have begun to respond to the changing status of women, especially the neces- sity for many families, however they are composed, to balance paid work and childrearing. In other words, local governments are reacting to comparable worth with threats to renege on the emerg- ing social commitment to policies addressing what have been traditionally defined as women's problems. Well-unionized localities would be better able to meet these threats (if they materialize) than would those without employee representation.

Many unions have faced these problems before in regard to janitorial work and garbage collection, mostly men's work. This time it is possible, however, for comparable worth raises whose beneficiaries are mostly women to become the scapegoat for un- popular management decisions to shrink the size of the labor force.

Some of the problems described above simply clarify the limits of comparable worth as a reform and in fact are problematic only if one expects comparable worth to solve problems it was never designed to address. The significance of any of these problems depends on the conditions under which a particular comparable worth policy is implemented. In Minnesota, for example, there has been a dramatic contrast between the apparent simplicity and ease of implementation at the state level and the technocratic com- plexities and conflicts in many local jurisdictions. An examination of the differences may clarify more of the strategic concerns that must inform any comparable worth campaign.

The state of Minnesota passed a law in 1982 establishing com- parable worth as a policy for state employees. Special appropria- tions in 1983 and 1985 provided funds for "pay equity increments" which were allocated through the process of collective bargaining. By 30 June 1987, the process was complete and a scattergram of female- and male-dominated job classes clustered neatly around what had formerly been the male wage line, as figure 3 shows. Several conditions contributed to this smooth process.

First, the state had a job evaluation system in place, completed by Hay Associates in 1979. Comparable worth advocates made a strategic decision that despite that system's biases (which tend to underrate typically female jobs), they would use it in their analysis and as the basis for changes in salaries. Thus, they avoided the disruption of job evaluation and the addition of managerial agen- das having to do with reclassification and rationalization of per- sonnel policies. Comparable worth in such a setting seems "cleaner," less likely to take the blame for other discontents.

Second, both managers and union representatives strongly sup- ported the policy and accepted the view that comparable worth was a necessary remedy for historic discrimination against tradi- tionally female jobs. Rothchild had been one of the prime movers behind the original legislation in her former role as director of the state Council on the Economic Status of Women. And AFSCME had been the most powerful lobby in its behalf. This basic com- mitment meant that neither side used comparable worth divisive- ly. No wages were lowered or frozen. Indeed, the separate ap- propriation of comparable worth funds left the impression that this was additional money that would not otherwise have been allocated to salaries. And the collective bargaining process remain- ed the central mechanism for allocating those funds within the established guidelines.

Finally, both sides downplayed the comparable worth process, seeking to avoid division and hostility. This was, of course, a double-edged policy. It succeeded in its purpose and was probably necessary. Yet, in a systematic telephone survey of state employ- ees conducted in 1985, we found that 43 percent of employees who actually received comparable worth raises were unaware of that fact, a finding with complicated implications for comparable worth as a policy of social transformation.27 What is the role of changed consciousness in social transformation? Are there dif- ferent long-term results in the autonomy and power of individuals who knew the origins of their comparable worth raises as well as spent them, in comparison with individuals who merely spent them?

At the local level, there was significant conflict over comparable worth. Of the 1,597 local jurisdictions in Minnesota, approximate- ly one-half used a simple job match with the state's Hay system. But most larger jurisdictions instituted new job evaluation systems. In many cases, managers, concerned with satisfying the letter of the law but unwilling to accept the underlying premise that women's historic wages are discriminatory, seized the oppor- tunity for rationalization and found ways to redistribute raises rather than augment pay levels.28 Constant discussion, with con- fusing and incomplete information, about a technology that even managers did not fully understand also fed employeesf fears and led to conflict.

Nevertheless, comparable worth does extract a price from pa- triarchy which makes it a strategy worthy of our best efforts. At the bottom line, comparable worth puts money in the hands and pockets of women and minorities who are at the lowest end of the pay scale. For example, the comparable worth analysis under- taken in Minnesota showed that the lowest-paid position in the state was Clerk 1. The implementation of comparable worth dramatically changed the compensation for this job. If comparable worth had not been implemented, the base pay of an entry-level Clerk 1 in contract year 1983-84 would have been $11,922. If, four years later, the position had received its general pay raises of $1,753 but had not received pay equity raises, the base pay of an entry-level Clerk 1 would have increased to $13,675. With com- parable worth, entry-level base pay for the Clerk 1 position actual- ly rose to $15,931 in contract year 1986-87, the year that com- parable worth was fully implemented. Of the $4,009 increase in base salary, $1,753 came from regularly negotiated raises, and $2,256 came from comparable worth raises.

Another perspective on these raises can be seen by comparing entry-level salaries for the Clerk 1 position to the poverty line for a family of four. In 1983, the poverty line for a family of four was $10,178, and the base salary for an entry-level Clerk 1 (before comparable worth increments were added) was $11,922 or 117 percent of the poverty line. If over the next four years this position had only received the pay raises negotiated between the union and the state, the base salary of $13,675 would have been 122 percent of the poverty line (which was $11,203). The actual salary of $15,931 -the salary that included the completed implementation of comparable worth raises as well as general salary raises -was 142 percent of the poverty line. Of course, not every Clerk 1 was the sole support of a family of four, but the salary change indicated the change in the capacity of people working as Clerk 1's to sup- port or help to support families.

The process of providing comparable worth raises is far from complete in Minnesota's local jurisdictions. Already it is clear that the financial benefit to workers in female-dominated classes will be muted by policies designed to contain costs. For example, in the Minneapolis School District the decision to use the pay line based on all jobs rather than one based on male-dominated jobs, com- bined with an "equity corridor" which claims that salaries within

7.5 percent above or below the line will be considered equitable, reduced the potential cost to the district by a factor of ten. At the same time, preliminary figures from a suburban school district il- lustrate the very real impact comparable worth policies can have. A Health Associate, in 1985-86, earned an entering salary of $6.21 per hour. If the associate worked full-time (forty hours per week for thirty-nine weeks), which many did not, the salary would be $9,687.60 for the year or 86 percent of the poverty line. Following the guidelines of its job evaluation and reclassification study, the district proposed that the minimum salary for health associates in

1986-87 be $7.75 per hour. That would raise a full-time associate's yearly salary to $12,090 or 108 percent of the poverty line for a family of four.

The point of our essay is that comparable worth is a long-term strategy, whose consequences, both good and troublesome, we will have to wrestle with for the foreseeable future. Comparable worth is a procedural reform-not the same as creating new in- dividual rights and leaving individuals to enforce them in the courts. Like environmental protection, where we can never stop testing air quality regardless of specific victories, comparable worth will require constant vigilance. If it validates hierarchy, we need to be thinking about other, simultaneous ways to press for participation and power. If comparable worth is incorporated in other political agendas, we need to understand the implications of those agendas and fight or ally with them consciously. If it means possible unemployment, particularly in women's occupations, we must debate among ourselves how the social costs of childcare and other traditional women's responsibilities should be borne. If com- parable worth legislation strengthens the hands of public managers, then seeking comparable worth directly through collec- tive bargaining may offer a different set of possibilities.

This prescription for vigilance, however, raises questions about how crowded the feminist agenda is and how we can gird our- selves to persevere when issues like comparable worth leave the front pages and the legislative hearing rooms. In Minnesota, the police and firefighter unions went to the legislature in the spring of 1985 to challenge and weaken the comparable worth law affecting local governments. At that time, the women's organizations that had backed the original legislation were absorbed in other issues on the presumption that the comparable worth battle had been "won." Suddenly, two committees approved the police and fire- fighters' bill, and women's groups found themselves on the de- fensive. Only strenuous efforts headed by the League of Women Voters and other women's groups saved the day. Comparable worth is not an issue that can be won as policy and forgotten in implementation.


We would like to thank Nancy Johnson for research assistance. This research was sup- ported by grants from the Northwest Area Foundation, the National Academy of Sciences, and the University of Minnesota. University support included grants from the Center for Urban and Regional Affairs, the Office of the Academic Vice-President, the Graduate School, the Hewlett Center on Conflict Resolution, and the Hubert H. Hum- phrey Institute of Public Affairs.

Eleanor Holmes Norton, quoted in Leslie Bennett's "The Equal Pay Issue," New York Times, 26 Oct. 1979, 20.

William T. Bielby and James N. Baron, "Undoing Discrimination: Job Integration and Comparable Worth," in Ingredients for Women's Employment Policy, ed. Christine Bose and Glenna Spitze (Albany: State University of New York Press, 19871, 211-29. See also Barbara F. Reskin, ed., Sex Segregation in the Workplace: Trends, Explanations, Remedies (Washington, D.C.: National Academy Press, 1984).

Donald Treiman and Heidi I. Hartmann, Women, Work, and Wages (Washington, D.C.: National Academy Press, 1981), 93.

See ibid.; U.S. Commission on Civil Rights, Comparable Worth: Issues for the Eighties, vol. 1 (Washington, D.C.: U.S. Commission on Civil Rights, 1984); Helen Remick, ed., Comparable Worth and Wage Discrimination: Technical Possibilities and Political Realities (Philadelphia: Temple University Press, 1984); Michael Evan Gold, A Dialogue on Com- parable Worth (Ithaca, N.Y.: ILR Press, 1983); Mark Aldrich and Robert Buchele, The Economics of Comparable Worth (Cambridge, Mass.: Ballinger Publishing Co., 1986); Phyllis Schlafly, ed., Equal Pay for Unequal Work (Washington, D.C.: Women's Research and Education Institute, 1984); Gender at Work: Perspectives on Occupational Segregation and Comparable Worth (Washington, D.C.: Women's Research and Education Institute, 1984); and Elaine Johansen, Comparable Worth: The Myth and the Movement (Boulder: Westview Press, 1984).

Using a natural experiment methodology, we chose local jurisdictions for careful ex- amination on the basis of type, size, and geographic locations. The University of Min- nesota's Comparable Worth Research Project collected data from local media, public documents, and interviews with key managers, elected officials, union leaders, and other employees. We also interviewed officials in the major jurisdictional professional associations (the Association of Minnesota Counties, the Minnesota League of Cities, and the Minnesota School Boards Association) to determine their influence on policy making. See Sara M. Evans and Barbara J. Nelson, "Mandating Local Change in Min- nesota: State Required Implementation of Comparable Worth by Local Jurisdictions," in Comparable Worth: A View from the States, ed. Ronnie Steinberg (Philadelphia: Temple University Press, forthcoming, 1989).

For a discussion of the passage of these laws, see Sara M. Evans and Barbara J. Nelson, "Initiating a Comparable Worth Wage Policy in Minnesota: Notes from the Field," Policy Studies Review 5 (May 1986): 849-62; Sara M. Evans and Barbara J. Nelson, "Comparable Worth for Public Employees: Implementing a New Wage Policy in Min-

nesota," in Comparable Worth, Pay Equity, andPublic Policy, eds. Rita Mae Kelly and Jane Bayes (Westport, Conn.: Greenwood Press, 19881, 191-212.

7. This definition of comparable worth draws on Treiman and Hartmann, chaps. 4 and 5; Helen Remick, "Beyond Equal Pay for Equal Work: Comparable Worth in the State of Washington," in Equal Employment Policy for Women, ed. Ronnie Steinberg-Ratner (Philadelphia: Temple University Press, 1980), 405-48; Ronnie J. Steinberg, "A Want of Harmony: Perspectives on Wage Discrimination and Comparable Worth," in Remick, Comparable Worth and Wage Discrimination, 3-27; and Barbara J. Nelson, "Comparable Worth: A Brief Review of History, Practice, and Theory," Minnesota Law Review 69 (May 1985): 1199-2000.

8. Aldrich and Buchele (p. 130 n. 3) derive the criteria for female- and male-dominated jobs in the following manner:

In 1980 civilian non-agricultural employment was 43.2 percent female. An occupation in which women were represented in proportion to their numbers in total employment would therefore employ 432 women for every 1,000 employees or 432 women for every 568 men. If women are over-

represented by a factor of three [the typical measure of overrepresentation], there would be 3 x 432

= 1,296 women for very 568 men-that is, the occupation would be 1,2961[1,296 + 568) = 70 per-

cent female. Likewise, an occupation in which men were overrepresented by a factor of three would

contain 3 x 568 = 1,704 men for every 432 women-that is, it would be 1,7041/1,704 + 432) = 80

percent male.

For research on the underpayment of equally valued jobs, see Treiman and Hartmann, 59-62; Ronnie J. Steinberg, "Identifying Wage Discrimination and Implementing Pay Equity Adjustments," in Comparable Worth: Issues for the Eighties, 99-1 16.

Heidi I. Hartmann, "The Political Economy of Comparable Worth (Paper presented at the Conference on Alternative Approaches to the Labor Market, University of Utah, Salt Lake City, 12-13 Oct. 1985).

Randall Rpley and Grace A. Franklin, Policy Implementation and Bureaucracy, 2d ed. (Chicago: Dorsey Press, 19861, 95-101, 115; Raymond A. Bauer, Ithiel de Sola Pool, and Lewis Anthony Dexter, American Business and Public Policy: The Politics of Foreign Trade (New York: Atherton, 1963); Theodore J. Lowi, "American Business, Public Policy, Case Studies, and Political Theory," World Politics 16 (1964): 677-715; Theodore

J. Lowi, The End of Liberalism: Ideology, Policy, and the Crisis of Public Authority (New York: Norton, 1969); Lewis A. Froman, Jr., "The Categorization of Policy Contents," in Political Science and Public Policy, ed. Austin Ranney (Chicago: Markham, 19681, 41-52;

E.E. Schattschneider, Politics, Pressures, and the Tariff: A Study of Free Enterprise in Pressure Politics As Shown in the 1929-1930 Revision of the Tariff (New York: Prentice- Hall, 1935) and The Semi-sovereign People (New York: Holt, Rinehart &Winston, 1960); Robert H. Salisbury, "An Exchange Theory of Interest Groups," Midwest Journal of Political Science 8 (1969): 1-32; Robert H. Salisbury and John P. Heinz, "A Theory of Policy Analysis and Some Preliminary Applications," in Policy Analysis in Political Science, ed. Ira Sharkansky (Chicago: Markham, 19701, 39-60; Michael T. Hayes, Lobbyists and Legislators: A Theory of Political Markets (New Brunswick, N.J.: Rutgers University Press, 1981), 19-39.

Ripley and Franklin, 72-73.

Janet A. Flammang, "Effective Implementation: The Case of Comparable Worth in San Jose," Policy Studies Review 5 (May 1986): 615-37; Helen Remick, "The Case of Com- parable Worth in Washington State," Policy Studies Review 5 (May 1986): 838-49.

See, for example studies in Remick's Comparable Worth and Wage Discrimination; Alice H. Cook, Comparable Worth: A Case Book of Experiences in States and Localities (Honolulu: Industrial Relations Center, University of Hawaii at Manoa, 1985); Rita Mae Kelly and Jane Bayes, eds., "Symposium: Implementing Comparable Worth in the Public Sector: Theory and Practice," Policy Studies Review 5 (May 1986): 769-870. See also Treiman and Hartmann.

The Swedish policy of solidarity wage setting, where lower-paid jobs get higher percentage raises, does address the salary problems of low-wage workers but is silent on the work value problems. See chap. 4 of Sara M. Evans and Barbara J. Nelson, Wage Justice: Comparable Worth and the Paradox of Technocratic Reform (Chicago: University of Chicago Press, 1989).

Nina Rothchild, Address to the Association of Minnesota Counties, St. Paul, 22 Jan. 1985.

See Lisa Portman, Joy Ann Grune, and Eve Johnson, "The Role of Labor," in Remick, Comparable Worth and Wage Discrimination, 219-37.

This was true in Oregon. See Joan Acker, "Comparable Worth: The Oregon Case," in Steinberg, Comparable Worth: A View from the States.

Jules Zimmer, quoted in Joel Stottrup "Equal-Pay-for-Equal-Work Idea Part of City Pay Hikes," Princeton Union Eagle, 18 Feb. 1984.

St. Paul Mayor George Latimer, panel presentation, conference on "New Directions in Comparable Worth: Minnesota and the Nation," Minneapolis, 18 Oct. 1985.

Larry Oakes, "St. Paul Firefighters Lose Pay Beef with City Council," Minneapolis Star and Tribune, 6 Aug. 1987, 5B.

Heidi I. Hartmann, "The Case for Comparable Worth," in Schlafly, 11-24.

See R.G. Gregory and R.C. Duncan, "Segmented Labor Market Theories and Australian Experience of Equal Pay for Women," Journal of Post-Keynesian Economics 3 (Spring 1981): 403-28; R.G. Gregory and V. Ho, "Equal Pay and Comparable Worth: What Can the U.S. Learn from the Australian Experience?" Centre for Economic Policy Research, Australian National University, Discussion Paper no. 123, July 1985.

Ronald G. Ehrenberg and Robert S. Smith, "Comparable Worth in the Public Sector," in Public Sector Payrolls, ed. David A. Wise (Chicago: University of Chicago Press, 19871, 279-80.

Commission on the Economic Status of Women, Pay Equity: The Minnesota Ex- perience (St. Paul: CESW, 1985), 15; also Bonnie Watkins, Pay Equity Coordinator for the State of Minnesota, "Pay Equity: The Minnesota Experience," Remarks to the Na- tional Committee on Pay Equity Senate Staff Briefing, Washington, D.C., 14 Mar. 1986.

David Lewin, "Public Employee Unionism in the 1980s: An Analysis of Transfor- mation," in Unions in Transition: Entering the Second Century, ed. Seymour Martin Lipset {San Francisco: ICS Press, 19861, 246.

Jeann Linsley, "Stillwater Schools' Equity Raises May Cost $510,000," St. Paul Pioneer Press, 9 Oct. 1985, 3NE; interview with David Griffin, Director of Personnel Ser- vices, Olmstead County, 13 Mar. 1987.

This figure is based on a random survey of state employees conducted in June 1985 by the University of Minnesota's Comparable Worth Research Project.

The latter was done by manipulating the statistical techniques used in a com- parable worth analysis. Hennepin County led the way with a policy that would 11) use the "balanced class" pay line rather than the male wage line as their goal, (2) place a 15 percent corridor on either side of that line within which wages would be considered equitable {in other words, wages could differ by 30 percent at the same point and still be nequitable"),and (3) conduct a market survey and pay no more than 10 percent above or 10 percent below the market rate regardless of what their study showed. Many jurisdic- tions also have frozen wages in higher paid occupations, offering cash bonuses instead of raises in base pay.

  • Recommend Us