LETTER (SANITIZED) FROM ROBERT M. GATES

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP89T01156R000100130003-6
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 42
• Document Creation Date: December 23, 2016
• Document Release Date: July 6, 2011
• Sequence Number: 3
• Publication Date: January 28, 1986
• Content Type: LETTER

Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
Central Intelligence Agency
28 January 1986
Palo Alto, California 94306
I have again looked at your article on Mexico and have
decided, with the Director's concurrence, that you may submit it
for publication as a"Hoover Monograph -- subject to suggested
fixes I mentioned on the telephone. Frankly, we are still a
little uncomfortable with this, as we would be uncomfortable with
any analyst writing on issues of contemporary significance.
However, given the changes you have already made and the means of
publication, we hope that any negative reaction will be
limited. Please be sure that the publication, if it identifies
you as an employee, explicitly also states that the views therein
are your own and not those of either the Agency or the US
government.
Siryeci)ely,
Robert M. Gates
Deputy Director for Intelligence
DDI/RMGates
DISTRIBUTIOiv:
0 - Addressee
1 - DD/OCR
1 - NIC Admin
1 - DDI Registry
1 - C/NIC Chrono, Watts.
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
STAT
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
Central Intelligence Agency
Office of the Deputy Director for Intelligence
17 January 1986
The DCI is thinking about letting
publish his Mexican paper in a Hoover
Press publication (see attached). This
publication has a limited dissemination
and the appropriate disclaimers would be
printed. What do you think?
Robert M. Gates
Deputy Director for Intelligence
STAT
STAT
STAT
STAT
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
Iq
Next 39 Page(s) In Document Denied
STAT
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
copyright
WOMEN AND
COMPARABLE WORTH
Rita Ricardo-Campbell
Hoover
Institution
Stanford
University
STAT
STAT
STAT
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
About the Hoover Institution
The Hoover Institution at Stanford University consists of a specialized library
and archival depository as well as a center devoted to advanced interdisciplinary
study on domestic and international affairs in the twentieth century. Since its
founding by Herbert Hoover in 1919, the Institution has become an international
center for documentation and research on problems of political, economic, and
social change throughout the world.
The Domestic Studies Program and the International Studies Program publish
not only the results of basic research but also current public policy analyses by
economists, political scientists, sociologists, and historians. Each year the Na-
tional, Peace, and Public Affairs Fellows Program provides about fifteen scholars
the opportunity to pursue advanced postdoctoral research.
The results of this research are disseminated through a variety of channels:
seminars, conferences, books published by the Hoover and other presses, journal
articles, lectures, and interviews and articles in the news media. In addition,
Hoover Institution staff members provide expert congressional testimony, consult
for executive agencies, and engage in a wide variety of other public service
activities. Some have joint appointments with Stanford University departments
and other universities, teach courses, and offer seminars.
Centrally located on the Stanford campus, the Hoover Tower and the Lou
Henry Hoover and Herbert Hoover Memorial buildings house a library of about
1.5 million volumes and one of the largest private archives in the world, consisting
of more than four thousand collections. In addition to Stanford students, faculty,
and resident staff, users of the library and archives include scholars from all over
the world who come to do research in the outstanding area collections on Africa
and the Middle East, East Asia, Eastern Europe and Russia, Latin America, North
America, and Western Europe.
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
WOMEN AND
COMPARABLE WORTH
Rita Ricardo-Campbell
Hoover Stanford
Institution University
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
The Hoover Institution on War, Revolution and Peace, founded at
Stanford University in 1919 by the late President Herbert Hoover,
is an interdisciplinary research center for advanced study on do-
mestic and international affairs in the twentieth century. The views
expressed in its publications are entirely those of the authors and do
not necessarily reflect the views of the staff, officers, or Board of
Overseers of the Hoover Institution.
Hoover Monograph Series
Copyright 1985 by the Board of Trustees of the
Leland Stanford Junior University
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted in any form or by any
means, electronic, mechanical, photocopying, recording, or oth-
erwise, without written permission of the publisher.
First printing, 1985
Manufactured in the United States of America
89 88 87 86 85 9 8 7 6 5 4 3 2 1
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
WOMEN AND
COMPARABLE WORTH
Despite the feminist movement and spread of affirmative action programs,
full-time working women have not substantially increased their earnings
relative to full-time working men in the United States. Why is this so,
especially when this sex-earnings ratio has narrowed significantly in other
industrialized countries?
The ratio of full-time working women's annual earnings relative to
men's increased to 62 percent in 1983 compared to 59 percent in imme-
diately recent years. However, that ratio has hovered around 60 percent
over the last several decades. The ratio peaked at 63.9 percent in 1956,
reached a low of 56.6 percent in 1973 and was 62 percent in 1983. A
similar ratio of median weekly earnings of experienced, full-time earners
was, during the second quarter of 1983, 66 percent, reflecting six consec-
utive years of a one percent rise. "The near stability of wage differentials
[by sex] in the U.S. ... turns out to be an exception in our [twelve
industrialized countries] international comparisons,"' states Columbia Uni-
versity professor Jacob Mincer in a 1985 article.
The use of average earnings, unadjusted for age, hides among full-time
workers that younger women earn more relative to men than do older
women. In 1983 the earnings of women 20-24 years were 78 percent of
average earnings of men who were in the same age bracket, and for those
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
25-34 years, that ratio was 73 percent.2 Younger women entering the labor
force have more education relative to men than older women in the labor
force. Women are also gaining in number of years of experience, because
most women are not leaving the labor force when they bear their first child.
The data indicate that earnings of all full-time working women will, in the
future, be rising faster than those of all full-time working men.
Moreover, if the earnings data are changed into wages per hour data
to reflect the fact that full-time working women average 9 percent fewer
hours than full-time working men, the 62 percent earnings ratio of 1983
becomes somewhat higher. If a man and a woman each earn $5 per hour,
and the man works 40 hours, he will receive $200. If the woman works 10
percent fewer hours, 36 hours, she will receive $180. There is no wage
discrimination involved, but the ratio of earnings is 90.
It is not that women work fewer hours than men on a given job for a
specific company with scheduled hours, but rather that women are concen-
trated in industries and occupations where a full-time work week is defined
as 35 hours. For example, common in clerical work is nine to five with one
hour for lunch or a 35-hour week versus the hours worked in production,
eight to four-thirty with a half-hour for lunch, or a 40-hour week.
The concept of "comparable worth" of a job apart from its market price
implies two things: that jobs have a fair or just price (an idea commonly
held in the Middle Ages for products and services) and also that it is
possible to measure the fair price or worth of jobs and thus compare them.
Adam Smith's "diamond-water paradox" illustrates the fallacy that
goods and services have an intrinsic value which should set the price that
society pays. Adam Smith asked "Why is a cup of water worth less than a
small diamond?" Water is necessary to life, but water is relatively abundant.
Diamonds are not necessary to life but they are very scarce. Similarly, a
nurse's services can be more valuable to a particular person at a specific
moment than a tree-trimmer's services. But the market price of the nurse
is determined by the value of the last nurse hired, not that of the first
nurse, or the average of all nurses hired. Nurses are hired until the last one
hired adds in value an amount equal to her additional cost. The supply as
well as the demand affects the price. But intrinsic value, just price and
comparable worth are all concepts that ignore the influence of supply on
market prices.
A common measure for the value of each job is sought under "com-
parable worth" to compare the relative values of different occupations.
Although Webster's New Collegiate Dictionary (1977 edition) defines
"comparison" in terms of "similar," "like another," and "equivalent," the
popular usage of comparable worth is to compare non-alike jobs by assigning
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
3
points for broad, common factors, such as skill, problem-solving, physical
effort, accountability and working conditions, and then totalling the
points. The broad factors chosen may not be mutually exclusive and thus
sub-factors common to more than one broad description can have a multiple
effect. If each broad factor has the same maximum number of points, then
each is artificially weighted as equal. Problems with this measure are
discussed later under the heading "job evaluation plans." To believe that
"comparable worth" is not a logical method to determine wage rates does
not rule out that discrimination exists. For example, a recent Roper Report
(83-10) asked individuals to respond to whether they think that a "specific
kind of occupation ... should be pretty much restricted to men, pretty
much restricted to women or equally open to both men and women" (my
italics). Twenty percent answered that the job of auto company executive
and 21 percent the job of electronic engineer should be restricted to men.
Thirty-eight percent believe that the job of airline pilot should be restricted
to men; 41 percent, police officer; 56 percent, construction worker; and
59 percent, firefighter. For these and other job categories persons 60 years
and older who were queried would restrict entry for women by even greater
percentages. Thirty percent would restrict the auto company executive job
and 32 percent the electronic engineer. Obviously, social acculturation acts
to bar women from entering specific jobs.3
Sex-Earnings Ratio in Other Countries
Other industrialized nations have already experienced a decreasing average
earnings differential by gender. The ratios of full-time women workers'
hourly earnings to full-time men workers' hourly earnings in manufacturing
have narrowed in Sweden and Italy to 90 and 83 percent, respectively.' In
Australia, the ratio of full-time working women's to men's earnings was
76.5 percent in 1980.5 Why do such substantially different experiences
from that of the United States occur?
In 1975, the Australian Federal Tribunal adopted a formal policy of
"equal pay for work of equal value." The Australian government sets,
through tribunals, minimum "award" rates for hundreds, even thousands,
of jobs. This is wage regulation. Australia's female to male earnings ratio
for full-time workers moved from 58 percent in 1969, when "equal pay for
equal work" was introduced, to 71 percent by 1974 and jumped to 76
percent in 1975, where the ratio hovers, even with the comparative worth
law. This ratio is very close to a projected ratio of 74 percent for the United
States by the year 2000,6 when current factors of more education, greater
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
4
commitment by women to the labor force and freer entry to male-dominated
occupations will have had a substantial effect. A correction for full-time
women workers averaging 9 percent fewer hours than full-time men workers
in the United States and 7 percent less in Australia does not substantially
reduce either country's earnings gap differential.
In Australia as in the United States, full-time, nonmanagerial women
workers work fewer hours than men: in 1980, 38.4 hours per week while
men worked 40.9 hours; the difference accounts for about 7 percent of the
Australian wage earnings gap.7 Occupational segregation continues in that
country as it does here, but there was in Australia after 1975 a noticeable
shift of women workers into government jobs and a substantial increase in
the number of female part-time workers. In some instances these two factors
are linked. For example, the government of New South Wales offers per-
manent part-time work to women returning from maternity leave. The
ratio of the hourly earnings of part-time female workers to part-time male
workers in private industry has been consistently higher in Australia than
the comparable ratio for full-time workers. The part-time hourly earnings
ratio by sex was 72 percent in the 1960s compared to 58 percent for full-
time workers. The part-time workers' earnings ratio by sex from 1975
through 1980 reached the low 90 percent range and then fell to 86 percent
in 1981.8 Ninety percent of the increase in the number of Australia's female
workers over the past two decades has been among part-time workers. In
Australia, where only 45 percent of women work, a low 16 percent of the
female labor force work part-time. Australia is a highly unionized country
and its labor unions are against extending shopping hours into the evening
and beyond Saturday mornings of the weekend. As one Australian econ-
omist has commented to me, "If women are assigned the chore of shopping
and shopping is allowed only during normal work hours, then women are
limited in their job opportunities."9 Illustrative of the perception of women's
role in the Australian economy is the denial of unemployment benefits to
women workers whose husbands are working.
In Sweden, 85 percent of women work and 55 percent of them work
part-time. United States data lie between, with 60 percent of women
working and 30 percent of these (including teenagers) working part-time.
For comparative purposes, it should be noted that primarily because of
retirement, only 78 percent of men work.
A recent narrowing of the wage gap by sex occurred in Sweden when
income taxes on a family basis were changed to an individual basis. Because
of very high marginal rates the second worker had been discouraged from
working. Today, second workers in families, and especially in those families
where the primary worker earns a middle or high income, have a strong
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
5
incentive to work, and they are entering the labor force in large numbers.
Wives of high earners usually earn more than the average for all wives.
The range of wage rates among full-time workers in Sweden is less than in
the United States and the unionization of occupations in which women
primarily work is greater. Because the U.S. income tax is based on family
not individual income, the marginal rate on earnings of the second worker
continues to be high.
In many industrialized countries recent increases in the number of
women workers have occurred primarily in part-time jobs, while in the
United States it has been mainly in full-time jobs. The reasons for this are
varied and some may be unknown. The expansion of government jobs is
greater in the welfare-state economies than in the United States and part-
time jobs are more common in their governments. The supply of full-time
women workers in Australia and Sweden has not greatly increased but that
supply exploded in the United States in response to the growing demand
by the service industries. This places downward pressure on full-time
women's earnings. In countries where the new additions to the female
labor force were primarily part-time workers the earnings of female workers
relative to males rose in both part-time and full-time worker categories.
Part-Time Work in the United States
In the United States, 32 percent of women workers (including teenagers)
in 1970, and 28 percent in the mid-1980s, work part-time and about 70
percent of them prefer part-time work. Among men 20 years and over, 8.0
percent worked part-time in 1982, compared to 4 percent in the mid-
1960s.10 Only half of male part-time workers prefer those hours to full-
time work. Among males, self-selection into part-time work may stem from
lesser human capital, education and experience, when compared to that
of male, full-time workers. In 1979 male part-time workers averaged $3.20
per hour, and full-time workers nearly double that amount, $6.25 per
hour.11 Among women, part-time workers earn about the same as full-time
workers, each close to the minimum wage. It is possible that some men
who work part-time may, irrespective of age, have lower energy levels or
more health problems than full-time male workers.
Young persons who work part-time often combine going to college
with work. The comparative number of younger men with lesser job ex-
perience may be disproportionately greater among part-time workers than
among full-time workers. That the woman who works part-time earns on
the average nearly as much as the man who works part-time does not mean
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
6
that part-time female workers earn more per hour than full-time female
workers. It primarily reflects that part-time male workers earn much less
than full-time male workers. In 1979, part-time women workers in the
United States, on the average, earned only $3.21 per hour and full-time
women workers an average (median) of $3.98 per hour.12
Part-time work in the United States does not pay well whether the
worker is a man or a woman. Part-timers are usually confined to the lowest-
paying jobs. Women who work part-time may have the same amount of
human capital as women working full-time, but many of them wish to
combine, more easily than full-time work permits, marriage, children and
a job. Comparative output per hour between part-time and full-time workers
is unknown.
Older women and older men may be more heavily represented among
part-time workers than among full-time workers. Although by 1980, 41
percent of women 55-64 years were working in the U.S. and 72 percent
of men in that age group, only 8 percent of women 65 years and older and
19 percent of men were working. 13 About one-fifth of part-timers are
"moonlighters," that is, have more than one job.
Economists Explain
Wage rates are determined by the supply of and the demand for a specific
kind of labor. The market wage rates being paid to women appear to be
high enough to attract sufficient numbers of workers. Women have entered
the labor force in record numbers during the past ten years. Yet, unem-
ployment rates among women are lower than among men. Women who
believe that they are underpaid can, and many do, seek other jobs. The
restrictions on entry to male-dominated occupations have lessened, but in
the past informal and formal barriers have kept some women out of higher-
paying jobs. To the degree that employers are more willing to invest money
to train men employees in-house rather than equally competent women
employees, the latter's promotion opportunities are limited. The effect of
past restrictions, which have greatly lessened, is still seen.
The traditional economist explains that although women earn less
than men that, in the long run, women as do men, receive the increments
in value of the product that their work produces. Because women on the
average bring less human capital in terms of education, skills and experience
to the labor force the value of their products is less. Additionally, women
work fewer hours even in full-time jobs and are concentrated in less pro-
ductive occupations than men. As women become better educated and
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
7
more enter higher-paid occupations the average wages paid full-time, work-
ing women will rise. A recent study estimates that by 2000 the sex ratio of
earnings will be at least 74 percent, reflecting the continuing increase in
the work experience of women. The estimate is conservative because it
does not reflect probable greater investment by women in market skills. 14
The marginal product line of argument assumes competitive labor and
product markets. But there are structural rigidities in labor supply that
affect the sex differential in wages. Unionization in the United States is
more prevalent in the male-dominated industries of construction, steel and
automobile production than among the female-dominated service occu-
pations. That some union wages have been artificially edged upwards above
competitive levels is supported by recent international trade data. U. S.
automobiles and steel are not competitive largely because of high unit
labor costs, both wages and fringe benefits. The male-dominated profes-
sions have, until recently, successfully discouraged entry by women, and it
is claimed that psychological barriers still exist. Hazing of women in male-
dominated occupations is still occasionally reported as being greater than
that of men. Some married women with children and some men prefer the
less demanding jobs, and thus lower-paying jobs, within a special occupa-
tion. There are no data to quantify to what degree preferences rather than
discrimination account for existing occupational segregation.
The health sector, our third largest industry, is illustrative. It has a
unique manpower ladder that restricts advances from lower to higher levels.
Educational credits earned for lower level jobs do not count towards edu-
cational requirements needed for higher level jobs. Nurses with five-year
degrees, for example, start at scratch in respect to formal education credits
if they wish to become physicians. The gap in economic rewards between
the female-dominated occupation of nursing and the male-dominated oc-
cupation of physician is very large. Physician organizations have been
labeled "guild-like" associations by some economists. While the interme-
diate area of health care occupations includes "physician assistant" (male-
dominated) and "general nurse practitioner" (female-dominated) the for-
mer, often with less training, receive more than the latter in civilian jobs,
although in the armed services this is reversed. The miliary uses what is
considered to be job evaluation. There is also historical precedent that sets
the pay of military nurses as commissioned officers and physician assistants
as non-commissioned officers. Barriers to entry have artificially reduced
the supply of women physicians. Although increased enrollments in med-
ical schools by women indicate that this is being gradually corrected,
anecdotal evidence substantiates the statement made in a study for the
Canadian Committee on the Healing Arts that "males normally dominate
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
the medical professions and seem to delegate to females jobs that require
less responsibility and competence than their training would justify.,, 15
Job Evaluation Plans
Large firms tend to use job evaluation plans that justify the structure of
rates within their companies. Many such plans compare unlike jobs. A job
evaluation plan describes factors common to all jobs and sets a range of
points for each factor. The points assigned to different factors such as
fatigue, mental composure, willingness to risk, accountability, education
required, experience, etc., involve value judgments. A job is rated by
summing up the total of points assigned to it for each broad factor, e.g.,
skill, effort, responsibility. Each broad factor usually is assigned the same
number of maximum points thus weighting each equally. This creates an
artificial scale that permits comparison of unlike jobs in terms of wage rates.
The use of points or numbers gives an unreal preciseness to job evaluations.
Individual companies set hundreds and some even thousands of wage
rates. Companies may use area wage surveys indicative of market supply
and demand levels to determine the going wage rates for different jobs in
their geographical areas. There is usually a lag between survey data and
practice. Many jobs are unique to a given company in an area and for these
jobs industry-wide surveys can be used to build rates into the wage schedule
that is loosely tied to a job evaluation plan's specific wage rates. Experience
with the number of applicants for a given job acts to check whether a wage
rate is approximately equivalent to the market rate. If too few applicants
of needed quality apply, the rate is increased; too many applicants, the rate
is reduced. In practice the market rates, not the rationalized job evaluation
rate, prevail over the long run.
Proponents of equal pay for jobs of comparable worth claim that many
job evaluation plans, by the method used to measure component factors,
favor male-dominated jobs. For example, skill is usually measured by ex-
perience. Skill could be measured by equivalent educational levels and/or
knowledge displayed by testing. Until recently most men have had more
experience than most women.
In some companies the government's Dictionary of Occupational Titles
(DOT), which is designed to be used in vocational counseling, has been
used to set relative wages. The fourth edition supplement (1982) of DOT
carries a special disclaimer: "The United States Employment Service has
no responsibility for setting wages or hours or for settling jurisdictional
matters. In preparing occupational definitions, no facts concerning such
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
9
matters were collected. Therefore, the data contained in this publication
cannot be considered as establishing standards for setting wages or hours
or for settling jurisdictional matters."16
That this formal note is considered necessary indicates that DOT has
been used to set wages and it probably influences their relative levels. DOT
assigns each position a number which indicates the difficulty or complexity
of a particular job with the lower the number the greater the complexity.
The first two numbers of DOT's three number code refer to "data" and
"people" respectively, and the third to the job's relationship to "things."
An example of the difficulties that DOT has in setting points using its
"data, people and things" criteria follows. (The lower the number the more
complex the job.) "Environmental analyst" (1982 supplement) has a "167"
rating and "parking analyst" a "261" rating. The last digit refers to "things";
a "7" is the least complex, merely "handling," while "1" is next to the most
complex, here "precision working." Both jobs are "analyst jobs" and al-
though the descriptions vary, the range between "1" and "7" for "things"
seems excessive. Obviously, job evaluation involves arbitrary decisions.
Whatever job evaluation scheme is used, the user has difficulties in assigning
points.
It is claimed that some existing government and company plans, many
originally derived from the third edition of DOT, are out-of-date. The
third edition (1965) of DOT has been charged with sex bias against
traditionally female jobs, and the fourth edition (1977) is claimed by a
National Academy Press publication to be free of sex bias. 17 However, state
and local governments and others that set rates in accordance with the
earlier edition may not have changed the structural relationship of their
wage rate structure to conform to the fourth edition.
An example of an outmoded definition of a factor is the one usually
used for "fatigue." Fatigue is often defined in terms of the physical effort
required in using the large muscles. Fatigue, however, also occurs with eye-
strain and concentration on detail. The female-dominated occupations
have a much larger component of these latter attributes than do the male-
dominated occupations.
Proponents of "equal pay for jobs of comparable worth" would weight
differently the factors so that female-dominated jobs would gain in wage
rates over those in male-dominated jobs. Because there are no fixed guide-
lines, job evaluation plans can be tailored to fit a specific goal. Hiring
experience acts to check the "correctness" of a wage rate. If, however, the
rate is too high, the excess of applicants will not necessarily set in motion
corrective factors as when the job evaluation rate is below the market rate.
The thousands of applicants who over the years have lined up for a chance
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
for a postal clerk job illustrate this. Government wage rates can be out-of-
line with market wages, but they do not fall. Higher government labor
costs result in higher taxes, not higher prices that slow demand for a
product and thus for labor. Government wage rates appear to be even more
rigid downwards than those in the private sector where labor unions may
be strong.
For example, airline companies have been lowering the salaries of their
pilots and flight attendants because their salaries are higher than what must
be paid to attract sufficient applicants of the necessary quality. However,
in many airline companies those with seniority are retaining their relatively
higher wages, creating a two-tier wage structure; apparently unequal pay
for the same work. Redefining of some jobs and seniority modifications may
rationalize this process.
The Equal Pay Act of 1963 requires employers to pay the same wage
to each person doing an identical job, but a bona fide seniority system
creates a permissible exception. Lags in enforcing the Act exist and addi-
tional lags in effectiveness occur because of the bona fide seniority system
exception. Private industry wage rates are probably more fluid downwards
than government wage rates. As society moves to an information, service-
oriented economy, claims that the existing job evaluation plans used by
various levels of government are outmoded are mounting. Women continue
to be concentrated in the service area occupations where growth is occur-
ring. The number of new types of jobs in these areas requiring new rates is
increasing. However, women are concentrated in the low-paying, com-
puter-related jobs of data entry and computer operator. Of the higher-paid
occupation of programmer one-third are women and one-fifth of the high-
est-paid, engineering-type, system analysts, are women. "The ratios of
women's to men's earnings [in these four computer-related jobs] generally
remained constant between 1970 and 1980" at 73 percent in the higher-
paying jobs and 65 percent for the others.18 Not many women take courses
in engineering.
Companies tend to promote from within and artificial barriers to in-
house training courses can limit women's opportunities. Highly unionized
industries such as steel, automobile, petroleum, and heavy equipment
manufacturing are male-dominated. Few women know about jobs in these
industries and high school counselors still do not inform them. Jobs in
these industries are customarily filled by men. Union and employer exclu-
sionary policies may be playing a decreasing role but seniority and company
pay structures reward the longtime employee while society's attitudes are
still influential. Causative factors become interrelated. For example, a
recent analysis of differences in occupational earnings, using "ridge regres-
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
sions" to reduce sensitivity to multicollinearity among the many variables,
comments that "What is striking ... is the influence of `business suit or
dress' on earnings differentials" and that there is a "fairly sizeable white
collar differential even in the presence of a host of variables designed to
capture white collar effects. It is difficult to attribute the outcome for a
variable such as this to market forces. Assuming that variables such as risk,
education, decision-making and job-related experience are capturing real
economic effects that tend to distinguish white collar from blue-collar and
other jobs, it is difficult not to reach the conclusion that the `class' influence
on white/nonwhite collar pay differentials may be sizeable. X19 The analysis
did not fully assess the influence of the "female variable." For many who
have tried to promote in a large bureaucracy a top-notch female secretary
whose work is largely administrative to an administrative assistant, this
subtle argument about the effect of `class' has overtones applicable to
earnings of women.
Governments and Comparable Worth
During the past decade a few states and some cities have been examining
their pay structures with the intent not to mesh them with market-deter-
mined rates, but to ask whether those occupations dominated by women,
pay women less than different occupations of "comparable worth" domi-
nated by men. Governments that use job evaluation plans are especially
open to challenge. Minnesota, as a result of a 1979 study, amended its
personnel law to award "pay equity" raises that average $1,600 over two
years to 8,000 employees. Moreover, that state requires all its local gov-
ernments to analyze their pay structures and correct any pay inequities
which they find. Minnesota involved trade unions in these changes, which
are expected to cost the taxpayers about $44 million. Although current
legal and union activity is confined largely to governments, over 25 states
are studying the comparative worth concept or have enacted legislation on
comparable worth. Six states have laws that either require, or their laws
have been interpreted to require, equal pay for work of comparable value.
Other states have completed their studies and one, Pennsylvania, has
proposed similar standards for private employers.' Bills and resolutions for
'Alaska, Arkansas, California, Florida, Georgia, Hawaii, Illinois, Indiana, Kansas,
Kentucky, Louisiana, Maine, Massachusetts, Maryland, Missouri, Nebraska, New Jer-
sey, New York, North Carolina, Oklahoma, Oregon, South Dakota, Tennessee, West
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
the federal government to make similar studies were introduced during the
1984 legislative sessions.
Implementation by local and state governments of "comparable worth"
designed to increase wage rates in female-dominated jobs is usually done
by increasing more the rates for those jobs than the rates for male-dominated
jobs. Personnel administrators prefer to keep everybody happy and not to
correct a perceived inequity by lowering wage rates in the other jobs, here
those held by men. Even so, correcting one perceived inequity creates new
inequities and over the years a wage structure which is thus manipulated
is likely to become increasingly out-of-line with the prevailing market wage
rates.
Private employers are disturbed by the potential spread of government
action that will interfere with the determination in the private sector of
wage rates in conformance with supply and demand. How can they pay
wages which are above the market rate? The federal government is a very
large employer. Many state governments are large enough so that their pay
schedules affect local market wage rates, for example, Sacramento, Cali-
fornia, and the indirect effect is obvious to employers. Further, under the
Davis-Bacon Act private industry holding government contracts must pay
the highest prevailing wage in a wide geographic area surrounding the
workplace.
Governments receive revenues from taxes; business from sales. When
government costs rise the demand for their services does not shrink, but
when business costs rise, either profits fall and/or their prices rise. Wage
increases to favor female-dominated jobs are unlikely to result in actual
decreases in male-dominated jobs but may result in the latter having lower
increases than would otherwise occur. In California, this has already been
occurring in a limited fashion.
A comparable worth version of affirmative action could result in fewer
but more highly paid jobs in the female-dominated occupations of the
private sector. More men will compete for these jobs and it is possible that
the percentage of such jobs held by women will fall. The number of
government jobs paid from tax revenues could remain unaffected, but more
men will apply if the wage rates are higher than supply and demand dictate.
Already the high number of applicants for a few government jobs testifies
to the fact that some government wage rates are higher than the market
requires.
Virginia, and Wisconsin are studying the concept. Connecticut, Iowa, Minnesota,
New Mexico, North Dakota and Washington are adjusting wage rates under pay-equity
plans.
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
Governments cannot ignore this political problem. Ten million women
are maintaining families today compared to 5.6 million in 1970. Sixty
percent of these women have children under 18 years of age and 60 percent
are working, nearly 80 percent of them full-time. Their earnings in March
1982 averaged $256 weekly, about 64 percent of earnings by male heads of
households. 20 Fifty-nine percent of children born last year are expected to
grow up in single-parent homes. However, many women who head house-
holds moved out of poverty and others into it during a single year as their
family structure changed.
To ignore the problem at the state level may result in lawsuits. Governor
Deukmejian vetoed in 1984 a $76.6 million comparable worth pay increase,
part of a five-year, $400 million legislative proposal and this was followed
by a class action civil rights lawsuit filed by the California State Employees
Association. About 37,000 state employees and hundreds of others who
have left the state payroll could get increases if the suit were successful.
Already California has given a ten percent increase to clerical workers and
only a five percent increase to those in male-dominated jobs during 1984.
What some males fear is already occurring: less pay in male-dominated jobs
to benefit those in female-dominated jobs.
Discrimination, Prejudice and
Statistically Significant Risk Groups
What is discrimination? Employers may attribute to a specific woman
perceived attributes of all women as a class (for example, married and in
childbearing years) which tend to lower women's marginal revenue products
and wages. A specific woman may not belong to that class, but if her pay
and promotion opportunities are limited by the perception, then that is
"discrimination." Here, discrimination exists even though each person
receives the same pay when performing the same job. Discrimination exists
because an individual is excluded from higher-paying jobs because of sex.
Artificial barriers such as rules of professional associations and trade
unions have been largely eliminated. It is impossible, however, to eliminate
fully by legal means prejudging by employers from stereotypes. There are
some academics who do not perceive this form of sex stereotyping that
limits job opportunities as "discrimination" but as "prejudice," a less derog-
atory term to describe more elusive behavior. It is agreed that stereotyping
by an easily determined characteristic, e.g., sex, can convey information
about the probable behavior of a particular person with that characteristic.
Economist Walter Williams, for example, states that:
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
in employment choice, the applicant's sex may convey information about
unobserved characteristics which pertain to the applicant's expected produc-
tivity. The employer may believe that women have higher turnover rates than
men. There are two good reasons why a woman's employment will be inter-
rupted to a greater degree than a man's. Women are more likely to bear
children than men, so that's one reason for absence from the labor force;
furthermore, women are more likely to care for babies once they are born.
Women are also more likely to move or change jobs to accommodate their
spouses' career changes. All this means that sex can be used as a valuable
proxy variable for turnover rate. Turnover rates have a lot to do with produc-
tivity.21
Although these observations may have been true prior to the 1980s
for all women on the average, they do not apply to each individual woman
and they are less true in the mid-1980s than in earlier years. The percentage
of women who do not bear children is increasing. Among women who
today bear children many do not leave the labor force. In the mid-1980s
the majority of women are combining homemaking, bearing children and
working for pay at the same time in their lives, rather than as in the 1970s,
going out of the labor force when they bore children and remaining out of
the labor force for a number of years. Additionally, family moves and
changing of jobs are usually made to accommodate the higher earner,
which increasingly may be the woman. In many families, not the majority,
the wife is the higher earner.
The traditional stereotype fits fewer women as time passes. It is se-
mantics to distinguish, as does Professor Williams, between "discrimina-
tion" and "prejudice" when the end result is the same. It is true that use of
stereotypes to convey information at minimal cost can be defended. Other
possibly more acceptable proxies in employment choices are a four-year
college degree, and for university teaching, a Ph.D. Obviously, not all
holders of these degrees are better employees for the specific job than all
nonholders of the degrees. Is the use of a college degree as a requirement
for employment acceptable because the proxy is not a physical characteristic
as is sex or race?
The concept of statistically significant risk groups with sex as the
delineating factor could, until a recent Supreme Court decision, be legally
used to set actuarial rates. However, when the stereotype of a statistically
significant risk group describes an increasingly smaller portion of the whole
group, protests against such informational shortcuts will increase.
Both "discrimination" and "prejudice" connote a degree of unfairness
because the individual is not judged on his or her merit, but on the group's
anticipated performance. Although it is illegal to discriminate by race and
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
15
sex it is not illegal in hiring to use credentials to predict performance of
an individual. The latter can be obtained by hard work, a highly regarded
value in our society. Race and sex are very difficult to change.
Occupational Segregation
It is virtually impossible to look at the distribution of women workers across
the spectrum of occupations to assess to what degree women choose to stay
out of certain occupations and to what degree prejudgments by employers
and artificial barriers have kept them out of higher-paying occupations.
Unions in the United States have been stronger in traditionally male
occupations, for example in the construction trades, than in traditionally
female occupations, for example, retail trade and clerical work. Many small
companies are prevalent in the latter areas and women tend to work for
smaller firms. Employees scattered across many firms are difficult to organ-
ize. Unions act to preserve jobs for their existing membership and this
discriminates against whoever is outside the union, whether women,
Asians, or other groups. The same has been true for professional associa-
tions. However, these barriers have greatly lessened.
Women workers are still concentrated in the relatively low-paying
service sector, including teaching, nursing, retail sales (especially apparel),
financial services and state and local governments. Seventy percent of all
jobs held by both men and women are in the service area, but not in the
same subsectors. For example, women are more likely to be elementary
school teachers; men, high school teachers. Jobs carrying the same job title,
even a very precise title, in a given industry are not always comparable.
In the manufacturing sector, women work primarily in the lower-paying
industries of apparel and leather products, not in high-paying coal and
metal mining, petroleum products or heavy construction. Manufacture of
apparel and shoes is shrinking in the United States because of effective
foreign competition stemming from lower-paid labor abroad and, some
believe, the more distinctive styling and possibly higher quality of foreign
products. As the demand for these U.S.-made products shrinks, unem-
ployment occurs and downward pressure against costs in these industries
makes these wages lose ground relative to all wages.
In the financial sector women have been moving into the higher-
paying jobs of banks, investment and insurance companies. Eventually the
average earnings of full-time working women will reflect that more women
are working in these and other higher-paid jobs. Women have just recently
been obtaining four-year college degrees at the same rate as do men and
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
women are earning nearly half of all master's and nearly one-third of
doctorate degrees. However, because women tend to take courses in the
more general areas of knowledge which may not require as much continuing
effort in order to maintain skills and possibly to gain excellence, the effect
of education on their earnings will be less than the above implies. More
women than men major in English and far more men major in engineering,
mathematics and the hard sciences for which expertise the pay is high.
The effect on earnings data of the recent substantial increase in women's
educational training and in holding higher-paying jobs than earlier is just
beginning to be recorded. In 1980, 14 percent of all lawyers were women,
13 percent of physicians and 38 percent of accountants and auditors-
much higher levels than in 1970, when 5, 10, and 25 percent, respectively
were women. In the traditionally male-dominated, executive, managerial
and administrative positions the number of women during the recent decade
has soared from 19 to 31 percent of such workers. There are very few
women CEOs and presidents of companies, but the number of women vice-
presidents has increased.
To what extent women's preferences and to what extent discriminatory
barriers are the explanation for women dominating some occupations and
having much smaller representation in others is unknown.
An analysis based on survey data22 examines a single company to
explore whether women and men in that company voluntarily sought out
different types of jobs or whether women were blocked at entry and pro-
motion from the higher-paying jobs. The unidentified Fortune 500 com-
pany, "XYZ," with about 6,000 employees, 5,500 in entry-level clerical
positions, was assessed by a consulting firm. The most pertinent fact found
was that although a relatively low proportion of women had been promoted,
"male and female clerks at XYZ were promoted in exactly the same pro-
portions as they expressed interest in promotion."23 Questionnaires revealed
that many more women than men chose not to seek promotion and some
women voluntarily chose demotion-usually because of family demands.
Consideration of women's occupational preferences, usually to better dove-
tail paid work with household and family demands, lessens the unexplained
portion of the unaccounted-for wage gap differential by sex.
A 1983 study by Randall K. Filer24 cautiously concludes, from an
analysis of one consulting firm's client data covering 8,359 men and 1,021
women, that omission of tastes and personalities has led to an overesti-
mation of the degree of discrimination against women who do not have
college schooling, and "that there remain large components of the differ-
ential in earnings between men and women" about which we know little.25
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
The men in Filer's sample earned 36 percent a month more than the
women. Filer's work supports the conventional wisdom that an additional
child in the family correlates with increased earnings by the man and
reduced earnings by the woman. The study implies that many women who
graduate from college knowingly trade off income against more congenial
work. Less educated women may not have such choices available to them.
A basic reason given for the sex-earnings gap is that women go in and out
of the labor force more than men. Women who re-enter the labor force do
not return at as high a level as they previously held, although their wages
do rebound somewhat after they have worked continuously over a period
of years. Persons with continuous work histories are more likely to have
the in-house training that opens up management jobs. Employers are less
willing to invest in training of those who, they believe, will not be steady
workers. A few employers place all women in this category. Is this sex-
stereotyping discrimination?
Full-time working women work fewer hours than their male counter-
parts. Full-time work is defined as 35 hours or more of employment for fifty
or more weeks a year, or 1,750 hours a year. In 1981, full-year, full-time,
male workers reported 163 more hours of work than full-year, full-time,
female workers.26 However, women in medicine averaged 2,050 hours and
in management, about 1,900 hours. Women may restrain their job com-
mitments in order to have more time and energy for commitments in the
home. Because the birth rate is still about 1.8 births per woman over her
lifetime as compared to 3.6 births in 1960, child care demands in the
household have fallen. Sharing between husband and wife of household
tasks appears, according to anecdotal reports, to be increasing. Addition-
ally, use of labor-saving household equipment is widespread. Washers and
dryers within the home and use of no-iron materials have greatly lessened
demands on time as also has the more recent microwave oven now owned
by nearly half of all households. With fewer household demands and greater
real wages than were paid in 1960, more women are making continuing
commitments over their lifetime to the labor force, and the wage gap
between women and men will narrow. By January 1983, the difference in
the median years of working for the current employer between men and
women was 1.8 years as compared to 2.4 years in 1968.27 Up to age 25
there is no longer any difference in seniority between men and women,
and for ages 25-29 only 0.6 years. Seniority for women lags men increas-
ingly in the older age groups being 5.1 years for ages 40-44 years and 6.9
years for ages 50-54 years.28
By 1984, 70.4 percent of women 20-29 years were working, as were
69 percent of women 30-34 years, and 70.1 percent of women 40-44
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
years. Even at ages 50-54 years, 59.4 percent of women were working in
1984.29 The bona fide seniority plan exception under the Equal Pay Act
will favor men over women far less consistently in the future than it does
now.
Not until women reach age 50 is their labor force participation rate
lower in 1984 than in 1979. Seventy percent of women age 40-44 years
in 1984 were in the labor force, whereas in 1979 only 64 percent of that
age group were in the labor force. Women who were 40-44 years in 1984
were born in the period 1941-45. In 1979, among this age cohort, then
35-39 years, only 63.4 percent worked but when they were 20-24 years
only 50 percent of them worked. Today, 70.4 percent of women 20-24
years are working. These are revolutionary changes. Younger women are
accruing seniority and human capital at nearly the same rate as younger
men.
Using late 1960s data, the 1973 Report of the Council of Economic
Advisers states that "a differential, perhaps on the order of 20 percent,
between earnings of men and women remains after adjusting for factors
such as education, work experience during the year and even lifelong work
experience. How much of this differential is due to differences in experi-
ence or in performance on the job which could not be measured adequately,
and how much to discrimination ?00 The Council did not answer. What
part may be due to differences in preferences was not addressed. The
Council's estimate of the gap reflects lower levels of education, higher birth
rates with time off from paid work as well as more rigid sex preferences
than exist in the mid-1980s. While the earnings gap has changed very
little, subsequent Council reports have not addressed this problem.
Because single women's median wage and salary income is closer to
that of single men-about 90 percent among those who never marry and
75 percent for all "unrelated" (in census terms) individuals31-it is some-
times stated that it is marriage and children and women's commitment to
the household above a paid job that basically accounts for their relatively
lower earnings.
However, "unrelated" males earned in 1981 only 57 percent of what
the male "householder" earned, while "unrelated" women averaged 85
percent of the amount earned by "female householders:' The claim that
the married male works harder because he has a family to support may be
true. But it is also probably true that never-married males on the average
have poorer health, less energy and drive and that these factors lower their
earnings.
Because all the quoted census data are unadjusted for age and occu-
pational distribution it can be safely stated only that there is little difference
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
19
in earnings between married and unmarried women, even though there is
a known sizeable difference in earnings between married and unmarried
men. The data do not support or oppose the assertion that is sometimes
made that married women work less hard than single women and single
men.
Also, there is no way of knowing the wage history and labor force
commitment over several years of the women who are working today in
comparison with those of men who are working. Comparative earnings by
last job held which are in the census data do not answer the question of
how many years a person has worked, although the data are suggestive. In
1981, about 25 percent of male workers and 32 percent of female workers
had been on their current jobs less than one year. Twenty-four and a half
percent of the men and only 16 percent of the women had been on their
current jobs more than 10 but less than 30 years.32 Women who bore
children prior to the mid-1970s usually left paid work for a few years. But
in the 1980s, this was no longer true. Labor force participation rates of
women in the mid-1980s are fairly constant at 70 percent from 20-45
years. Over 40 percent of both men and women 18-45 years had changed
jobs during the year preceding the January 1981 survey, because of the offer
of "better pay, full-time work" and about 20 percent for "other" reasons.
Differences by gender occur in the responses only in the 18-24 age group
where, for example, 15.2 percent of the men but only 9.2 percent of the
women cite "lost job, laid off" as their reason for changing jobs. 33
It may be that the overall higher mobility rate of women reflects greater
comparative upward mobility in that period as women moved into mana-
gerial jobs.
Should The Federal Government Act?
There is general agreement that differences in occupations, education,
continuity of employment, training on the job, perceived risk, working
conditions and other similar identifiable variables do not explain all of the
wage gap between women and men. However, there is also near agreement
that the unexplained remaining differential does not prove discrimination.
Women's preferences and "cultural factors" play a role but the degree of
their impact is unknown. Some argue that cultural factors per se are a form
of discrimination. That part of the wage gap may be due to discrimination
has neither been proved or disproved. Disagreement exists about whether
there is a need for the federal government to do anything beyond what it
does now and if it is believed that it should act there is no agreement on
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
what the government should do. Many states have comparable worth laws
and also many European countries. To the degree that wage rates are not
fluid downwards it may be that individuals in traditional male occupations
in those firms where longevity of service acts to create routine annual
increases are paid above the market rates. It is unlikely that wage rates paid
to women are below market rates because shortages of labor would have
forced them up, unless there are still some barriers to entry which may be
unrecognized.
Even without unassailable proof that discrimination exists, the status
quo is not politically acceptable to all. "Comparable worth" has become a
political question with emotional involvement. A legal remedy that would
not hurt employment opportunities of women is not obvious. My proposals
to continue to eliminate artificial barriers to jobs as they become known
and widen part-time job opportunities will not satisfy proponents of com-
parable worth. I know of no direct way to increase women's earnings relative
to those of men which a government can enact and which will not reduce
employment opportunities for some women. The instances where men in
male-dominated jobs who are paid more than market rates have accepted,
without demonstrable protest, lesser increases so that rates to women in
female-dominated jobs may be increased without an increase in total labor
costs are very few. The introduction of a two-tier pay system in the airline
industry indicates, however, that restraint on increases and downward
pressure on some existing wage rates exist.
The federal government has few options. It should continue to enforce
equal pay for equal work and to eliminate entry barriers to jobs and to the
required education and training for jobs.
The federal government should not be involved in setting wage rates
except for its own employees. Comparisons of wage rates in government
jobs to identical jobs in the private sector should include adjustments for
fringe benefits: health, life insurance and pension benefits. Costs of fringe
benefits are part of labor costs. Increases in fringe benefits are in lieu of
increases in wage rates and dollar amounts can be placed on them.
Enforcement of existing laws that prohibit barriers to jobs based on
sex or race needs to be continued. It is sometimes difficult to prove that
exogenous barriers exist and that apparent occupational segregation by sex
is not basically the result of women's preferences.
Economist Gary Becker writes in his well-known book, The Economics
of Discrimination, 34 that actual discrimination occurs when persons "pay or
forfeit income" for the "privilege" to discriminate. An employer who
discriminates by not hiring a woman whose addition to the product which
the employer sells would be valued as greater than the cost of hiring her,
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
loses money. This fact needs to be widely publicized. The employer who
discriminates acts "as if he incurs non-pecuniary, psychic costs of produc-
tion." He has a "taste for discrimination" because of "race, religion, sex,
color, social class, personality or other non-pecuniary considerations."35
Although an employer may lose money by discriminating against
women in promotion and hiring, the employer may simultaneously gain
from social approval by his peers, and this has a value to him. Because of
the cultural, traditional values engendered in childhood, some men do not
want to accept women at higher levels of authority, especially higher than
their own. In this regard, Gary Becker has speculated that the greater
amount of discrimination "against older and better-educated non-whites"
probably reflects the resentment of whites in lower occupations because of
the non-whites' "relatively large amount of authority and decision-making
powers" and their relatively large income. 36
Even if employers perceive that the promotion of a woman, say to
executive vice-president, is in the economic interest of their companies,
they perceive individually that they will lose not only approval by some of
their colleagues, but also by some employees at lower occupational levels.
A promotion which on the surface appears to be in the best economic
interests of the company may not occur because this subtle form of discrim-
ination can result in loss of productivity from supervised males and loss of
existing or potential customers. With the tremendous increase in the
numbers of women in high level jobs in finance and accounting and their
great visibility in managerial decision-making jobs, subtle discrimination
should be declining in importance, but it still persists.
All-male luncheon clubs are self-advertised as social in nature, but
only the naive would believe that no business arrangements result from the
informal contacts from which women are excluded. Self-employed lawyers
and real estate sales persons are only two common occupations which
benefit from such social contacts. A 1984 letter written by a male chief
executive officer reads:
Unfortunately, women are still excluded from informal social networks where
trusting and comfortable business relationships are formed. Two male execu-
tives can have a private lunch together (and even look as if they're enjoying
each other's company), an afternoon of golf, or an after-dinner drink while
strolling the lakefront at a planning retreat-without raising a single eyebrow.
Women will only be able to exercise their considerable talents at the top
when society understands that informal relationships between the sexes don't
necessarily mean sexual relationships.
I know our women managers are grateful that my wife understands the
difference. 37
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
There is subtle discrimination resulting from married, non-working
women who object to their husbands working in a day-to-day relationship
with other women on an equal basis. This non-measurable factor has often
surfaced in those jobs which require normal "sleep time on a stand-by basis"
at a place of work; for example, firemen. Although this is an extreme case,
there are nuances of this sort of discrimination scattered throughout busi-
ness, occuring at upper levels of hierarchical companies as well as at lower
levels among blue-collar workers. As the percentage of married women
who work increases and ignorance of workday activities decreases, this type
of subtle influence may decline.
These non-measurable, and sometimes subconscious forms of discrim-
ination do not always appear in the rigid market models and quantitative
analyses used by economists. Gary Becker, whose model of non-pecuniary
variables is the pioneer work in the field, believes however, that market
forces will eliminate even these forms of discrimination which still persist.
In a May 13, 1985, Business Week article, Becker states that "Between 1979
and 1984, the earnings of women compared with those of men rose at
probably the most rapid rate in our history" and this is "explained by mostly
market forces" (p. 16). Nowhere in the brief article are actual data given
and historical comparisons made. In 1956, the ratio of full-time working
women's annual earnings to men's was 63.9 and in 1983, 62.0. The ratio
rose from 59.7 in 1979. The more recently compiled series of median
weekly earnings of experienced, full-time workers reports an increase in
its ratio of one percent a year for six years reaching 66 percent in 1983.
Increases in market demand for women workers have helped absorb
the great inflow of women workers and helped push their wages up, but
early childhood acculturation of both men and women will have to change
before subtle discrimination and its effects will erode. Five years experience
is a very short period from which to draw conclusions. Government during
1979-83 probably influenced the earnings ratio by helping to remove
artificial, easily assessed barriers to entry, including barriers to the required
educational programs. There are multiple forces at work.
The more difficult cases of subtle discrimination are best left to the
courts. A recent example, which entails some of the complexities of as-
sessing an individual's quality and productivity in a professional field and
the subtle, men's club form of discrimination is briefly mentioned to illus-
trate the difficulties in proving or disproving discrimination of this nature.
Former associate professor of the Harvard Graduate School of Business,
Barbara B. Jackson, is suing Harvard University upon denial of tenure after
eleven years on the faculty. The Harvard Business School, or "B" School,
has a ten-year, not the usual seven-year, requirement for tenure. There is
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
23
currently only one woman among ninety tenured professors at the "B"
School, where I once worked many years ago, and in its 76-year history
only one other woman was a tenured professor, Henrietta Larson, who was
primarily a researcher. Any federal government interference in tenured
appointments of universities in the name of fairness should be assessed
against other worthwhile goals, such as protecting the freedom of private
university education. The courts can better handle this type of issue than
the federal bureaucracy.
Pay increases in line with seniority on the job are common in unionized
and non-unionized industry. As women commit at earlier ages to the labor
force they will, as do men, enjoy the higher wages that occur with seniority.
The recently published data (May 1985) on airline pilots' salaries, by years
of service are revealing. United Airlines pilot salaries averaged $22,452,
first year; $43,752, second year; $60,456, fifth year; $74,232, tenth year;
and $91,752, twentieth year,38 or more than four times as much as in the
first year. In this occupation, experience increases skill, but more than
four times as much seems an excessive estimate of the degree.
Although years of experience with one company probably increase an
employee's productivity, beginning workers appear to earn less in many
occupations than their marginal revenue product, while some longtime
employees appear to earn more than their marginal revenue product. Large
companies encourage their employees to retire early for this reason. Women
who do not enter or re-enter the labor force until middle age do not have
long seniority. Because bona fide seniority systems are an exception to the
Equal Pay Act, they permit unequal pay for equal work. Until women on
the average have the same seniority as men on the average, women will,
because of this exception alone, earn less than men. This does not argue
for changes in seniority rules which are best left to union/company nego-
tiations where they originated.
Sixty percent of women in the United States are working or looking
for work.34 Society cannot turn back the clock to a rural way of life as in
an agricultural society where both husband and wife worked in and about
the home. Today, the loss of income to a family is potentially so great if
the wife does not work that most wives are working.
The federal government can help working women in ways other than
setting wage rates for government jobs in order to lessen wage differences
between female and male-dominated occupations. However, despite the
U.S. Civil Rights Commission's April 1985 rejection, based on economic
reasoning, of "comparable worth" as a remedy, at least six states have been
increasing wage rates for female-dominated jobs faster than for male-dom-
inated jobs, and over twenty-five states are studying the matter of compa-
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
24
rable worth. The wait for market forces to narrow the earnings ratio is
perceived by many to be too long. This is an emotional political problem
and recent court decisions do not assure employers that they are exempt
from a comparable worth legal challenge despite the self-correcting market
forces which are at work.
The government should recognize that society has an interest in the
number of hours women work and how much they are paid. About twenty
percent of all families with children are maintained by women.40 About
half of these women work and their earnings are substantially lower than
in families maintained by men, about 80 percent of whom work. Of the
larger population of all mothers, sixty percent with children under 18 years
are working (80 percent of divorced mothers) and 71 percent of them work
full-time. 41 In 1983, among all women 25-44 years who work part-time,
54 percent "wanted [to] or could only work part-time." This compares to
15 percent of men in that age group who work part-time.42
It is reasonable to assume that there are nonworking mothers and full-
time working mothers who would prefer to work part-time if they could
find part-time work which pays the same wage rate that they could earn
doing identical work full-time. This high a wage level for part-time work
is unlikely and some trade-off between paid work in the marketplace and
time freed for other pursuits occurs. This is also true for men. The accepted
amount of wages traded for more leisure or other activities differs for
different people. Society has an interest in encouraging employers to offer
part-time jobs. Part-time work is not as readily available in the United
States as in most other industrialized countries. In some of these countries
women are a higher percentage of government employees and government
employees are a greater percentage of the labor force. This is not a solution
which I advocate.
In the 1970s the U.S. federal government encouraged implementation
of "flex-time," which permits individual scheduling but does not satisfy the
demand for part-time work. Flex-time rather than part-time job opportu-
nities develops more easily if shopping hours are extended beyond normal
working hours and Saturday mornings. Large stores in the United States
are commonly open evenings and weekend hours. Where this pattern does
not exist part-time work may be more common. Deviations from normal
working hours occur usually under pressure from labor shortages. Flex-time
redistributes within a 24-hour period the hours of paid work, but it does
not reduce the total number of the individual's hours worked. Part-time
work reduces the number of individual's hours worked and also the income
earned. Flex-time permits working mothers and others to dovetail home
duties with other family members, but the total work burden on the mother
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
25
may remain the same. Flex-time and part-time are not complete substitutes.
Society can benefit if women and older workers who might otherwise have
retired can choose from a variety of full- or part-time jobs.
The service sector of the economy is expanding and it lends itself more
easily than manufacturing to use of part-time workers. Economic incentives
in part discourage this practice because the employer has increased paper-
work and has to work out acceptable pro-rating of fringe benefits. However,
there will be a labor shortage by 1990 because new entrants to the labor
market in the 1980s will be reflecting the low birth rates from 1960 on.
Flex-time may not attract a sufficient supply of labor to satisfy the demand.
Moreover, flex-time is better fitted for some types of jobs than others. This
is also true for part-time work. Both should be available to yield an optimum
supply of labor in a period of labor shortage. Hewlett-Packard in its Palo
Alto plants continues to use flex-time, which it started in its Waltham,
Massachusetts plant in 1972, in order to attract more women workers.
Documentation of productivity per hour worked may show that in some
cases those who work part-time produce more per hour than those who
work full-time. Tax incentives might be considered to offset additional
costs, if there are any. Too little is known about part-time work. The
government could create an economic climate through provision of infor-
mation for private employers to provide part-time jobs and work-sharing
by two persons of one job.
Societal Concerns
There may be a multiple public policy payoff if this approach can be
realized. If some married mothers of young children switch from full-time
to part-time work the stability of family life probably increases. With more
part-time jobs available more older persons will work over a longer period
of their lives and lessen the drain on the social security system. Recent
surveys indicate that three-quarters of persons aQe 55 years and over also
"prefer part-time work to full-time retirement."4 Increases in life expec-
tancy make clear that retirement at age 62 or 65 may no longer be desirable
for many individuals. For the years up to age 75 or 80, most people have
years of relatively good health. At age 65, white men on the average can
expect to live until eighty years and black men to 76 years. The 1983 social
security legislation will eventually encourage later retirement and increase
the supply of part-time older workers. Older people often seek second
careers that are less demanding than their early, primary careers. Part-time
work would help to fill this need.
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89T01156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
It is in society's interest that some women who maintain households
and carry the emotional burden of day-to-day, sole decision-making in
respect to their children's lives be able to earn income sufficient for minimal
needs. Society's mores have in the past restricted job opportunities for
women. It will take years for their increasing education, greater commit-
ment to the labor force and lessening of the earlier severe restrictions to
act to increase women's earnings in the marketplace. Even then the earn-
ings gap will not disappear unless women's tastes for specific occupations
change to be almost identical to those of men. This is unlikely. The
country's future depends on its children, one-fifth of whom are being
brought up in households headed by women, half of whom work. It should
improve the fate of these children if their mothers had more choices in the
labor market.
The courts have not generally accepted an alleged "historical bias"
argument as the determinant of wage discrimination. The supply as well
as the demand for labor determines its market price and if there are no
formal entry barriers then it is presumed by many that rates are generally
what the market system dictates. Women have entered the labor force in
record numbers during the last ten years thus acting to depress their wage
rates. A recent article in the American Economic Review answered "are
workers paid their marginal products?" as follows:
For a limited and probably not very representative sample of occupations for
which coherent measures of marginal products could be devised, we may
conclude that the answer to this question seems to be no; in each instance
examined, the most productive members within an organization appear to be
paid substantially less than their marginal products while the least productive
members appears to be paid substantially more.44
Although this observation is based on a limited number of companies'
internal wage structures, it still weakens the unqualified assertion that in
normally competitive markets each worker, including each woman worker,
receives the value of his or her marginal product. Among currently working
women are many who entered the labor market in their middle years and
who, therefore, have on the average less seniority than working men.
Customary annual wage increases eventually push many individuals' wage
levels beyond the value of their marginal product. This is the primary
reason why companies encourage early retirement. This reflects a lifetime
pattern of earnings. In the early years with a company an individual's
marginal product may exceed his or her wages. This is made up for in later
years. Many currently working women will never benefit from this phenom-
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
27
enon. Younger women today are expected to have lifetime work patterns
similar to men and will benefit.
For the individual woman and for society, a major question is "where
the balance will be struck between the roles of women as wives and mothers
and women as workers and economic providers for their families.i45 Enact-
ment of government regulation of wages in line with comparable worth is
not the answer, but society cannot ignore the problem.
1. Jacob Mincer, "Inter-Country Comparisons of Labor Force Trends and of Related
Developments: An Overview" (Mimeo), pp. 26-27; Journal of Labor Economics 3,
no. 1, pt. 2 (forthcoming).
2. Ellen Sehgal, "Work Experience in 1983 Reflects the Effect of Recovery," Monthly
Labor Review 107, no. 12 (December 1984), p. 19.
3. Public Opinion (Washington, D.C.: American Enterprise Institute, February-
March 1985), p. 40.
4. Mincer, "Inter-Country Comparisons;' Table 3.
5. "Working Party No. 6 on the Role of Women in the Economy. Review of Policies
for Equality of Opportunity," National Report, Australia, November 1981. Mimeo,
p. 37.
6. James P. Smith and Michael P. Ward, Women, Wages and Work in the Twentieth
Century. (Santa Monica, Calif.: Rand, October 1984), p. 82.
7. R. G. Gregory, P. McMahon, and B. Whittingham, "Women in the Labor Force:
Trends, Causes and Consequences: Australia" (National University of Canberra,
Australia, 1984, Mimeo, forthcoming), Table 4.
8. Ibid.
9. Peter L. Swan, Professor of Management, University of New South Wales, Aus-
tralia, and Visiting Scholar at the Hoover Institution, Stanford, California, 1985.
10. U.S. Bureau of Labor Statistics, Handbook of Labor Statistics, Bulletin no. 2175,
December 1983, pp. 23, 24.
11. U.S. Department of Labor, Women's Bureau, Time of Change: 1983 Handbook on
Women Workers, Bulletin No. 298 (Washington, D.C.: Government Printing Of-
fice, 1984), p. 92.
12. Ibid.
13. Constance Sorrentino, "International Comparisons of Labor Force Participation,
1960-81," Monthly Labor Review 106, no. 2 (February 1983), p. 30.
14. Smith and Ward, Women, Wages and Work.
15. R. D. Fraser, Selected Economics of Health Care Sector in Ontario: A Study for the
Committee on the Healing Arts (Toronto, Canada, 1970), p. 226.
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
16. Dictionary of Occupational Titles, 4th ed., (Washington, D.C., 1982), p. ix.
17. Ann R. Miller et al., eds., Work, Jobs and Occupations: A Critical Review of the
"Dictionary of Occupational Titles" (Washington, D.C.: National Academy Press,
1980), p. 11.
18. Myra Strober and Carolyn L. Arnold, "Integrated Circuits and Segregated Labor:
Women in Three Computer-related Occupations," Project Report 84-A27 (Stanford
University, School of Education, November 1984, Mimeo), pp. 20, 44, 45, 46.
19. R. C. Duaffenbach and R. C. Greer, "Economic Variables and Occupational
Earnings Differentials," Journal of Economics and Business 36, no. 1 (February 1984),
p. 58.
20. Beverly Johnson and Elizabeth Waldman, "Most Women Who Head Families
Receive Poor Job Market Returns," Monthly Labor Review 106, no. 12 (December
1983), p. 31.
21. Walter Williams, "Discrimination, Productivity and Marriage," in Phyllis Schlafly,
ed., Equal Pay for Unequal Work (Washington, D.C.: Eagle Forum, 1984), pp. 96,
97.
22. Carl Hoffman and John S. Reed, "Sex Discrimination?-The XYZ Affair," Public
Interest, Winter 1981, pp. 21-39.
23. Ibid., p. 26.
24. Randall K. Filer, "Sexual Differences in Earnings: The Role of Individual Person-
alities and Tastes," Journal of Human Resources, Winter 1983, pp. 82-99.
25. Ibid., p. 97.
26. Shirley J. Smith, "Estimating Annual Hours of Labor Force Activity," Monthly
Labor Review 106, no. 2 (February 1983), p. 20.
27. Edward S. Sekscenski, Job Tenure Declines as Work Force Changes, Special Labor
Force Report 235, prepared for the U.S. Department of Labor, Bureau of Labor
Statistics (Washington, D.C.: Government Printing Office, 1980), p. 48.
28. U.S. Bureau of Labor Statistics, "Employment in Perspective: Working Women,"
Report 712, First Quarter, 1984.
29. Unpublished data, Bureau of Labor Statistics, San Francisco, 1985.
30. Annual Report of the Council of Economic Advisers (Washington, D. C.: Government
Printing Office, February 1973), p. 106.
31. US. Census, Money, Income of Households, Families and Persons in the U.S., 1981,
Series P-60, no. 137 (Washington, D.C., 1983), pp. 102, 105.
32. Francis Horvath, "Job Tenure of Workers in January 1981," in Job Tenure and
Occupational Change, U.S. Bureau of Labor Statistics Bulletin 2162, January 1983,
33. Nancy F. Rytina, "Occupational Changes and Tenure, 1981," in ibid., p. 6.
34. Gary S. Becker, The Economics of Discrimination, (Chicago: University of Chicago
Press, 1971).
35. Ibid., p. 153.
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
29
36. [bid., pp. 155, 156.
37. D. Ward Fuller, "Letters to Fortune," Fortune Magazine, May 28, 1984, p. 13.
38. San Francisco Chronicle, May 21, 1985, p. 6.
39. Sehgal, "Work Experience in 1983:' Monthly Labor Review 107, no. 12 (December
1984), p. 19.
40. U.S. Bureau of the Census, Census Report P-20, no. 388, p. 7.
41. Howard Hayghe, "Working Mothers Reach Record in 1984," Monthly Labor Review
107, no. 12 (December 1984), p. 31.
42. Ibid., p. 23.
43. "Congressional Report Proposes Republican Agenda for Women," The House
Wednesday Group, Washington, D.C., June 21, 1984, News Release, Mimeo, p.
22.
44. R. H. Frank, "Are Workers Paid Their Marginal Products?" American Economic
Review, September 1984, pp. 369, 370.
45. Suzanne M. Bianchi and Daphne Spain, "American Women: Three Decades of
Change," prepared for the U.S. Bureau of the Census, Analyses DS-80-8 (Wash-
ington, D.C., August 1983), p. 26.
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
About the Author
Economist Rita Ricardo-Campbell, Ph.D., a senior fellow of the Hoover
Institution at Stanford University, is an expert on the health care sector, the
Social Security program, and drug industry regulations. In 1981, she became
a member of the President's Economic Policy Advisory Board, and in 1982 she
was appointed a member of the National Council on the Humanities.
Dr. Ricardo-Campbell is a Director of the Watkins-Johnson Company
and of the Gillette Company, whose Finance Committee she chairs. She
served eight years as a member of the President's Citizens' Advisory Council
on the Status of Women. She has testified before the House Ways and Means
Committee and the Senate Finance and Budget Committees.
Dr. Ricardo-Campbell's latest book, The Economics and Politics of Health,
was published in 1982 by the University of North Carolina Press (second
edition, 1985). She has also written Social Security: Promise and Reality, Drug
Lag: Federal Government Decision Making, and Food Safety Regulation: A Study of
the Use and Limitations of Cost-Benefit Analysis.
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
STAT
STAT
STAT
STAT
STAT
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
HOOVER INSTITUTION
ON WAR, REVOLUTION AND PEACE
Stanford, California 94305-2323
Thanks for your recent letter
and for spending so much time consider-
ing my Mexico draft. I am disappointed
not to be able to offer it to a journal,
but I do understand your concerns and
the Director's.
If there is also no possibility
of offering it to Hoover for limited
distribution in their monograph series
(we discussed this over the phone after
you wrote), then I would of course be
pleased to have it published as a NIC
Memorandum as you proposed.
Meanwhile, best wishes for a happy
holiday season.
Sincerely,
STAT
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
HOOVER INSTITUTION
ON WAR, REVOLUTION AND PEACE
Stanford, California, USA 94305-2323
Robert M. Gates
Deputy Director for Intelligence
Central Intelligence Agency
Washington, D.C. 20505
Itiitiiiitttltlllliittltltlitiil
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
STAT
 Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6
Iq
Next 1 Page(s) In Document Denied
STAT
STAT
STAT
STAT
STAT
Sanitized Copy Approved for Release 2011/07/06: CIA-RDP89TO1156R000100130003-6