TESTIMONY OF JUNE O'NEILL ON THE FEDERAL EQUITABLE PAY PRACTICES ACT OF 1985 BEFORE THE SUBCOMMTTEE ON COMPENSATION AND EMPLOYEE BENEFITS, COMMITTEE ON POST OFFICE AND CIVIL SERVICES

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP87M01152R000100050011-9
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 20
• Document Creation Date: December 22, 2016
• Document Release Date: January 4, 2010
• Sequence Number: 11
• Publication Date: July 23, 1985
• Content Type: MISC

Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
Testimony of June O'Neill on The Federal Equitable Pay
Practices Act of 1985
Before the Subcommittee on Compensation and Employee
Benefits, Committee on Post Office and Civil Services
July 23, 1985
Madam Chair and members of the committee, my name is
June O'Neill. I am an economist and the Director of the
Program of Policy Research on Women and Families at The Urban
Institute in Washington, D.C. I am also a consultant for The
U.S. Commission on Civil Rights. The views that I am
expressing today, however, are my own and are not necessarily
the views of The Urban Institute or its sponsors or of the
Commission on Civil Rights.
It is an unassailable fact that in the federal workforce
women and men tend to be employed in different occupations,
that women are disproportionately concentrated in the lower
grade levels and that the average pay of women is considerably
below that of men. I believe that it would clearly be
worthwhile to investigate the reasons for these differentials.
However, while I share the broad goal of the Federal Equitable
Pay Practices Act of 1985, to conduct studies that aim to shed
light on the observed sex differentials, I do not believe that
the particular study mandated by the bill would be a
constructive step toward addressing this goal.
Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
My objections are threefold. First, the bill sets up the
expectation that the discriminatory component of the wage gap
can be measured with precision. In fact, however, such a
measurement can not be made with the existing tools of social
scientists. It would be a complete misuse of economic or job
content analysis to identify discrimination with the portion of
the wage gap that can not be explained by the crude measure of
skill and job characteristics that are currently available to
analysts.
Second, job content analysis, which is one of the two
mandated methodologies for studying sources of the wage gap,
should not be used for that purpose. Job evaluation may be
useful to management for some purposes. But it is inherently
subjective and is therefore likely to provide particularly
misleading results.
Third, the legislation sets up a partisan commission to
direct the study, thereby politicizing what should be an
objective, technical investigation.
In the testimony that follows I elaborate these points and
present suggestions for other studies that I believe would
provide useful information about the sources of pay and
employment differentials in the federal government and for
formulating policies to address them.
Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
SOURCES OF THE WAGE GAP
The wage gap - measured as the ratio of women's to men's
hourly earnings--is currently about 72 percent in.the United
States. This ratio is close to 90 percent comparing 20-34 year
olds and 80 percent at ages 25-34 years. However, when those
aged 35 and older are compared, the ratio is not much higher
than 65 percent. All of these ratios are higher than they were
in the 1970's; so there has been an impressive narrowing of the
wage gap in the past few years.
Theories
What accounts for the pay gap? Labor market
discrimination could play a role, but non-discriminatory
factors have been shown to be important. Non-discriminatory
factors would be those related to differences in worker
productivity or to differences in job characteristics.
Worker productivity is exceedingly difficult to measure
directly. A large volume of economic research has, however,
identified many of the characteristics associated with higher
wages, and, presumably, higher productivity. Many of these
characteristics involve investments in "human capital" which
improve the worker's skills and market value. These
investments include formal schooling and training as well as
the training and learning acquired on-the-job. Other
investments include the time spent searching for a job with
higher pay and migration to a different geographic region, if
that is where better opportunities are found.
Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
Human capital theory provides an explanation why women may
have lower earnings than men and pursue different occupations
than men even in the absence of any labor market
discrimination. The theory stresses that individuals make
schooling, training and other investment decisions on the basis
of their perceptions of the costs of the available options and
of the expected benefits associated with each. Women may
evaluate these options differently than men because they
anticipate spending less time over the years working at a job
outside the home. Although the average woman of today is much
more likely to work in the market after marriage than was the
case in the past, it is still atypical for a woman to work
continuously from the time of leaving school, without taking
breaks to care for children or their homes. Data from the
National Longitudinal Survey show that working women in their
forties have worked only about 60 percent of the possible years
they could have worked since leaving school.. Moreover, many
women haved worked part-time or part-year during some of the
years counted as work years. Data on job tenure from the
Current Population Survey for 1981 show that at ages 45-54
working women have accumlated only 53 percent as many years of
work with their current employer as working men.
Schooling and training are typically undertaken at younger
ages. Many women, however, have underestimated the extent to
which they will be working over their lifetimes and this factor
Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
5
may well have influenced early occupational decisions.
According to the National Longitudinal Survey, among women who
were in their early twenties in 1968, only one-third said that
they expected to be working when they reached their thirties.
Yet, when followed up by the survey ten years later, more than
60 percent of these same women were actually found to be
working.
If market work is expected to be secondary to home
responsibilities, and of uncertain duration, an occupation
requiring many years of investment in schooling and training is
less likely to be chosen. Training with carry-over to the home
would be more likely to be chosen, while training to learn a
skill which depreciates or may become obsolete during periods
of labor force withdrawal would be less likely to chosen. The
subjects that women have traditionally chosen in high school
and college reflect these factors. Subjects such as education,
nursing and home economics enhance productivity in the home as
well as in the market. Subjects such as the sciences,
engineering or business are more narrowly applicable to the
market and provide skills that depreciate rapidly when they are
not used. Studies have shown both that women's choice of
college majors is related to future career plans and that
occupational choices are related to career expectations. It
should be noted, also, that the young women of today have high
expectations of working careers and this has been reflected in
Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
6
a dramatic chance in schooling patterns. Within just the past
decade, women's enrollment in the sciences, in engineering, in
medical and law schools and in business courses has escalated
sharply.
Other aspects of investment that increase productivity and
earnings may also differ between men and women. Women may be
less willing to invest considerable resources in job search if
they are uncertain about the number of years they will work;
and women may be less free to move to-other regions if their
careers are secondary to their husbands' careers.
Since many women continue to be responsible for a
disproportionate share of household maintenance and child care
even after they enter employment, they may also evaluate
certain job characteristics differently than men and this could
have implications for the occupations and industries they work
in as well as for their earnings. For example, some women may
place a premium on a job with a work day and calendar year
corresponding to the time children are in school or on flexible
hours in more informal work settings. Jobs requiring
considerable overtime, on the other hand, would be avoided. A
shorter work day may make a long trip to work a poor
investment; women working part time may therefore be willing to
give up some pay for a job located closer to home.
Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
7
The burden of household responsibilities may affect more
than desired hours on the job. Married women holding full-time
jobs have been shown to spend about 25 hours a week on
household work--twice the hours spent by their husbands. Some
analysts believe that the energy drain of home responsibilities
make it infeasible for many women to take jobs requiring
intense commitments of time and effort. One recent study of
highly successful women executives found that this group of
women were much more likely to -be -unmarried or-childless than_
the average woman.
Empirical Evidence
A growing body of literature has investigated the effect
of women's human capital investments and other factors on
women's earnings and on the male-female pay gap. These studies
use different data sources, refer to different populations and
control for many, but not always the same, set of variables.
Even the unadjusted wage gap ---that is, the ratio of women's
to men's earnings before adjusting for skill and other
factors - varies from study to study, depending on the type of
population considered. Studies based on heterogeneous samples
including women and men of all ages and in all occupations and
industries tend to have lower ratios (larger differentials).
Studies based on more homogeneous populations tend to start out
with higher ratios even before adjustment. Thus, unadjusted
differentials among single men and women, among college faculty
or among doctorates in particular fields are relatively small.
Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87MOl152R000100050011-9
8
The proportion of the differential that is attributed to
various factors also differs from study to study. Since women
have complex and varying work histories, the adequacy of
information on work experience is crucial to the findings of
the studies. Typically, studies that do not have a direct
measure of women's work experience and try to infer it by
estimates of potential experience (e.g., by using age) find
little effect of work experience on women's earnings or on the
wage gap. When data on women's actual years of total work
experience, current job tenure and the pattern of work breaks
are used, a substantial portion of the gap is explained. In
fact, sex differences in work experience and schooling have
been found to account for roughly half of the wage gap.
Does the remaining differential reflect discrimination in
the labor market? There is simply not enough information to
answer the question. The unexplained residual will
overestimate discrimination if all the relevant variables have
not been accurately identified and measured. Clearly, many
variables of possible importance have been omitted from
analysis because the data are not availalbe. Among the
variables usually omitted are the subjects studied in school,
the extent of training outside school, or the extent of job
search. Details about actual hours worked in the past are
seldom available. Moreover, what may be the most important
factor of all--ties to home responsibilities and career
Approved For Release 2010/01/04: CIA-RDP87MOl152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
9
motivation --is difficult to quantity and has as a result been
omitted from most studies. Whenever.a researcher has been able
to introduce into the analysis a rough proxy for any of these
additional variables, an additional amount has been explained,
however. For example, one recent study finds that difference
in college major can account for one-third of the male-female
wage gap. Women who had early intentions of a work career have
also been found to have higher earnings later on than those who
expected to be homemakers but later revised their-plans and
went to work. Married women who migrated to another geographic
region to follow their husbands have been shown to have lower
earnings than would otherwise be the case.
On the other hand, some analysts have also pointed out
that variables used to measure skill may themeselves reflect
discrimination. For example, women may work fewer years
because they anticipate lower earnings due to discrimination.
Statistical efforts to account for this reverse causality have
generally not found it to be important. But these suppositions
illustrate how controversial the interpretations of the
unexplained residual can be.
To the exptent the unexplained residual does reflect
discrimination it is important to recognize that discrimination
in the labor market can take several different forms. Although
it is illegal to pay women and men different wages for equal
Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
10
quality work on the same job, it is still possible that such
cases exist, although most analysts do not seen to believe they
are widespread. A second and more likely form of
discrimination may arise in hiring and promotion. Although it
is prohibited by Title VII of the Civil Rights Act women may
have had more difficulty getting hired in certain jobs than men
or in obtaining promotions.
Studies of the wage gap have seldom been able to-include
direct measures of discrimination. Court cases have provided
evidence of unequal treatment in employment which is sometimes
found to be discriminatory and sometimes not. An article in
the Public Interest described one case involving the "XYZ
company" (so named to protect the identity of the company)
which was accused of denying women promotions on an equal basis
with men. It was found, however, that women turned down
promotions for family reasons, often because a promotion
involved moving to a different region. When women themselves
are asked if they have experienced sex disrimination, surveys
indicate that only a small fraction report that they have. For
example, 6 percent of women responding to the National
Longitudinal Survey reported having experienced-discrimination
in the previous five years attributable to their sex. These
women may have forgotten or failed to recognize discriminatory
behavior; nonetheless these data do suggest that discrimination
may not be as pervasive as some believe it to be.
Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
A third type of discrimination --the type addressed by
comparable worth policy --refers to the possibility that wages
in occupations that are predominantly held by women are lower
because employers systematically downgrade them. Moreover,
they are lower for both the men as well as the women who hold
them. This argument differs from the idea that pay in women's
occupations is depressed because of an oversupply to these
occupations. An oversupply could arise either because large
numbers of women entering the labor force choose these
occupations (which is compatible with no discrimination) or
because women are barred from entering some occupations causing
an oversupply in others (a discriminatory situation). The
argument made by supporters of comparable worth, however, is
that discrimination goes beyond restrictions on entering
certain occupations and includes the ability of employers to
pay less to women's jobs regardless of supply considerations,
simply reflecting prejudice against such jobs because they are
mostly held by women.
The ability of firms to wield such power, however, is
highly questionable. If a firm underpaid workers in women's
occupations, in the sense that their wages were held below
their real contributions to the firm's receipts, other firms
would have a strong incentive to hire workers in these
occupations away, bidding up the wages in these occupations.
Competition is a strong force curtailing employer power and
Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9 0
 Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
12
prejudice. This process could only be thwarted by collusion and
conspiracy among employers, an unrealistic prospect considering
the hundreds of thousands of firms.
Some researchers have attempted to test the validity of
this argument by adding a variable measuring the percent female
in the worker's occupation to a standard equation for
estimating the effects of different variables on earnings. The
reasoning behind this exercise is that, holding other things
the same, a finding of lower pay in typically female
occupations would be evidence that something is holding down
pay in these occupations. There are two problems with this
interpretation. One is that there is no way to distinguish
whether lower pay, if it is found, results from supply factors
or "crowding" (whether due to choice or to--discriminatory
restrictions) or whether the lower pay results from wage
discrimination. Another problem is that "other things" may not
be the same. That is, women (or men) who work in predominantly
female jobs may differ in ways that are not accounted for by
the measured variables. The findings, however, although
erratic, have generally been that while pay in typically femlae
occupations is lower, it is not much lower, once work experiene
and schooling variables are included. The variable appears to
explain only a small portion of the gender gap in wages.
Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
Implications of Economic Research for the Proposed Study
In sum, while economic analysis has proven to be valuable
in providing insights about the sources of the wage gap, it is
inconclusive about the portion that can be attributed to
discrimination because of the difficulty of quantifying many
variables believed to be important. The variables likely to be
omitted include non-discrimintory factors,--i.e., unmeasured
differences in attributes that would affect job assignment and
pay--as well as discriminatory factors. Discriminatory factors
could relate to unequal pay for equal work' to discriminatory
hiring and promotion practices' or to discriminatory wage
schedules that underpay occupations with a predominance of
women and overpay occupations held predominantly by men. The
proposed legislation has predetermined that the unexplained
residual in an analysis of the wage gap shall be deemed to be a
reflection of discriminatory pay setting practices. Yet
non-discriminatory factors may account for most of this
unexplained portion. Moreover, to the extent the residual is
due to discrimination, it may well take the form of
discriminatory hiring and promotion practices and not
discriminatory rate setting.
JOB EVALUATION
The proposed legislation mandates that a job-content
analysis also be conducted and gives equal weight to the
results of such an analysis. However, job evaluation has no
Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
14
place as a method for objectively determining the sources of
the pay gap, and it is wholly inappropriate as a tool for
determining discrimination.
There are two attributes associated with job evaluation
that particularly limit its usefulness as a tool for analyzing
sex based wage disparities. One is the inherent subjectivity
of the concept of value. The other is the confusion over the
proper role of "value" in determination of wages in a modern
market economy.
Most job evaluations attempt to rate jobs according to
their knowledge and skills, mental demands, accountability and
working conditions. These broad categories, however, have no
single objective definition. There are numerous possible
attributes that could in principle be related to skill,-"for
example, but there is no scientific way to select precisely
which attributes they should be; and once enumerated, there is
no scientific way to measure the relative importance of the
attributes. Is the knowledge of calculus worth more or less
than the knowledge of anatomy, or the knowledge of wiring an
electrical system, or the skill of typing or the knowledge of
speaking French? For the category "working conditions", is
there any single bias-free way to determine the relative value
of lifting 100 lb. weights, working with a noisy drill, working
with noisy children, cleaning a bedpan or working outdoors in
Chicago in the winter. And how then should it be determined
Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
15
whether, and by how much, skill is worth more or less than
working conditions? People, of course, may have opinions about
these matters and a committee could vote on the worth of the
various attributes. But the outcome would reflect the
subjective views of the majority of the committee and could not
in any way be regarded as objective truth.
The difficulty of interpreting the results of a job
evalution are aptly illustrated by the evaluation conducted in
Washington State. In the 1970's the State hired a
job-evaluation firm, Norman Willis Associates, to assist a
politically selected committee in rating the jobs used as
benchmarks in setting pay in State civil service employment.
The Committtee's task was to assign points on the basis of
knowledge and skills, mental demands, accountability, and
working conditions. In the 1976 evaluation a registered nurse
at level IV was assigned 573 points, the highest number of
points of any job - 280 points for knowledge and skills, 122
for mental demands, 160 for accountability, and 11 for working
conditions. A computer systems analyst at the IV level
received a total of only 426 points - 212 points for knowledge
and skills, 92 points for mental demands, 122 points for
accountability, and no points for working conditions.
These ratings likely differ radically from the values
other committes might assign. They clearly differ radically
from the market's evalution of these jobs. In the market,
Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
16
systems analysts earn about 50 percent higher pay than
registered, nurses. Such differences between the Willis
assignments and market wages are found throughout the
Washington State job evaluation. A clerical supervisor
received a higher rating than a chemist, yet the market rewards
chemists with 41% higher pay. The evaluation assigned an
electrician the same points for knowledge and skills and mental
demands as a beginning secretary and five points less for
accountability. Truck drivers were ranked at the bottom,---
receiving fewer points than telephone operators or retail
clerks. The market, however, pays truck drivers 30% more than
telephone operators and the differential is wider for retail
clerks.
The Willis study found large disparities in pay between
"male" and "female" jobs and this disparity formed the basis of
the well known case, AFSCME v. State of Washington. Should the
Willis findings be taken as evidence of discrimination and
should the state pay according to the Willis scale, or is the
Willis scale faulty?
The Willis methodology has been criticized in a recent
article by two professors who believe that the results were
subject to the biases of the evaluation committee (See the
article by David Baumer and Robert Moffie in the North Carolina
A Review of Business and Economics, Spring 1985). They noted:
"The Willis study not only used a subjective methodology, but
Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
17
also their use of statistics was primitive and transparently
biased." The state called upon another job evaluator, Paul
Jeanneret to conduct an independent analysis. The Jeanneret
study, however, found no disparity in pay among comparable male
and femele jobs. In comparing the two evaluations Prof. Arthur
Paddilla finds that: "In every instance where the Willis
method of job evalution rated the job lower than Jeanneret's,
the job was predominantly male." The Judge, however, refused
to admit the results of the Jeanneret evaluation which, in the
opinion of some experts, was more soundly based.
The two evalutions mentioned studied the same jobs and
came up with different answers. The firms involved were each
acknowledged experts. The two studies did utilize different
methods. However, widely different results can also be
produced from the same method when different job
characteristics are assigned different factor weights. the
Washington State case illustrates the virtual impossibility of
conducting a job evalaution study regarded as an objective
standard for assessing pay differences between women and men.
Although Judge Tanner who presided over the Washington
State case was willing to accept the Willis study as evidence
of discrimination, the case is on appeal. Moreover, in a more
recent case in the State of Illinois, Judge Kocoras refused to
accept the findings of a job evaluation as evidence of
discrimination. Judge Kororas' decision aptly expresses the
point:
Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
...any critieria of pay equity ultimately rests
on value judgments. Although some correlation
between sex-segregated jobs and lower wages no
doubt exists, the job characterization factor
is undeniably only one among many of the
determinatives of pay scales. Because jobs do
not have an intrinsic value that can be
scientifically measured, the limitations
inherent in job evalution techniques prohibit
the proposed extension of Title VII.
It is on this fundamental point that I depart
from Judge Tanner's reasoning in the one case,
to date, which has recognized a broad based
comparable worth claim. In AFSCME v. State of
Washington,. 578 F. Supp. 846 (D. Wash. 1983),
currently on appeal, Judge Tanner concluded
that the State of Washington violated Title VII
by compensating women employees in
female-dominted job classifications at levels
below those paid to employees in male-dominated
job classifications that had been rated
comparable in State-sponsored job evalution
studies. The State had commissioned studies
that placed numerical values on State jobs, but
no agreement had been rached, at least until
1983, that the State would adopt a new pay
scale on the basis of those numbers. To reach
the conclusion in the Washington State case,
the court had to asume that the results of the
State's job evaluation studies were valid
measurements of the relative "worth" of the
jobs in question - more valid, in fact, than
the value placed on those jobs by the market
which, the State alleged, formed the basis for
the State's existing pay scales. I, unlike
Judge Tanner, am unwilling to treat the results
of the State's job evaluation studies as the
true and reliable measurement of the inherent
worth of the surveyed jobs."
The proposed legislation, H.R. 3008, mandates that a job
evaluation be conducted under the aspices of a commission that
appears partisan in its makeup. Because of the inherent
subjectivity of the job evaluation methodology, the results
will not be bias-free, but will instead reflect the opinions of
the commission and the consultants chosen to execute the study.
Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
ALTERNATIVE STUDIES
As economic analysis of the earnings of male and female
federal employees might provide considerable insight into the
factors underlying the differential. It would be important, of
course, to obtain detailed information on work histories prior
to federal service' on years of federal service, years of
schooling, college major, and additional training. It would
also be useful to try to incorporate information on job
characteristics. Although the bill specifies that the studies
should use occupations as the unit of analysis, it would be
important to conduct an economic analysis based on individual
workers. No matter how detailed the analysis, it is likely
that a portion of the wage gap will remain unexplained. This
unexplained residual should be taken for what it is--not as
evidence of discrimination.
There are two other types of studies that I believe could
provide useful insights into the treatment of men and women in
federal employment. One study would use survey data to compare
the pay that workers with similar characteristics receive in
the private sector, in state and local employment and in
federal employment, to determine if the federal government is
paying on a different basis for men and for women, and if those
in female or male dominated occupations are rewarded similarly
in federal employmen relative to the market.
Approved For Release 2010/01/04: CIA-RDP87M01152R000100050011-9
 Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9
Another study would examine why women and men are in
different grades and in different occupations in federal
employment. Such a study might help in identifying whether the
federal govenment does in fact erect barriers to qualified
women with respect to entering male dominated jobs such as the
construction crafts, engineering, or the law. Are women with
the relevant skills who apply for these jobs steered into
clerical work? Are women who desire promotions to higher level
management posts denied equal treatment? A.carefully designed
questionnaire might provide information that would help
determine whether the observed employment patterns reflect
discrimination by the government or the choices and preparation
of male and female employees. Certainly the policy remedies
undertaken should be guided by studies such as these if they
are to address real problems of unequal opportunity and not
simply fulfill the policy agenda of those who seek to legislate
equal outcomes for women and men.
Approved For Release 2010/01/04: CIA-RDP87MO1152R000100050011-9