"NET ASSESSMENTS" IN NATIONAL INTELLIGENCE

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP84B00049R001403580012-3
• Release Decision: RIFPUB
• Original Classification: S
• Document Page Count: 10
• Document Creation Date: December 20, 2016
• Document Release Date: April 16, 2007
• Sequence Number: 12
• Content Type: REPORT

Approved For Release 2007/0 REYIA-RDP84B000449 001403580012-3
"NET ASSESSMENTS" IN NATIONAL INTELLIGENCE
A. Problem
1. In the debate of the last several years about "net assessments" in
national intelligence issuances, the differing views among departments and
agencies have focused on the quality and thus the value to consumers of speci-
fic types of analyses and on bureaucratic prerogatives of CIA and the Depart-
ment of Defense. Resolution of the issue has been hampered by the lack of a
common understanding among agencies of the definition of net assessments and
the relationship of such assessments to the estimative process. The purpose
of this exposition is to examine the several aspects of the issue; to delin-
eate the points of contention, providing a better basis for the continuation
of the debate; and, hopefully, to stimulate some accord about the proper
content of National Intelligence Estimates. (C)
2. Objections have been raised about two types of analyses appearing
in National Intelligence Estimates on Soviet strategic programs.
--The Presidents' Foreign Intelligence Advisory Board (PFIAB) and
the Senate Select on Committee on Intelligence characterized as mis-
leading subjective judgments about some aspects of Soviet capabilities,
such as low altitude air defense, because they gave the impression that
they were based on quantitative interaction analyses when in fact they
were not.
--Department of Defense intelligence agencies have objected to
interaction analyses involving the use of data on US forces and capabili-
ties as being "net assessments" beyond the purview of intelligence. In
practice, DoD agencies have been inconsistent. They have not taken
issue with all such interaction analyses as net assessment. They ob-
jected when they have disagreed with the substantive findings of the
assessments, or when they felt certain assessments had been given undue
emphasis.
Thus, the objection to assessments in the NIEs involves issues of both
substance and bureaucratic turf. (S)
3. Despite the contentious nature of this issue, the National Security
Council staffs of three administrations did not raise objections to any of
the assessments in the NIE. On occasion, moreover, the Intelligence Community
was tasked by the NSC staff to perform the type of assessments which DoD
agencies find objectionable. Also President Carter, indicated that he found
the interaction analysis in the NIE useful. (S)
B. Net Assessments in NIEs
4. The proscription of all "net assessments," that is, all types of in-
teraction analyses, in NIEs would be contrary to the nature of the estimating
process.
--All estimates of future developments, like forecasts of any kind,
are the end product of interaction analyses. Estimates are predictions
based on a systematic evaluation of the likely interactions of key deter-
minants of future developments. US policies, intentions, military forces
 Approved For Release 2007/04/16: CIA-RDP84B00049R001403580012-3
? SECRET ?
-2-
and capabilities are among the important determinants in the evaluation
process. The US element of the analyses may be revealed in intelligence
estimates only by oblique references or may be expressed by explicit
references to US forces.
--The estimates or judgments which result from interaction analyses
are conveyed in NIEs as quantitative expressions of probability ranging
from terms such as "almost certianly" to complex statistical presentations.
The estimating process is therefore one of narrowing uncertainty--of
assigning relative probabilities to future events from among many that
are feasible. (S)
5. As far as can be determined there are no comprehensive studies
which catalog all the analytical techniques used in the intelligence process
or which differentiate between permissible and impermissible types of estimates.
The categories of net assessments discussed below are not all inclusive of the
many hundreds of individual analyses involved in even a single NIE, but they
serve to isolate the interaction analyses that have drawn objections as "net
assessments" from those that have not.
--Political Assessments. Estimates in NIEs of the future policies
of foreign nations are the product of complex interactions with those of
the United States. The quality of a forecast will be adversely affected
if the estimators are unclear about the US side of their analytical
equation. The US element in political assessments can be highly visible
such as in an estimate of likely foreign reactions to specific US courses
of action or it may be implicit.
--Weapon Characteristics. The most prevalent assessments in military
NIEs that involve interaction between US and foreign forces are estimates
of the technical characteristics and performance of weapon systems of an
adversary. The performance of individual weapon systems can be expressed
parametrically to encompass a range of targets, but it is usually ex-
pressed in terms most meaningful to intelligence users, namely, against
known US weapons. After all, many Soviet weapon systems were designed
with US targets in mind. DoD intelligence agencies as well as CIA
perform this type of assessments.
--Subjective Assessments of Military Capabilities. National Intelli-
gence Estimates have included assessments which were based on the estima-
tors' judgments about Soviet capability to perform a specific mission,
such as defense against US low altitude bombers. Even though these type
assessments, as noted by the President's Foreign Intelligence Advisory
Board, were not based on simulations of Soviet and US force interactions,
they involved valid analytical techniques. They were based on evidence
that Soviet forces could not perform at least one of the sequential
functions required for successful mission performance. DoD intelligence
agencies have not disagreed with this type assessment.
 Approved For Release 2007/04/16: CIA-RDP84B00049R001403580012-3
? SECRET ?
-3-
--Quantified Assessments of Military Capabilities. For use in NIEs,
the CIA has conducted computer simulations of the interaction of Soviet
and US strategic nuclear forces. These analyses have been intended to
show trends in Soviet forces relevant to certain types of military mis-
sions, such as the potential of Soviet forces to destroy US ICBMs or to
survive a US counterforce first strike or the potential of Soviet civil
defense to limit damage from a US nuclear attack. These assessments
provide insights into the US-USSR strategic force relationship, but they
do not show the likely outcome of a nuclear exchange nor are they net
assessments of the strategic balance. The DoD intelligence agencies have
only objected to quantified assessments depicting the size and destructive
potential of the two sides following a counterforce attack by one side on
the other, singling out these interaction as "net assessments."
--Comprehensive Net Assessments of the Strategic Balance. The
Department of Defense conducts what are referred to as comprehensive
net assessments of the interaction of US and Soviet forces for the
purpose of measuring the capability of the two sides to conduct several
types of attacks, the damage and casualties sustained under the several
attack situations, and the residual forces of the two sides after a
full nuclear exchange. Intelligence has not conducted this type of
assessment for use in NIEs. The Soviets apparently conduct such assess-
ments to measure what they call the "correlation of strategic nuclear
forces," but intelligence has not yet been able to replicate them.,
--Net Assessments of the US-USSR Military Balance and Balance of
National Power. National Intelligence Estimates have sought to identify
the main trends in Soviet forces and capabilities for military operations
and conflicts threatening to US security and national interests. As such,
the NIEs have constituted a set of terms of reference for futher considera-
tion by national leaders responsible for security policy, but have not
contained assessments of the overall US-USSR balance of military or na-
tional power. Such assessments, embodied in the grand strategy and percep-
tions of Administrations, include judgments about US public attitudes,
politically feasible programs and policies, and other factors well beyond
the limits of national intelligence.
6. In summary, assessments in national intelligence employ analytical
models similar to those used in policymaking and defense planning, but they
differ in purpose. In concept and practice, intelligence evaluates alternative
policies, forces and conduct of foreign states using as a given, US policies,
forces and plans. Policymakers and defense planners evaluate alternative US
policies, forces and actions using as a given intelligence estimates of likely
foreign developments. Producers of national intelligence have stopped far
short of making net assessments such things as the balance of US and Soviet
military power or the NATO-Warsaw Pact military balance. The limits of assess-
ments in national intelligence has been the use of computer simulations of the
interaction of USSR and US strategic nuclear forces to measure trends in Soviet
capabilities to perform generic military missions.
Approved For Release 2007/04/16: CIA-RDP84BOO049ROO1403580012-3
 Approved For Release 2007/04/16: CIA-RDP84B00049R001403580012-3
SECRET ?
-4-
C. Genesis of the Net Assessment Issue
7. Early 1950s -- The question of the proper role of intelligence in
assessing threats to the United States is not a new bureaucratic issue. In
1951, the DCI proposed to the National Security Council that CIA prepare an
appraisal which integrated national intelligence on the USSR with military,
political and economic operational data dealing with US capabilities--in the
nature of a "commanders estimate" of the situation. The Joint Chiefs of
Staff objected on grounds that preparation of the estimate proposed by the
DCI was not a proper function of intelligence, rather, that it was within
the responsibilities of the JCS. (S)
8. Early 1960s -- During the first half of the 1960s, none of the
assessments in NIEs on Soviet strategic nuclear forces was based on the
type of computer simulations then being used for cost-effective analyses
by the Department of Defense. Intelligence estimators were informed by the
OSD cost-effectiveness analyses, but there was no analytical requirement or
demand from consumers that intelligence estimates compare US and Soviet
forces or use computer simulations to depict changes and trends in Soviet
capabilities to perform strategic missions. The lack of consumer demand for
these comparisons in NIEs was mainly because, in strategic nuclear power, the
US was widely perceived to be superior to the USSR in most measures of the
balance and was expected to remain so throughout the decade. (S)
9. Late 1960s -- Following Secretary McNamara's tenure, cost effec-
tiveness analytical techniques lost their central role in defense planning.
Instead, planners at several levels of the bureaucracy made "net assessments"
of US options for deliberation and selection at the NSC level. The term "net
assessment" cropped up in the titles of DoD positions, organizations and
defense planning studies and in justifications for contractor analyses. "Net
assessments" were not defined as any particular form of analysis; they ranged
from net assessments of policy options to net technical assessments. Despite
the proliferation of the net assessment function and the plethora of types of
analyses, there was no periodic, national level, comprehensive net assessment
of the US-USSR balance of power, of the NATO-Warsaw Pact military balance or
of the US-USSR strategic nuclear balance.
10. Early 1970s -- In the last half of the 1960s, CIA began using computer
simulations of Soviet and US force interations to evaluate the implications of
changes in Soviet forces, to quantify uncertainties, to illuminate force ef-
fectiveness and to assist in understanding Soviet perceptions. The first de-
piction in an NIE of the results of CIA's force interaction analyses appeared,
without objection from other intelligence agencies, in NIE 11-8-73 to show the
potential of Soviet ICBMs to destroy US Minuteman silos. (S)
--The "new look" in decisionmaking during the 1970s also resulted in
requirements for NIEs on Soviet military developments that were much more
detailed. They contained several alternative future force projections,
more explanation of their findings, and explicit depiction of uncertainties.
These requirements stimulated greater use of computerized force inter-
actions to assess Soviet strategic nuclear capabilities and to quantify
Approved For Release 2007/04/16: CIA-RDP84BOO049ROO1403580012-3
 Approved For Release 2007/04/16: CIA-RDP84B00049R001403580012-3
SECRET
? -5- ?
uncertainties. In part for this purpose, the DCI established the Strategic
Evaluation Center within the Office of Strategic Research, CIA.
--Also, there were cogent substantive reasons for increased CIA
use of computer simulations. With the Soviets having attained a strate-
gic posture recognized as equal to that of the US, even modest changes
in Soviet strategic programs took on greater significance. Interaction
analyses to depict the Soviets' growing threat to the US Minuteman re-
ceived more emphasis in NIEs. The analyses became more complex with
Soviet deployment of MIRVed missiles and the need for calculations of
ranges of uncertainty in NIE findings. These assessments were expected
by the users of NIEs and were not objected to by DoD intelligence agencies.
--As the Soviet potential to destroy the US Minuteman force grew,
the number of missile weapons remaining to the two sides following first
strike attacks was of increasing interest to intelligence analysts and
to NIE consumers. Accordingly, NIE 11-3/8-74 and NIE 11-3/8-75 depicted
the number of ICBM and SLBM weapons (excluding bomber weapons) remaining
to the two sides following countersilo first strikes. The analyses were
included in the NIEs without objection by Intelligence Community repre-
sentatives. (S)
11. NIE 11-3/8-76 -- The next step in assessing the implications of the
growing Soviet counterforce potential was to depict the size and composition
of all the strategic offensive forces remaining to the US and USSR--missile RVs
and bomber weapons and the equivalent megatonnage carried by these weapons--
following first strikes against offensive forces. These analyses of "residuals"
of offensive forces provided some insights about the Soviets' incentives to
strike first, their ability to maximize damage to US offensive forces and their
potential to attack other targets. For the first time, the 1976 NIE on Soviet
strategic capabilities depicted the size and composition of Soviet and US re-
maining and surviving forces following counterforce first strikes. The Senior
Intelligence Officer of the Air Force (joined by the intelligence representative
of the Department of Energy) objected to the "residual" analyses, stating that
the simulated attack used was overly simplistic, operationally unrealistic,
potentially misleading and did not belong in an NIE. (S)
12. NIE 11-3/8-77 -- For the 1977 NIE, the new DCI, Admiral Stansfield
Turner, directed that greater prominence be given to the analysis of US and
Soviet residual weapons and that measures be devised to show the destructive
potential of the residual weapons of the two sides. For this purpose, two
measures were used:
--Lethal Area Potential - a measure of the potential of the residual
weapons of the two sides to destroy soft area targets, that is, to sub-
ject targets to a given overpressure.
--Hard Target Potential - a measure of the potential of the residual
weapons of the two sides to destroy point targets of a nominal hardness.
These measures of residual weapons appeared in the summary volume of NIE 11-3/8-77
and received even more emphasis in the DCI's briefing of the US-USSR strategic
 Approved For Release 2007/04~ERECIA-RDP84B00049R001403580012-3
? + -6-
force relationship to Congressional committees. As a result, the Director, DIA
and the Senior Intelligence Officers of all the military services joined in
the Air Force position. They noted that interaction analyses falling short
of a comprehensive net assessment of a full-scale, two-sided nuclear exchange
probably did not provide useful insights into how the Soviets might perceive
the strategic balance, but that conducting such assessments was outside in-
telligence responsibilities. (S)
13. NIE 11-3/8-78 -- The DCI's emphasis on the residual analysis was
continued in the 1978 Estimate. The DoD intelligence agencies expanded their
dissent by citing the findings of a JCS study depicting the results of a
full, two-sided US-USSR nuclear exchange. The purpose was to show in the NIE
the type of net assessment the dissenting agencies regarded as the only valid
measure of the strategic balance. While the purposes and thus the results of
the NIE and JCS analyses differed, the trends depicted in the relative capabil-
ities of the two sides were consistent. The dissenting agencies continued to
hold that the residual analyses were misleading and constituted a net assessment
outside the purview of intelligence. (S)
14. NIE 11-3/8-79.
--Following publication of NIE 11-3/8-78 the DCI accepted an offer
from the Chairman, JCS, for his respresentatives to participate in a
net assessment of the US-USSR strategic balance to be conducted by the
Studies, Analysis and Gaming Agency (SAGA), Organization of the Joint
Chiefs of Staff. It was the hope of the Director, DIA, and the Chairman,
JCS, that, with the DCI's participation in the SAGA analysis, the as-
sessments in the NIE to which DoD intelligence agencies objected could be
deleted from the Estimate. The SAGA analysis did not become an agreed
DoD or national level net assessment. The DCI consented, however, to in-
clude the results of both the SAGA and the CIA analyses in the main text
of the NIE. The findings of the SAGA analysis were removed from the Esti-
mate prior to its publication because Secretary of Defense Harold Brown
informed the DCI that he did not regard the SAGA conclusions as suitable
for use in an NIE. Secretary Brown believed that force exchange calcula-
tions provided insights into the relationship in strategic forces but they
did not, by themselves, depict the strategic balance. The DoD intelligence
agencies, conforming to the desires of Secretary Brown, did not cite the
findings of the SAGA analysis in their dissenting position as they did the
previous year.
--In the Summary Volume I of the 1979 NIE, however, the DCI made the
results of the residual analysis a centerpiece. The DoD intelligence chiefs
thus lost ground in their disagreement with the DCI on the emphasis given to
the results of the residual analyses as key measures of the US-USSR strategic
relationship. Consequently, they disassociated themselves from the Summary
of the Estimate, contending that its judgments were unduly shaped by US
perceptions and strategic thinking and were not properly refective of Soviet
strategic objectives. (S)
15. NIE 11-3/8-80 -- The Summary of the 1980 NIE gave balanced coverage
to all of the important findings of the more detailed Estimate. The DoD in-
telligence chiefs did not disassociate themselves from the Summary, but they
disagreed with the DCI on which of the Estimate's findings should be included
in the Key Judgments. As a result, the DCI's version of the Key Judgments
Approved For Release 2007/04/16: CIA-RDP84B00049ROO1403580012-3
 Approved For Release 2007/04/16: CIA-RDP84B00049R001403580012-3
? SECRET ?
-7-
were followed by Key Judgments that had been coordinated among agencies of
the Intelligence Community. While the coordinated Key Judgments section
contained dissenting views, it was acceptable to DoD intelligence chiefs
because it did not contain the residual analysis as a centerpiece.
16. While the debate about net assessments had its origins in a parti-
cular type of analysis in NIEs on Soviet strategic programs, the DoD intel-
ligence agencies have applied their concept about the proper functions of
intelligence to other estimates:
--In what was probably an overzealous application of a bureaucratic
position, a DIA officer refused to provide data comparing the thickness
of the frontal armor of a Soviet T-72 tank with the penetrability of US
antitank weapons. He contended that he did not have access to "blue"
(US) data and that making such a comparison would amount to a net assess-
ment outside the responsibility of intelligence. In general, however,
DoD agencies have not objected to assessments which depict the charac-
teristics and performance of individual foreign weapon systems.
--The DoD intelligence agencies objected to the assessments ap-
pearing in NIE 11-12-80, "Soviet Military Research and Development,"
comparing Soviet and US technology applicable to military programs,
although the NIE was requested by the Department of Defense. They
refused to participate in production of an NIE on the NATO-Warsaw
Pact military balance, and insisted that US forces be omitted from an
NIE titled "The Military Balance on the Korean Peninsula." (S)
D. Summary and Recommendations
17. The history of the debate about net assessments in national intelli-
gence shows that, to reach a solution to the problem, two issues must be ad-
dressed. The first concerns DCI and CIA prerogatives for producing national
intelligence since it has been contended that net assessments are not their
proper function and should not be included in NIEs. The second concerns the
substantive value of the analyses that have been criticized as being mis-
leading net assessments. (C)
18. The Issue of DCI Prerogatives. While there are no extant statements
of policy concerning the proper content and scope of national intelligence,
the producers of NIEs have conformed to well understood limits on their pro-
ducts. National estimates on Soviet strategic programs, and, so far as can
be determined, on other subjects have not contained suggestions, recommendations
or evaluations of US policies, weapon systems, forces or plans. They have
not contained "estimates of the situation" which take into account domestic
as well as international developments. At the same time, the producers of
national intelligence have not believed they were foreclosed from using any
particular type of analysis even if it was also being used by another depart-
ment or agency. (S)
Approved For Release 2007/04/16: CIA-RDP84B00049R001403580012-3
 Approved For Release 2007/04/16: CIA-RDP84B00049R001403580012-3
0 SECRET ?
19. The argument that "net assessments" of any kind that analyze the
interaction of US and foreign policies, programs and forces, are beyond the
purview of intelligence is patently without merit.
--To proscribe these types of assessments for NIEs would be contrary
to the nature of the estimating process since all forecasts are the result
of interaction analyses in which US policies, programs and forces are
important elements. The analysis of US and Soviet residual weapons
employing computerized simulations to which DoD agencies have objected
is merely one of the many such assessments used in producing NIEs.
--The DoD agencies have been inconsistent in their position on this
issue. They have not objected on any grounds to other types of interaction
analyses appearing in NIEs on Soviet strategic programs which fall within
their definition of net assessments.
--Finally, the DoD intelligence agencies have concluded incorrectly
that the prerogatives of the DCI and CIA for conducting computerized
simulations of US and Soviet force interactions are the same as those
of departmental intelligence agencies. The responsibility for con-
ducting these types of analyses within the Department of Defense and
the Military Services has been assigned to staff organizations other
than intelligence. It does not follow that the DCI and the CIA should
be constrained in producing national intelligence by a particular alloca-
tion of staff functions which the Department of Defense finds suitable to
its needs. (S)
20. The Substantive Issue. The challenge to DCI prerogatives for con-
ducting net assessments came about during intensification of disagreements
between the DCI and DoD intelligence agencies on a substantive issue. On
substantive grounds, the objections of the DoD intelligence agencies to the
analysis of residual weapons in NIEs had greater merit than their objections
related to bureaucratic "turf." (C)
21. A further indication that the crux of the problem was substantive
was in the fact that in NIEs on Soviet strategic programs the DoD intelligence
agencies invoked their challenge to DCI prerogatives for conducting net assess-
ments only in the case of the analyses of residual weapons and destructive
potential of the two sides following first strike counterforce attacks. They
did not take issue with assessments in the NIEs depicting Soviet potential
to destroy US ICBMs, the vulnerabilities of Soviet ICBMs to a US attack, the
potential effectiveness of Soviet air defenses against a US bomber and cruise
missile attack, the potential of Soviet ASW forces to destroy US ballistic
missile submarines, or the potential effects of Soviet civil defense in pro-
tecting the USSR from a large US nuclear attack. (S)
Approved For Release 2007/04/16: CIA-RDP84B00049R001403580012-3
 Approved For Release 2007/04/16: CIA-RDP84B00049R001403580012-3
SECRET
-9- ?
22. In part, the substantive objections to the presentation of the
residual analyses in the NIEs were justified in that the calculations are
simplified and do not necessarily reflect a Soviet point-of-view. When used
as the sole basis for an assessment of a Soviet perception of its strategic
sufficiency in the late 1980s, these measures are inadequate and could be
misleading. The DoD intelligence agencies believed such assessments of the
balance would require comprehensive net assessments of a full, two-sided
US-USSR nuclear exchange. (S)
23. On the other hand, the DCI's position had merit. He pointed out
in the NIE that the purpose of the residual calculation was to put into broader
perspective some of the qualitative attributes and asymmetries which af-
fect Soviet and US capabilities and options, to illustrate trends, and to
display these trends in a manner most relevant to deterrence in its most
elementary form--that is, assured destruction. While the measures of
residual weapons were simplified ones, they depicted the same overall
trends in the relationship of US and Soviet forces as the more complex,
so called net assessments of full two-sided nuclear exchanges. Moreover,
he noted that such DoD assessments supposedly gave more attention to opera-
tional considerations than the NIE analysis, but in fact, they also used simpli-
fying assumptions to treat the most important operational variables. (S)
24. In recognition that any assessments of the present and especially
the future relationships between military forces will always be subject to
uncertainties and alternative views it would be useful to agree on certain
guidelines for the presentation of such analyses. The advice of Lieutenant
General Paul Gorman, Director of Plans of the Joint Staff, concerning intelli-
gence assessments of military force relationships would be appropriate for
adoption by the Intelligence Community:
--Compare opposites. (Compare numbers of tanks with numbers of
anti-tank weapons or numbers of bombers with numbers of air defenses.)
--Measure output rather than input. (Measure the results of inter-
action rather than the forces interacting.)
--Show trends (Show changes in numbers of forces and their capabilities
over time rather than the current situation.)
--Be selective (Look for key indicators rather than comprehensive
measures of force relationship.)
--Go Gra hic (Show visually the key findings of the analysis of force
relationships.)
--Think Soviet (Try to replicate Soviet measures of force relation-
ships.)
Approved For Release 2007/04/16: CIA-RDP84BOO049ROO1403580012-3
 Approved For Release 2007/04/16: CIA-RDP84B00049R001403580012-3
? SECRET ?
--Deal with perceptions (Try to judge how the relationship of forces
is perceived by the USSR and other nations.) (S)
25. Recommendations. The DCI should resolve the problem of net assess-
ments by enunciating policies substantially as follows:
--Affirm that net assessments of US-USSR military balances will not
appear in NIEs and reaffirm restrictions on NIEs containing suggestions,
proposals or evaluations of US policies, programs, weapons or plans.
--Acknowledge that producers of national intelligence are not fore-
closed from using any type of analyses.
--Prescribe that interaction analyses should be conducted by intelli-
gence to assist in evaluating alternative policies, forces and conduct of
foreign nations using as a given, US policies, forces and plans.
--Reserve judgment of the DCI on the appropriateness and substantive
value of the interaction analyses used in any particular NIE.
--Emphasize that the national intelligence production process permits
and encourages dissenting and alternative views competing with key findings
in NIEs, including findings based on force interaction analyses.
--Require that the Intelligence Community make an even greater ef-
fort to replicate Soviet assessments of the US-USSR force relationships.
(S)
Approved For Release 2007/04/16: CIA-RDP84B00049R001403580012-3