Production of National and Other Intelligence Estimates

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP78Z02997A000100180003-7
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 7
• Document Creation Date: December 9, 2016
• Document Release Date: August 31, 1999
• Sequence Number: 3
• Publication Date: November 4, 1975
• Content Type: MF

Approved F4elease 2001/08/25: CIA-RDP78Z02?iA000100180003-7
o 4 0V 97:
DCI/IC-75-39L11
MEMORANDUM FOR: Director of Central Intelligence
SUBJECT Production of National and Other
Intelligence Estimates
1. Problem
How can national and other intelligence estimates
best be produced.
2. Position of Congressional Committees
a. The Senate Select Committee Staff has been.
reviewing NIEs over a long period, and exploring the
history and problems of the production-of estimates.
Several former ranking CIA officials (e.g., John McCone)
have argued for a BNE in interviews with the Senate .
Staff. Senator Mondale has publicly argued for recreating
the BNE, reflecting some Senate Staff sentiments.
b. Position of the House Select Committee is not
known. The inference from numerous questions addressed
to General Wilson, Mr. Hyland and Mr. Clark was that
several members admire the old-style NIEs and methods
of producing them, believe the NIOs are too engaged in
day-to-day business for systematic long-term thinking,
are skeptical that NIOs have sufficient authority in
the Agency/Community to get the job done, and wonder
about their ability to withstand policy pressures on
estimates.
c. On the other hand, two House Staffers who
have been briefed on the NIO system and are studying
estimates have professed themselves impressed with it
and with the steps being taken to improve it.
Approved For Release 2001/08/25: CI,A-RDP78
2,..IA?^i1_T CL F3Y.2.. 6 3 2
.
--. t
 Approved Fd elease 2001/08/25 CIA- DP78Z0299 000100180003-7
3. Discussion of Options
a. This question tends to be debated in either/or
terms, i.e., "Are NIOs better than BNE or visa versa?"
This is probably a false dichotomy, certainly an
unnecessarily restrictive one. One may wish to choose
either an NIO or a BNE system; one does not have to.
In terms of symbols, there is a tendency on the Hill
and in parts of press to glamorize the BNE as synonymous
with objectivity and wisdom, NIOs with the obverse.
Whatever his personal inclinations, no knowledgeable
person thinks in these simplistic terms. The matter
is further complicated by charges (Mondale speech) that
Nixon caused the abolition of BNE for ideological reasons--
charges which are not true.
b. President Nixon had nothing to do with the
abolition of the BNE. Ideology was not a factor in the
decision to replace the Board with the NIOs, nor was any
alleged dissatisfaction with the content of the BNE's
analytic judgments. The schematic outlines of the NIO
system were * developed by Dr. Schlesinger while he was
DCI and conceptually elaborated by Mr. Colby after
Dr. Schlesinger's departure.
c. Mr. Colby felt he needed an NIO structure to
assist him in the discharge of his substantive responsibilities
as head of the Intelligence Community. He was sensitive to
the view widely (often strongly) held outside of CIA that
under the BNE/ONE system, the drafting of national estimates
was .a monopoly closely guarded as such by a single Community
component.
d. At the time Mr. Colby decided he needed an NIO-
type structure, the BNE was under-strength. The DCI decided
not to appoint new members to the BNE and thus risk estab-
lishing two entities with overlapping functions but, instead,
to fold the NIE production supervision function into his
new NIO concept, hoping that the new structure could also
serve as a device for doing a number of things the BNE had
never been asked to do and a device for knitting the Community
together.
e. In actuality, therefore, the NIOs were set up in
1973 to perform a cluster of functions deriving from the
DCI's mandate to lead the Community and from Mr. Colby's
own felt need for close substantive support in certain
geographical and functional areas. The NIOP responsibilities
include production of NIEs and other estimates, but many
other things as well. Their NIE responsibilities were
assigned partly to offset charges that ONE had too monopo-
list -on Por el st f s,1t8ht4CIA-RDP78Z02997A000100180003-7
 Approved Fw Release 2001/08/25 : CIA-RDP78Z0 'A000100180003-7
f.. The BNE was established in 1951 to produce
National Estimates period. As it evolved, and as it
developed a sophisticated staff, it broadened its
activities to include a variety of analytical and
estimative functions and speculative work in addition
to NIEs--both coordinated within the Community and
unilaterally for the DCI or at his request.
g. Thus, NIO and ONE functions overlap in part
but only in part.
(1) NIOs do a variety of staff support jobs
for the DCI, look at substantive affairs from his
point of view, are active in consumer relations and
in grey area between substance and management (KIQ-
KEP, etc.). The NIOs and their assistants are
specialists in their subjects of responsibility--
through broad-gauged specialists. They employ
regular production offices in CIA and elsewhere
to get jobs done. Personnel are drawn from senior-
and middle-level substantive experts of CIA, State
and the services, with some infusion of talent from
outside Government.
(2) ONE was based on the concept of a small group
of very senior generalists, civilian and military,
removed from current and administrative responsibilities,
charged with thinking ahead substantively. They served,
individually and collectively, as the DCI's executive
agents for producing National Estimates and other
estimative products. BNE was supported by a small
substantive staff which did the drafting, helped by
departmental contributions. As ONE evolved, relative
importance of staff grew.
h. Whether considered empirically in the light of
experience or conceptually, the NIO and BNE systems each
show a mixture of advantages and disadvantages, with
preponderance depending on which cluster of functions
one considers most important.
(1) Little doubt exists in light of two
years' experience, that the DCI, charged with
his present mandate to lead the Community, needs
something like NIOs as substantive staff officers.
He would probably need them all the more if he or
a super-senior intelligence officer receives a
strengthened mandate for leadership of the Community.
NIOs, as a channel between policy consumers and
intelligence producers, have also proved their
usefulness in making intelligence more responsive to
consumer needs and in helping the DCI prioritize
Community's substantive effort.
Approved For Release 2001/08/25: CIA-RDP78ZO2997A000100180003-7
 Approved P[elease 2001/08/25 CIA-RDP78Z02+1`A000100180003-7
(2) There is also little doubt, in light of
experience, that NIOs do not have enough time and
staff support to deal systematically with long-term
problems. They have had difficulty getting quality
estimates drafted--especially when problems cut
across geographic specialities, when outside agencies
are asked to do the job, or when time pressures are
urgent. The press of daily responsibilities is not
compatible with the sustained attention needed for
thoughtful long-term estimating. Sustained proximity
to high policy makers could interface with rigorous
independence and objectivity. (In practice, there
has been little evidence of improper policy pressures
and NIOs independence has been vigorously asserted
when the need arose.)
(3) In summation, NIOs have clarly proven
useful to the DCI and have clearly also helped in
fine-tuning the intelligence effort to policy makers'
needs. These improvements have been made at some
cost in systematic provisions for long-run, pressure-
free estimating and for handling problems across
geographic and functional lines.
The Range of Options
a. Conceptually, options range from the extreme
of keeping the present NIO system without a staff to
abolishing it in favor of an enhanced BNE. The sensible
options probably lie between these poles and admit many
.design variations in terms of composition, function,-
administrative location, subordination,, etc. Illustratively,
these variations include:
(1) Clarify and reinforce each NIO's authority
to commandeer drafting and other substantive support
from all appropriate CIA offices and other agencies.
(2) Organize a drafting staff for the NIOs--
either qua staff or attached as needed to particular
NIOs.
(3) Make the NIOs a more collegial substantive
body to strengthen capacity for estimates which cross
geographical or functional lines.
(4) Add three or four generalist NIOs without
specific geographic or functional responsibilities,
14
Approved For Release 2001/08/25 : CIA-RDP78Z02997A000100180003-7
 Approved pri Releaste-2001;108125 -. CIA- RDP78Z027A000100180003-7
(5) Establish a substanative review group.
drawn from CIA, State, DoD and outside convening
regularly or on call--to review estimates prepared
under NIO auspices.
(6) Establish an office of estimates apart
.from, but parallel with, the NIOs. It would be
composed of a small group of senior generalists
with a small substantive staff'for drafting and
specialized support, both elements-drawn from
within and outside the Community, to act as
executive agent for estimates' production, a
quality control mechanism and long-range think
tank. It would free NIOs for customer relations,
substantive and quasi-management duties in support
of the DCI or'other senior officials, etc.
(7) Establish an enhanced Board of Estimates,
complete with "eminent and highly qualified
intelligence analysts, diplomats and statesmen...
not susceptible to White House or parochial agency
pressure.". (Mondale) Give it a strong staff which,
inter alia, would perform many of the present NIO
functions,
Recommendations
recommendations on this issue must take
An
B
N
y
.
.
account of the DCI's needs, the climate in the White House
and in the Congress, and past experience and anticipated
requirements. No two people will see these in precisely the
same light. Considering all the factors, two alternatives
seem sensible:
a. 1a(6) above (an office of estimates parallel
with the NIO structure). This has the merit of real and
visible change (intrinsically desirable in present
context), but not too much change. It is sufficientlyi
flexible to meet all or most of the desiderata associated
with other options. It would cost some money and slots
not presently available, and these would have to come out
of other resources being cut back. It should be headed
up by someone,kideally an outside civilian, not identified
with BNE or NIO system, who could claim the confidence of
Congress and the academic/think tank community as well
as of the DCI and Executive Branch. Its working relation-
ship with the NIOs would need precise and thoughtful
clarification.
Approved For Release 2001/08/255 CIA-RDP78Z02997A000100180003-7
 A mixture of Lc(2) (3) (4) and (5) above
two or three generalist NIOs and/or a review group for
(drafting staff -or NIOs, a more collegial NIO process,
quality control and generalist perspective.) This
would build on the present structure and help repair
present deficiencies. It would cost somewhat less than
the other recommended option. It would help free the
present NIOs for what they do best. It would be
.criticized as cosmetic, tinkering rather than real change,
but that criticism would not be fatal.
`Neither of these pteps would require legislation
and either could be accomplished by Executive Order or DCI
maximized and.especially on'whether a new senior intelligence
officer is created, where he would sit, etc. Unless and until
directive. But much depends on what functions are to be
sriall task force should be established to blueprint a workable
plan. The group should initially include representatives of
another line of approach, there is no basis for action. If
he should approve either of the recommended approaches, a
the DCI concurs in one of tl-ese. recommendations, or directs
NIO office, ICS, DDI, and perhaps DDA. A second phase would
.be 'accompli shed in a , couple of weeks.
Involve elements of the Comri_unity. The planning phase could
~5X1B
Abbroved*oYReledse;200i`1y/O8/25 6CIA-RDP78Z.02997A0001
i
 'ase 2001/08/25 CIA-BQPr7. 02 99710100180003-7
Approved For Release 2001/O,8/25 CIA-RDP78Z02997A00010018}0003-7
Approved For
Distributio,:
Orig. - Adsee.
l - ER
1 - D/DCI1'TC
3 - CS/ICS
1 - IC Registry
DCI/1110 originally