IMPROVEMENTS IN NATIONAL ESTIMATING: A SUGGESTION

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP79M00467A001300180027-0
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 3
• Document Creation Date: December 16, 2016
• Document Release Date: July 13, 2005
• Sequence Number: 27
• Publication Date: March 1, 1976
• Content Type: MF

Approved Fo elease 2005/07/20: CIA-RDP79MO01V 001300180027-0
1 March 1976
MEMORANDUM FOR: Director
FROM . National Intelligence Officer
? fo-r USSR-EE
SUBJECT . Improvements in National Estimating:
A Suggestion
1. Herewith a specific response to your invitation
to suggest improvements in the process of producing
national intelligence.
2. On the estimative side, the NIO system emphasizes
the participation of specialists, whereas the former Board.
of National Estimates system emphasized the role of gen-
eralists. If the NIO system is retained in some form, we
ought to try for a better balance between the two..
3. One way to do this would be to set up a Review
Panel of non-specialists for individual estimates. This
Panel -- three.to five people -- would read the draft at
an early stage and comment upon its organization, its
coherence, its success at defining the right questions,
its success in answering them. It would then be up to the
NIO, the chairman of the paper, and the drafter to consider
these reactions and make appropriate changes.
4. Different Review Panels should be recruited for
different estimates. Otherwise -- if this kind of review
becomes someone's fulitime job -- he goes stale. The
task is essentially critical; it is not creative enough
to attract and hold top-flight people.
5. Panel members would not be specialists in the
subject of the estimate; specialist input would come
earlier (in the drafting) and later (in the community
review) in the production process. Rather they should be
chosen because they were well versed in related fields,
offered a different perspective on the subject, or were
simply very smart. They would be recruited from various
places:
Approved For Release 2005/07/20 : CIA-RDP79M00467AO01300180027-0
 Approved Fo lease 2005/07/20 : CIA-RDP79M004 001300180027-0
-- inside the intelligence community. These
people would be the easiest to commandeer
in some respects, the most experienced.
-- in the policymaking departments. These
would be harder to get, but their input --
particularly on the matter of the draft's
policy relevance -- would be unique and
important. One would have to depend upon
the integrity of the producers to keep the
assessment. from being distorted by the
policy preferences of these men.
non-governmental analysts. Our experience
with su:h types has been erratic; carefully
chosen, they can be quite useful. It would
be complicated, however, to schedule their
input into an estimative process that is
often conducted on short notice.
6. If we went down this road, we would have to
keep flexible, recognizing that each piece of national
estimating has peculiarities calling for a particular in-
put from any Review Panel. A wide-ranging estimate with
a long lead time could accommodate, and profit from, careful
and lengthy review (or reviews) involving outsiders as well
as insiders. A highly focussed and time-urgent SNIE would
probably get only half a day, in which the reviewers
assembled quickly, read the draft, and immediately discussed
their reactions with the NIO and drafter. We might decide
that some papers would not benefit enough from this kind of
outside review to make it worth while. Thus any "system"
set up along these lines should have maximum flexibility,
and in general, the less system the better.
25
Approved For Release 2005/07/20 : CIA-RDP79M00467AO01300180027-0
 P'. mc f;ve. R*Ei ft
Room. 7,~E-12
#.adqu?rt.n,
Approved For Release 2005/07/20 : CIA-RDP79M00467A0013001800