GAO BRIEFING REPORT TO CONGRESSIONAL REQUESTERS RE NATIONAL DEFENSE STOCKPILE

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP90B01013R000200050004-7
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 27
• Document Creation Date: December 27, 2016
• Document Release Date: July 18, 2013
• Sequence Number: 4
• Publication Date: August 4, 1986
• Content Type: LETTER

Declassified and Approved For Release 2013/07/23 ?
CIA-RDP90B01013R000200050004-7 UM*
111A0110 I WIWI ruin.p nriameom nu owe- 18 Aug 86
TO:Name, office symbol, room number,
. tinging, Agency/Post)
i
A/DCI
Initials
Date 't
1../
2. il-DDI
& D/OGI 2 6
AlC/L
e
4. In Turn Please)
/....5.--,
Action
File
Note and Return
Approval
For Clearance
Per Conversation
As Requested
For Correction
Prepare Reply
Circulate
For Your Information
See Me
Comment
Investigate
Signature
Coordination
Justify
REMARKS
NIO/ECON and D/NIESO have copy each.
2--
/7)API\ 11/1
/
DO NOT use this form as a RECORD of approvals, conc disposals,
clearances, and similar actions
,poi (Name, org. symbol, Agency/Post)
go ES
? Room 14o.-131dg.
7als, 64
Phone No.
5041-102 OPTIONAL FORM 41 (Rev. 7-76)
Declassified and Approved For Release 2013/07/23: _11.206
CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
GAO
United States General Accounting Office
Briefing Report to
Congressional Requesters
August 1986
NATIONAL DEFENSE
STOCKPILE
Adequacy of National
Security Council Study
for Setting Stockpile
Goals
GAO/NSIAD-86-177BR
- Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
United States
General Accounting Office
Washington, D.C. 20548
National Security and
International Affairs Division
B-223657
August 4, 1986
The Honorable James A. McClure
Chairman, Committee on Energy
and Natural Resources
United States Senate
The Honorable Charles E. Bennett
Chairman, Subcommittee on Seapower
Strategic and Critical Materials
Committee on Armed Services
House of Representatives
and
\ Executive Regi
In July 1985, you asked us to evaluate the National Security Council's
(NSC's) study of national defense stockpile ,goals, the results of which
were announced on July 8, 1985, and to obtain participating agencies'
views on the study. In subsequent discussions with .your office, we were
asked to provide this interim briefing report assessing whether the NSC
stockpile study is-a sufficient basis for U.S. mobilization planning,
including the proposed changes in national defense stockpile goals.
Materials, such as cobalt and titanium, are stockpiled to meet increased
defense demands expected at the beginning of a wartime mobilization.
Which materials, and the amounts to be stockpiled, can vary signifi-
cantly depending on the assumptions used. Assumptions need to be made
about issues such as the anticipated defense demand, the capability of
U.S. industry to surge to meet demand, sacrifices in consumer-goods
production to reallocate resources to mobilization needs, and the
availability of materials from foreign sources in times of conflict.
Because of the assumptions it used, the NSC study recommended a stock-
pile goal of $0.7 billion, which is much lower than the previous goal of
$16.1 billion, established based on a 1979 study. Of $10.1 billion in
stockpile inventory on hand against the previous $16.1 billion goal, the
NSC study recommended selling $3.2 billion and holding a $6 billion
supplemental reserve, at least temporarily, of materials, already on
hand.
Our preliminary assessment is that the NSC study does not appear to
provide a sufficient basis for setting stockpile goals or for other U.S.
mobilization planning. Although the NSC study methodology was similar
to the methods of past studies and made some improvements, the assump-
tions used were very different,. and the study report did not adequately
reflect major disagreements among study participants with regard to key
assumptions. Furthermore, the study did not adequately show that its
results could vary greatly with changes in its assumptions. Such ranges
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
B-223657
of results, which can be quantified by doing sensitivity tests on the
assumptions used, were a key part of the prior study, and provided
decision makers a basis by which to assess the study's conclusions.
We caution that this preliminary assessment is based on a partial
analysis of unclassified material. In our ongoing evaluation of the NSC
stockpile study, we are reviewing the stockpile study report and
supporting classified documentation, examining past stockpile studies
and critiques of those studies, and discussing stockpile issues with
representatives of each of the agencies that participated in the NSC
study. We are also obtaining the views of outside experts. In our
remaining work, one of the key steps is to determine which of NSC's
assumptions have the most impact on the level of the stockpile. In
order to accomplish this task, sensitivity tests need to be conducted
for each assumption.
MAJOR QUESTIONS ARISING IN OUR EVALUATION
OF THE NSC STOCKPILE STUDY
Although our work is not complete, it raises questions about whether the
NSC study adequately supports its recommendations to (1) significantly
reduce stockpile-goal levels and (2) use the study's planning assump-
tions for other mobilization-preparedness areas. Specific questions
include whether NSC study assumptions are consistent with defense
planning assumptions and data, with past U.S. economic experience and
economic projections by other groups, and with estimates of supply and
demand by industry and other experts.
Study participants and industry and economic experts have expressed
serious concerns about study assumptions, methodology, and results. For
example, key study participants said that they did not agree with NSC's
assumptions and that the NSC study did not adequately show the impact
of alternative assumptions. Our initial tests confirmed this, and
showed that stockpile-goal levels can vary widely as assumptions change.
The NSC study group initially computed a stockpile goal of $230 million.
NSC then modified assumptions by making adjustments to increase material
requirements for the defense and industrial sectors and reduce world
supply, which increased the goal to $691 million'. However, NSC's
adjustments were limited. For example, study participants reported that
no changes were considered for such factors as oil availability and
essential civilian requirements. Also, the reported adjustments for
such assumptions as defense-sector 'requirements covered only part of the
assumptions' plausible ranges.
We believe that analyses on several additional assumptions are needed,
which could provide the basis for decision makers to choose a different
goal than the $691 million NSC proposed. The analyses would involve key
assumptions, such as for defense expenditures, nonresidential investment
in equipment, the degree of civilian austerity, availability of oil,
2
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
B-223657
wartime production capabilities of the critical materials mining and
processing industries, and the availability of critical materials
imports to the United States.
Plausible changes in many of the study's assumptions could cause
computed stockpile goals to vary widely. For example, outside experts
have suggested that, in a major conventional war, the U.S. wartime
economy and associated defense expenditures could far exceed the levels
assumed in the NSC study. If the increased expenditures caused a 50-
percent increase in material requirements for the defense sector, this
one change alone could cause the overall stockpile goal to increase to
almost $1 billion--well beyond the $691 million proposed by the NSC
study. As another example, NSC's study places much greater reliance on
foreign sources of supply than was done in prior stockpile studies, or
was recommended by some study participants. The impact of these kinds
of assumption changes needs to be clearly identified through additional
analyses.
BASIS FOR INTERIM STOCKPILE
PURCHASES OR DISPOSALS
While we believe that final congressional action on approving a stock-
pile goal should not be made until we have completed our evaluation, and
the Administration has responded to our findings, there appear to be
some low risk interim actions that can be taken based on areas where the
NSC and 1979 studies are in agreement.
Both the NSC study and a 1979 interagency study, coordinated by the
Federal Emergency Management Agency, agree that about $3.4 billion worth
of materials on hand are excess to national security needs, and could be
sold or bartered. On the other hand, the current inventory of at least
one material--germanium--falls short of both its current and NSC-
proposed goals. Furthermore, material experts among the study partici-
pants and advisory committees say that some materials being stockpiled,
such as cobalt, may need to be upgraded. Your Committees may wish to
discuss with one or more of these organizations, such as the National
Materials Advisory Council or the General Services Administration, the
desirability of using proceeds from future disposal sales, or moneys
already in the National Defense Stockpile Transaction Fund, to upgrade
such materials.
We discussed our preliminary results with NSC and Office of Management
and Budget (OMB) officials who coordinated the NSC study. They told us
that stockpile goals were driven primarily by defense planning assump-
tions, and that they believed the NSC study's assumptions to be consis-
tent with defense planning. A detailed discussion would involve
classified information; however, we can point out that the defense
guidance addresses a likely range of wartime effort including levels
3
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
B-223657
greater than assumed by NSC. Also, unlike the NSC study, which accepts
increased reliance on foreign sources of material supply, the defense
guidance indicates that a growing reliance on foreign sources poses a
threat to national security.
Analyses of different assumptions than those used in NSC's proposed
$691 million stockpile goal would, in our opinion, show a much broader
range of stockpile goal options. Differing assumptions for a variety of
factors have been suggested by top study participants and other experts.
In response to our request for further analyses of defense and other
assumptions, NSC and OMB officials said that they would consider
analyzing key assumptions on a case-by-case basis, but they have not yet
begun such analyses.
We are sending copies of this briefing report to the Chairmen, Senate
and House Committees on Armed Services, the Senate and House Committees
on Appropriations, the Senate Committee on Governmental Affairs, and the
House Committee on Government Operations; to the Assistant to the
President for National Security Affairs; and to the heads of the 12
agencies which participated in the NSC study. Copies will be made
available to other interested parties upon request.
If you have any questions, please call Martin M Ferber, Associate
Director for Manpower, Reserve Affairs, and Logistics, at 275-4001.
Frank C. Conahan
Director
4
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
CONTENTS
APPENDIX
Page
I RESULTS OF NSC STOCKPILE STUDY 6
Background 6
NSC study results 6
II EVALUATION OF METHODOLOGY AND ASSUMPTIONS USED TO
SET STOCKPILE GOALS IN THE NSC STOCKPILE STUDY 14
NSC methodology 14
Preliminary GAO results 14
Basis for interim stockpile purchases or disposals 21
FIGURES
1.1 Total Existing and Proposed Stockpile Goals 7
1.2 Existing and Proposed Stockpile Goal
Values for Major Materials 8
1.3 Stockpile Inventory Status Under Existing
and Proposed Goals 9
1.4 Planned Disposition of Stockpile Materials
Excess to NSC's Proposed Goals 10
1.5 Amount by Which Six Materials in the Supplemental
Reserve Exceed Higher of the Existing or the
Proposed Goals 11
11.1 Gross National Product Increases 16
11.2 Defense Expenditures as a Percentage of GNP 17
11.3 Impact of Changes in Defense-Sector Demand
on Stockpile Goals 19
11.4 Stockpile Materials That Are Excess Under Both
Existing and Proposed Goals 22
TABLES
1.1 Stockpile Material Values and NSC's Proposed
Adjustments 12
ABBREVIATIONS
DOD Department of Defense
GAO General Accounting Office
GNP Gross National Product
NSC National Security Council
OMB Office of Management and Budget
5
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX I APPENDIX I
RESULTS OF NSC STOCKPILE STUDY
BACKGROUND
The objective of the Strategic and Critical Materials Stock Piling
Act, as amended (Public Law 96-41, 50 U.S.C. 98 et seq.), is to
ensure that materials are stockpiled to minimize dependence on
foreign sources of supply in times of national emergency. The Act
requires that the stockpile inventory be sufficient to cover U.S.
needs for not less than 3 years of a national emergency.
The President approves stockpile policy, which is to be followed
in determining the stockpile goals. The goals are the difference
between emergency supply, and estimated 3-year requirements for
each strategic material, such as cobalt and titanium.
Since the passage of the Stockpiling Act in 1946, stockpile goals
and actual inventories have varied widely. For example, the
stockpile was initially expected to support requirements for 5
years; later the time was reduced, first to 3 years and then to 1
year, and--in 1976--was increased again to 3 years. Goals have
been relatively stable since the major reassessment of stockpile
policy and goals which President Ford approved in 1976, and Presi-
dent Carter reaffirmed in 1977. Stockpile goals were recalculated
in 1979 using then-existing policy guidance, with some changes to
improve methodology.
NSC STUDY RESULTS
The most recent NSC stockpile study--begun in June 1983--proposed
major reductions in the stockpile. The Administration announced
the results of the study on July 8, 1985. The study considered 45
stockpile materials, which constitute $15.6 billion (97 percent)
of the total current stockpile goals of $16.1 billion. The study
recommended a stockpile goal of $0.7 billion to meet national
security needs.1 If the study recommendations are adopted:
Two materials (germanium and one form of mica) would have goals
increased a total of $125 million (from about $51 million to
$176 million).
Thirteen materials would remain in the stockpile, but have
their goals reduced by $3.9 billion (from $4.4 billion to $0.5
billion).
lUnless otherwise noted, all dollar values are as of May 31,
1985, the reference date used in the NSC study report.
6
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX I APPENDIX I
-- Thirty materials, with a stockpile value of $11.1 billion,
would be eliminated.
Of $10.1 billion worth of materials now on hand in the stockpile,
about $9.5 billion would be excess to the proposed new goals. Of
the excess, the study proposed that $3.2 billion be declared
surplus and sold, and $6 billion would be retained as a supplemen-
tal reserve. (See P. 11 for a further discussion of the reserve.)
The remaining materials, valued at about $0.3 billion, were not
studied, but the Administration has since also proposed the sale
KN of about $37 million of such materials.
11
The Administration proposed that receipts from the sale of materi-
als are to be used to fill stockpile shortfalls or to be returned
to the Treasury. With regard to purchases, only the germanium
inventory is now below proposed goals, but more materials might be
later proposed for purchase because over 20 materials (some not
included in the NSC study) are being examined to determine
possible need.
Figure 1.1 compares total current and proposed stockpile goals.
Under both goals, the current inventory has too much of some
materials and too little of others. Figure 1.2 compares the value
of major items under goals now in effect with NSC's proposed
goals.
Figure 1.1: Total Existing and Proposed Stockpile Goals
IC3citatirag ifigcortl
Fot-c.p.oa.etcl ag.oa.1
1 6 . 1
a 4 6 a to 1 14
DsDlIc1rs3 billion
16
1 6
20
I r-Ivn tory
Iltetzt.ca
Sttcrt.fatll
On hand inventory in the above chart consists of only those
materials needed to meet existing and proposed goals. Excess
inventories are not shown in this chart.
7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX I
Figure 1.2:
Aluminum metal
Titanium sponge
Copper
Zino
Nickel
Cobalt
Chromium (a)
Rubber
Tin
Lead
Platinum
Palladium
Manganese (a)
Tantalum.
Bauxite (b)
Beryllium
Turiguten
Fluorspar (c)
Fluorspar (d)
Diamond (e)
APPENDIX I
Existing and Prqposed Stockpile Goal Values
for Major Materials
Goal:
23132. 9
2090.5
.ezzzzzzi 1879
1350.3
1007.
926.6
822
618.1
451
361.6
345.5
331
318.9
293.8
283.4
273.5
242.2
6--1212.5
if- A leg
0
Aluminum Oxide CO
Vanadium
Antimony
Chromite (b)
Graphite i,g)
Iridium
Manganese (h)
Columbium
Rutile
Mica (I)
Iodine
Germanium
Mica (j)
Silicon Carbide
13iernutla
Graphite (k)
Cadmium
Quartz
Mercury-
AU cAter
Sc
000 700 1000 1000 3.800 1700 2000 2200 2000
(Note change in scale)
0
11:sr
112.4
101.8
99
84,8
80
42
.hT= 43.7
395
.i2F=r2= 30.0
.r.erc,Lc.c..z.e 35
fi2.3=223 33'1
1:767ZZ= 32.8
461811111V153
.9:013.1
-Z6:8 12.3
01.6'3
11
ledzw-
154.9
0 SO 40 60 80 100 120 140 160 180 ZOO
Goal value (dollars in millions)
8
Existing
Proposed
Goal:
Existing
Proposed
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX I APPENDIX I
a Chemical and metallurgical group
h Refractory
? Acid grade
d Metallurgical grade
e Industrial stones
f Abrasive grain group
g Natural Malagasy
h Battery grade
i Muscovite block
J Muscovite splittings
k Natural Ceylon
Figure 1.3 shows the status of existing inventory held toward
meeting the goal versus that which is excess.
Figure 1.3: Stockpile Inventory Status Under Existing
and Proposed Goals
Dollars in billions
22
10 -6-
a
4
10.1
Under existing goals
10.1
Under proposed goals
9
Inventory
Excess to goal
MHeld to meet goal
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX I APPENDIX I
Figure 1.4 shows which materials are considered excess to national
security needs under NSC's proposed goals. The excess, which
totals $9.5 billion, is further broken down in the chart as to how
much is surplus (eligible for sale within 5 years), and how much
is to be held at least temporarily in a supplemental reserve. For
more information related to this chart, see table 1.1, which lists
stockpile materials in order of greatest excesses under NSC's
proposed goals.
Figure 1.4: Planned Disposition of Stockpile Materials
Excess to NSC's Proposed Goals
Tin
Chromium (a)
Silver
Aluminum Metal
Manganese (a)
Tungsten
Zino
Titanium
Diamond (b)
Cobalt
Lead
Beryllium
Nickel
Fluorspar (o)
Palladhun
2254
12=2=1011989
SEMMEMOIMM1853
828
-t7 A4111491
-ZZIN 411
ill= 359
-02111 352
1EMMO 350
iiMMI 330
24
-Mil 232
-1.1 187
MIN 155
MI 146
Aluminum Oxide (d) 129
Platinum
Rubber
Antimony
All other
IN 125
-0 118
11 93
Immom1551
0 860 600 760 MOO 1E60 1600 1760 8000 2260 8600
Inventory value (dollars in millions)
a Chemical and metallurgical group
b Industrial stones
Acid grade
d Abrasive grain group
10
Type of excess
Reserve
Surplus
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX I APPENDIX I
NSC does not clearly define the rationale for holding supplemental
reserve commodities as part of the national defense stockpile.
Also, $1.9 billion (almost one third of the entire proposed
reserve) exceeds national defense requirements as computed by both
the 1979 and NSC studies. For example, the NSC study proposes to
retain $1.8 billion worth of tin in the supplemental reserve,
although the study's computed requirement for tin was zero, and
the current goal (based on the 1979 study) is only $0.5 billion.
Figure 1.5 shows the.extent to which the proposed reserve's
holdings of six materials would exceed the higher of either the
current or the proposed goals.
Figure 1.5:
Tin
Silver
Manganese (a)
Tungsten.
Quartz
Diarn.can.d
(b)
Amount by Which Six Materials in the
Supplemental Reserve Exceed the Higher of
the Existing or the Proposed Goals
37.6
4.8 4.8
I 73
6.4
542.5
1298.1
150 300 450 890 750 800 1050 1200 1350 /500
Material value (dollars in millions)
a Chemical and metallurgical group
b Industrial stones
11
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX II APPENDIX II
Table 1.1: Stockpile Material Values and NSC's Proposed Adjustments
Material
Existing
goal
Proposed
EL
Variance Inventory
.in goals on hand
Existing
shortage
Proposed excess
Excess
Surplus to NSC
Reserve sell study
Tin
516.1
0.0
516.1
2253.7
-1737.6
2253.7
1814.2
439.5 Yes
Chromium, Chemical &
822.0
84.9
737.1
1043.7
-221.7
958.8
755.5
203.3 Yes
Metallurgical Group
Silver, Fine
0.0
0.0
0.0
852.5
-852.5
852.5
542.5
310.0 Yes
Aluminum Metal Group
2382.9
0.0
2382.9
827.9
1555.0
827.9
827.9
0.0 Yes
Manganese, Chemical and
331.1
0.0
331.1
490.6
-159.5
440.6
368.7
121.9 Yes
Metallurgical Group
Tungsten Group
273.5
0.0
273.5
410.5
-137.0
410.5
298.3
112.2 Yes
Zinc
1350.3
0.0
1350.3
358.5
991.8
358.5
80.5
278.0 Yes
Titanium Sponge
2090.5
43.3
2047.2
394.8
1695.7
351.5
232.6
118.9 Yes
Diamond, Industrial,
151.0
0.0
199.0
349.9
-150.9
349.9
205.4
144.5 Yes
Stones
Cobalt
926.6
245.0
681.6
575.2
351.4
330.2
65.1
265.1 Yes
Lead
451.0
0.0
451.0
246.4
204.6
246.4
123.0
123.4 Yes
Beryllium Metal Group
2133.4
0.0
283.4
232.0
51.4
232.0
164.4
67.6 Yes
Nickel
1007.0
0.0
1007.0
187.4
819.6
187.4
24.2
163.2 Yes
Fluospar, Acid Grade
242.2
0.0
242.2
155.0
87.2
*155.0
0.0
155.0 Yes
Platinum Group Metals,
345.5
0.0
345.5
145.6
199.9
145.6
0.0
145.6 Yes
Palladium
Aluminum Oxide, Abrasive
112.4
0.0
112.4
128.6
-16.2
128.6
65.0
63.6 Yes
Grain Group
Platinum Group Metals,
361.6
0.0
361.6
124.9
236.7
124.9
0.0
124.9 Yes
Platinum
Rubber
785.4
0.0
785.4
115.9
669.5
115.9
115.9
0.0 Yes
Antimony
99.0
12.6
86.4
105.3
-6.3
92.7
0.0
92.7 Yes
Vegetable Tannin Extract,
19.0
0.0
19.0
86.8
-67.8
86.8
0.0
0.0
Quebracho
Tantalum Group
318.9
72.1
246.8
156.5
162.4
84.4
84.3
0.0 Yes
Jewel Bearings
134.8
0.0
134.8
82.1
52.7
82.1
0.0
0.0
Bauxite, Refractory
293.8
0.0
293.8
57.7
236.1
57.7
57.7
0.0 Yes
Fluorspar, Metallurgical
212.5
0.0
212.5
51.5
161.0
51.5
0.0
51.5 Yes
Grade
Mercury
3.1
0.0
3.1
51.4
-48.3
51.4
0.0
51.4 Yes
Copper
1579.0
0.0
1579.0
45.9
1533.1
45.9
45.9
0.0 Yes
Iodine
32.6
0.0
32.8
41.6
-8.8
41.6
31.0
10.6 Yes
Chromite, Refractory
84.8
0.0
84.8
39.1
45.7
39.1
18.0
21.1 Yes
Grade Ore
Diamond, Industrial,
38.5
0.0
38.5
38.5
0.0
38.5
0.0
0.0
Crushing Bort
Silicon Carbide, Crude
13.1
0.0
13.1
36.2
-23.1
36.2
0.0
36.2 Yes
Mica Muscovite Block
33.1
1.3
31.8
27.8
5.3
26.5
1.1
25.4 Yes
Asbestos, Amosite
11.9
0.0
11.9
23.8
-11.9
23.8
0.0
0.0
Morphine, Refined
70.3
0.0
70.3
21.3
49.0
21.3
0.0
0.0
Asbestos, Chrysotile
5.5
0.0
5.5
19.6
-14.1
19.6
0.0
0.0
Thorium Nitrate
1.7
0.0
1.7
19.6
-17.9
19.6
0.0
0.0
Manganese, Dioxide,
39.5
0.0
39.5
19.5
20.0
19.5
0.0
19.5 Yes
Battery Grade Group
Columbium Group
36.9
0.0
36.9
18.4
18.5
18.4
18.4
0.0 Yes
Mica Muscovite Film
1.1
0.2
0.9
13.8
-12.7
13.6
0.0
13.6 Yes
12
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX II
APPENDIX II
Material
Existing
goal
Proposed Variance Inventory
pal Irliali on hand
Existing
shortage
Proposed excess
Excess
Surplus to NSC
Reserve sell
Platinum Group Metals,
43.7
0.0 43.7 13.2
30.5
13.2
,study
0.0 13.2 Yes
Iridium
Rutile
35.0
0.0
35.0
12.9
22.1
12.9
0.0
12.9 Yes
Bismuth
12.3
0.0
12.3
11.7
0.6
11.7
0.0
11.7 Yes
Graphite, Natural,
60.0
42.0
18.0
53.6
6.4
11.6
0.0
11.6 Yes
Malagasy, Crystalline
Quartz Crystals
3.6
0.2
3.4
11.1
-7.5
10.9
10.9
0.0 Yes
Ricinoleic/Sebacid Acid
18.7
0.0
18.7
10.7
8.0
10.7
0.0
0.0
Products
Vegetable Tannin Extract,
10.6
0.0
10.6
10.6
0.0
10.6
0.0
0.0
Wattle
Vegetable Tannin Extract,
3.4
0.0
3.4
8.7
-5.3
6.7
0.0
0.0
Chestnut
Quinidine
34.2
0.0
34.2
8.4
25.8
8.4
0.0
0.0
Vanadium Group
101.8
0.0
101.8
8.4
93.4
8.4
8.4
0.0 Yes
Quinine
9.5
0.0
9.5
6.8
2.7
6.8
0.0
0.0
Cadmium
11.0
0.0
11.0
6.0
5.0
6.0
0.0
6.0 Yes
Morphine, Crude '
0.0
0.0
0.0
5.3
-5.3
5.3
0.0
0.0
Mica Phlogopite Split.
1.9
1.0
0.9
3.0
-1.1
2.0
0.0
2.0 Yes
Diamond Dies, Small
2.7
0.0
2.7
1.1
1.6
1.1
0.0
0.0
Graphite, Natural,
12.3
9.9
2.4
10.7
1.6
0.8
0.8
0.0 Yes
Ceylon, Amorphous Lump
Graphite, NaturallOther
2.0
1.6
0.4
2.0
0.0
0.4
0.0
0.4 Yes
Mica Muscovite Split.
18.9
21.6
-2.7
22.0
-3.1
0.4
0.0
0.4 Yes
Talc, Steatite Block &
0.0
, .0.0
0.0
0.4
-.4
0.4
0.0
0.0
Lump
Mica Phlogopite Block
1.1
0.5
0.6
0.7
0.4
0.2
0.0
0.2 Yes
Sapphire and Ruby
0.0
0.0
0.0
0.2
-.2
0.2
0.0
0.0
Cordage Fibers, Abaca
82.2
0.0
82.2
0.0
82.2
0.0
0.0
0.0
Cordage Fibers, Sisal
22.7
0.0
22.7
0.0
22.7
0.0
0.0
0.0.
Molybdenum Group
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0 Yes
Natural Insulation Fibers
16.5
0.0
16.5
0.0
16.5
0.0
0.0
0.0
Pyrethrum
72.1
0.0
72.1
0.0
72.1
0.0
0.0
0.0
Germanium
31.8
154.8
-123.0
0.0
31.8
-154.8
0.0
-154.8 Yes
** TOTAL **
16117.6
691.0
15426.8
10057.0
5959.7
13
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX II APPENDIX II
EVALUATION OF METHODOLOGY AND ASSUMPTIONS USED TO SET
STOCKPILE GOALS IN THE NSC STOCKPILE STUDY
NSC METHODOLOGY
The NSC study methodology followed the basic approach of prior
studies but used different assumptions and different models (for
example, the Wharton econometric model to determine the Gross
National Product [GNP], rather than the Chase model).
Key participating agencies included NSC; the Central Intelligence
Agency; the Council of Economic Advisors; the Departments of
Commerce, Energy, Interior, Defense (DOD), State, and Treasury;
the Federal Emergency Management Agency; the General Services
Administration; and OMB.
The fundamental approach of the NSC study was to estimate material
demand and supply for 3 years of war and compare the two. The
study estimated material demand by (1) using macroeconomic models
to estimate industry-output dollar levels for a wartime economy,
and then (2) converting these industry-output levels into demands
for critical materials expressed in physical units. The study
estimated material supply by (1) estimating world production
capacity of raw materials; and (2) subtracting some, but not all,
materials demand from foreign countries; and (3) reducing overseas
supply for such reasons as war damage, attrition during transpor-
tation, and reliability of foreign sources of supply.
The NSC study projected significantly reduced material demands and
some increased supply relative to the prior study, and initially
computed a stockpile goal of $230 million, using October 1984
prices. The goal was subsequently increased to $691 million,
based on limited assumption changes and May 1985 prices. The
limited changes increased material requirements for the defense
and industrial sectors and reduced world supply.
PRELIMINARY GAO RESULTS
We found that the methodology used in the NSC study was similar,
though improved in several respects, to that used in interagency
studies in 1976 and 1979. However, some study participants
expressed serious concerns about the reasonableness of the NSC
assumptions, and about the NSC study's not adequately showing the
impact of choosing alternative assumptions.
Our review raises questions about whether NSC study assumptions
adequately reflect (1) defense planning assumptions and data, (2)
past U.S. economic experience and economic projections by other
groups, and (3) estimates of supply and demand made by industry
representatives.
14
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX II APPENDIX II
Inadequate analyses of the impact of
assumption changes on stockpile goals
The NSC study did not include adequate sensitivity analyses to
show the impact of changes in major assumptions. Key assumptions
questioned by study participants included minerals-production
levels, materials-use for a given level of industrial output, and
selection of reliable foreign sources. Stockpile goals are
extremely sensitive to changes in such assumptions.
Our review suggests that one of the most critical overall assump-
tions is the war scenario and associated wartime demand. Our
tests of plausible upper limits of wartime demand showed potential
goals to be far greater than the NSC's $0.7 billion, even if the
other NSC assumptions remain unchanged. These upper limits were
determined using input from study participants, and from industry
and economic experts. Another assumption, dealing with the
capability of the United States to increase material supply, has
less impact on stockpile goals, but may have significant impact
when combined with the impact of other assumptions.
DOD and NSC wartime scenarios differ
Defense policy guidance as to the likely range of wartime effort
differs from NSC's assumptions. Defense policy guidance to the
military services instructs them to plan for a range of defense
expenditures, including levels greater than those assumed by,NSC.
Also, the defense guidance indicates that a growing reliance among
industrial democracies on foreign sources of supply poses a threat
to national security, while the NSC study accepts increased
reliance on foreign sources. Our ongoing work also indicates that
some major categories of defense requirements may have been
omitted from NSC's study. Specific data concerning the additional
requirements is classified.
The NSC and OMB officials who coordinated the NSC study said that
they believed that the NSC study's scenario was consistent with
defense plans, and that any differences must involve the views of
low-level DOD analysts. We will explore this issue further in our
ongoing work, but the differences are not low level. They involve
statements by top study participants and specific guidance by top
DOD policy makers, including the Secretary of Defense.
Differences between NSC assumptions and other
available data for U.S. economic activity
NSC's economic assumptions include small increases in GNP, which
have the effect of reducing stockpile goals. The following two
figures show data for wartime growth rates in GNP and the defense
share of GNP. Included for comparison with NSC's estimates are
actual data from World War II and the Korean War (to show actual
U.S. wartime experience), and estimated data from the 1979 study
(to show one other projection).
15
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX II APPENDIX II
Figure 11.1 shows that NSC assumed cumulative GNP increases during
the warning year and 3 war years to total 17.8 percent. These
increases were less than those' assumed by the 1979 study (33.3
percent), and also less than those that actually occurred in the
Korean War (25.7 percent) and World War II (65.4 percent). If the
higher GNP increases favored by some study participants are
assumed, it results in greater levels of production and a greater
need for stockpiled material.
Figure 11.1: Gross National Product Increases
(4-Year Compounded Increase)
70-
BO
W
4.4
1
& act
A T
w
1
A so
c.,
J.?
I
,
1
17.8
23.3
68.4
i
/
i
/
..,* "
,.."....
47.- .....
?..'
?--..
,-
/
_././..
./..'//
./".7
-,...
19E14 Study KO re WM WEIS' 1079 Study WW H
Study a.ssuirriptiorxe and wartirne actual
Figure 11.2 shows that NSC assumed that a greater portion of GNP
was devoted to defense expenditures during the war period than
occurred in the Korean War or was assumed by the 1979 study.
However, the NSC assumption was still far below actual experience
in World War II.
16
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX II
APPENDIX II
Figure 11.2: Defense Expenditures as a Percentage of GNP
(4-Year Average)
40
aaao
a
as
32.a
,
lis
15.7
07'
N
10.5
3.cp
r-
9.3
a
":?.....?......?.
Are:A
...0"..:
---? --i'e...-.-''''''':"...:.?
1
NSC Study Korcari Wa..r 1979 study WW II
Comparing the NSC estimates with other estimates helps identify a
reasonable range for analyzing assumptions, but is not intended to
suggest that the NSC assumptions and resulting estimates should
duplicate past experience or projections. However, comparative
data are helpful in evaluating the economic conditions that are
possible during wartime mobilization.
Impact of assumptions about defense-sector
demand on national defense stockpile goals
To determine a plausible range of defense expenditures for
sensitivity testing, we consulted with economic-modeling experts,
and estimated defense expenditures using World War II experience
with adjustments for changes since World War II. We found that,
for sensitivity-testing purposes, the U.S. economy could operate
with projected wartime defense expenditures 50 percent greater
than the amount NSC used to compute stockpile goals, and that an
upper limit could be about twice the amount NSC used.
We requested data on sensitivity analyses done by study partici-
pants, and were advised that nothing was documented beyond the
limited tests described in the NSC report. The NSC tests were
limited in that the tests for some assumptions covered only part
of the possible ranges. Also, the tests simply adjusted output
data for material supply and demand. The proper method of chan-
ging assumptions would have been to enter new defense expenditure
data in the initial stages of the model-estimation process, and
17
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX II APPENDIX II
allow the models to determine what material demand changes
resulted. The method used did not measure possible secondary
effects, such as production bottlenecks.
We believe that additional sensitivity analyses are needed to
better-determine the impact of different assumptions favored by
key study participants and outside experts. We are attempting to
reach a :mutually satisfactory arrangement with NSC to perform the
additional analyses using the NSC model. In the meantime, we are
proceeding with limited and simple analyses of our own.
For example, the limited analyses described below address an
expanded range for just two (defense-sector demand and programs to
increase U.S. material supply) of numerous interrelated factors.
We have been unable to do a full analysis because we do not have
access to all of the models used in the NSC study. The data below
is intended to show a need for accurate analyses of apparent wide
swings in stockpile-goal levels, not to predict actual goals for a
given set of assumptions.
Effect of defense-sector material demand on stockpile goals
We found that different assumptions about defense-sector material
demand within a plausible range (as suggested by our analyses and
the opinions of study participants and outside experts) produced
significant changes in stockpile goals. The significant changes
occurred even though other NSC assumptions (such as those about
U.S. production capacity and reliable foreign supply) were left
unchanged. A 50-percent increase in defense-sector material
demand more than quadrupled total stockpile-goal value. Doubling
the defense sector material demand resulted in more than a ten-
fold increase. Some materials, such as titanium, appeared parti-
cularly sensitive to changes in defense-sector demand.
Figure 11.3 shows the sensitivity of total stockpile goals to
changes over a range of demand from a reduction of 50 percent to
an increase of 100 percent.
In commenting on our preliminary results, the NSC and OMB offi-
cials who coordinated the NSC study said that the U.S. economy is
much different now than in the 1940s. They said that World War II
should not be used a basis for projecting increases in defense
spending. We agree. For that reason, the upper limits of our
sensitivity test for the defense sector assumed lower GNP growth
(about 10 percent) than the 15 percent annual growth during World
War II. We recognize that specific estimates of the size of the
defense sector would likely fall below that upper limit. Nonethe-
less, we believe that it is useful to demonstrate ranges of
stockpile goals and the assumptions associated with those ranges.
18
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX II
APPENDIX II
Figure 11.3: Impact of Changes in Defense Sector Demand
on Stockpile Goals
9.00
-BOX -40X -3011 -90X -LOX No Me +20X +0011 +30% +413X +00X +00X +70X +BOX +00X +100X
Percent change in expected demand
The NSC and OMB officials also commented that the highest goal
resulting from this sensitivity test was still within the stock-
pile value recommended by the NSC study ($0.7-billion goal plus
$6-billion supplemental reserve) and well below the current goal
of $16.1 billion. While we are performing additional analyses of
how sensitive the stockpile goals are to changes in assumptions
about various factors, the comments of NSC and OMB officials do
not change our preliminary conclusions that:
Stockpile goals are more sensitive to assumption changes than
NSC concluded. The test described above shows that changes in
just one of many possible factors caused goals to vary beyond
the upper limit reached in the NSC's tests of several factors.
19
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX II APPENDIX II
The goals for some material, such as titanium, are particu-
larly sensitive to assumption changes. In our test described
above, the upper limit for titanium need was greater than NSC's
proposed goal and supplemental reserve combined. Further
analyses could show that other materials likewise exceed both
NSC's proposed goal and supplemental reserve.
The combined effect of several key factors could be signifi-
cantly greater than the effect of one factor alone. This is
indicated by our test of the assumptions regarding one aspect
of supply, which is discussed below.
Effect of programs to increase U.S. material supply
To alleviate shortfalls in materials available from reliable
sources worldwide, NSC explored ways to increase domestic availa-
bility of such materials through such programs as opening new
mines or reopening closed mines. Such programs are intended to
raise supply above the levels achievable through surging existing
facilities. Such additional programs alone did not appear to have
great impact on stockpile goals, but the effect became significant
when considered in conjunction with defense-sector demand.
For its base case, NSC assumed that such programs could increase
supply for 29 different stockpiled commodities, generally during
each year of a war. Some study participants and industry experts
believed that such a large number of programs was not feasible
simultaneously, saying that supply for perhaps only two or three
materials could be successfully increased. The 1979 study assumed
that no more than three materials could be increased through such
programs, and those mainly in the third year of the scenario.
Also, industry representatives told us that material supply would
not be likely to increase as much or as quickly as NSC assumed.
For example, copper industry representatives told us that the U.S.
smelter industry may not be able to handle the domestic ore
production increases that the NSC study projected. Opening new
smelters could raise capacity, but industry estimates of the time
required to build a smelter ranged from 2 to 4 years.
If it is assumed that supply was increased for only three materi-
als in the second and third years of the scenario, NSC's initial
stockpile goal almost doubles (from $230 million to $445 million).
However, when combined with increased defense-sector demand, the
impact was substantially greater.- With a 50-percent increase in
defense-sector demand, the lower assumptions for material supplies
raised the stockpile goal by about $.6 billion (from $1 billion
for increased defense-sector demand alone to $1.6 billion). The
reason for the increased impact is that, in many cases, a
material's stockpile goal was zero, both with and without programs
to increase supply. Thus, the factor had no impact by itself in
those cases. When combined with increased demand, however, the
20
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX II APPENDIX II
supply changes were sufficient to affect more materials goals.
Effect of NSC assumptions on other
mobilization preparedness issues
The Administration proposal stated that the NSC study assumptions
"also will be used for other appropriate mobilization preparedness
areas." Although official documents do not clearly identify the
specific areas, NSC study participants stated that two potential
uses were in deciding on requests for protection of U.S. indus-
tries endangered by foreign competition,2 and on providing direct
assistance to U.S. industries under the Defense Production Act of
1950.
The use of the NSC study assumptions could affect the implementa-
tion of the above Acts. For example, under the NSC assumptions, a
lower than previous level of wartime material needs for manufac-
turing resulted in less need for stockpiled materials. By the
same assumptions, it could be argued that with less manufacturing
need, U.S. industry would need less assistance or protection from
foreign competition. Conversely, assumptions that would require
more manufacturing and stockpiling could be used as the basis for
more assistance to, or greater protection of, U.S. industry.
Thus, our concerns about the effect of assumptions used to set
stockpile goals in the NSC study, also apply to the effect of the
assumptions on other mobilization preparedness areas.
BASIS FOR INTERIM STOCKPILE PURCHASES OR DISPOSALS
The Administration's Fiscal Year 1987 Annual Materials Plan
proposes purchasing 30,000 kilograms of germanium for $30 million,
and selling 37 materials for $303 million. It also proposes
selling 9 additional materials with no specified value. The plan
notes that changes to existing legislation are required to
implement it.
Because of congressional concerns about the adequacy of the NSC
study as a basis for setting stockpile goals, decisions on
identifying surplus materials for disposal could be further
delayed. If some acquisitions and disposals were deemed neces-
sary, however, alternatives may be available. For example,
disposals could be considered from the about $3.4 billion worth of
2Section 232 of the Trade Expansion Act of 1962 provides that
the head of a federal agency or some other interested party may
ask the Secretary of Commerce to determine the impact on national
security of importing products to the United States. The Secre-
tary may investigate and report his findings to the President, who
may take such action as he deems necessary.
21
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX II APPENDIX II
material where both the 1979 and NSC studies agree that current
inventories exceed stockpile needs.
Figure-H.4 focuses on the similarities between old and new
goals. It shows the materials that both studies concluded were
excess to U.S. national security needs. Only about $146 million
of the $303 million the Administration proposes to sell are
surplus under existing goals. (However, we are exploring initial
indications that both the NSC and 1979 studies may have under-
stated the requirement for tungsten).
Figure fI.4:
Stockpile Materials that Are Excess
Under Both Existing and Proposed Goals
(Total Agreed Excess--$3.4 Billion)
-.71737.6
Tin
Chromium iiiiMMOMOMMI958.8
852.5
Saver 852.5
Manganese (a) 159.5
490.6
137
Tungsten 410.5
Diamond (b) 349.9
Aluminum (c) h2
128.6
rZ
(1? Antimony 1.6392.7
48.3
Mercury 151.4
8.8
Iodine j4lM
23.1
Silicon Carbide 36.2
Mica (d) _c 12.7
13.6
7.5
Quartz T
Mica (e)2
3.1
Mica (f) -.4
LEGEND
2253.7
Existing excess
Proposed excess
4
0 250 500 750 1.000 1250 1500 1750 2000 2250 2500
Excess value (dollars in millions)
22
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
APPENDIX II APPENDIX II
a Chemical and metallurgical group
b Industrial stones
Abrasive grain group
d Muscovite film
e Phlogopite splittings
f Muscovite splittings
Although germanium is the only material that the current inventory
falls short of both its current and recommended goals in terms of
quantity, other inventory items may need improvements in quality.
Commodity experts among the study participants and advisory
committees say that some materials, such as cobalt, may need to be
upgraded in quality to meet the needs for which the materials are
being stockpiled. Using proceeds from the sale or barter of
excess materials, or of moneys already available in the National
Defense Stockpile Transaction Fund, would seem an appropriate
means of accomplishing this upgrading and to purchase any needed
materials.
(391553)
23
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Requests for copies of GAO reports should be sent to:
U.S. General Accounting Office
Post Office Box 6015
Gaithersburg, Maryland 20877
Telephone [phone]
The first five copies of each report are free. Additional copies are
$2.00 each.
There is a 25% discount on orders for 100 or more copies mailed to a
single address.
Orders must be prepaid by cash or by check or money order made out to
the Superintendent of Documents.
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7
United States
General Accounting Office
Washington, D.C. 20548
Official Business
Penalty for Private Use $300
Address Correction Requested
First-Class Mail
'Postage & Fees Paid
GAO
Permit No. G100
Declassified and Approved For Release 2013/07/23: CIA-RDP90B01013R000200050004-7