Foreign Relations of the United States, 1969–1976, Volume VII, Vietnam, July 1970–January 1972

203. Memorandum From K. Wayne Smith of the National Security Council Staff to the President’s Assistant for National Security Affairs (Kissinger)¹

Washington, May 20, 1971.

SUBJECT

• Friday (May 21st) SRG Meeting on Vietnam Assessment

Status of Vietnam Assessment

All the Vietnam assessment studies requested in your April 15 and 28 memoranda have been completed with the exception of the air interdiction study.²

The logical sequence for SRG consideration of these studies would be:

(1) May 21 SRG: Military Assessment

• —CIA papers on manpower requirements for alternative strategies and Hanoi’s capacity to continue to wage war.³
• —JCS assessment paper on prospects for 1972.⁴
• —OSD Assessment and RVNAF improvements paper.⁵
• —ROK alternatives (OSD paper and Saigon cables).⁶

(2) Political and Economic Assessment (not yet scheduled)

• —SNIE (political and economic prospects)
• —economic stabilization paper
• —regional cooperation issues

(3) Two other meetings should be scheduled on Cambodia (DOD team report and CIA memorandum) and ceasefire (VSSG paper will be ready by June 6).

You will need to ask DOD to do an air interdiction study.

Meeting Strategy

Goals—You want to come away from this SRG meeting with agreement that:

• —(1) We need an integrated short paper encompassing the contributions in the CIA, JCS, and OSD papers. The framework for this summary paper should be an estimate of most, best, and worst probable 1972 situations based on differing assumptions on:

  • —the enemy threat
  • —the required main force ratio in MRs 1 and 2 to maintain the main force balance
  • —the requirement for RVNAF forces in Cambodia.

  The analysis should be done for a 100,000 and 50,000 U.S. force level and on the assumption that the two ROK divisions remain in MR 2. The concluding section of the paper should be a discussion of how the gaps and surpluses might be equalized with the currently planned RVNAF force. It should identify net shortfalls for the most, best, and worst probable cases.

• —(2) A commitment from DOD to develop alternative RVNAF improvement packages for consideration by the SRG. DOD has given you one package consistent only with the best probable estimate. The basis for the other two packages already exists in the NSSM 99 work, a summary of which you have seen in my earlier Vietnam assessment paper.⁷ We need to force OSD to produce a similar paper.
• —(3) A commitment from DOD to do the interdiction study. This study should encompass alternative sortie level/mix packages and alternative air/ground interdiction packages. This is a sensitive issue with Laird on which no new DOD analysis has been done since the fall of Sihanoukville and Lam Son 719. Your request for such a study in the April 15th memorandum (tabbed separately)⁸ has gone unanswered.
• —(4) A decision in principle to keep at least two ROK divisions in Vietnam through FY 73 and a decision to cable our Saigon and Seoul Missions to that effect and solicit their views on how best to approach the GVN and ROKs.

Approach—You should not show your hand at the outset of the meeting. Rather you should review in turn the JCS assessment paper, the CIA memoranda on new manpower estimates and on Hanoi’s capacity to wage war. Such a review, with short JCS and CIA briefings will provide a basis for turning to the OSD paper and showing that the calculations therein are based on a “best probable” set of assumptions.

Then you can ask for the best, worst, and most probable analysis, the assessment of what this analysis would imply given the presently planned RVNAF force and 50,000 or 100,000 U.S. forces.

Finally, you will want to remind DOD that they owe us an air interdiction study.

Your talking points are designed to achieve the goals set out above using the strategy I have described. JCS and CIA are prepared to give short briefings on their papers.

Your talking points are tabbed separately.⁹

The balance of this memorandum reviews the papers to be considered

JCS Assessment Paper

The JCS paper is more optimistic than the earlier CIA analysis¹⁰ on the prospects for 1972. However, it is also vague on the key issues and on the basis for its conclusions. Thus there is a basis for closing the gap between its conclusions and those of CIA. The significance of the JCS paper is that:

• —the JCS prepared it,
• —the JCS paper is well presented and considers a full range of alternatives.
• —the JCS paper spells out the military’s judgment on how alternative mid-1972 U.S. force levels (150,000; 100,000; and 50,000) will affect the enemy’s options.
• —the JCS conclusions, while vague, can probably be integrated in a manner consistent with the CIA’s estimates of what the enemy can do.

The JCS estimate of the probable enemy strategies is shown in the table below.

	1971			1972
	May Sep	Oct Dec	Jan Feb	Mar Apr	May Sep	Oct Dec
Case I (Prot)	X	X	X	X	X	X
Case II (MR 1)		11	X	X	12	X
Case III (MR 2)			X	X	14	X
Case IV (MRs 1&2)				13	14	13
Case V (Camb)				13	14	13
Case VI (MR 1/Camb)	- - - - - -- - - - - -Not Feasible- - - - - -
Case VII (RVN/Camb)	- - - - - -- - - - - -Not Feasible- - - - - -

Major substantive points in the JCS paper are:

• —“It is expected that Hanoi will attempt to gain at least one dramatic tactical victory in RVN [e.g., seize Hue or Pleiku] or Cambodia in 1972 to improve NVA morale and diminish GVN and U.S. will to continue.”
• —Possible fixes for territorial forces not already in train are not likely to have a major impact on the control situation and thereby the main force balance in 1972.
• —The enemy threat section (annex A, p. 14) concludes that “Hanoi will continue its protracted war strategy (Case I) through 1972 with some surges from time to time . . .”
• —At a 150,000 mid-72 U.S. force level friendly forces could cope with a major enemy offensive in MR 1, MR 2, Cambodia, or MRs 1 and 2 simultaneously.
• —At a 100,000 mid-72 U.S. force level, a permanent redistribution of GVN forces to MR 1 would be required to permit friendly forces to cope with a MR 1, MR 2 or Cambodia offensive. A simultaneous MR 1 and MR 2 offensive would exceed friendly capabilities, but the military losses would not be great (a district town occupied). The enemy would achieve a propaganda victory.
• —At a 50,000 U.S. force level, the GVN capacity would be exceeded by an enemy offensive in MR 1, MR 2, or Cambodia or combination attacks in any two of these areas. ARVN would lack the combat support capability to “maintain a significant force in Cambodia.”

Issues—The JCS analysis is thorough and generally very competent. I recommend you read the first 14 pages and Annex B, pages 9–13. The following issues merit discussion:

• —Referring to the same table on page 4,¹⁵ what does “at the margin” mean with regard to the MRs 1 and 2 and Cambodia offensives? If the enemy is not constrained by logistics in MRs 1 and 2, why can’t he mount an MR 1 and 2 offensive? On what basis can it be said the enemy has only a marginal capability to mount an offensive in Cambodia if he sets out to do only that?
• —Again referring to the table on page 4, what prevents the enemy from attacking in Cambodia and MR 1 simultaneously (Case VI)? Wouldn’t a smart enemy attack in Cambodia in order to keep ARVN from moving forces to MRs 1 and 2? On page 13 it is said the enemy can accomplish more with less in Cambodia than in South Vietnam.
• —Is there an inconsistency between the conclusion of Annex A, page 14, that protracted war is the most probable 1972 enemy strategy and the analysis of the friendly situation in Annex B, pp. 9–13? Namely, if our forces drop to 100,000 and ARVN does not permanently deploy additional forces to MR 1, won’t the enemy change his strategy? Or if our forces fall to 50,000 and ARVN does what it can, won’t the enemy still have the capability to mount a major offensive. Why won’t he do so?
• —The assessment of the implications of a 50,000 U.S. force level (Annex B, p. 13) becomes awfully vague. This point is brought home on pp. 12–13 of the summary when we are told that at none of the force levels will enemy attacks be “militarily decisive” but “they could lead to a deterioration in morale and confidence in RVN and GVN control of its population and territory.” How would the JCS describe what occurred in South Vietnam in 1964–65?

CIA’s Revised Estimates of Manpower Requirements for Alternative Enemy Strategies

CIA has revised its manpower estimates to account for the effect of U.S. redeployments on: (a) force augmentation requirements in the preparation phase for an offensive and (b) enemy KIA estimates in the execution phase of an offensive.

	Manpower Requirements (000)
Strategy	Augmentation Req.			Total Infiltration
	Old	New	Old	New A	New B
Protracted	N/A	N/A	100	70	70
MR 1	20	0	120	85	65
MR 2	16	14	115	80	78
MR 1 and 2	36	14	135	95	83
Cambodia	20	20	150	105	105
MR 1/Cambodia	40	20	180	125	115
RVN/Cambodia	66	28	250–300	175–210	137–172

The new A estimate adjusts the “old” downward by 30% to reflect the reduction in KIA due to U.S. redeployments but does not reflect the lower augmentation estimate. CIA prefers this column because: (a) the enemy will probably plan against some U.S. force above 44,000 and (b) there may be some double counting in adding together the reduced KIA and augmentation estimates.

The new B estimate risks the double counting problem by adding both columns together. It is probably closer to the 1973 estimate whereas new A is the best approximation of 1972 infiltration requirements.

CIA concludes that “the principal effect of the reduced infiltration requirement is to bring all strategies into the range of feasibility during the 1971–72 period, though the general offensive strategy would continue to pose considerable strains on North Vietnam’s military conscription and training organization and result in some draw-down in manpower reserves. The MR 1/Cambodia strategy, which in our original analysis was probably only marginally feasible, would be well within the realm of possibility under the reduced casualties variant.”

North Vietnam’s Capacity for War

CIA has produced an excellent memorandum on North Vietnam’s capacity to wage war. Unfortunately this memorandum is based on the “old” manpower requirements estimates just reviewed and discarded above.

The analysis is nonetheless illuminating. I urge you to read pages 1–8, and pages 25 and 26.

The major points in the memorandum are:

—(1) The table below on estimated inductions by year:

	Thousand men
1965	155
1966	260
1967	185
1968	215
1969	80
1970	200 (preliminary)

Given CIA’s estimates that 100,000 inductions per year will sustain a protracted war and that a lead time of more than a year is required for a major mobilization, does the 200,000 estimate for 1970 imply major offensive activity in 1971–72?

—(2) Hanoi would appear to be hard pressed to repeat its 1967–68 performance of two successive years of inductions at the 200,000 rate. The manpower pool has dropped from 1.6 million in 1964 to 800,000 in 1970 (17 years to 35 years). CIA estimates that two years at a 300,000 induction rate would run this pool down to 600,000 in 1971 and 400,000 in 1972. Thus there appears to be a limit to what Hanoi can do. However, this limit is above the 125,000 infiltration requirement for a MR 1/Cambodia strategy that we fear most for 1972. The enemy is more pressed logistically than he is in terms of manpower, if one dares to compare these very different types of constraint. CIA should be asked if this judgment is a fair conclusion based on their analysis. [Such a conclusion would represent a fundamental departure from CIA’s long-held view that Hanoi was concerned for manpower rather than logistics. U.S. redeployments and the loss of Sihanoukville may have changed CIA’s viewpoint.]

—(3) CIA presents an interesting table comparing North Vietnam’s mobilization with South Vietnam’s. According to the table (p. 4), in 1970 46% of South Vietnam’s 18–39 year old males were in service versus only 21% for Hanoi.

CIA may have neglected to include Hanoi’s paramilitary forces although the GVN’s RF and PF forces are included. Your talking points raise this issue.

OSD Paper

OSD circulated to the SRG yesterday the same Vietnam assessment Secretary Laird sent to the President. [A DOD messenger retrieved one page late yesterday and provided a substitute that blanked out the fixed date/POW proposal.]

In addition OSD circulated a JCS paper on modifications in the RVNAF which in Nutter’s words “is in accordance with the [OSD] assessment.” In actual fact the JCS paper proposes only minor fixes for FY 72 and their impact would not in the judgment of DOD’s own analysts, close the gap described in OSD’s assessment.

OSD Assessment

The crux of OSD’s assessment lies in the assumption behind the table reproduced below (from p. 3 of the OSD paper).

	Ratio of RVNAF Main Force Bn Strength to VC/NVA		RVNAF Surplus/Deficit
	Projected	Desired	Troops	Bns
MR I	0.6	1.1	-26,500	-44
MR II	2.8	2.5	+ 4,700	+ 8
MR III	3.0	2.5	+ 6,500	+11
MR IV	3.7	2.5	+10,100	+17
Reserve			+ 9,100	+12
RVN Total			+ 3,900	+ 4

The points I make below have been fully checked with the OSD/SA analyst (Lt. Col. Eddins) who put the table together. He has worked closely with us in the MR studies. Unfortunately ISA has filtered his analysis to Laird and ISA does not understand it fully.

The analysis is based on the following assumptions:

• —The most probable threat to South Vietnam in 1972 is an MR 1 offensive. This is what is termed by Laird “CIA’s estimate which we consider maximizes both the size and supportability of possible VC/NVA offensive operations.”
• —50,000 U.S. forces in South Vietnam at mid-72.
• —ROK’s remain in MR 2 until FY 73.
• —A “desired” main force ratio (MFR) of 1.1 for MR 1 and 2.5 for MR 2 would maintain the main force balance in these areas with the risk of only slight control losses.
• —There is no requirement for the permanent stationing of RVNAF forces in Cambodia or for operations against South Laos of long duration, i.e. ARVN forces would be withdrawn from Cambodia.

These assumptions are reasonable. In combination they represent a best probable outcome. Most and worst probable outcomes need to be estimated as well.

But even accepting OSD’s best probable assumptions, the RVNAF improvements called for by the analysis far exceed what in OSD’s view are required. For example, there is a 44 battalion deficit in MR 1 that can only be filled in part with ARVN’s present 31 battalions of mobile forces (marines, paratroopers, and the 9th Division). Thus another division must be added in MR 1.

Other assumptions would considerably magnify the task of restructuring RVNAF to meet the 1972 threat. I state below the key assumptions in the best, worst, and most probable variants.

Threat	Best (Laird) MR 1 Offensive	Most Probable MR 1 and 2	Worst MR 1 Cambodia
Desired MFR
MR 1	1.1	1.5	2.0
MR 2	2.5	3.0	3.0
Main force requirements for Laos and Cambodia	0 bns	10 bns	15 bns

All cases would assume the ROKs remain in MR 2 and the U.S. force level is 50,000 in mid-1972.

In your talking points you ask the SRG to examine best, worst, and most probable cases.

Points you make to convince them that this is necessary are:

• —The OSD analysis assumes there will be no requirement for RVNAF forces in Cambodia or Laos in 1972. Ask OSD if this assumption is consistent with the findings of the DOD team.
• —Whereas the OSD analysis assumes a MR 1 offensive in 1972, the CIA and JCS/DIA analysis of the threat make it probable that the enemy can exceed an MR 1 offensive. More likely is a MR 1 and 2 offensive. Worst probable but within Hanoi’s capability is an MR 1/Cambodia offensive.
• —The MFR’s used in the analysis for MRs 1 and 2 are below what to date have been considered necessary to maintain the main force balance, even with a substantial U.S. force component included. In the most probable case these ratios should be increased from 1.1 to 1.5 for MR 1 and from 2.5 to 3.0 for MR 2. You can illustrate this by asking OSD if it believes the GVN will be able to take 8 battalions (almost a division) from MR 2 in 1972 as the analysis indicates. GVN control in MR 2 has fallen from 54.6% to 52% since December and in the northern part of the MR it has dropped about 10%.

If these modifications are made in OSD’s analysis, all within the best, worst and most probable framework, the countryside RVNAF 4 battalion surplus will vanish and the fixes required will become of a greater and more realistic magnitude.

JCS RVNAF Fixes Forwarded by OSD —The OSD paper includes as Tab B a list of JCS proposed improvements for RVNAF.

These are minor and focus on the combat support and logistics capabilities of RVNAF that must be expanded in 1972 because of the accelerated rate of U.S. redeployments.

• —The current FY 72 goal of a 1,092,087 man RVNAF is accelerated to 1,100,000 or the FY 73 goal. This represents a 7,913 increase.
• —ARVN grows by 3,219. One air defense artillery weapons battalion and three station hospitals are added. An additional armored cavalry squadron is added in MR 2, 10 M–106 mortar platoons are added to division armored cavalry squadrons, 10 military police companies are added, and 17 RF battalion headquarters and 219 PF platoons are added.
• —The Vietnamese Navy grows by 1,070. The Coastal Surveillance Radar System is activated and 29 river patrol boats are added.
• —VNAF grows by 1,539. No new VNAF units are created. The logistics and maintenance forces are strengthened.

ROK Forces

There are two issues. The timing of ROK withdrawals and deployment within SVN. The ROKs have indicated an intention to begin withdrawal of one division equivalent after the SVN presidential elections. The GVN wants all ROKs to stay until CY 73 because they believe earlier withdrawals would expose the critical coastal areas of MR 2 where the ROKs operate.

The interagency paper produced 4 sets of options which are reduced to table form in your book for ready reference. MACV favors option 4 which would involve withdrawing ROKs in parallel with U.S. forces, with the last combat ROKs leaving in December 1972. This option is the most convenient in terms of phasing out present U.S. logistics for the ROKFV. It does not take into account GVN desires or the possible political and military desirability of continued ROK combat presence in 1972 and into CY 73.

The Washington Agencies take no explicit position on the options although they believe that there should be no increase in support costs for the ROKs and that present funding arrangements should continue to January 1, 1973, with follow-on arrangements to be examined at a later time. (This position effectively precludes redeployment of ROKs within SVN.)

In the cable traffic Ambassador Bunker originally favored the MACV option for ROK withdrawal in phase with our redeployments. In a subsequent message to the President, however, he appears to have shifted to the GVN point of view.

The other issue is deployment of ROK forces within Vietnam. The GVN, MACV and JCS favor continued utilization of ROKs in MR 2 without redeployments to MR 1. The rationale is twofold: (1) the need for main forces in critical MR 2 coastal areas and (2) the quid pro quo of 80–100 million in equipment likely to be demanded by the ROKs for shifting forces to MR 1.

Your purpose at the meeting should be to ensure that we do not foreclose the possibility of continued ROK combat presence if that is what the GVN and ROK agree upon and also to keep open the possibility that at some point ROK redeployment to MR 1 may be desirable depending on the situation there. You may also want to point out that the MACV position is based on an assumption that the enemy continues his protracted war strategy. We can all agree that no one would advocate continued ROK combat presence if it were only to meet a protracted war threat.

1. Source: National Archives, Nixon Presidential Materials, NSC Files, NSC Institutional Files (H-Files), Box H–56, SRG Meeting, Vietnam Assessment 5-24-71, 1 of 2. Top Secret. Sent for urgent information. All brackets are in the original.
2. A list of Kissinger’s requested studies is in Document 179.
3. CIA Intelligence Memorandum ER IM 71–84, May 10, “Manpower Considerations and Hanoi’s Strategic Options,” is in the National Archives, Nixon Presidential Materials, NSC Files, NSC Institutional Files (H-Files), Box H–56, SRG Meeting, Vietnam Assessment 5–24–71, 2 of 2. CIA Intelligence Memorandum ER IM 71–88, May 14, “North Vietnam’s Capability To Continue To Wage War,” is ibid.
4. Knowles forwarded the report, “RVN Assessment May 1971,” to Kissinger, Johnson, Packard, and Helms under covering memorandum CM–921–71, May 20. (Ibid.)
5. Laird forwarded a paper entitled “Assessment of the RVNAF Improvement and Modernization Program” to Nixon under a May 1 covering memorandum. (Ibid., Box 154, Vietnam Country Files, Viet 9 Apr 71–30 Apr 71)
6. Not further identified.
7. See Documents 39 and 102.
8. Attached but not printed.
9. Attached but not printed.
10. See footnote 2, Document 192.
11. Logistically feasible, but weather probably will not permit a major offensive in northern MR–1 until late 1971 or early 1972. [Footnote in the original.]
12. Weather favorable in northern MR 1, but logistic support would be difficult, unless supplies were prestocked, because of rains in Laos. [Footnote in the original.]
14. Although logistic situation might support a major offensive, the weather probably will not permit execution. [Footnote in the original.]
13. Considered to be at the margin of the enemy’s capability, since execution would depend on the enemy increasing substantially his logistic throughput in spite of weather and the assumed continuation of allied efforts to interdict. [Footnote in the original.]
14. Although logistic situation might support a major offensive, the weather probably will not permit execution. [Footnote in the original.]
13. Considered to be at the margin of the enemy’s capability, since execution would depend on the enemy increasing substantially his logistic throughput in spite of weather and the assumed continuation of allied efforts to interdict. [Footnote in the original.]
13. Considered to be at the margin of the enemy’s capability, since execution would depend on the enemy increasing substantially his logistic throughput in spite of weather and the assumed continuation of allied efforts to interdict. [Footnote in the original.]
14. Although logistic situation might support a major offensive, the weather probably will not permit execution. [Footnote in the original.]
13. Considered to be at the margin of the enemy’s capability, since execution would depend on the enemy increasing substantially his logistic throughput in spite of weather and the assumed continuation of allied efforts to interdict. [Footnote in the original.]
15. The reference is to the table printed above.