THE SECRETAFIY OF DEFENSE WASHINGTON SUBJECT Rec ended FY l966-1910 Programs for Strategic Offensive Forces Continental Air and Missile Defenae Forces and Civil Defense U I bave ccupleted my reviev of the three major cc mponents of our General Nuclear War posture the Strategic Offensive Forces the Conti Dental Air and Missile Defense Forces and Civil Defense ·•-·- -- -· ru·s memorandUQ suzmnarizes ni c -ac eris ics ol ' our current strategic posture the major progra ms proposed by tbe Services ll '·recamnended program and the ration ale for choice ainoog these aJ ternatives The esti llated costs excludiDg R D and reserve forces or the previously approve Service propooed and recommended progrB IIIS are presented belO'II Tot al FY 65 FY 66 FY 67 FY 68 FY 69 FY 70 FY 66-FY 70 --rTotaI obliga tionaI'autbori ty $ millions Previously Approved 11l9 Service Proposed 8237 SecDef Rec iended 7184 54l3 5024 10597 9063 5190 4 46 S 15 26 850 Tbere are six major issues 1 nvolved 1D our FY 1966-1970 program for the General Nuclear War Forces These issues concern l The develop ient and deployment of a nev manned bDl lber estimated 5-year systems cost for a force of 200 aircraft -$8 9 to $ll 5 billion 2 The size of the strategic missile force estimated 5-year cost for an e dd tional force of 200 MINU'JlMUi II missiles -- $1 3 billion 3 The overall level of the anti-bOI lber defense program estimated 5-year cost if units rec ended for phaseout are retained -- i300 to $350 mill 1011 4 The production and deployment of a nev manned interceptor estimated cost for a force of 216 IMI aircraft -$4 bill on - --P_re_l_1_m_1_n_e ry __ - ' - t t s to be revised at'ter c pletion af budget reviev FJI CASE -J ' ID · 1- 7 7 Doc i ent_ _'t__ of__ f_ _Doc ents - I • -• ' C •• 2 5 production and deployment of the NIKE X antimissile system estimated 5-yea r systems cost depending upon the mode ot deplo 'l lent 1number11 ot radars and numberG ot cities covered 11 to 47 J-- $8 to $24 bill1on 6 The construction ot te ll out shelters for the e11tire population estima ted cost -- $5 2 billion Bef' re I discuss these major issues and my other recoJJm1endations to improve our general nuclear var· capabil1 ties I believe it vould be use to -ev1ev the nature of tbe general nuclear var problem itsel f the characteristics of proper 'balanced general D Uclear var forces and the capabilities ot the presen -programmed forces A HA'roRE OF THE GENERAL iUCLEAR ilAR PROBLEM By general nuclear var ve mean a var ill vbich strategic nuclear yea pons are directed a gaillst the homelands of the U S or the U S S R Such attacks might be directed against lllillta ry targets only cities only or both either simultaneously or vith a delay they might be selective ill terms of targets or they might be general The fol loving types ot strategic forces are illvolved l Strategic Otfensive Forces l-anned bombers ICEM9 and submarine-launched missiles together vith the associated camnand and control systel is • 2 Strategic Defensive Forces Allti-aircra tt defenses manned interceptors surface- to-air missiles ard their associated varning a nd control systems Anti-ballistic missile defe115es varning systems and active defense Gystems 3 Civil Defeos e Programs FaJJ out shelters 'Wal Iling etc It may be assumed that both the United States a nd the Soviet Union have the same general Gtrategic objectives l To deter delibe -a te nuclear attack by maintaining a clear and convincing capability to illf'lict severe dw nage on the attacker even after a n enemy first s1 rike and 2 In the event suer a var should n vertheless 3 occur to limit d ama ge to capacity its cr m population and industrial The first of tbese objectives ve call Assured Destruction i e tbe capability to destroy both the Soviet Union and Ccmmmist Cbina as viable societies even after a veil planned and executed surprise attack on our forces Or in tbe vords or tbe Joint Chiefs or Staff • • • • tbe assured capability of destroying singly or in canbination tbe Soviet Union and the Camir mist satellites 1n Europe as national societies In canbination v1 th tbea tre nuclear forces • • • tbe ability to impose adequate punishment on Red China ror 1IUclear or non-1IUclear aggression The second capability ve call Damage Li mitation i e the ability to reduce tbe veigbt of tbe enemy attack by botb offensive a nd defensive measures a nd to provide protection for our population a ainst the effects of nuclear detonations Vieved 1n this light our assured destruction forces vould include a portion of tbe ICI Ms the submarine-1 allllched ballistic missiles SLB - s a nd the lllBJllled bombers The damage 11 miting forces vould include tbe remainder or the strategic offensive forces ICJ3Ms SLBMs a nd lllBJllled b llll ers as veil as area defense forces manned interceptors and termina l defense forces anti-bomber sur race-to-a ir missiles and anti-ballistic missile and passive defenses fallout shelters warning etc The strategic orrensive forces can contribute to the d ama ge limiting · objective by attackillg enemy delivery vehicles on tbeir bases or lallllch sites provided tbat our forces can reacb their targets betore the enemy vehicles are lallllched Area defense forces can attrit the enemy's forces enroute to their targets and betore they reach tbe target area s Terminal defenses can destroy enemy veapons or delivery vehicles vithin tbe target areas before they impact Passive defenses can reduce the vulnerability o r our population to the weapons tbat do impact Since each of the three types o r Soviet strategic offensive systems land-based missiles s lhma rine•laUilcbed missiles a nd bCX lbers could by itself inflict severe damage on tbe United Sta tes even a very good defense against only one type of system bas 11 i 1ted value A very good defense against 'l o i bers ror eX lmple could be outflaz ked by targeting missiles 14 ainst those areas defended solely 4 by anti-bCl lller systems 'I'h1s 1s the principal reason vby 7 today 1 ill the absence of a n effective defense aga 1 Dst missiles the large U S outlays 01' the la st decade for mamied ba iber defense by themselves nov contribute little to our real strategic defense ca pa billty reover the a nti-banber defense system designed a decade ago is itself' vulnerable to missile a-t ta ck Thus a significant capa bilt t- '' to l1m1t the damage of a determined Soviet a tta ck requires lll illtegrated balanced ccmbination or strategic offensive forces area defense forces term1na l defense forces and passive defenses Such a be la need cCl lbills tion creates a defense in depth Yith each type of force ta king its toll of the incOIUilg veapons 7 operating like a series of filters or sieves vhich vould progressively reduce the destructive potential of the attacking Soviet nuclear forces B CHARAC'lZRISTICS OF PROPERLY BALAl'iCED GENERAL NUCLEAR WAR FORCES It is generally agreed that a vital first objective to be met ill full by our strategic nuclear forces is the capability for assured destruction Such a capability vould 7 Yith a high degree o r confidence ensure tba t ve could deter under all foreseeable conditions a calculated deliberate nuclear attack upon the United States What amounts and kinds of destruction ve vould ba ve to be able to deliver in order to provide tb s assurance cannot be ansvered precisely but it seems reasonable to a as that the destruction of say 5 percent of its population 55 million people a nd more than two-thirds of its industrial cape city vould mean the destruction of the Soviet Union as a oational society Such a level of destruction vould certainly represent intolerable punishment to any industrialized oation and thus should serve as a n effective deterrent Once an assured destruction ca pa bility has been provided any further increase 1n the strategic offensive forces must be justified on the basis of its contr bution to limiting damage to ourselves Here certain basic principles should be noted First against the forces ve expect the Soviets to have during the next decade it vould be virtually impossible for us to be able to provide anything approaching perfect protection for our population no matter how large the general nuclear var forces ve provide even vere veto strike first Of course the number of survivors in a general nuclear var depends OD Soviet forces as veil as ours The Soviets have the technical and econonic capacity to prevent us ran assuring that more tha il 8o percent of our population vould survive a determined attack possibly less They can do th s by offsetting any increases 1n our defenses by in reaeee 1n their missile forces If ve vere trying to 5 protect a high percent e g 8o or more of OU population and if the Soviets were to choose to frustrate this attei Ipt possibly because they vieved it as th -eatenillg their assured destruction capability the extra cost to them appears to be substantia ll y less than the extra cost to us The question of ball' much we should spend on damage lilll1ti ng prograz i s can be decided only by carefully ve1ghing the costs against expected bene1'1 ts The second basic principle vhicb must be borne in mind is that for any given level of enemy o1'fens1ve capability successive additions to each of our various systems and types of defenses have d minishing marginaJ value While it is true that in general the more 1'orces ve have the better ve can do beyond a certain point each increment added to the existing forces results in less and less additional e N'ectiveness Thus ve should not expand one element of our damage lilll1ting forces to a point at vhicb the extra survivors it yields per dollar spent are fever tha l for other elements Rather any given amount of resources ve apply to the damage limiting objective should be allocated Bl IOilg the various elements of our defense forces in such a vay as to max1rn1ze the population surviving an enemy attack This 1s vhat ve mean by a balanced damage l i1 l1 ting force structure The same principle holds far the damage limiting force as a vhole as additional forces are added the incremental gain in effectiveness diminishes When related to our other national needs both military and non-m1l ita ry this tendency for dirn inishiilg marginal returns sets a p --a ctical lilll1t on ball' much ve should spend for damage 11 d ting progr • Then there 1s the factor of uncertainty of vhich there are at lea st three t 18 jor types -- technical operational and strategic Technical uncertainties stem ran the question of whether a given system can be developed vith the performance characteristics required Operational uncertainties stem fr im the question of whether a given system will actua ll y perform as planned ill the operational environment This type of uncertainty 1s particularly critical with regard to general nuclear var since so little is actually kllovn about the kind 01' operational environment such a var voul d create The third type of uncertainty 1s perhaps the most pervasive since it stel ls frCJ l the question of vbat our opponent or opponents vil l actually do -- vhat k nd of force they vill actua ll y build vhat kind of attack they vill actually lallllch and ball' effective their weapons 6 vill actually be etc Wbat may be a 1 optimum defense against one kind o r attack may not be an optimul L ue fense against a different kind of -attack For e Ple Vi thin a given budget a IIXE X defense optimized for an attack by 200 ICJ Ms vou1d defend more cities vi th fever interceptor missiles than a defense optimized for an attack by 6oo ICEl-l s S1m1Jar y a NIKE X defense optil lized against an attack by ICBMs vi th simple penetration aids wuld have rever high cost radars tba D one optimized against an attack by ICllMS Vith more advanced penetration aids ID the same vay the effectiveness or our strategic missile forces in the damage limiting r le vould be critically dependent on the timing o f a Soviet attack on U S urban targets 'nlese forces vould be most effective against the Soviets'' bCl lbers and ICBMs if they withheld their attack on our urban targets for an hour or more Our IIIBJllled bomber force 'll'Olll d be etfective in the damage limi tillg role only it the Soviets Withheld their attacks against our urban centers for eight hours or more To reduce the technical uncertainties ve rely on paillf'teking studies and research and devel opnent tests and to hedge agllin st the risks ot technical t 'ailure ve may support parallel developnent approaches 'We try to cope Vi th the operational uncertainties by repeated testing in a s1mulated operatiODal environment but this approach bas sane very definite limits for general nuclear var tn es of operations 'We hedge against the stre tegic uncertainties tor e mmple by accepting a less tba D optimum defense against any one form of attack in order to provide some de t 'ense against several forms or attack and by purchasing insurance by keeping open various options -- to develop and deploy a nev bCl lber a new interceptor an nti-missile de t 'ense system etc Hw tar ve should go in hedging against these various uncertainties is one of the most difficult judgments vhich bas to be made Analytical techniques can focus the issue but no mechanical rule can substitute tor such Ju dginent C CAPABILITIF 5 OF 'IHE PRESENTLY-PROGRAME r FORCES FOh ASsum D DES'IBUCTION In order to assess the capabill ties of our genere J nuclear var forces c ver the next 11everal yea rs ve mu st also make BCllle estimates of the siz e a nd character or the Soviet forces during the same period The table belw sU lllll ' riz es current estimates of Soviet strategic offensive forces for· the mid-1965 -1967 and -1970 periods United States forces for the same time periods are sbmm tor cc mpa rison - ·- · ·· · - U S 2'S SOVIET STRATEGIC NUCLEAR FORCES · U S f I ' - · _ - Mid 1970 ·'· · _' ' U S USSR 1· _ · i ·• -- -- _ _· · - ·· · · • ·•• O _ 146 854 91-ll6 g - - · - ·· · ' _- J· · ' f · · -· So f 't La uncbers Ha rd Launchers Total ··•-- Mid 1967 U S • USSR USSR ' - · _ ·Mia ·1965 7 ' · · · · ' §2 2 35-266 - · O 147 -15 9 1054 181-237 ·33o-395 O 138-162 - · · · 105 gr2-_ 537 · ··· ·· · · · · · 1 05 410 1____ 700 · · - _ · t ' Ha rd Launchers Total o · 144-147 - - - 756-7b3 __o 756-763 •· · · - ··· - _ _ _ · 756-763 ·· · _-· B bers Ta nkers Heavy Medium Tota J • ··'·125 190-220 1205 17Q-210 770-850 76 5lio-755 l 75 960-1070 l28J 710-965 -· · · · · '· t·· ·· - _· H _ _ •--· -- _ · · - -· · -·· -_ _ - - _ - - -_ --- · - • - _ · · • -- _ Bo -- i 1205 72 290-510 - 0 •• • JZT7 430--690 • · _· r - ···· ' - '_ ' _· -- ·_ - ··-- • ·_· · - - -· _ __· · 2 _ _ · ' _ _ · •-_ - 1 ·_· -· · _· · ·_ · -· -· - · ·· -_·_ - __ ·-' • • •· _ - ·_ · ·_• · ·· -· ·- - -- ExcJ udes tes_t range launchers ha ·- op_ ational capability or vhich · tbe Soviets are estimated to bave1· - ·· 1 ll the mid-J 965 to mid-1970 · period _· · _ -·--c · · · '• ·'-'-··' -'_·· - ' · ·· · · · fJJ 1 El Di diti on SLBMs · the the Soviets vill possess ub bi laUDc d cruise missiles -whose pr ilnary targets v-e believe are DB val and · · merchant vessel s but -which may al so be used for sballO'II' penetrations of land areas mid-1965 175-207 mid-1967 247-3ll mid-1970 · - 244-388 · _ · -·- • ' · - • · ••• r · - • _ - •• · - • M- •o• _ _ ·-- ' - · ' ·- ------·--- ··- · · p Tl Blftl ·· C ·' 1 Character o Soviet strategic Forces _ • 8 ·- · · ·11 ' -_ •• · · · _ · · - ' ' ' Although projections of Soviet forces iJ I the lete 196os and eiu-ly · 1970s are necessa rily only iJ lto ed eS ti tes development and ·deploy- · · ment patterns al ready apparent have mede it possible to identify some · · - '- i broad trends - · ·· _ • - _ · - ' · At present about 85 hard SS-7 and SS-8 launchers con f'igured · -- - · · - - nth three silos per site have been identified as operational or _ ·• · __ • • under construction and the deployment ot the ss-77ss-8 in a sort · ··· ····· · contiguration nth two leimchers per site appears to be leveling · · · · '' ' · · oft at about 14o launchers For the sort sites one additional missile is probably available to each launcher alloving a re-fire capability but there is no evidence that this capability exists for hard_ '· la c ers · r t e _bard con f'igurations _silo desi h exdnes_ s _i - ' ' ' - ' - _ esti ted to be in the range o 200 to 400 psi • · _ · '-' - -- ·- - _ · ·· _ ·- ·' ' _ __ _ __ ________ - t _ · ·-• -_ · r-· · - · ··• - The deployment o the ss-8 at one t 1 me suspected to have been a very large peyload missile has been curtailed Analysis has indicated that the payload of the ss-8 missile is similar to that of the SS-7 approximately 4500 lbs · 'Host SS-7s probably have three MT vaxheads Hovever a nev nosecone nth si c Mr is probably avail able or missiles entering service this yea r · and some portion ot the existing forc·e 'Will probably be retrofitted 'ldth higher yield -warheads The developnent o a nev nosecone 'ldth arhead yields higher t three Mr for the SS-8 is considered unlikely A nev missile development beyond the successf'ul SS-7 progrem e Dd the not-so-successful SS-8 program has been con1'i ed T cis follov-on to the SS-7progra m esignated the SS-9 is expected to become operational in 1965 Probably larger than the ss-7 ss-8 the SS-9's payload is est l ulated at beween 8 000 end 13 000 pounds nth the yield possibly as high as 12-25 Hr We esti at that this missile 'ldll be deployed in a hard contiguration one launcher per site · · · · ' _ - The Soviets appea r to have leveled off their HRBM 102 -mi BIJd IREM 2200 n mi progr s _Tb is force is cieployed in a four launcher per site soft con figuration plus a re- fire ce pebility a thre launcher per site contiguration for the hardened IR3Ms e nd a four launcher per site · contiguraticn · for the hardened MRBMs We expect that the va rhee d yields of• Soviet MR 1 RBMs rlll be in the 25 KT to 6 r range There is no evidence o a foliov-on MR IRBH development · - - -- · · · · -· _ _ _ C -·- ___ ·_· •· - -- · s··· - • ·· _1 The trend 1n· Soviet submar iJ le construc'c1 n is not very clear There is s e evidence that the construction o f the ballistic missile G- and E class submarines hes stopped_ Allllost a11· Soviet ballistic missile sub ines a re eg_uipped 'ldth the 350 n cl bal 1 istic missiile hich hes a yield of 2 to 3 5 Hr Moreover the su bl mr'_ne must surface to · r1re • •' _ -------------------------·----------- ·i J LG_nC t - • • l _- ' -•v '•• • 9 •'•••C • ' t••-• • • µ - - - •••a•• _ • • ' •- i • • • _- ·••• ' •••v ' 't•• •••• • • 1 i I k1 i 11 J l t i i 14wiI3i 1 - - By mid-1970 Soviet sub r Des cou1d have the capability of be een 194-249 ballistic missiles __ _ · -· r · ·• · There is no evidence that the Soviets are developing a nev heavy bomber during the late sixties Barring this poss bility the projec'ted reduction in both the heavy e nd medium bomber forces 'I ill ·continue into the 1970s ·· Heavy b ber training in the Arctic hes emphasized extended navigational -- - --- -·--· flights into the polar basin BISON training is oriented tcn ·ard s those · activities no a lly associated 'I 1th a strike bamber role and BEAR tre ining has the added feature of reconnaissance specifically oriented against ships in the Atlan l ic e od Pacific The training of the medium bomber force bas been increasingly oriented tcn -ard continental or ni val rather then intercontinental operations The increasing age of· the heavy bomber and the continued phase-out of the BAIGER medium bomber 'I ill reduce both the heavy end medium bo nber c PQnents of Soviet Long Range Aviat on The output o r BLINIER medium bombers 'I ill probably continue to be· shared between long ra 'lge and naval aviation and it is believed th et in 1970 there 'llill be· s e 200-300 of these bombers in Long Range Aviation · Currently it is e sti mated that BArGER medium b bers do not figure prollrl nently in Soviet pl ens for an inititl b ber attack i gainst North America · llevertheless considering the requirements tor Arctic stl ging and re f'ueling as veil as noncombat attrition factors it is believed that at pre ent up to 150 BA ' ER s could arrive over North American target areas on tvo-vay missions The combat radius of these bombers would limit such attacks to targets in Greenlend Canada Alaska and the extreme northvestem U S • The short range of the BLINDER medium bomber lllUes it even less suitable ·· than the BArGER for atte cks against North America At present it is estimated that the Soviets could put s ewhat over 100 heavy bol lbers over target aree s in the U S on tw-vay clissions HOll'ever the use of Soviet hell vy bombers in me rltime reconnaisse oce roles leads to the belief that t ' ev of these aircraft might be diverted to this mission We had previously estimated that the Soviets ere constructing an anti-missile defense s- rstei i at Leningrad · ihich might be operational as ea 'ly as mid-1965 and one at Mosco- r to be opero t1onal about mid-1967 'i-lhile there is still considerable uoaertainty evidence since early s er indicates thn t the Leningrad syste i may be redirected vith pri ry capability i gainst aircratt and tadi l cissiles but little capebility age inst ICBMs S1m1J ar cooi'igurations have also e ppeared at several othe r- locetioi · -hich rould support the viev tMt if longer re ige interceptor missiles ere associated · 1th these sites this system is prl --ily designed to cope vith our s tr tegi 10 aircra 't tbxeat Radars at M91cov which ve believe are phased ll Tay rsde -s and ver previously associated vith 11 Ilti-missile defense may be ssociated vith the Soviet r pace tracking e N'ort11 The SA-2 missile system a high- and medium-altitude anti ai rcre rt defense 16 alrei tdy extensively deployed The SA-3 vith a 5UPPOeed lOll'-altitude capability vill probably be less extensively deployed thA l previously estimated Adequacy of Our Progre med M ' 11sile Forces tor 2 Assured Destruction In evaluating our assured destruction capability it is important to note that as sh by the table below euccesstul- attacks on a relatively small number ct targets e g 100 v1ll kill large numbers of people and destroy a high percentage ot the 1Ddustrial base Cul rulative Distribution ct Population and Indust bl Size ot Ci u s USSR Ind trial City P EulatiClll C J21 Ci I Rank Millions of Total ot Total ci l 2 3 10 20 50 100 150 200 7 3 11 l 12 6 20 3 28 8 44 7 58 7 67 0 73 4 3 0 4 5 5 2 8 3 11 8 18 3 24 o 27 4 30 0 0 2 13 1 14 8 25 0 36 0 52 0 64 o 69 0 73 0 P Eulation Millions or Totalj 12 4 21 4 28 6 52 8 70 1 97 5 112 0 130 0 136 0 5 9 10 4 13 6 25 1 33 5 46 5 5T O 62 0 65 0 Industrial Ca 11citz % ot Total 6 6 12 5 17 5 33 l 44 2 58 0 69 6 75 8 80 3 Note The total population base tor the Soviet Union vas taken to be the projected 1970 population ot 240 million vhereas the total population base tor the U S vas the 1970 projected base or 210 million ll ' 'he destructive potential of various s1z e U S attacks on Sov1et cities is shown in the folloving table assuming both the ex1st1Dg fallout protection in the Soviet Union vhich ve believe to·be m1nil lal and a new Soviet natir- • -vide fallout shelter program For purposes of this table it is assumed that delivered varheads have a vield of one megaton vhich is the approx imate size of both the • varheads Soviet Pop1 llation and Indust Z Destr ed AB a Fwlction of Delivered Warheads Assumed total popul ation of 24o m1lli n urban population of 14o million Delivered Ltd Urban Fallout Protection Megatons Total Urban Warheads Millions Millions 100 200 400 800 1200 1600 20 4o 15 57 77 90 41 97 29 56 65 70 25 46 68 94 109 118 Nation-'W'ide Fallout Program Urban Total Millions Millions i 11 J 6 19 28 39 45 49 71 84 92 12 21 35 52 61 67 · 17 32 51 74 7 13 21 31 95 39 87 36 Ind 50 65 74 77 79 Bo The point to be noted rrom this table ill that 4oo one megaton warheads delivered on Soviet cities so as to JMX1mize fatalities wuld destroy 4o percent o r the urban population and nearly o· percent of the population o r the entire nation If by the 1970s the Soviets ve i -e to provide a t'ull fallout shelter progrem for their entire population these percentages vould be reduced to about 35 and 21 respectively In either case IUlllOSt tbree- rourths o-r the illduatria l cspacity o r the Soviet Union vould be destroyed If the number of delivered varheads 'llere doubled to Boo the proportion of the total population destroyed 'IIOuld b increased by oozy about ten percentll8e points and the 1ndustl1 a l capacity destroyed by onl v three percentage points Furthtir increases in the nuober of varheads delivered produce smaller and sn aller increases in the percentage of the population destroyed and ne ligible increases in the industrial capacity destroyed This is so becawe we vould have to bring tlDder attack smaller and smaller cities each Ieq_uiri Dg one delivered varhead In ract vhen ve go beyond about 85c deli'V red varheads ve are attacking cities of less than 20 000 population 1 1 t · •' · 1- 12 · ·• Based on the projected Soviet threat f r the ee rly 1970s - the most likely planning factors calculations show that even e tter absorbing a Soviet first strike Jere Je to te rget all of our el ready authori ed strategic missile force ageJ nst population centers it could cause 105 million fatalities and destroy about ·78 percent of their industrial capacity _· even vithout employing our manned bomber force Indeed the use of the bombers for this mission about 600 ad ditioneJ veapons delivered vould increase fatalities by only 10 to 15 million eJJd industrial destruction by only a percent or-t IJo · And the bombers would be taking u oder atta ck cities of only 10 000 to 20 000 population The retention of the ATLAS and -TlTAN I through the early 19J0s which for reasons I discuss on Page 6 of Appe ldix A of this memorandum I recommend phasing out during the current · ·· fiscal year would increase the number of delivered weapons by less than 50 and · the assured destruction· capability by only a negligible emount - · ' ' c_ - · • - - · I 1 I I I I i- i J -- - _i i i I I I j_ The fact that the progrBIDed missile force alone more than provides an e deg_u ate capability for assured destruction does not imply that the job might not b_e done mbrc e t ricicntly by bo mber11 only or nth higher assurance by a mix ·of' bambers and mis-siles To test the first possibility i e ·using bombers a lone _ ' line examined the comparative 0 I 1· -· -- ·---·-·--------- • 13 cost and effectiveness of four a lteroative strategic systems IIJ N Jr - 1-' A l POLARIS B-52 SRAM and A l-BA SRft M is a proposed oev air-to-ground missile All SA is tbe oew bomber proposed by the Air Force Each system vas separately targeted to the Soviet urba iodustria l complex so as to briDg about 150 cities vith one-quarter of the population and tvo-thirds of the industrial capacity UDder attack Any one of the folloving forces alone could achieve this objective a MINUTEMArl Using expected operational factors 540 operational laUDchers vould be required total 5-year systems cost vould be $2 6 billion If the Soviets deploy an anti-missile defense system around 15 cities and if the Soviets assigned 300 of their ICBMs to attack MINUTEMAN 950 operational launchers vould be required 5-year systems cost of $4 5 billion b POLARIS With expected operational factors 640 POLARIS A-2 A-3 missiles vould be required 5-year systems cost as defined vould be $4 0 billion If the Soviets deploy an Ilti-missile de- fense system around 15 cities an additional 10 POUIRIS submarines carrying an improved missile proposed by the Navy vould be required the 5-year systems cost for the entire force vould be $6 2 billion • ' c B-52 SRAM Using expected operational factors 160 operational deployed aircraft vould be required tota l 5-year systems cost vould be $1 8 billion If the Soviets deployed an improved antibomber defense vith the same effectiveness the U S Army estimates for a U S advanced anti-bomber d fense currently under study then 500 deployed aircraft vould be required at a 5-year systems cost of $5 4 billion J y In this comparison MINlJ I'EMAN and POLARIS 5-year systems costs consist of the remaining R D and investment costs including missile replacement for Ff 1966 through 1970 plus five full years of operating cost 2 B-52 SRAM 5-year costs consist of all modification osts - including life extension of the B-52G and H from Fi 1966 through 1970 the development and procurement of SRAM and five full years of operating cost3 C -t I 14 d AY A Usins projected operatiooal factors 100 operational deployed aircraft vould be required total 5-year systems cost vould be $6 0 billion per Air Force estimates or $7 2 billion per OSD cost reV iev If the Soviets deploy the improved anti-bomber defense cited above lild if only 50 p rcent of the aircraft could be maintained on alert then 350 operational deployed aircraft vould be required at a 5-year systems cost of $16 billion per Air Force estimates or $18 billion per OSD cost reviev The 5-year systems costs of the 1·equired deployments of these four syste ls are sUll Cari z ed belO w · ID Billions Existing SoViet Defenses MThv r' ·lAN POLARIS B-52 SRAI All SA $ 2 6 4 o 1 8 6 0 - 7 2 Improved SoViet Defenses $ 4 5 6 2 5 4 16 - 18 It is clear that AY SA vould be the most expensive vsy of accomplishing the task There are several arg llllents sometimes used to support the case for a missile-bomber mix a Complicating the Enemy's Defensive Problem - As long as ve have strategic aircre rt the eneJllY C IDllO- effectively defend against ballistic missiles vithout concurrently defending against aircra j't B lld their air-to-surface missiles ASX Conversely defense age inst aircre rt 'W'ithout concurrent defense against ballistic missiles also leaves him vulnerable At present the Soviets appear to be devoting the equivalent of $6-8 billion per year including 500 000 men on their anti-bomber defenses Without a bomber threat these resources could be reallocated to their strategic retaliatory fo -ces anti-missile defenses or some other military p -og -am that might cause us more trouble Calculations suggest that by continuing to r iaintain a bomber ASX threat le can force the Soviets tc spend about 15-25 cents or more on tem inal ba lber defense for every dollar they vould spend on ABM 15 · ' • ·· ·- - _ - - -· However this f_actor does not necessarily argue for a le rge · bo nber force Most of the major elements cif cost in an anti-eircre f't defense system e g the ground enviro ent and pert of the 1 ntercepto force are quite insensitive to the size of the opposing bomber ' force The requirement for surface-to-air missiles is a f'unction of the number of targets to be defended Since the Soviets vill not knew in edvance which targets our bombers would attack they have to continue to defend· eJi of them end their expenditures for air defense ere likely to be ebout the same whether we have a relatively smaJJ or a large_force of bombers • ' _ · ·- · Hedging Uncertainties In the Dene dability of Our· · · ·· · Strategic Offensive Forces - There are four releva it factors which·_ · determine the dependability of our strategic offensive forces the e lert rate pre-launch survival rate reliability and penetration The lert rate is the proportion of the operational force which can i edil' tely respond to en execution order the pre-launch survival rate is the proportion of the alert operational fore which is expected to survive eneIIzy attack in operating condition the reliability rate is the probability thet the system 'ldll launch proceed to target areas as pl2 D led and detoDB te its weapon exclusive of enemy defensive action and the penetration rate is the probability that a reliable system vill survive enemy defenses to detonate its warheed The·· reediness alert rate and reliability of our strategic missile forces is good and improving We are providing substantial emounts of money for en extensive testing progrem There can be no reasonable doubt that for the time period in question the reediness and reliebility of our MIIIDI'D-iAN end POLARIS systems will be fu1J y satisfactory · · · b · - I · 11th reg to survival it is highly unlikely that the Soviets even by the early 1970s would be eble to destroy a ny significant mmiber of POLARIS submarlJ ies at sea I em convinced that they do not· · · · have this capability IlDW Nor is it likely that they would be 'Willing· to co i t the large amoUIIt of resources required to achieve e n effective capability in the future especially in view of the range of our POLARIS missiles Recognizing t t the Soviet missile fore es imated at liioo-100J· latinchers in the ee rly 1970s will face over 11 000 ·hardened e '1Ji dispersed U S ICBMs I believe that our land-based inissiles elso h l Ve high ' · survival potential · - r -• ···------------•-• • 16 On the other hand I am not coDVinced that the surv val potential of aircraft is as good as POLARIS or MINUI'EMA N If for any of a number of reasons they are not launched within the warning time they vould be caught on their hC1111e bases by an enemy missile attack If the bombers are not to be completely dependent on varn1ng they must be 1dely dispersed Today B-52s and B-58s are dispersed only to a squadron level 15 aircraft because in part greater dispersal is both difficult and expensive Furthermore the extent to vhich assured com nand control and col IDunications is possible under widespread dispersal remains to be determined The Air Force proposal to disperse a force of 200 AMSAs to 4oo bases vould still represent a far lesser degree of dispersal than that achieved by MINUl'EMAN -- measuring degree of dispersal by the amount of our investment in veapon systems per independent aiming poirrt presented to the Soviets Leaving aside 1 the fact that the Soyiets vould vant to target marzy- of these bases anyvay because they contain our defensive and C't e torces 2 our investment other than Af 'SA 1n these bases end 3 the undesirability of dispersing strategic bC1111bers to civil airfields near cities1 he 5-year system cost of AJ SA per soft point vould be $22 to m1lJ ion •_ 'Which i three or four times the cost of an individual MINUI'EMAN hard point 'With regard to penetration the deployment of an effective Soviet anti-ba lllstic missile system could degrade the capability of our missiles Hovever it appears unlikely that the Soviets vill deploy in this decade or the early 1970s a system having the potential effectiveness of NIKE X And even if they vere to deploy ABM defenses our penetration aids and multiple warheads should keep the entry price of missile attacks against defended targets vithin · tolerable limits Price is defined as the number of missiles that must be placed over the defended target area to ensure that the target is destroyed Aircraft vill also face penetration difficulties Many studies have shown that an effective anti-bomber defense is a necessary ingredient to an anti-missile defense and that the tvo should have an inter-locked deployment to avoid obvious vulnerabilities The cost of effective enti-bamber defense appears to be ellout 01 1e-1'ourth of the cost of en enti-m1ssile defense In s e -y I see little merit to the argument that a nev aircraft develo i n ent is requ1rcci to hedge uncertainties in the dependability or· our i iesile force 17 Other arguments are also frequently advanced in favor of the borober -- flexibility reuseability sho- of force in a crisis etc Each of the1 1 has some merit but ve vould not support a bomber force for those reasons alone I am not convinced that further large investments in this type of insurance for exa ' Ple $8 9 to $11 5 billion for the Air Force proposed AJJ SA program can be justified for assured destruction D CAPABilTI'IES OF THE PRESENl'I Y-PRCXiRAM D FORCES FOR DAJI AGE LIMlTATIO l The ultimate deterrent to a deliberate Soviet nuclear attack on the Ullited states is our clear and Ulll listakable ellility to destroy them as a viable society But once deterrence has failed vhether by accident or l liscalculation a cboice must be made as to h0' 1' our forces should actually be targeted in order to reduce dsmage to ourselves to the max imur l extent possible I believe it evident from the preceding discussion that the ei ployment of our entire strategic offensive force so as sil lply to meximize Soviet urban damage vould not represent an opt1J rum use of this capability in the light of our objective to limit da mage to the U S As noted earlier vhen tbe ber of 'llarheads delivered on Soviet cities passes beyond about 400 ve begin to encounter rapidly d minishing returns in the amount of additional destruction achieved For example if ve had fired our strategic missiles age inst Soviet cities our bo iber force directed against Soviet military targets vould produce throu h fallout simply as a by-product of their attack about the same m imber of fatalities as they vould produce if targeted againzt the remaid ng Sov et cities The utility of the Gtrategic offensive force in the dama e limiting role hCl' lever is critically dependent on the timing of the Soviet attack on U S urban targets For exa -nple if the Soviet missile attack on U S cities vere to be delayed for one hour or more after the attack on U S military targets our strategic missiles vhich can reach their targets in the Soviet Union in less than one hou - could significantly reduce the veight of that attack by destroying a large a -t of the vithheld Soviet forces before they vere launched 18 I ' the Soviet attack on cities vere to be delayed fo - cir ht hours or more after the Soviets attack our military targets our bo ber force could also contribute to this objective Ho1 ever if the Soviets vere to launch their attack against our urtan a ·cas at the beginning of a general nuclear var our strategic offer ive forces vould have a greatly reduced value in the damage lir it1nc role Their contribution in that case vould be llmted to destM1ctior of Soviet residual forces -- ur launched strategic missiles and hoober re-fire missiles and a ny other strategic forces the Soviet rr ir ht vith ' iold for subsequent strikes Since· ve have no vey of kncr ting hcr the Soviets ould execute a nuclear attack upon the United States ve must also intensively explore defensive systems as means of limiting demaec e to ourselves Conversely because of the critical nature of this un ertainty ve should also hedge against the possibility that ve ma be presented vith an opportunity to destroy at least some of the Soviet offensive forces before they are launched and this means that ve must include in our strategic offensive forces some capability for this purpose The problei i here is to achieve an optimum balance mnonc all the elements of the general nuclear var forces particularly in their damage limiting role This is vhat ve meBll by balanced defense Although a deliberate nuclear attack upon the Unite States by the Soviet Union mey seem s highly unlikely contingency in viev of our unmistakalile assured destruction capability it t receive our first attention because of the enonnous consequences it 10'-lld have To appreciate the implications of a Soviet attack en our cities it is usetu l to examine the assured destruction objective from the Soviet point of viev since our damage lilllitin problem is their assured destruction problem and our assured destruction proble m is their da mage limiting problem The follcr ins table is sil lilar to the one used earlier in this memorandum to illustrate the assured destruction problem from our point of vie - It sh01ts the potential ber of Americans killed as a function of the number of varheads delivered on the United States in a Soviet assured destruction effort The yield of each varhead is assumed to be 10 MI' As ill the case of the couot 'rpart table U S fatalities are calculated UDder conditions of a limited as vel as a tu ll nation-Vide faJJ out shelter program ' United States Population and Industry Destroyed As a Func'tion of Delivered Warhea c s Assumed t rtal 1910 population ct 210 million urban population ct 150 million Delivered Ltd Fallout Protection Na ti m-'llide Fallout Proga i Ind 'llarbea ds Urban Total Urban Tote l Ca-o 10 MI Millions Millions Millions Millions m Hi 100 200 400 Boo 88 79 93 110 53 62 73 l 16 143 l2l 81 164 42 55 49 64 68 8o 78 90 33 43 53 6o 53 74 95 118 25 35 4 - 56 39 50 61 7l Several points are evident rrcm the above table First it 1s clear that '111 th limited ta l lout protection a Soviet attack consisting ot even 100 delivered '118 I'b eads each vith a ten-mega ton yield 'llould cause great lo is ot lile -- 79 million f'atallties in the cities attacked and 88 million tatalities or e llllost 4 percent of the total popu la tion nation-'lrt de The high level ot tatalities frcm 100 delivered '118 I'heads is more a function ct the heavy concentration ct population in Olll' large cities than ct the greater yield assumed tor the Soviet va rheads The d minuhing return 11imply retlect s the 'act that 11maller and mall er citie11 would bave to be targeted as the scale o ' the attack 1s raised Second the table clearly demonstrates the dutinct utility o ' a nation-'llide tallout 11helter program at all levels ct attack Third 100 delivered w rhea ds ' 10 l d destroy about 39 percent ct our 1lldustrial cape city Ea ch successive doubling of the nwnber ot delivered varheads ot this size 'IIOuld increase the deatruction ct our indU1Strial c a pa c ity by t 10 percentage point In order to assess the potential of various damage l jmti Jg programs 'lie have tested a n1 llllber ot bal snced defense postures at ditferent budget levels These postures are designed to defend 18ainst a Soviet threat in the 1970s consistillg of l6o sof't ICBM launchers 46o hard ICBM launchers 230 s lbmarine-launched ballistic missiles 140 heavy ba nbers and 300 medi1 llll banbers These tigures lie 'llith in the rmige o t the e atilnates tor mid-1970 ahovn an Page 7 of the memorandum O We examine1 the total deetruction potential of the Soviet inventory assuming that their eoft ICBMs and b bers are assJ sned against OW military targets and their hard ICBMs SIJ Ma and • ome bombers are assigned against our cities In order to illustrate the critical nature of the ti ming of the Soviet attack ve used tvo limiting cases First ve assumed that the Soviets initiate nuclear var vith a siam ltaneous attack against our cities aod military targets Second ve assumed that they delay their attack e gainst our cities until atter the U S retaliates against their military targets we hsve assumed solely for the PUipOse of this analysis that the presently programed U S stretegic retaliatory forces vould be earmarked for the assUJ ed destruction objective end that only the additional forces vould be used for damage limiting Obviously these are tvo extreme cases end do not reflect all of the other more co nplex end more l Utely possibilities vhich lie betveen Finally ve assumed that all nev systems vill perform essentially es defined since our main purpose here is to gain an insight into tbe overall probleI l of limiting damage The results of this analysis are presented in the table belmt Estimated U S Fatalities for Several Damage LiJD iting Prog U S Da nage Limiting Progrer is Buaget $ 0 billion 5 billion Civ Def Only 10 billion Millions of U S Fatalities Based on 1970 population of 210 million Eerly Urben Attack Delayed Urben Attack 163 163 l20 l20 118 82 20 billion 59 30 billion 41 Balanced al locations of expenditures among the several cc Iponents of a damage lillliting posture for the four illustrative budget levels are sh in the next tahle 21 IDcre1 1ental invecti ient plus cost of 5-years' operation in billions Total Bud11 et Civil Defense Missile Defense Bomber Defense Subc a ine Defense $ 5 2 $5 2 $ 0 $ 0 $ 0 10 0 5 2 0 1 7 20 0 5 2 8 8 30 0 5 2 17 1 Cow rt erfor Missiles a Counterforce BD lbers 0 $ J 1 3 0 0 2 8 2 3 0 0 4 4 3 3 0 0 $ Existing programed forces can probably meet this req_uire lle lt For a budget level of $5 2 billion a complete fallout shelte system 'Would be the most effective component of a balanced damage limiting progre i against large attacks At none of the budget levels exar iined 'Would it pay to spend less for al lout protection Indeed a transfer of resources from the fallout shelter system to other defense systems 'Would result 1n a substantially less effective defense posture This is borne out in the foll O'W ing table Millions of U S Fatalities U S Dama e Lil liting Pro5 em Cost in Bill ioi s Based on 1970 population of 210 million Civil Defense Early Urban Attack Delayed Urban Attack Total Bue get $ 0 5 $0 163 5 120 10 10 0 5 162 ll 8 126 20 20 0 5 142 96 84 59 30 30 0 5 126 63 41 1 53 120 · · - 78 82 The foregoing table indicates that for the same level of su -vivors any dBl lllge limiting progra n 'Which excludes a fllJJ out shelter system 'Would cost at least t'Wo or three ti'lleS as much as a progra 'Which includes such a system even on the favorable assU1 tion that the Soviets vould not exploit our lack of failout protection r 22 by surface bursting their veapo is upvind of the defended areas Fe llout shelters have the highest priority because they decrease the vulnerability of the population to nuclear veapon detonations under all types of attacJ - s in luding collateral damage by fallout from attacks limited to U S military targets Age inst a vide range of urban milita ry attacks a complete fallout shelter system alone vould save 20 to 25 percent of our population and should therefore be a first cor iponent of any larger damage liJniting progrB l · At the $20 and $30 billion budget levels the bci of the additional funo s go to missile defense Hovever a high confidence in the effectiveness of the missile defense system must be assured before COlll litment to such large expenditures vould be justified Moreover at the higher budget levels missile defenses must also be interlocked vith local bomber defenses in order to avoid having one type of threat undercut a defense against the other The exact combination of these tvo defense systems requires further study At each budget level above $5 2 bill ion about $3 billion 'WOUld be allocated for strategic missiles targeted against Soviet offensive forces presently programed forces are probabzy sui'ficient to provide these n1issiles United States missiles vhich destroy Soviet vehicles before launch shov a very high utility for their cost in the dBJDage limiting role up to the point vhere one reliable missile has been targeted against each Soviet Long Range Aviation base and missile site Nev missile systems vhich vc believe could be developed for deployment in t -Je 1970s shov even higher utility The utility of this type of force in limiting de mage depends entirezy o vhether or not our forces arrive before the enemy's nuclear delivery vehicles are -launched aga nst our cities But in this respect missiles have a better chance than aircra 't Nevertheless ve have carefully examined the effectiveness of bombers in destroying various classes of eneey targets In one a alys is ve compared tvo strategic aircraft the W SA m'ld the B-52 SRA ·' a '1d tvo strategic missiles MIN Jl'DIAN II and an improved missile for the 1970s This improved missile vhich could be developed and deployed vithin the SBJ le t Jlle frame as the AMSA and vhich is al ready under stu iy by the Air Force vould be able to carry multiple independer zy-directed re-entry vehicl s enabling a SiDGle missile to attack several differ nt targets Tbe results of this ans zyds are shcr n in highly sur ma ry form in the folloving table 23 THE EFFECTIVElltSS ArrD C 0S' IS OF AL' ER lATIVE STRAT GIC WEAPON SYSTEM 3 Al£A Force Level 200 Five Yr CostsY $ Billions Weapons per Carrier Bombs Missiles Weapons on Ta - et 8 9-ll 5 4 25C 1000 6oo 3 0 4 5 10 0 0 18 4 C 9 9 lo l 7 ll40 1476 820 ll34 675 2520 3 3 6 4 12 0 6 7 6 7 7 2 4 o 4 o 4 5 Cost Target Des $ Millions Soft 8 9-ll 5 100 psi 8 9-ll 5 300 psi 9 4-l2 l y MM II Imp Cap le u Gu1d Missile B-22 5 AJ 6 1-8 6 6 7-8 6 7 0-9 l 4 4 6 3 9 l The five-year systecs costs consist of the RDT E and invesilnent beyond FY 1965 and the f'ul l five-years' operations Throughout this analysis ve bave used essentially the Bl le planning factors used by the Air Force i e alert rates survival rates CEP etc The assumptions underlying the table were chosen· to be representative for most mill tary targets For elCBZllple at this ti Cle ve e ti ma te tbat most nuclear target threats 1n the U S S R vill not be protected by an anti-ballistic l lllsile defe l6e during the next ive to ten years Recognizing that there are uncertainties in all or the assumptions as veil as L'l the plann1ng actors used I believe that this ts ble does tlei ion 5trate clearly at least o ie important point namely that ther are less costly vays -- including other aircraft -- o destroying ilitary targets than by de ·eloping and deploying a nev AMSA The B-52 SRAM for example is much m re co IIpetitive '111th missiles than Al£A a ainst soft targets Moreover the advanced avionics proposed for the ''£A could also be employed '111th the B-52 SRAM increasing the accur cy of the mia Ue delivery eyste by about tl reefold i e to the CEP aesumed or the AM A This wuld 24 cost an additional $1 2 billion But a ainst the 300 psi ha -dened targecs the cost per target destroyed for a B-52 SRAM vould be reduced to betveen $4 5-$6 5 million compared vith the $7 to $12 l million shovn for Af lSA With regard to the SW three only nominal funds vere al located to extra ru ti-submarine defense for dw ia e lii iiting at each budget level Full advanta e vould be taken of the- ASw capabilities ve already have for defense of the fleet and shipping The currently projected Soviet SLBM threat vill not be particularly effective in comparison vith our OVD POLARIS Deployme it of e n improved SLBM force by the Soviets need not mean that ve should necessarily respond Vith improved anti-submarine forces since a terminal anti-ballistic missile defense could also deal vith a SI BM attack There remains the possibility of a m nall nuclear attack on the United States either accidentally or deliberately possibly by a nation other than the Soviet Union Since the nert decade vill probably see a proliferation of nuclear veapons and strategic delivery systen s and remembering that a single thermonuclea1· veapon could kill as many Americans as vere lost in the entire Second World War this may become a o important problem Accordingly ve have undertaken a n llber of studies in that area OUr preliminary conclusion is that a small balanced defense program involving a moderate civil defense effort and a very limited deployment of a lc t cost configuration of the NIKE X system vhich is technically feasible vithout comnitment to a full-scale deployment could indeed significantly reduce fatalities from such an attack In s e ry several important conclusions may be dravn from our analysis of the d e limiting problem l With no U S defense against a nuclear attack in the early 1970s the Soviet strategic offensive forces vould be able to inflict a very high level of fatalities on the United States -- about 160 million or 75 percent of the total population 25 2 A nat n-wide civil defense roGr costing about $5 billion could reduce these fatalities to abo' lt 120 million 3 A le -ge bale need de nage lil itii g program for a $30 billion 5-ye cost could reduce fatalities associated with an eB ly urban attack to about 80 million 4 There is no defense program within this genere l range of expenditures vhich w could expect vith confidence to reduce the fatalities to a level much belo- 30-4-0 million even it the So dets delayed their attack on our cities or much 'belcv 6o-75 million if they attack our cities en the first strike H01Jever - c nave thus far not takei into acco lllt a most importsnt factor -- possible Soviet reactioi s to our damage limiting initiatives which could serve to offset their benefits For example assume that we had a ready spent $20 billion for a balanced damage limiting posture as described above expecting it - uld ensure survival of 54 percent of our population in the event of a Soviet first strike a c ainst our cities Assume further that ve then decided to spend another $10 billion to raise the proportion surviving to 62 percent If the Soviets choose to offset this in crease 1n survivors they should be able in the 1970s to do so by adding about 250 illlproved ICE s - nh penetration aids at a cost of perhaps allout $6 billion Similarly if we increased our damage limiting expenditures by still another $10 billion to $40 billion in order to raise the proportion of the population surviving from 62 to 68 percent the Soviets could offset ou action by adding another increment of 200 improved ICBMs to their force at a cost of perhaps another $5 billion Thus at each successively higher level of U S vors the ratio of our costs for damage limitation to their costs for assured destruction becomes less and less favorable for us Indeed at the level of spending required to assure ourselves 80 percent survivors 1n a large Soviet first strike against our cities ve vould have to spend on damage limiting forces abo1 four times vhat the Soviets vould have to spend on damage creating forces i e assured destruction forces This does not necessarlll' met Il that the Soviets l'Ould actu all y react to our damage lilDiting initiatives bu' it does underscore the tact that beyond a certain level ot population GUrV Lving the cost advsntage lies 26 in reasingly vith the offense and this fact must be taken into account in e ny decision to CO II ' it ourselves to large outlays for additional defensive measures There is little doubt that it is technically and econor ucally feasible for the Soviets to defeat our attam ts to achieve high percentages of survivors in a large nuclear attack If ve vere to choose to aim for a high percentage a level at vhich the cost leverage is quite UDfavorable and if the Soviets vere to choose to run the race then ve might find ourselves devotin very large amounts to dama e limiting mee Sures and realizing very little in retu -n as far as an effective defense against a large deliberate Soviet attack is concerned E RECCM -ENDATIONS ON MAJOR ISSUES IN THE GENERAL NUCLEAR AR PRCXiRAM3 In this section I shall attempt to s arize ey vievs on the six major issues involved in the general nuclear var progrei s A more detailed statement of rrzy vie'Js plus those of the Joint Chiefs of Staff and Service Secrete ries may be found in Appendix A l Develop ier t and Deployment of a Ne t Me ruled Bomber I believe it is clear from the foregoing discussion that it is difficult to make a good case at this time for the development and deployment of a very expensive De'J manned bomber such as the AMSA proposed by the Air Force Although the destructive potential of our missile forces alone provides a most persuasive deterrent to a Soviet attack on the United States it may nevertheless be vise for the reasons I have already discussed to provic e a 'l option for maintaining some manned bombers in our forces indefinitely This ve propose to do There are at least three other alternatives available to us n addition to the development of the JiJl SA vhich voulc preserve thr option to mainta n a force of strategic bombers into the 1970s T nese are a the retention of late model B-52s and the improvement of their attack capa ilities b the procurement of a strategic version of the F-lll B-lll and c the initiation of advance development vork on long lead time ca iponents of n combat aircraft With appropriate maintenance and modification est of the current B-52s can be maintained in safe effective operation at least throug the early 1970s I reco inend that $339 million be included in the 27 Fi 1966 budget for this purpose and that another $930 million be approved for planning purposes in the Fi 1967-1970 programs These f'un is vould pen iit us to continue our program of structural modifications for the B-52s and vould make it possible to keep the B-52Cs throu h Fs cun-ent total Jlventory numberi lg 336 aircref't 1D the l'orce until 1970-1972 and the B-52Gs and He current total inventory numbering 287 aircraft beyond end FY 1975 The 41 B-52Bs still in the force should beccmpletely phased out by the end of fiscal year 1966 and the force structure reduced by one ving These are the oldest active B-52s and ve vould have to spend about $70 million over the next fe l years to keep tbe n in safe operating condition Including operating costs their phase out could produce a saving of about $200 million during the F' 1966-1970 perioc vithout a -zy significant effect on our strategic offensive capability Tne latest series of B-52s the Gs and Hs could also be modified to incorporate the Short Range Attack Missile SR AM proposed by the Air Force for the Af iSA Without extensive De l avionics the S AM ca -ried by a B-52 vould have an accuracy approaching feet against kDO ln fixed targets and could be launched as far ir ay from the targets as 60 n mi outside the range of local defenses Preli l lina ry estimates shO I that the costs of developmeIIt and the additional structural modifications required for SRA deployment vith the B-52s vould amount to about $3 million per e ircrai't Altho igb these aircraft have some limitations in dispersal capability speed damaae assessment and ride quality vhen compared vith a B-111 or an AMSA I believe that for the next ten years this option vould provide at the 10 lest possible price adequate insurance as a hedge against unforeseeable degradations of our assured destruction capability Accordingly I reccmm end approval to initiate a project definition phase for SRA V at a cost of $5 million in FY 1965 and about $15 million in FY 1966 an additional $14 million vill be required for development in FY 1966 a total of $29 million and $67 million in Fi 1967-1970 A strategic version of the F-111 vith but minor modifications carry up to five SRA • an equivalent loading of bombs or a cccbination of both Its speed oYer eneey territory could be supersor ic at high altitudes and high-subsonic at 10 I altitudes While a B-111 f rce vould have to place greater reJ iance on tankers than an Af-'SA f rce its range considerably better than the B-58 its r ould target coverage a 'ld its payload-carrying capability vould be sufficient to bring under attack a very large percent of the Soviet urban industrial cc cplex Since this aircraft is already nearing pr duction a strategic version could be IDB de available within tvo or th 'ee years after approval Therefore no decision is necessa --y at this t llle The AYSA as presently envisioned by the Air Force vould incorporate the payload-carrying capabilities of the·B-52 and the speed altitude characteristics of the F-lll Its takeoff gross veight vould be in the 350 000 pound class and it vould require the development of a nev engine and nev avionics as veil as the SRAM Considering the other alternatives available I do not believe ve are DOV ready to go ahead with development Y But I do believe it is desirable to keep open the option tor a nev heavy bomber ill the strategic forces atter the retirement o the B-52s y Secretery Zucke -t in his memorandum transmitting the AMSA proposals to noted that the Air Force intends • • • to complete prior to the initiation of the Project Definition Phase a prerequisite phase vhich vill f'urther refine our systems evaluation This phase vill include further evaluation of an advanced strategic aircraft against the TFX the stretched TFX and a growth version of the TFX · incorporating advanced engines In addition AJ GA vehicles in the 200 1 000 to 300 000 pound veight class vill be further investigated Aircraft con f igured for subsonic penetration only vill be CCllll ared vith designs having supersonic high altitude performence as veil as lov-level capability Each systei i configuration vill be assessed in terms of performance cost schedule military effectiveness complexity and development risks I - ' · J t--i ' f '- ' J' _ 30 silos CO lllllencing in July 1966 instead of January 1966 as previously approved in order to reflect a six month slippage in the progrm and to smooth out the early buildup rate The total cost of the retro-fit progrwi throu h 1970 vill amount to $1 3 billion 550 silos by end F'Y 1970 in addition to the $1 1 billion spent on MINUI'DIAN II development The MINUrEMA11 II vith all the improvements I am recormnending could increase target destruction capabilities by at least a factor of tvo compared to a MIN JI'EMAN I force of the same size The recor cended improvements include a nev guidance improvement pz ogram the development of a nev re-entry vehicle the · vhich vould have much smaller re-entry errors as veil as a larger yield varhea d and a precise varhead election i yst vhich vould permit a single MIN1 1 l'EMA N II to deliver three · · re-entry vehicles to geographically separated targets The guida ce improvement program and the DE ' l re-entry vehicle promise to reduce the overaU CEP of the MlNVl'EMt N II to around feet half the present CEP and give the missile a 90 percent probability of destroying targets hardened up to psi The post boost control system vould greatly increase the kill capability of the rec ended MIJWI'EMAN force against sof't targets many of vhich require no more than for their destruction The R D and investment cost of the guidance 1mprovement program is estilllated at $35 r illion the RIY l' E cost of the nev · re-entry vehicle at $89 million exclusive of the flight test program and the precise varhead ejection JYStei at $125 million exclusive of the flight test program A version of this system is al ready under develqpment for the ejection of penetration aids as part of a $31 million program in FI 1965 and $52 million 1n Yi 1966 Along vith· MINUTDlAN ve should also consider the other strategic ssile progra is To prepare for the possibility that the Soviet Union may deploy an effective anti-missile·defense system arOUDd its urban industrial areas I rec nend the inclusion in the FY 1966 budget of $35 million to begin development of a ne' t POLARIS B-3 e intend to initiate a project definition for this missile during FY 1965 The B-3 vould incorporate 1mproved accuracy and payload flexibility pennitting it to attack a single heavily defended urban industrial target or a single hardened point target or several undefended targets vhich might be separated by as much as 75 miles Since ve are uncer ain about beth the ultimate shelf life of the present POLARIS missiles and the schedule of deployment of a Soviet ABM system the 31 pa ce o f' the B-3 develoµnent has not been precisely established et this til e Total developcent costs of the B-3 missile mey approximate $900 tn llion and the total cost of a 41 Polaris subl lBrine force including for example 22 submarines carrying the B-3 missile could total $2 5 billion Finally in view of the fact that le ill have 800 MINUI'D' Ali and 416 POLARIS in the operational forces by the end of the current fiscal year t believe le can safely phe se out the KrLAS Es and Fs and TITAN Ls by that t ll e at a saving of about $515 million in the FY 1966-1970 period These older liquid fuel missiles are very costly and difficult to maintain on a i alert statUS Moreover on the basis of their present operational factors they represent less than 50 delivered larheads 3 The Overal l ·Level of the Anti-Bomber Defense Progrem Our present system for defense ainst manned bombers las desie ned a decade ago hen it vas estimated that the Soviets vould build a force capaole of attacking the United States vith maey hundreds of heavy bc mber aircra i't This threat did not develop as estimated Instead the major threat nO I confroIIting the United States is the Soviet ballistic missile With no defense against the ballistic missile and only the beginning of a viable civil defense posture our anti-bomber defenses could operate on only a smal l fraction of the Soviet offensive forces in a determined attack A balanced defense requires a major reorientation of our effort -both vithin anti-bomber defenses end betveen anti-bomber and antimissile defenses The characteristics of a balanced defense have already been discussed For defense against the diminishing bomber threat ·our present forces are que 'ltitatively excessive in relation to their cost end effectiveness I therefore recomnend a The -phaseout of 9 National Guard F-89 squadrons along vith the transfer of 9 active F-101 squadrons to the Air National Guard by end F' 1967 end the phaseout of 9 active F-102 squadrons by end FY 1969 l in FI 1965 4 1n FI 1968 8 4 4 1n F'i 1969 - for a FY 1966-70 saving of $300-$350 million • Studies mede by the North American Air lle ense Command indicate that in 1970 the fatalities from e Soviet attack a - ter vithdraval of these squadrons vould be no more than l 5 to 5 million higher than they vould be if the squadrons wre retained--i e the fatalities might be 48 to 50 percent of th population instead of 47 percent Y The Jo1Dt Chiefs of Staft' less Chief of staff J nrq recamnend that the intercept force be retained as previously approved 32 b The phase out of the Devlioe extension aircraft aod the offshore radar picket ships beginr iog in FY 1965 as proposec by the Navy -- for a FY 1966-1970 saving of $266 million $69 million in Fi 1966 y c The reorga nization of the air defeose surveillance system as proposed by the Air Force entailing the phase out of 16 pr1me rade s 32 height fioder radars and 9 gap filler radars by eod FY 1967 -- for a FY 1966-1970 saving of $111 millioo y The i'unds saved by these actions cao be better ap_plied tc the improve nent of the qualitative effe tiveness of our anti-bomber defense forces To this end I recommend a The initiation of develop i ent of an i mprovemeot to the HAwK syst and continued advaoced development of a nev improved surface-to-air missile system for both cootinental and overseas t q tre air defeose at a Fi 1966 cost of $24 5 million The inclusion oi' about $28 million in the FY 1966 budget tor SAGE BUIC Ill an improved gTOUnd envirooment system b for air defense control 3 c Continued syst study of an Airborne Warning and Control Systei and component development in an over-land Radar Technology progrw n to augment land-baaed surveillsnce and control syete ms for oth continental and tactical sir defense y 4 The Production and Deployment of a NEN Manned Interceptor On the basis of the aoalysis in the preceding sections of this memoraoduc it is clear that the production and deploymeot of a nEN manned interceptor in a balanced defense progra should be considered only if ve vere to increase significantly our damee e limiting program including the deployment of an anti-missile defense system aod a nation-vide fallout shelter system Indeed it is not at all clear at this time that a nev manned ioterceptor vould be preferazle to a nEN advanced surface-to-air missile system the continued development of vhich I have recommended above Nor is it clear that the F-l2A already developed is preferable to an interceptor version 'y The Joint Chi fa of Staff less the Chief of Naval Operations do not concur in t is reco endation 2 ·rile Joint Chiefs of statt concur in this rec endation This plan meets the objectives sought 1n the JCS reco endation on this subject • i • I 33 of the F-llJ Our analyses indicate that against liUbsonic banbers the F-lll vould be preferclile at smaller budget levels vhile the F-12A vould be preferable only at high budget levels In any event at higher levels of dEUna ge limiting expenditure the anti-bomber and anti-missile defenses must be interlocked and proceed in parallel At this time I recon nend the provisicn of $5 million in the FY 1966 budget for the further development of electronics equipcent for the YF-12A and the deferral of a decision on the production and deployment of either the F-12A or the F-111 for the iriterceptor r ussion 1 The recommended program vill retain the option of future deployr ient of either or both of these interceptors 5 The Production ·and Deployment of the NIKE X Anti-Missile System During the past year ve have greatly expanded our knowledge of a ti-missile defense '111th regard to both the cost and effectiveness of alternative deployr ients and the technical aspects of the system The Army has developed three basic systems configurations vhich differ primarily in the nu iber and kind of radars utiliz ed a The so called HI-l' AR configuration vhich includes one high cost Multifunction Array Radar MAR and about tvo siDgle-face l cw cost Missile Site Rada MSR for each urban area defended This configuration provides the most effective defense against a large technologically sophisticated attack per urban area defended but it is the most costly for a given number of areas b The LO-MAR configuration vhich includes on the average one MA for every three urban areas and one double-face MSR and tvo single-face MS l for each urban area defended For a given level of expenditures recent Army studies indicate that the LO-MAR configuration vould possibly maximize survivors against a moderately sophisticated attack and vould be clearly superior to a HI-MAR configuration ae ainst a small er or less sophisticated attack c The NO-MA configuration vhich includes only MSR radars in the same combination as the LO-MAR con figuration This vould be the lcwest cost configuration per urban area defended but it vouJ d not be effective against a large sophisticated attack iJ The Joint Chie f's o f' Stafi rec0111Dend finding 1n FY 1966 procuremen't o f' either 18 F-l2As or 18 F-llJ s to ritain the option f'or future deployii ent of an advanced interceptor A co u-i cr of representative deploy ients of the tr e confi u -atior -- the n- ber of urba 'l B eas protected population in the p ·otected a reas anrl development and production costs -- is shc- -n in t e table belOI ' SELECTED l XI X DEPLOYME lT ALTE l IATIVES Defended Urban Areas P D Pree 13 23 30 10 9 17 7 25 4 Sep 69 Mar 72 Dec 73 ll 6 8 ll 7 19 8 Sep 69 Mar 71 Mar 73 Costs $ Bil Initial Operational Cenacilitv HI-HAR I II III LO-MAR I I II IV 20 47 NO-MAR I IV 11 4 5 Sep 69 10 CJ Y ar 73 VI 102 14 6 Mar 75 Other a ternative deployments rnd details on costs and configun tions are s· ow in Appendix A 50 If le ished to start deployment at the earliest possible date f r3t qua -ter EY 1970 e would have to include about $200 million in t le i 1966 budget for production in addition to more then $liOO million for continued development However in vie of the continuin uncertai lties concernille the preferred concept of deployr ient the relationship of the Tr E X system to other elements of a bala 'lced damage limiting effo -t the prospect for an effective nation- ride fallout helter system and t ie m t'l-rt 11 ml effect of the Soviet reaction to a rme X deploy ient do net elieve a decision on production shouJ d be made at this tir ie 3' l°' 3o rcccr nend the t a total of$400 0 million be provided for TC in the 'FY 1966 budect $390 c r illion 1 0 continue development · t e y te at a - o tir rate and $10 tillicn for prod ction ul i u 1 The question or production and deployment of the NJXE X y - The Joint Chiefs of -aff recoI llend that $200 million pre-productiiz funds be allocated in FY 1966 to protect the option to achieve an initial operational capability in October 1969 35 ' a 1 te -' 2 'l b ree - ine l next yeax Def'c - icnt of the deci ion p• - i t -t of deployment in lnte FY l' i7C• t' l I l ' j • l T ie · -t io r rallout Shelte -s for the Lnti e opule tion C- ir r i li• ' - t 1 ler c liDitinc proble t i ke it crystal · ea - · - i c ·i - l ·e nation-· ide fallout heltc-r yste i voulil _ l -o- icl th _ r ·c ca - - -P t i -n for the money ei per de The Executive J -a h n r o cr led such El p -ograi i to the C ong ·cz three yea -s ru -_ 1 ' n b Jt 1 · i r• q_ red legislation authcriz ir £ the shelter develop - t p -- - r - l i L 1i h 'e CWU ot prcv le a co i letc n a tion- ridc · y t i h j - cc c er1act c i Accordlngly l r o r a That the Executive Branch undertake a major effort to inform the CO r ress of the relationship betveen a shelter development program J -OYiding full fa1lout protection for the population and the other elements of a da nsge limiting _program before such legislation is ae a n tra '1sr itted to the Congress ' 1 ·· 1 1 be ir cluded ir ti' c · 1 l u i et to e pe l i i c '1·c ent shelter survey pro r-·c tc in lude c ve · · h c i i other small p -ivetc b 1 ltl i n5z ar d to fi m 1c c - ire thoroUGh evaluation r f exi ting i helter chc a- er t tic art supplies h c Th lt $j_5 i llion be included in the ri 1966 b· id et tc inr rc e th Civil Defense R D pro r i prim ri to cval· 1at shcj_te · construction teclmi -ue s I tr dcve i op a the 71c cot nte ·-r ie i zure systei and t e 1a Llisr o t e h ical l asis -or post-e tteck recovery u Th t othe - ele ients of the pre e tly l J '- -ove proerem be cont n Jea t IT 1966 level to be deter i i ed durina the current bu1gct -evi e'-1 l•t• reco nendetion i on O'ther issues in the f eneral nuclea - ' proi rcr z e re included in Appendix A Append contain selected ·iscal a d force structure s aries of the recomnended progrBl S ' 'able l ir r lecliately 'oll ing s arizei the Strateg c Offensive Forcez '1hich I l recom nending TABLE 1 RECOMH NDED Al D SERVICE PRorosr rJ 'P STRATEGIC OFFEIISIVE FORrn Eod Fiscal Year 36 1961 1962 1963 1964 1965 1956 1967 1958 1970 B x be- -s- B-52 555 B-EB-47 B-58 Total B imbers Air-Launched Msls H und DOG 615 630 630 900 810 585 450 4o Bo Bo 80 1495 1505 1295 u6o 216 46o 580 58o 630 6oo 6oo 28 Titan 57 126 126 2l 67 l 08 l 6o J linute a l I inute n II Polaris MU' Polaris A-3 Total I_CBl-l Pol Othe- °Quail ' 1 Tog 96 l 44 6oo 56o 540 540 99 99 68 54 54 54 l o8 108 lo8 6oo Boo Boo 100 750 610 Bo 300 200 390 224 416 448 656 'i'fl 497 1058 1270 1382 1710 540 520 520 25 25 25 10 10 10 35 35 35 68 54 550 480 450 620 656 8 68 68 54 54 4oo Goo 800 656 48 250 300 750 900 656 128 rrffil 71 175 l 3 l 419 16ol 1832 l878 1978 1978 224 4oo 6oo 17 KC-135 KC-97 ReguJ us PACCS KC-135 B-47 Alert Force Weapons Bo 6oo 225 Bo Bo 78 76 74 72 935 o8o b'7E b 7b ti'f i b72 710 708 706 7o4 702 Strate ic Reconnaissance SR-71 25 RB-47 45 30 90 30 30 10 10 RC-135 Total 90 45 30 30 30 7 5 35 Surface-Surface 1-lsls Atlas 6oo 630 630 630 630 630 392 440 580 17 392 500 34o 17 392 58o 24o 7 392 620 120 390 620 390 620 390 620 390 620 390 620 17 36 17 36 18 24 24 24 24 24 18 l lj MeGatons Footnotes on ce rt page 836 1551 2071 2689 26o1 2535 2715 2722 2732 2775 28ol 2798 2896 2938 30l5 3015 1651 3382 3976 5835 5041 494o 5128 5128 5129 5195 5383 536o 5367 5681 5758 5781 EXCLUDED l''RO l AU'IDMATIC REGRADDlG DOD DIB 5200 10 DOES NOT APPLY 37 1 The forces proposed by the Secretary of the A 1 r Force and the Joint Chiefs of Staff less Chief of Staff Air Force vhere different fro n the Rec ended Forces are sh n in parentheses £ Possible assignment to NA'IO of UK or other nuclear veapons including the UK Polaris force in accordance vi th the ten s of the Nassau Pact have not been taken into account in the rec ende U S force structure ' Nu ibers of aircraft do not include c nd support or reserve aircraft The Multi-lateral Force consisting of the Polaris A-3 on surface ships is included under the assumption that formal agreements wuld exist b · July 1965 The cost of this force is not included in the costs of the Strategic Offensive forces The proposed force of 200 missiles in 25 ships v rul d be achieved by mid-l97l 'lf Excludes I atio w l Emergency Airborne C mcand Post and Post Attack C nd and Control SysteI aircraft f The alert force veapons and megatons are based on actual· data tbrO gh end FY 1964 except for end FI 1961 vhere the actual data are based on ·Aprill 1961 position On July 15 1961 about 50 percent of the stratei ic aircraft vere on alert c011 pa Xed vith about 30 percent previously Beyond FI 1964 the extrapolations are based on most recent data The aver8 8e numbers and yields of aircraft veapons are as follO'llS B-47s 1 75 veapons and B-52 3 32 veapons and ' exclusive of the Hound Dog missiles B-58s five veapons and • For the FI 1965 period and beyond 90 percent of the ICE-ls are assumed on alert except Mi nutel ao I for vhich an 85 percent alert rate vas assumed during the period of missile retrofit In addition about 53 percent of the Polaris force is assumed to be on-station vhile an additional lO percent of the force vould be in-transit to pa trol areas