Document Number (FOIA) /ESDN (CREST): 
Document Page Count: 
Document Release Date: 
July 30, 2014
Case Number: 
Publication Date: 
September 1, 1993
PDF icon DOC_0000622826.pdf788.95 KB
Approved for Release: 2014/07/29 000622826TITLE: Soviet Military Power in the 1980sAUTHOR:(b)(3)(c)VOLUME: 37 ISSUE: Fall YEAR: 1993Approved for Release: 2014/07/29 000622826 pproved for Release: 2014/07/29 000622826TUDIES IINTELLIGENCEA collection of articles on the historical, operational, doctrinal, and theoretical aspects of intelligence.All statements of fact, opinion or analysis expressed in Studies in Intelligence are those ofthe authors. They do not necessarily reflect official positions or views of the CentralIntelligence Agency or any other US Government entity, past or present. Nothing in thecontents should be construed as asserting or implying US Government endorsement of anarticle's factual statements and interpretations.Approved for Release: 2014/07/29 000622826 Approved for Release: 2014/07/29 000622826Evaluating predictionSoviet Military Power in the 1980sJames A. BarryIn 1987, John Lewis Gaddis, the eminent diplomatichistorian, published an article entitled "How theCold War Might End." Four years later, he repub-lished it, with comments and the subtitle, "AnExercise in Faulty Prediction." In this evaluation,Gaddis concludes that his predictions were at bestmixed, and makes several recommendations to hisfellow historians on how to improve their under-standing of events and processes.In 1981, the CIA',s National Foreign AssessmentCenter (NFAC?subsequently renamed theDirectorate of Intelligence) published a study entitledThe Development of Soviet Military Power: TrendsSince 1965 and Prospects for the 1980s. It was theculmination of a two-year, multidisciplinary researcheffort aimed at describing and evaluating the stra-tegic, technological, political, and economic factorsthat would influence Soviet military forces and poli-cies in the coming deeade.a The NFAC effort was in-tended to provide an intellectual foundation for intel-ligence support for the national security policies of anew administration.Like Gaddis's article on the Cold War, the NFACstudy had a mixed record. And no one who partici-pated in the research effort would have predictedthat, barely 10 years after the study's publication,Soviet military power as such would cease to exist.Thus, it seems worthwhile that we, like Gaddis,reexamine our assessment. What did we get right?What did we get wrong? What did we miss? And itseems appropriate as well to ask: what could wehave done to improve our record and what differencewould it have made?This article tries to probe these questions by adopt-ing Gaddis's method. In the sections that follow,excerpts from the 1981 paper are reprinted in italics,followed by comments made with the benefit ofhindsight.' In the conclusion, an attempt is made todraw some lessons about the capabilities and limita-tions of intelligence analysis and its influence onpolicy decisions.31The CIA's Projections_Seergr"(b)(3)(11)As the Soviet leaders formulate their defense plansfor the future, they face major external and domesticuncertainties:? The fluid international situation dictates a pru-dent defense posture and the Soviets perceptionsof emerging military threats argue especially forcontinued qualitative improvement in forces.? On the other hand, to maintain even a modestrate of economic growth, those leaders must allo-cate more resources to capital investment andmust improve labor productivity, in part byproviding a rising standard of living.This dilemma could cause political tension, particu-larly at a time of leadership transition.This was an accurate statement of the problem. TheSoviets and Western analysts were facing whatGaddis calls paradigm fratricide. They were subjectto countervailing pressures and could not predictwhich would win out. The dilemma certainly didcause political tension!In the international arena, the Soviets are concernedby the prospect that the US will augment its defenseeffort, by China 'c opening to the West, and by thepossibility that US opposition to Soviet glopal aspi-rations will increase. They are troubled by instabilitv.on their borders?an insurgency in Afghanistan thatthey have been unable to suppress, an unpredictableregime in Iran whose fundamentalist Islamic ideol-ogy could spread to Muslim minorities in the USSR,and a major threat to Communist Party control inPoland. They probably view the 1980s as d decadeof heightened competition, in which they will run agreater risk of military confrontation with the USand of actual combat with major powers.Approved for Release: 2014/07/29 000622826ecret ?stragr(b)(3)(n)  Approved for Release: 2014/07/29,000622826All of the Soviets' worst fears (except the last) cametrue!As they attempt to react to the wide array of situa-tions they perceive as either promising or threaten-ing, Soviet policymakers will face afar more con-strained resource picture than in the 1960s and1970s.? Soviet economic growth, which has been decliningsince the 1950s, has slowed to a crawl in the pastseveral years. The real average annual growth in GNPin 1979 and 1980 was a little over 1 percent?theworst in any two-year period since World War II.? In the 1980s, developing energy and demographicproblems probably will hold GNP growth to anaverage of 2 percent or less?only half the rate atwhich defense expenditures have been growing.? If military spending is allowed to follow its pasttrend, its share of economic output could increasefrom about one-eighth now to over one-sixth in1990.? More importantly, this increased military burdenwould reduce significantly the share of the annualincrement to GNP that can be distributed amongcivilian claimants to ease the political tensionsthat arise from competition for resources. Militaryprograms?especially those for nonstrategicforces?divert key resources from the productionof critically needed equipment for agriculture,industry, and transportation.This forecast was essentially accurate, perhaps evena bit optimistic. According to later CIA estimates,economic growth did hover around 2 percent formost of the 1980s. (Some Soviet economists, as wellas Western critics of CIA analysis, had even lowerfigures.) Military spending continued to rise through1988, when Gorbachev announced unilateral militaryreductions and a reduction in the defense budgetby some 14 percent over a two-year period. Hisunilateral reductions concentrated first on conven-tional forces, to release manpower and productionresources for the civilian economy.SecSovietThe problems of Soviet leaders in allocatingresources could be further complicated by a politicalsuccession. Soviet President Brezhnev is 74 and inpoor health, and most of his colleagues are also intheir seventies, many of them also ailing. The depar-ture of these men could affect military policy butprobably not immediately. The process of Sovietnational security planning and decisionmaking ishighly centralized, secretive, and resistant to fun-damental change. It is strongly influenced by mili-tary and defense-industrial organizations,represented by men who have held their positions formany years, providing a continuity of plans and pro-grams. Because of this momentum, and the politicalclout of the men and institutions that support defenseprograms, we doubt that Soviet emphasis on militarypower would decrease in the early stages of leader-ship succession.Not a bad call. The momentum continued for severalyears after Brezhnev's death, through the tenures ofAndropov and Chernenko. It was the third politicalsuccession, from Chernenko to Gorbachev, that madethe big difference. Gradually, the defense policyprocess began to open up and the previouslysacrosanct defense-industrial sector of the economyhad to shoulder part of the burden of diminishedeconomic performance.In contrast to the imponderables of the economicand political environments, we have a good capabil-ity to identify most future Soviet weapon systems.The forces of the 1980s will be equipped primarilywith systems already in the field and secondarilywith those now entering production or in late stagesof development. (Because it takes a decade or moreto develop and test modern weapon systems, few ofthose now in early stages of development could beintroduced in significant numbers in the 1980s.) Webelieve that we have identified about 85 percent ofthe new systems likely to be introduced in thisdecade. Knowing Soviet military requirements andthe amount of available development and productionresources, we can postulate others. These identifiedand postulated systems, plus existing systems, willmake up well over 90 percent of the weapons in thefield in 1990.32Approved for Release: 2014/07/29 000622826 SovietApproved for Release: 2014/07/29 000622826We did get most of the weapon systems right. Wesometimes were overly optimistic, however, aboutwhen they would become operational and about theirrates of deployment. Of the systems that weprojected, more than 90 percent were deployed, andmost within a year or two of the estimated date.Those that we missed fell into three categories: sys-tems in early stages of technology demonstration thatwe misinterpreted as prototype weapons; high-technology weapons that the Soviets found unusuallychallenging to develop and produce; or systems notscheduled for production until the late 1980s, wheneconomic and political turmoil had begun to disruptmilitary programs.Because changes in political and economic condi-tions could lead to discontinuities in policy, wepresent three alternative projections: two that requirean acceleration inIhe growth of military spendingand one that requires an absolute reduction. We con-sider all these to he less likely than the baselineprojection but present a discussion of them intendedto suggest reasonable limits to the options open toSoviet policymakers.This was a good idea. It is beneficial to acknowledgeour uncertainty and explore scenarios that couldmake a difference to US policy. Most NationalIntelligence Estimates (NIEs), and many DI papers,now give "alternative futures." The problem is, asdiscussed below, that we were not quite adventure-some enough in our alternative projections and failedto follow up with a systematic effort to see if ourbaseline projection was consistent with evolvingSoviet reality. The fact that we included twoscenarios involving an increased military programand only one that postulated a reduction seems inretrospect to have been misleading.Baseline Projection. For our baseline projection weestimate?on the basis of the weapon production anddevelopment programs we have identified?that theSoviets will continue their policy of balanced forcedevelopment. Within the outlines of this continuity,however, we expect them to increase their emphasison strategic forces that can survive a US attack, onstrategic defense, and?to a lesser extent?on forcesfor the projection of Soviet power to distant areas.Manpower constraints will limit increases in the sizeof forces, but improvements will continue rapidly asnew weapons become available. Improvements in33Soviet military forces will lead to growing capabili-ties in many areas, including some areas of tradi-tional Western strength.Yes on balanced force development and survivablestrategic forces, yes and no on strategic defense, andWRONG on power projection forces, which werebuilt up to a lesser extent than we anticipated. Weunderestimated the influence that the Afghan imbrogliowould have on Soviet ambitions in the Third World,and misread some of the military programs as sug-gesting greater interest in projecting power overseas.We expect the Soviets to carry out programs aimedat maintaining or increasing their lead over the USin most measures of intercontinental nuclear attackcapability and at upgrading their nuclear war-fighting capabilities. They will continue to improvethe accuracy of their ICBMs and will develop a vari-ety of payload options for responding to US deploy-ment of new ICBMs. As a result, the Soviet 'ICBMforce?with or without the SALT II Treaty?will havethe theoretical potential to destroy most of the war-heads on US land-based missiles throughout thedecade. This potential will be greatest in the early1980s, before the US can deploy a new ICBM. Buteven in that early period, US forces could conduct amassive retaliatory strike.Pretty good. By 1989, according to NIEs, the USSRhad the capability to launch preemptive strikes or to"launch on warning" against a comprehensive setof targets in North America and Eurasia, includingattacking US missile silos with two warheads each.(NIEs of the early 1980s had enormous detail aboutthis potential vulnerability of US retaliatoryforces?an obsession of US military planners of thatperiod.) But the Soviets still could not destroy USballistic missile submarines, bombers on alert or inflight, or dispersed mobile ICBMs.To maintain survivable strategic forces in the faceof a potential threat to their own fixed, land-basedmissiles, we expect the Soviets to increase the capa-bility of their submarine-launched ballistic missilesApproved for Release: 2014/07/29 000622826Sccrct Approved for Release: 2014/07/29 000622826(b)(3)(n)SovietSelected Soviet Weapon Systems:Projected and Actual Initial Operational Capability (IOC)Weapon SystemProjected IOCActual IOCCommentsMedium solid ICBMEarly 1980s1987Designated SS-24Small solid ICBMEarly 1980s1985-Designated SS-25SS-N-20 SLBMMid-1980s1983New long-range bomberLate 1980s1988Nicknamed"Blackjack"Wide-body cruise missileaircraftLate 1980sNot deployedSA-I0 SAMEarly 1980s1980New airborne warning andcontrol aircraftEarly 1980s1984Nicknamed"Mainstay"Two new interceptorsMid-I980s1984 and 1986Nicknamed -Flanker" and "Fulcrum"Two new ABM missilesMid-1980s1989Nicknamed "Gorgon" and "Gazelle"Possible ground-based laser airdefense systemLate 1980sNot deployedSA-II SAMEarly 1980s19801-80 tankEarly 1980s1981T-80 follow-on tankMid-1980s1984Upgraded version rather than new designNew ground support aircraftEarly 1980s1984Nicknamed aircraft "Frogfoot"New attack helicopterLate 1980s1991Nicknamed "Havoc"SS-23 SRBMEarly 1980s1985New SRBMLate 1980sNot deployed: SRBMs limited by INFTreatyNew cruise missilesubmarineEarly 1980s1982Oscar-classNuclear-powered aircraftcarrierLate 1980s1991Admiral Kuznetsov-class; not nuclearpoweredNew heavy transportaircraftMid-I980s1986Nicknamed "Condor"34Approved for Release: 2014/07/29 000622826 SovietApproved for Release: 2014/07/29 000622826(b)(3)(n)and possibly (especially in the absence of SALTconstraints) to deploy land-mobile ICBMs. They mayintroduce a new strategic bomber or an aircraft tocarry long-range cruise missiles, and they mayalready be testing a sea-launched strategic cruisemissile.All of the above. Clearly, the Soviets perceived anemerging threat to their own forces?a threat thatwas embodied in collection tasking to their ownintelligence services. For their strategic forces, theSoviets deployed the rail-mobile SS-24 ICBM andthe road-mobile SS-25. By 1989, there were sixTyphoon-class ballistic missile submarines, eachcarrying 20 ballistic missiles with multiple warheads.Other modernized submarines were equipped withnew missiles, some of which had a limited capabilityto attack hard targets such as US missile silos. Anew long-range, supersonic bomber was introducedin 1988 (although the Soviets apparently scaled backthe number produced) and the Soviets developed air-launched and submarine-launched long-range cruisemissiles.Should strategic arms control negotiations beresumed, -these weapon developments could compli-cate monitoring an already difficult US intelligencetask. Land-mobile strategic weapons and cruisemissiles cannot be counted with high confidence.As a result, monitoring strategic arms control agree-ments will be much more difficult in the 1980s thanit was in the 1970s.Yes. We did not envision, however, that the USSRwould tolerate intrusive on-site inspection that hasmade monitoring the INF and START treaties morefeasible.The Soviets continue their antiballistic missile (ABM)programs, but the technical difficulties of detecting,identifying, and intercepting ballistic .missiles havekept progress slow. Moreover, the deployment con-straints of the 1972 ABM Treaty severely limit theeffectiveness of defenses against missiles. (Should theSoviets abrogate the treaty, they could deploy ABMdefenses widely in the latter half of the decade.) We. expect continuing Soviet interest in antisatellitedefenses and in high-technology systems for strategic35defense. Possible developments in the late 1980scould include a space-based antisatellite laser sys-tem and a few laser air defense weapons. Continuingcivil defense efforts will improve protection for theleaders and essential work force, but not for thegeneral population or for military or economic facil-ities. Soviet capabilities against ballistic-missile-launching submarines will remain poor.This projection was accurate, except in the case ofthe laser programs. The Soviets continued throughthe 1980s to devote about as much investment tostrategic defenses as they did to trategic offensiveprograms. They modernized the ABM system aroundMoscow. They did not abrogate the treaty and incurthe expense of a nationwide ABM system. The smallnumber of ABM launchers and associated radarsmade the system highly vulnerable and ineffectiveagainst large-scale attacks, and in retrospect we prob-ably overestimated the possibility of an "ABMbreakout." They continued to work on laser systemsfor defense, but they did not deploy operationalweapons. Although a major effort to provide protec-tion for the leadership continued, civil defense pro-grams for the general population and the economyremained selective and ineffective. An extensiveantisubmarine warfare (ASW) effort continued, butby 1989 the Soviets remained incapable of threaten-ing US ballistic missile submarines in the openocean.We project that, despite the widespread Westerndeployment of counterforce weapons in the 1980s,the Soviets will maintain the capability to destroymost of the US population and industry in a retalia-tory strike. Conversely, despite their own growing,counterforce and defensive capabilities, they will notin the 1980s be able to prevent a devastating retalia-tory strike by remaining Western ICBMs and air- andsubmarine-launched weapons.A net assessment. This is not our job, strictly speak-ing, but it is generally accurate. The "balance ofterror" remained a central feature of the US-SovietApproved for Release: 2014/07/29 000622826 cprrPt(b)(3)(n)Approved for Release: 2014/07/29 000622826relationship throughout the decade, although by theend of the 1980s the Soviets had scaled back someof their strategic nuclear programs. As recently as1990, analysts were debating whether this presaged anew doctrine of "minimum deterrence." Debates onthe role of nuclear forces continue in Moscow.Programs for theater nuclear weaponry will furthererode NATO's nuclear advantage in Europe unlessNATO takes action to offset them. The Soviets haveprograms under way to improve the accuracy andflexibility of nuclear delivery systems at all ranges.These include the introduction of new tactical air-craft and short-range ballistic missiles, the continu-ing deployment of nuclear-capable artillery, and fur-.ther improvements in the number and quality ofweapons long-range theater nuclear delivery vehicles(missile launchers and aircraft) based in the USSR.At the time of writing, NATO had already takenaction?notably the 1979 "Dual-Track" decision todeploy the US Pershing II and Ground-launchedCruise Missile while simultaneously negotiating toconstrain Soviet forces. In the event, the INF Treaty,and later the collapse of the Warsaw Pact, led to afundamental change in the European nuclear balance.Neither was obvious to the authors of the 1981study, though we did recognize the increasing pres-sures on the Soviets to pursue negotiations, as wellas the strains in the Pact.Our baseline projection includes improvements inSoviet Ground Forces. They will continue to empha-size the central role of armor; by the end of the de-cade most major Soviet units (and some units oftheir allies) will have tanks with advanced armorthat provides good protection against current NATOweapons. The introduction of new artillery and airdefense systems, as well as organizational changesthat involve the addition of combat units andweapons, will increase the capabilities of Soviet divi-sions to respond to rapidly changing battlefield con-ditions. New fixed-wing ground attack aircraft andhelicopters, with increased ranges and payloads andimproved munitions, will increase the vulnerability ofNATO's installations and forces and improve Sovietcapabilities for close support of ground operations.SovietThese judgments were generally accurate, but sug-gested a greater Soviet confidence in their forcesthan they actually had at the end of the 1980s. AnNIE written in 1989 stated that "the Soviets have ,been able to match or exceed NATO's capabilities innearly every ground forces' weapon category." But italso noted that "The Soviets assess NATO to be atougher military opponent on the conventionalbattlefield today than in past decades." This wasbecause of improvements in NATO doctrine and itsability to integrate land and air forces, as demon-strated so vividly in the Persian Gulf war.With these new systems, we expect Soviet theaterforces to keep pace with NATO's modernizationprograms. The East European forces of the WarsawPact will improve less rapidly, however, because eco-nomic constraints will limit the amount of modernSoviet equipment they can afford to acquire andmaintain.They did not keep pace, as production rates dropped.Moreover, the non-Soviet Warsaw Pact states ex-perienced not only economic strains but also politicalupheaval.- While we knew that the Eastern Europeanforces would diminish in terms of comparative effec-tiveness, we did not expect that the Pact wouldevaporate.Soviet naval programs will continue to emphasizeopen-ocean forces and the deployment of air powerto sea. These programs will improve the Navy'scapabilities to contest areas of the open ocean withthe West. Ships and submarines with a new, long-range cruise missile are being introduced to offsetWestern gains in shipborne defenses. The Soviets areproducing nuclear-powered attack submarines at anincreasing rate, and the submarines introduced inthis decade probably will be quieter (and harder todetect and track) than current models.Something of an overstatement! (Could it have beendue to the fact that the principal author of the 1981study was a former naval analyst?) By the end of the1980s, the Soviet Navy had been modernized, but36Approved for Release: 2014/07/29 000622826 SovietApproved for Release: 2014/07/29 000622826lacked adequate ASW, air defense, sea-based tacticalair support, and naval amphibious lift to sustainlong-range operations. The long-range cruise missileswere slower in coming than we anticipated, and sub-marine production rates were falling at the end of thedecade.Another naval development has important implica-tions for Soviet military power?we have evidenceof activities that probably are related to a programfor a new aircraft carrier It could be introduced inthe late 1980s and probably would carry standardfighter or attack aircraft and be nuclear powered.(The Soviets have helicopter carriers and ships thatcarry short-range, vertical and short takeoff andlanding aircraft, but this could be their first attackaircraft carrier.) !t would improve the Navy:c airdefenses and?more important?it could inauguratea capability for projection of air power in distantarea. The USSR could not achieve a large-scalecapability in the 1980s?only one or two carrierscould be available?but this could emerge as amajor theme in the 1990s and laterAn exaggeration! We correctly identified the con-struction program for aircraft carriers, but greatlyunderestimated the construction time. The firstcarrier was still not finished at the end of the decade,and it now appears that projection of power to dis-tant areas was not the objective of the limited air-craft carrier program.With these new forces and capabilities, we expectthe Soviets to maintain a high level of activity in theThird World to achieve both military and politicalgoals. They may be willing to use their own forcesmore actively in the Third World, even if the activitybrings a greater risk of confrontation with Westernpowers.WRONG! The Soviets came to the end of their ropein Afghanistan, withdrew from Cam Ranh Bay, andgenerally hunkered down. A case of trying, unsuc-cessfully, to infer political intentions from militaryprograms.37?Sower?(b)(3)(n)Alternative Projections. More radical changes inSoviet military policy are possible. Currently availa-ble evidence provides no clear indications that theyare in the offing, but the interaction of political, eco-nomic, and technological forces in the 1980s couldconceivably lead to major discontinuities.Acknowledging that we could have been wrong wasa positive step. But we felt compelled to excuse our-selves on the basis that there was no current evi-dence.One possibility is that the Soviets will reduce thelevel of military expenditures absolutely (rather thanmerely reducing the rate of increase). We believe thisis to be unlikely in the near term. Their dim view ofthe international environment would argue againstsuch cuts, and the guidelines they have published fortheir next Five-Year Plan imply continued growth indefense spending. We have not detected any evidencethat the Soviets are considering reductions.We correctly acknowledged that what really hap-pened was possible.Nevertheless, reductions cannot be excluded as along-run possibility; and, as one alternative projec-tion, we have examined the consequences of a cut indefense expenditures. We believe that to reduce ex-penditure levels in real terms the Soviets would haveto alter the roles and missions of some of theirarmed forces. They probably would spread the cutsamongst all the military services?making themsomewhat deeper in general purpose forces, espe-cially ground forces. General purpose forces arelarger than strategic forces, and they take up moreof the defense budget and use more of the energy,manpower and key material resources needed by thecivilian economy. Production of general purposeweapon systems competes directly with production ofequipment for transportation, agriculture, andmanufacturing. (The resources devoted to productionApproved for Release: 2014/07/29 000622826 ?Seeret--(b)(3)(n)Approved for Release: 2014/07/29 000622826of strategic weapons, on the other hand, are morespecialized and less readily transferable to importantcivilian uses.)This proved to be accurate. In 1988, when Gorbachevannounced a unilateral reduction in defense spendingand forces, he articulated a new doctrinal conceptof "reasonable sufficiency" that involved changesin the roles and missions of the services. His cutsaffected all the forces, but impacted more heavily ongeneral purpose forces, especially the ground forcesin Central Europe and the portions of the USSR westof the Ural Mountains.Another alternative projection considers the possibil-ity that the Soviets will increase defense spendingmore rapidly than in the past to support astepped-up military comfietition. This effort (focusedon either strategic or conventional forces) could ex-pand the forces and improve capabilities morerapidly than is forecast in our baseline projection.The range of program options is broad enough topermit a major increase in defense spending, andSoviet military-industrial capacity is large enough tosustain it. Such an increase would affect the distribu-tion of economic resources significantly, however(especially if it were in conventional forces), and itspolitical consequences could be extremely serious:? The Soviets' ability to increase investmentresources critical to long-ternz economic growthwould be reduced substantially.? Per capita consumption might decline in realterms late in the decade.? Key sectors of the economy would be disrupted.In retrospect, we should have realized that these con-sequences were so serious that an increased militaryeffort was a real nonstarter.We do not know at what point the Soviets would findan increased defense burden to be unacceptable. Thiswould depend on the international environment andthe outlook of the leaders in power. Judging by theirpast behavior, we believe that they would prefer, if__Seerer?possible, to keep defense expenditures within theircurrent growth rate, while still pursuing their mili-tary goals:? The Soviets probably will seek to constrain USprograms and to reduce their uncertainty aboutfuture US capabilities by urging further arms con-trol negotiations.? They will also attempt, through propaganda anddiplomacy, to undermine Western cohesiveness onsecurity issues and to slow the pace of WestEuropean defense programs.They certainly did both of these. But eventually thepain became so great that they took dramatic,unilateral actions.The Soviets' incentives for such actions will increaseas their economic growth slows in the 1980s. ButSoviet leaders place a high premium on militarypower and will not, for economic reasons alone,accept constraints on defense programs that theyconsider vital to their interest.A truism! And a lousy way to end an analysis thatwas, all things considered, pretty good. Of course,the Soviets would not forego programs they consi-dered vital. But what Gorbachev and his colleaguesconsidered vital was somewhat different from whatwe projected. In sum, we were generally accurate inour projection of Soviet military capabilities in thelate 1980s, but underemphasized the possibility of a"paradigm shift"?a dramatic change in the underly-ing factors that shaped Soviet policy.Indicators of ChangeIn its concluding sections, the report explicitly ac-knowledged the possibility that its assessments wereincreasingly subject to uncertainty. It described boththe conditions that could lead to an accelerated orreduced Soviet military effort and the specific evi-dence or indicators that might accompany a policychange.38_Approved for Release: 2014/07/29 000622826 Approved for Release: 2014/07/29 000622826This notion of "indicators" is one that analysts hadconsidered for some time, partly at the urging of theDCI's Military-Economic Advisory Panel, which wascharged with overseeing CIA estimates of Sovietmilitary spending and economic performance. A fewmonths after publication of the study, the Office ofSoviet Analysis was formed as part of a large-scalereorganization of CIA analysis. The office estab-lished an "Indicators Project" to monitor evidencethat might lead to a new assessment of Soviet mili-tary intentions. A senior analyst was placed incharge, and an annual assessment was planned.Regrettably, because of personnel changes, a sys-tematic effort at monitoring indications of changewas never institutionalized.In retrospect, the indicators described in the 1981study as signaling a reduced military program werequite accurate. The relevant portions of the report arereproduced below in italics, together with commentson actual developments:We consider an absolute reduction in military spend-ing unlikely... Under some circumstances, however,the Soviets might feel impelled in that direction.These include:? Economic conditions poorer than those we cur-rently project?for example, a series of disastrousharvests causing an actual reduction in economicoutput.Poor harvests occurred, further depressing economicgrowth. Some Soviet economists calculated an abso-lute decline in the economy during this period.? The spread of popular unrest front Eastern Europeto the USSR, coupled with the rise to power ofpolitical figures sympathetic to the consumers'plight.Both happened.- A Sino-Soviet rapprochement, a general lesseningof tensions with the West, and a move by WestEuropean countries closer to the Soviet orbit andaway from US influence.39__secra(b)(3)(n)Tensions lessened, largely because Gorbachev tool(b)(i )steps to moderate them.(b)(3)(n)?... We would expect to receive evidence that would alert us to thechange ... But no single clue would be adequateto identify a policy shift. We would have to detectat least several indicators, and evaluate them overa year or more, before we could be confident ofidentifying an actual change...It took us several years to recognize that majorchanges were under way. This was because the"soft" evidence?political and economic indicators?preceded the "hard" evidence of actual changes inmilitary programs and forces.Political. Political evidence of a reduced militaryeffort could include:? Reports of greater optimism in the Kremlin on theprospects for detente.Gorbachev proved much more willing than hispredecessors to engage with the West on arms con-trol and foreign policy issues, but it took analystssome time to understand this. (A 1989 NIE containeda section entitled "Is Gorbachev's 'Detente'Different?") After meeting initial US skepticism, bythe late 1980s Gorbachev found a willing partner fordetente in Ronald Reagan.? The accession to leadership of political figures into support civilian economic activity?perhaps even at the expense of the military.In the mid- and late-1980s, Soviet leaders evincedgrowing concern about the costs of their militaryeffort and the need to divert resources to rebuildthe civilian economy. In 1988 and 1989,Gorbachev announced the first unilateral reduc-tion in military forces and budgets since the1950s.? The admission of additional civilian participantsto the defense decisionmaking process.By 1989, the USSR had established a defenseoversight committee in the Supreme Soviet, andcivilian academics increasingly challenged themilitary and served as alternative sources ofanalysis for political leaders:Approved for Release: 2014/07/29 000622826 Approved for Release: 2014/07/29 000622826(b)(3)(n)? A more flexible Soviet stance on arms control;in particular, movement on MBFR could signala desire to improve economic pezformance byreducing expenditures on theater forces.In the mid-1980s, the Soviets became more flexi-ble across the board on arms control, whileanalysts debated whether their new stance wascosmetic or serious. By the end of the decade,new agreements were in place covering the fullrange of conventional and nuclear forces.? Signs of greater tolerance for experimentation inecononzic management and more sympathy forconsumer complaints.In the mid-I980s, Gorbachev introduced the newconcepts of glasnost, broadening the limits ofpolitical and economic debate, and perestroyka,major restructuring of The economic system, in aneffort to stave off economic decline. A principalobjective was to improve living conditions andarrest social decline.Economic. Economic information that mightreflect a reduced defense program could include:? Pessimistic Soviet forecasts of economic growth.In the late 1980s, Soviet economic performancecontinued to slide, with agricultural output declin-ing in several years. As early as 1986, Gorbachevexpressed the fear that the USSR could become athird-rate power.? Energy shortages developing early in the 1980s.Many of our critics charged that we were un-necessarily pessimistic on Soviet energy resour-ces.' However, significant energy problems diddevelop in the late 1980s: oil production fell by2 percent in 1989 and 6 percent in 1990, and netexports of oil fell by 2 percent in 1989 and some15 percent in 1990.? Major shifts in plan targets toward increasedinvestment or consumption at the expense ofdefense programs.Against the backdrop of worsening economic per-formance, the Soviets announced in the late 1980sa 20-percent cut in spending for weapons procure-ment._Secret--Soviet? Reduction or halting of construction activity atdefense plants.In 1989, the Soviet leaders endorsed conversionof some military production capability to civilianpurposes, causing dislocation in military pro-grams.? Sharp increases, actual or planned, in the outputof civilian transportation or agricultural equip-nzent.This equipment was produced at plants that alsomanufactured tanks and armored vehicles.Gorbachev's 1988 decision to cut back generalpurpose military forces was an admission thatmilitary production was curtailing needed civilianprograms.Military. If the military effort were being cutback, we might also see:? Cuts in weapon testing levels and productionrates.By 1989, the number of strategic delivery systemswas decreasing and the number of strategic mis-sile test launches had decreased by 50 percent.? Dissolution of military units and reorganizationor consolidation of forces.This began to occur in the late 1980s in conven-tional forces and in the early 1990s in strategicforces.? Releases of men from active duty and reduceddraft calls.In December 1988, Gorbachev announced aunilateral personnel reduction of 500,000 in theSoviet armed forces, and by the 1990s the con-scription system had begun to collapse.? Evidence of debates on the roles and missions ofthe military services and on the nature of a futurewar and the goals of military strategy.40Approved for Release: 2014/07/29 000622826 SovietApproved for Release: 2014/07/29 000622826These debates were in full swing by the mid- tolate-1980s, and led to a major redefinition ofSoviet military doctrine under the concept of"reasonable sufficiency."Grading Our PerformanceGaddis gives himself a grade of C+ for his predic-tions regarding the Cold War. The NFAC analystshad a somewhat less daunting, but still challeng-ing task: to make predictions about the future ofSoviet military power. To do so, they had to con-sider both "soft" issues like leadership views andintentions and "hard" issues like weapon systemdevelopments. Not surprisingly, the analysts didbetter at the latter.The NFAC analystS seem to merit a grade similarto the one Gaddis gives himself for their forecastsof Soviet political developments; we did not doany better or any worse than historians and politi-cal scientists outside the Intelligence Communityin predicting the end of the Cold War and the col-lapse of Communism. But we merit a much bettergrade (say an A-) for having an excellent under-standing of weapon systems and programs and foraccurately describing most of the capabilities ofSoviet military forces nearly a decade in advance.Moreover, we had a good list of "leading indica-tors," although we certainly could have monitoredthem more systematically. Overall, a high B!At the conclusion of his essay, Gaddis asks whyhistorians and social scientists do not do a betterjob at prediction.6 Was the NFAC analysis ofSoviet military power guilty of the same errors ofjudgment as Gaddis and his colleagues? The fiveweaknesses that Gaddis noted were:I. The assumption that the future will resemblethe past.Guilty! Although we realize that changes in policywere possible, we characterized them as "unlike-ly." And even when we explored alternativescenarios, we modeled them on historicalevents?the reduction in military spending underKhrushchev in the late 1950s and the rapidbuildup of strategic military forces in the early1960s.41-SUrr71?(b)(3)(n)2. The temptation to return nations intoabstractions.Not guilty! One of the strengths of the NFACstudy was that it dissected in minute detail theinner workings of Soviet defense, as we under-stood them at the time. Even more than a decadelater, this is an impressive accomplishment, andthe study was genuinely multidisciplinary inbringing together the perspectives of political,military, technical and economic analysts.3. Neglecting points of intersection betweenshort-term and long-term phenomena.Guilty of a lesser included offense! We clearlyknew what the short- and long-term phenomenawere. (They included growing societal ills, afaltering economy, and an uncertain political suc-cession on the one hand; and the momentum ofmilitary programs, Soviet desire for internationalpower, and the threat from a rearming West on theother.) And we knew that they were intersecting.We did not, however, know how to weight thevarious phenomena, which were pushing Sovietdecisionmaking in different directions. In the end,we opted for the conservative assumption that theforces for continuity would outweigh those forchange.... We believe that we have identified most ofthe weapons ... that will shape the evolution ofSoviet military power over the next decade.That knowledge, plus our understanding ofSoviet decisionnzaking and of the military, po-litical and economic environment in which ittakes place, leads us to believe that Sovietforces and doctrine will develop much as out-lined below.There followed a series of projections that gavethe greatest emphasis to continuation of currentforces and policies. This proved to be accurate formost of the decade, with the major changes notbeginning in earnest until 1988.4. Ignoring the role of personalities.Guilty, but with mitigating circumstance(b)(i )  (b)(3)(c) (b)(3)(n)Approved for Release: 2014/07/29 000622826 Approved for Release: 2014/07/29 000622826(b)(1)(b)(3)(c)(b)(3)(n)We also co-vered the implications of both the Soviet politicalsuccession and possible changes in the militaryleadership. On page 57 of the study there are pho-tographs of six potential successors to Brezhnev,including one labeled "M S. Gorbachev, PartySecretary." The text reads:Gorbachev, 50, is the youngest of the top lead-ers and is responsible for agricultural policy;his views on defense and national security is-sues are unknown. He reportedly has opposedliberalization in Eastern Europe.On the other hand, there is no photo of YuriyAndropov. During the coordination process, oneof the analytical offices insisted that he be deletedfrom the list, because the Soviets would never ap-point a KGB Chairinan as General Secretary. Inshort, we knew that personalities mattered, butknew too little about both the individuals and theselection process.5. Chance. Unpredictable events, by definition,defy prediction!Nolo contendere. We did not predict the unpredic-table, but we did say:We cannot rule out the possibility that Sovietmilitary programs in the 1980s will differ fromour baseline projection because our evidence,though fairly definitive on future weapon sys-tems, is less conclusive in such areas as thepolitical succession and is too general to pin-point the foreign policy environment and eco-nomic peiformance in any given year But it isprecisely these less predictable factors thatcould make the greatest difference in Sovietpolicies.In retrospect, we might have noted more force-fully that the increasingly complex environmentfor Soviet decisionmaking on military programsheightened the possibility that chance eventscould result in sharp discontinuities. "We cannotrule out the possibility" is a real cop-out!Certainly, however, the role of unpredictable42Sovietevents in the actual development of Soviet mili-tary power should make us more humble aboutour powers of prognostication.So What?In another essay, Gaddis asks whether good intel-ligence in fact makes any difference in policy.' Hequotes Yale historian Robin Winks as asking "So,what difference does it make that ... Hitler hadone testicle, that Sicilians still use sixteenth-century vulgarisms, that narrow-gauge track is notthe same in New South Wales as in the Sudan."8The same question has been raised in a series ofcase studies conducted by the Kennedy School ofGovernment, Harvard University, under theCIA-sponsored Intelligence and Policy Program."The answer seems to be that the influence of in-telligence on policy depends more on the policyprocess itself than on the quality of intelligence.In the case of analysis of Soviet military programsin the early 1980s, it is difficult to imagine thatbetter analysis would have made much differenceto policy. To assess this we have to indulge inwhat historians call "counterfactual" analysis.Hold everything else constant and alter the varia-ble of interest, then try to imagine what mighthave happened. For example, if Cleopatra's nosehad been two inches longer.Let us imagine that CIA discerned as early as1983 that a significant unilateral reduction in theSoviet military effort at the end of the decade wasincreasingly likely."' Would this have altered keyelements of the Reagan Administration's defensestrategy such as the buildup of American strategicforces, the 600-ship Navy, the "Zero-Option"'proposal for the Intermediate-Range NuclearForce negotiations, or the Strategic DefenseInitiative (SDI)?Frankly, it seems highly unlikely. The buildupwas managed by officials whose views of theUSSR were shaped by the Committee onthe Present Danger and the so-calledB-Team?defense analysts who regarded theApproved for Release: 2014/07/29 000622826 Approved for Release: 2014/07/29 000622826SovietIntelligence Community's views with skepticism,if not contempt." The 600-ship Navy was thebrainchild of a forceful Secretary of the Navywith strong support from the President. TheZero-Option was the outcome of political bargain-ing among presidential advisers, and intelligenceplayed essentially no role in the policy decision(although it was useful in policy implementa-tion).12 And SDI resulted from the coincidence ofPresident Reagan's vision of a nuclear-free worldwith the technological hopes of a few key scien-tists, especially Edward Teller. The president'sprincipal national security advisers (let alone in-telligence analysts!) knew next to nothing about ituntil the last minute."Perhaps over time, as the indicators becamestronger, intelfigence on changes in Soviet mili-tary programs might have made a greater differ-ence at the margin. But, again, the influence ofpolitical processes and personalities seems to begreater. Would President Reagan have been evenmore adventuresome at Reykjavik if he had beenarmed with intelligence showing that Gorbachev'sreforms would lead to loss of Communist Partycontrol? It is hard to imagine that he would."Moreover, since Reagan became an ardent arm'scontroller, Gorbachev fell for domestic politicalreasons and the West eventually won the ColdWar, it is tempting to task again, "So what?"Explanation Versus PredictionPerhaps the most important "so what" is that theIntelligence Community was essentially correct inits understanding of the factors that influencedand explained Soviet military policy, and in its as-sessment of Soviet military capabilities nearly 10years in advance. This is an achievement of greatimportance, unprecedented in the history of intel-ligence. It resulted from a confluence of sophisti-cated technical collection capabilities, old-fashioned espionage, creative analytical methodol-ogies and sound scholarship. That we did not do abetter job of predicting the fall of Communism isregrettable but not surprising. After all, as Gaddishas noted more recently, social science failed inpredicting the end of the Cold War, and intelli-gence is in the final analysis dependent on themethods of the social and physical sciences:543?Ste trer(b)(3)(n)What is of more concern is how well theIntelligence Community?and specifically NFACand its successor organization?capitalized on itsunderstanding to help policymakers understandthe dynamics of Soviet military policy and the in-fluence their own actions might have on it. Againthe record is mixed. Many of the participants inthe, 1981 study went on to positions that gavethem direct contact with policymakers and theability to provide intelligence support to negotia-tions and policy decisions. But others took assign-ments that were at best peripherally related to theexpertise they gained and?as we have seen?the"indicators" effort foundered because of person-nel reassignments. MI of this suggests that theinfluence of the analysts' understanding on thepolicy process was less purposeful than might behoped for.The key lessons in this experience seem to be:? Intelligence has a comparative advantage whenintelligence sources or methodologies make amajor contribution to analysis, and does betterthan academia in integrating the expertise fromdifferent disciplines. This permitted us to forecastSoviet military capabilities with good accuracy.? Ultimately, intelligence insights on the softerareas of political trends depend on the methodsof the social sciences, which have proved to beineffective in predicting major changes in theinternational system. Hence, our performance waspoorer in forecasting Soviet political choices thanin projecting military forces.? Intelligence, like the social sciences, is likely tobe better at explaining than predicting, because ofthe expertise of its people and the high quality ofits written products and briefings.? Intelligence analysis should discuss alternativefutures, together with the conditions that couldlead to discontinuities abroad and indicators ofimpending change. Equally important, the sys-tematic monitoring of key indicators should beinstitutionalized, and bureaucratic procedures orpersonnel systems that interfere with this taskshould be reexamined.Approved for Release: 2014/07/29 000622826 --ssr frer(b)(3)(n)Approved for Release: 2014/07/29 000622826? Expertise is difficult to build and should not bedissipated for reasons of personnel management ororganizational convenience. Intelligence organiza-tions should foster and protect expertise, and en-sure effective interaction between intelligence andpolicy officials.? No matter how accurate the explanations orpredictions of intelligence analysts, in the finalanalysis policy depends more on political factorsand the values and mind-sets of leaders than onintelligence.NOTESI. The article appeared in The Atlantic, CCLX(November 1987), pp. 88-100.2. John Lewis Gaddis, The United States and theEnd of the Cold War (Oxford, Oxford UniversityPress, 1992), pp. 133-154.3. The research process was described in "AProcedure for Managing InterdisciplinaryIntelligence Production," Studies in Intelligence,fall 1981, p. 37.4. The comments are based either on well-knownhistorical facts or on the descriptions of Sovietmilitary forces in NIEs from the late 1980s andearly 1990s.5. See, for example, US Congress, Joint EconomicCommittee, Gorbachev's Economic Plan(Washington, US Government Printing Office,1987) p. 494.6. Gaddis (1992) pp. 153-4.7. "Intelligence, Espionage and Cold War History,"in Gaddis (1992), p. 94.8. Robin W. Winks, Cloak and Gown: Scholars inthe Secret War, 1939-1961 (New York, Morrow,1987), p. 63.Soviet9. Case Book: Intelligence and Policy Project(Cambridge, Harvard University, 1991).10. This would have been difficult to predict, asthere was at that time no clear evidence of sig-nificant changes in Soviet military programs,although we did see a slowdown in Soviet mili-tary procurement. More definitive evidence didnot become available until after Gorbachev tookoffice.II. See Anne Hessing Kahn and John Prados,"Team B: The Trillion Dollar Experiment,"Bulletin of the Atomic Scientists, April 1993,pp. 22-31.12. L. Keith Gardiner, INF Deployment: The Roleof Intelligence Analysts in a Policy Success(Washington, Center for the Study ofIntelligence, 1993)13. George P. Shultz, excerpts from Turmoil andTriumph, lime, 10 May 1993, pp. 50-51.14. "The Unexpected Ronald Reagan" in Gaddis(1992), pp. 119-132.15. "International Relations Theory and the End ofthe Cold War" International Security. winter1992/93, pp. 5-58.This article is44Approved for Release: 2014/07/29 000622826(b)(3)(n)