Secret Leviathan

7. SECRECY AND THE UNINFORMED ELITE

In 1989 the Soviet Union’s leaders finally let go one of their most valued secrets, one that Western agencies had devoted tremendous efforts to guessing or uncovering over a quarter of a century. This was the true size of the USSR’s military budget. The Soviet leaders agonized over this decision: what might they gain and lose by disclosing the truth? Was continued secrecy helping them achieve their goals—or had it become a hindrance to safeguarding the Soviet Union’s vital interests? Would the disclosure be enough to satisfy Western curiosity—or would it open the Soviet government to demands for more revelations?

Underlying these dilemmas was a still more agonizing question: Could the leaders themselves lay hands on the information some of them now wanted to disclose? Did anyone know the truth? If not, how could they disclose to the West something that none of them knew, even if there existed a compelling national interest in doing so?

From the point of view of secrecy, the subject of this chapter is the effects of compartmentalizing secret government information. The scale of a country’s economic commitment to military power is valuable information. The economy’s overall supplies must be divided between military uses on one hand, and civilian uses on the other—primarily, consumption and investment. The more is committed to defense uses, the less is available to others. The division can be an index of national priorities. The government of a country may have an idea of the influence it seeks in the world

and the role of military power in supporting its ambitions. But it also needs to know the economic costs and whether they can be sustained. Domestic opponents and foreign rivals can exploit the same information. If they know the costs and sustainability of a country’s military power, they can second-guess the government’s true intentions and criticize them at home or work to frustrate them abroad. In turn, that becomes the ruler’s reason to lock the information in a secret compartment reserved for a few trusted subordinates.

At this point, the ruler’s problem becomes: whom and how many can I trust? On the most charitable reading of the evidence, for much of the duration of Soviet rule, the size of the military budget was regarded as so sensitive that barely a handful of people could be trusted with it. In Stalin’s time, the fixing of the defense budget in the Politburo was a key moment in the annual cycle of economic policy. By Brezhnev’s time the defense budget was so secret that most Politburo members were excluded. The decision (if it was made at all, an issue raised below) could not be debated anywhere, not even in the hypersecret council of the Politburo.

From the point of view of state capacity, therefore, the subject of this chapter is the damage done by such secrecy. For the sake of keeping a few figures away from domestic critics and foreign adversaries, the Soviet rulers lost their own capacity to decide the balance of resources between military and civilian uses. Whether this had anything to do with the collapse of the Soviet system is debatable, but it is certainly an indicator of the system’s decline.

In this chapter we will see measures of Soviet and US military spending compared. The comparisons were controversial almost entirely because of doubts over whether defense activities were properly costed and whether the costs were properly attributed to defense. Mercifully for the reader, there was hardly any disagreement about how to define “defense activities.” Both sides of the Cold War shared a broadly common conception of what items ought to appear in the military budget. This was the government money allocated to each country’s ministry of the armed forces (the Soviet Ministry of Defense or, in the United States, the Department of Defense). On both sides the ministry was expected to pay for the military means of the country’s external security. So, the main line items of the military budget on both sides should have been the outlays of the armed forces on the pay and maintenance of servicemen and women; the procurement of weapons and other equipment; the construction of military bases and fortifications; the costs of military transportation and operations; and outlays on military research, development, testing, and experimentation (RDTE).

Both sides accepted that there was a grey zone. In the grey zone were military pensions—a legacy cost of military activity in the past rather than in the present.^[372] Also in the grey zone were the costs of building or maintaining various capacities that could be switched between civilian and military uses.^[373] For one example, the Soviet Union maintained “internal” troops that were charged to the Ministry of the Interior, not the Ministry of Defense, because their primary use was in domestic security operations. But they were fully capable of military action, and in the Soviet-German war, they took part in battles. Another example is the costs of building defense factories: in both countries these were not charged to defense on the grounds that building a new plant adds generally to a country’s productive capacity, not just to its defense capacity. The same applied to industrial mobilization capacities, such as war factories built in excess of the country’s immediate needs for defense goods, which were used in peacetime to produce civilian goods while being kept in readiness for a rapid switchover to munitions when conflict threatened. The costs of keeping them ready for war were paid out of civilian, not military funds. Likewise, both sides excluded outlays on civil defense from the defense budget because equipping and training civilians to deal with emergencies could be just as useful in the case of natural disasters and pandemics as for war contingencies. It is plausible that communist regimes sank more resources into dual purpose capacities than Western governments in the Cold War, relative to the size of their economies. But the grey zones on either side played little part in the issues discussed below.

On that basis, defining and measuring the costs of Soviet defense activities in a way that was roughly comparable to the United States should have been a simple matter. Instead, it became a mare’s nest of equivocation and disinformation.

As a point of departure, Soviet leaders seldom published exact figures that they knew to be false. The Russian archives have confirmed that most economic statistics that were published were the same as those available to government and party officials in secret. While the figures might have been biased by the underlying methods, or distorted by the incentives of those who reported them, the biases and distortions were similar for everyone.^[374] In other words, the system did not generally provide for two different sets of accounts: a false set for public consumption, and a true set for the informed elite.^[375] For most purposes, the limited information that the public saw was just a highly restricted sample of the mass of data on which the government relied for its secret business.

As the economist Peter Wiles described it, the Soviet authorities followed a rule of “minimal untruthfulness.”^[376] When the leaders wanted to conceal the truth of some quantity, they did not usually invent an outright lie. Rather, they preferred to mislead by giving out a half or a quarter of the truth, or else to say nothing at all.

There were exceptions, and the exceptions could concern matters of historic importance. On a few occasions, the Soviet government published figures that amounted to deliberate fabrications. The lies included the size of the Soviet harvest in 1932 (disastrous), the size of the Soviet population in 1939 (decimated by famine and terror), human losses in the Soviet-German war (catastrophic), and property losses in the same war (overstated). In all these cases, the false figure was adopted everywhere, in secret councils as in the public sphere. No secret compartment was set aside where the truth was retained.^[377]

Then, there was military spending. In the 1920s the Soviet government’s official figures for its annual outlays on defense were truthful, at least in the sense of a lack of intention to deceive. In 1930, however, the League of Nations opened disarmament negotiations in Geneva among the great powers. The Soviet government decided to take part. At home, rearmament had begun, and the Soviet military budget was growing rapidly. With the aim of supporting their negotiating position in Geneva, Stalin’s Politburo decided to conceal the increase in defense outlays by publishing false figures. The actual and false figures are shown in Table 7.1.

For a few years, the published figure fell below the actual one by a wide margin. By 1932, when the USSR published a defense spending figure of 1.3 billion rubles (less than 4 percent of total government outlays) to the League of Nations, the actual figure was 4.3 billion (nearly 11 percent of government outlays).

In defense accounting, therefore, for several years there were two sets of books, one for consumption by the public along with the broad mass of less-privileged party members and government officials, and another for the Politburo alone that showed the true situation. This position changed in 1935. The ruling circles in Berlin, Rome, and Tokyo were now openly expansionist. Disarmament hopes had collapsed. Stalin decided there was nothing to lose from resuming truthful publication. The following year, as shown in the table, the Soviet government published a figure for defense outlays of 14.8 billion rubles (now 16 percent of the state budget and rising).

Having returned to the path of relative truthfulness on this matter, the government remained on it until the outbreak of the Soviet-German war in 1941. While the war continued, Soviet defense outlays were kept secret, but the wartime budgets were disclosed soon after the war ended.^[378] With the return to peace, the published Soviet budget once more gave the actual picture, and this went on for a few years, at least—probably into the 1950s.

By the 1960s, however, the defense outlays disclosed to the public had departed once more from any sensible idea of the actual total. This time there was no early reconciliation, but a lasting separation.

As Table 7.2 shows, in 1950, the reported defense outlays of 8.3 billion rubles accounted for 20 percent of total Soviet government outlays. Over the following years, the ruble figures increased absolutely, standing at 17.1 billion rubles in 1980. But the size of the economy in rubles was now so much larger that the defense share of government outlays had fallen to less than 6 percent. That share continued to drift downward through the 1980s.

These figures made a puzzle. In 1980 the Soviet national income on the official basis (the net material product, which excluded outlays on final services) was 454 billion rubles—probably, more than 600 billion if adjusted to GDP.^[379]

Soviet defense spending as published, therefore, would come in at less than 3 percent of Soviet GDP. In the same year, the United States spent $134 billion on national defense, around 4.8 percent of its much larger GDP. Taken together, the figures suggested that the Soviet economy ought to be able to support a military establishment around one-quarter the size of America’s.^[380]

Western observers also approached the problem from the other end: what was the actual size of the Soviet military establishment? This was a state secret, but it could be gauged from intelligence sources. As Table 7.3 shows, Western estimates around 1980 indicated that the Soviet military establishment of the time was numerically larger than that of the United States in most respects.

Was Soviet military power at least equal to that of the United States, as the numbers of personnel and hardware suggested, or was it one-quarter the size, as the budget figures implied? There was a fourfold discrepancy.

The Soviet media argued that the Western estimates of their military power were overstated. On one argument, they maintained, an impartial observer would discount the numbers. The Soviet Union, unlike the United States, had to defend long continental borders. In many aspects of military technology, moreover, the Soviet Union was at a qualitative disadvantage that mere numbers could not capture. On another argument, the Western estimates of Soviet military power were lies intended to stoke fear. All such arguments were weakened, however, by the fact that the Soviet government presented no credible alternative figures.

It might be asked why the Soviet military budget mattered to anyone. After all, Soviet defense rubles were just a means to an end, an intermediate step to a greater goal, which was the production of Soviet military capabilities. Evidently it was possible to build capabilities without a transparent budget. Regardless of how the rubles were found to pay for them, Soviet missiles armed with atomic warheads could be deployed in silos and submarines. The Soviet Army could wargame battles with NATO forces on the Central European plain; it could invade Hungary, Czechoslovakia, and Afghanistan. All these capabilities could be evaluated independently of the budgetary channels that paid for them. Why then did Soviet defense costs matter?

The importance of the question can be seen from both sides of the Iron Curtain. While they largely understood Soviet military capabilities, Western leaders also wanted to understand Soviet intentions. Soviet military power rested on an economic foundation. The demands it placed on the economy could be used to measure the willingness of Soviet leaders to sacrifice other goals—for example, the modernization of the civilian economy and the living standards of the people—for the sake of military power. Thus, the size and rate of change of the military budget would have provided the Western powers with information about Soviet goals and intentions that could not be obtained just by counting troops and missiles. For Western leaders, these concerns became especially salient in the 1960s, as American defense accountants came to rely increasingly on measures of cost-effectiveness in deciding how to allocate defense dollars.^[381]

Western leaders also wanted to understand whether the Soviet defense burden was sustainable. Consider two ratios. Apparently, the size of the Soviet military establishment relative to that of the United States was roughly 1 to 1. But the size of the Soviet economy relative to that of the United States was perhaps 0.4 to 1. Per unit of military capability, the relative sacrifice of Soviet civilian needs, the loss of provision for present and future living standards, must have been several times greater. Could such a situation continue indefinitely? This depended on the trend of the share of overall production taken up by military demands. If the Soviet defense share was stable, it was sustainable. If the defense share was rising, it could not rise indefinitely.^[382] Stability was easier to maintain if the Soviet economy kept pace with American growth. If it was falling behind, a point might come when Soviet civilians would resist continually increasing hardships.

If the military burden on the Soviet economy grew indefinitely, something must give—but what? Conceivably, the Soviet system itself might give way. Writing about the rise and fall of great powers, the historian Paul Kennedy concluded that overambitious military commitments and the associated burdens are precursors of economic decline.^[383] That was one possibility—from the Western perspective, a hopeful one. A greater worry was that a rising defense burden on the Soviet economy, being unsustainable in the long run, could imply preparation for imminent war.

If Western leaders asked themselves these questions, they were also excellent questions for the Soviet leaders, who should have been capable of providing clear answers, at least to each other in private, if not to the public. But they could not do this without a reliable measure of the defense burden on the economy.

Soviet leaders had further reasons to need to know the size of their own defense budget. Defense outlays should have been an important control variable for economic management. This might not be obvious in the context of a textbook stereotype of the Soviet planned economy. According to the stereotype, it was the Soviet planning system that allocated supplies of materials such as steel, fuel, and machinery to their final uses through the planning. Money was supposed to follow the plan. It was not clear what role was left for demand factors such as budgetary assignments in rubles. But the stereotype is misleading: the real working of the Soviet economy was different.^[384]

Most importantly, the Soviet planners were much less all-powerful than the stereotype supposed. Centralized plans were not detailed enough to link particular products to particular users, and many products did not have a deterministic use. In practice, the final allocation of commodities was closed on the demand side. Aggregate demand was dominated by three monetary aggregates, all of them supposedly controlled by the government: the defense budget, the investment budget, and the annual bill for wages, pensions, and other benefits, from which households made their consumption decisions.

The decisive importance of aggregate demand is supported by the observation that Stalin and his immediate associates gave much more time and thought to fixing the amount of money for the annual budgets for civilian construction and military procurements than to cursory scrutiny and approval of the annual “control figures” for material supply plans. Seen in this light, Soviet money was not just a token; it was valued for its purchasing power. While the ruble’s purchasing power was much less flexible than money in a market economy, it was still decisive.

A related mistake of the textbook stereotype of the Soviet economy is to suppose that prices were fixed, and all decisions were made in quantities of output. In reality, most Soviet supply plans were denominated not in tons or physical units (as the stereotype claimed) but in rubles. To translate plan rubles into quantities required prices. The prices were supposedly fixed from above, but in practice they could be pushed up by suppliers. The result was a tendency to inflation, sometimes hidden, sometimes open. Prices could be pushed up more easily in some sectors than others; complex products like machinery and construction were particularly vulnerable. As a result, an increased budget for capital construction or defense might not result in more real investment or more real military activity. It might just lead to more inflation. In the case of defense, enlarged budgets were not always spent, perhaps partly because of this. Here was another reason for the Soviet government to keep a watchful eye on the aggregate of defense outlays.

To summarize, everyone had powerful reasons to want to know the Soviet government’s military budget. How important was Soviet military power, and what were the Soviet rulers willing to sacrifice to get it? What did the costs of military power imply for the present and future of the Soviet economy? Could the economy sustain Soviet defense costs, not just today, but tomorrow?

Western observers found themselves starved of the clear, credible official information that might have supported the answers they sought. In that situation, they turned to alternative sources.

Over several decades, Western governments and scholars puzzled over the Soviet defense budget. The puzzle was a rather obvious one. The official figures were impossibly small, and they failed to increase in step with the visible expansion of Soviet military capabilities.

The baseline for the alternative Western estimates was set by the US Central Intelligence Agency. The CIA maintained an Office of Soviet Analysis charged with evaluating the military activities and overall economic potential of the Soviet Union. Working out the likely Soviet military budget in both rubles and dollars was an essential part of this work.

In addition to the deceptive Soviet published statistics, the CIA had access to its own intelligence data. At infrequent intervals, its few human sources provided what purported to be inside information about the Soviet Union’s factual defense costs. The “benchmark data,” as William T. Lee, a former CIA analyst, called them, are listed in Table 7.4. They painted a picture of Soviet defense outlays rising rapidly from 50 billion rubles in 1970 (12.5 percent of GNP) to 100 billion in 1980 (16 percent) and to 150 billion in 1982.^[385]

To provide cross-validation without undue reliance on Soviet official statistics, the CIA developed a direct costing approach or “building block” methodology. The building blocks were estimates of quantities of Soviet defense resources used or stockpiled in each year. These were based

on another kind of intelligence data: estimates of the scale of the Soviet armed forces’ acquisitions, deployments, and operations. The building blocks were valued by working out how much it would have cost to provide them in the United States in some base year, such as 1960. For each block of a given resource, the quantity multiplied by the base-year cost gave its dollar value, and the sum of values of the blocks gave the CIA an estimate of overall Soviet defense outlays in constant dollars.^[386] Dollars were converted to rubles by applying either ruble prices of the base year (where known) directly to the same blocks of quantities, or by converting dollar values based on estimated purchasing-power parities. Finally, the current ruble value of Soviet defense activity for a given year could be derived by applying estimates of ruble price changes since the base year.

On that basis, the CIA estimated that the Soviet defense outlays at around 47 billion rubles in 1970, and 94 billion in 1980. These figures, included in Table 7.5, were close to the “benchmark data” if slightly below them, and they showed a similar trend over time.

At first the CIA had a monopoly in the field. Before long, outsiders’ doubts about CIA analysis developed into keen rivalry. Other players included the Pentagon’s Defense Intelligence Agency (the DIA, established in 1961) and the Stockholm Independent Peace Research Institute (SIPRI, established in 1966). Outside the institutions worked a handful

of independent scholars and other mavericks. Among them were Lee (a former CIA analyst who was hired by the DIA), Steven Rosefielde and Peter Wiles (university-based economists), and Dmitri Steinberg (an independent economist). A bewildering range of competing estimates emerged, as shown in Table 7.5.

While admitting that its own figures were of uneven reliability, the CIA maintained that they were corroborated by other sources and accepted by other intelligence services.^[387] But this self-defense was challenged, most fiercely by Lee and Rosefielde. They charged that the CIA’s direct costing procedures were incomplete and insufficiently transparent; they were insufficiently robust to the incorporation of new information and understated the growing quality and cost of Soviet weapons.^[388]

In the CIA’s exchanges with Lee and Rosefielde, there was bad feeling on each side. Lee and Rosefielde both charged that the CIA had colluded in a Soviet strategy of disinformation, resulting in understatement

of the Soviet military-economic effort. The implications that they drew diverged, however. According to Rosefielde, the outcome was insufficient United States preparedness for war.^[389] According to Lee, the outcome was excessive United States engagement in arms control. If (as he assumed) the Soviet economy was already at the limits of its military-industrial potential, restraining the arms race would one-sidedly advantage the Soviet Union since only the United States was capable of further economic mobilization.^[390]

In his own working, Rosefielde hewed closest to the lines of the CIA’s direct costing methodology. He started from the same quantity building blocks as the CIA, supplemented or corrected by those of the CIA, and from base-year prices or costs of i960 that the CIA first adopted, then later abandoned. In addition, Rosefielde claimed a superior methodology of price adjustment for quality change in later years. The result was a measure of defense spending in rubles that approximately matched that of the CIA, but Rosefielde claimed the match was no accident: the CIA estimates were being forced to adapt to realities, but the adjustment was being done dishonestly under a pretense of scholarship. Moreover, the implications of the CIA’s and Rosefielde’s estimates for military stock building in real terms were quite different: in Rosefielde’s view the Soviet Union had substantially overtaken the United States in its military capabilities by the 1980s.

The rest, including Lee, worked directly with Soviet data and did not attempt to build from blocks in physical units. If they estimated defense activity in real terms, it was usually at the final stage of deflating nominal values. Lee, for example, began from the residual of the nominal gross output of the machine-building industry in each year after subtracting reported or estimated net intermediate and final deliveries to other sectors and inventories. He attributed what was left over to military procurement, and this became the keystone of a calculation of the wider military budget. As Table 7.5 indicates, Lee’s result was a range of figures that approximately straddled the central estimates published by the CIA.

Rosefielde also worked with the machine building residual and approved of this method. Against it may be made a general point: working with small residuals left after subtracting one large number from another is always hazardous because small errors in large numbers lead to large errors in small ones. But there was also an important point in favor of it, one that could not have been made at the time. A document in the Soviet archives for 1937 shows that officials understood exactly the danger of publishing information that might permit computation of a heavy-industry residual: it could reveal the value of secret military procurement.^[391]

Beyond Lee and Rosefielde, several others attempted to estimate Soviet defense spending by decoding Soviet official data in different ways. Among these were both academic scholars and staff researchers at the DIA. They generally began from the official military budget figures, which they understood as most likely representing a subtotal of outlays on the pay and subsistence of troops. To cover other military items such as procurement, operations, construction, and RDTE, they then added varying shares of other items taken from published information concerning purportedly civilian outlays on the national economy or the accumulation fund or additions to state reserves. (We will see that this line of enquiry more or less predicted the efforts of Russian insiders in post-Soviet times to reconstruct the Soviet defense budget.) Working on these lines, DIA estimates generally turned out larger than those made by the GIA, but estimates made on similar lines by other Western scholars were sometimes smaller and occasionally much smaller.^[392]

Some estimates of this general type were very elaborate. Most complex was the apparatus developed by Steinberg, who began from the hunch that most Soviet defense production and activity was excluded from the official national accounts.^[393] Steinberg’s framework aimed to account for hidden flows, both real and financial, and to distribute them among producers and users in such a way as to consistently reveal the true scale of Soviet defense.^[394] But the hunch itself remained untested and has not been confirmed since. Unlike those of others in this group, as Table 7.5 shows, Steinberg’s estimates were relatively large and close to those of the CIA.

Cleverest was the method of Peter Wiles, who began from the idea that weapons are a form of capital asset that depreciates rapidly because of technical obsolescence. He called this depreciation the “weapons writeoff.” He argued that the Soviet defense accountants had hidden most of military procurement by reporting it only in net terms, after subtracting the weapons write-off.^[395]

There was a promise in the methods of Wiles and Steinberg that Rose-fielde for one acknowledged at the time.^[396] But the promise remained unfulfilled, and the underlying intuitions have not been substantiated.

While the rivalry was often heated and the range of estimates showed little sign of converging, it is important to note that there was more agreement among the rivals than might have appeared at the time. There was universal acceptance that Soviet official budget figures were deceptive. Alternative estimates in rubles—the currency that is of interest for this chapter—were mostly clustered in a fairly narrow range that was clearly separated from the Soviet official figures of the time. Thus, the acrimony arose more from the mutual questioning of motives than from radically different outcomes.

The range of alternative estimates for 1970 and 1980 can be judged from Table 7.5. For 1970 the mainstream lay somewhere between 42 and 49 billion rubles (10-13 percent of Soviet GNP); only Wiles stood outside (and far below) that range. By 1980 SIPRI had left the mainstream with a greatly reduced estimate, and Lee had gone to work for the DIA, which now set the upper limit. As a result, the mainstream had lifted and widened to between 81 and 107 billion rubles (13-17 percent). An error margin of several percentage points of GNP could not be grounds for complacency. But the problem was of enormous scope, and the uncertainties were intrinsic. For these reasons, demands for greater precision might well have been unreasonable.

Mikhail Gorbachev became the leader of the Soviet Communist Party on 11 March 1985. At that time, the Soviet Union was already in a “precrisis” situation. That is, the country was visibly beset by problems, none of which was critical in that moment; but all of them were growing.^[397] Gorbachev was determined to break out of the “era of stagnation”—a term that he coined to describe the outcome of his predecessors’ attitude of persisting with past policies while waiting for the country’s luck to change.^[398]

Among Gorbachev’s priorities was a comprehensive arms control agreement with the United States. The United States was engaged in a decade-long rearmament and modernization of atomic weapons. Rearmament began during the administration of President Jimmy Carter (1977-1981) and was pressed forward by President Ronald Reagan (1981-1989). Between 1979 and 1986, the GNP share of US defense outlays grew from 4.5 to 6 percent—the biggest slice since the end of the Vietnam War.^[399]

Meanwhile, thinking about arms control on both sides was being complicated by the widening range of new weapons that were entering the balance. In addition to nuclear missiles to be launched against the adversary’s cities from underground silos and underwater submarines, there were now mobile land-based missiles armed with miniaturized nuclear warheads for use against ground forces in Europe, and a US project for space-based antimissile defense. Both sides felt a growing threat.

Gorbachev wanted to persuade the West that the Soviet Union was neither a strategic threat nor a growing one. In public, he announced a unilateral nuclear test moratorium and renewed proposals to limit the deployment of theatre-based mobile nuclear weapons. In private, he had become convinced that the Soviet Union’s longstanding secrecy over the size of its armed forces, and the deception over its military spending, were now damaging its external security.

Meeting with Reagan at Reykjavik, Iceland, in October 1986, Gorbachev felt he had come close to agreement, and the failure made him all the more aware of the longstanding mistrust of Soviet official claims and assurances.^[400] The Soviet government’s initial silence concerning the Chernobyl nuclear disaster a few months before the Reykjavik meeting had done fresh damage to confidence in the Soviet willingness to negotiate frankly over military affairs and to confidence in Gorbachev personally.^[401]

What happened next is told here from the personal papers of Vitalii Kataev, a senior Soviet defense industry manager and arms control specialist. On 21 November 1986, the party Central Committee approved a resolution “On increasing the openness of the military activity of the USSR.” Formally, the initiative came from Vladimir Petrovskii, deputy foreign minister for disarmament (the foreign ministry was led at the time by Eduard Shevardnadze, later president of Georgia). We don’t have the exact words of the decree, but the Kataev papers include a commentary by Oleg Beliakov, Central Committee secretary for the defense industry.^[402] “In the opinion of the USSR Foreign Ministry,” Beliakov wrote, increased openness “would allow the Soviet side to take up an offensive position in questions of military activity in international forums, and not allow speculation that is inimical to us about the lack of sufficient information on these questions.” A handwritten addition, part of it torn away, goes on to describe the then current figure of 19 billion rubles as “unconvincing” from the standpoint of Soviet claims to strategic parity with the United States.

The Politburo commissioned a group of four to report back on ways and means of implementing greater openness. The four were Yurii Mashukov and Valentin Smyslov (deputy heads of Gosplan), Nikolai Yemokhonov (deputy head of the KGB), and Marshal Sergei Akhromeev (chief of the armed forces general staff).^[403] We’ll come across Akhromeev and Mashukov several times more.

Six months passed without progress. In April 1987, Central Committee secretary for the defense industry Beliakov reported the reason for the hold-up: “Gosplan considers it impermissible to publish the factual size of outlays on the defense activity of the USSR, but it does not propose any way out of the situation created as a result.”^[404]

There must be some uncertainty as to the original source of the resistance to disclosure. While Beliakov blamed Gosplan, a note by Kataev blamed the military on the grounds that “traditionally the military determined the level of secrecy in the country.”^[405] Whether or not this was generally true, Kataev’s recollections make clear that on this issue, a chief opponent of disclosure was Marshal Akhromeev, the country’s senior serving military officer.

According to Kataev, Akhromeev based his resistance on several arguments. His starting point was that, if the published figure for Soviet defense outlays was not credible, the available secret figures were little better. Because of omissions, they too understated the real cost of Soviet military activities by a large margin. To publish them would not help; they would raise many more questions than they would answer. Akhromeev feared the public reaction within the Soviet Union:

Who would believe that the USSR had such a small military budget? Among ordinary people, among the Supreme Soviet deputies, questions would immediately arise for the country’s leaders and for the military: how, on such a small budget, was it possible to secure parity with the Americans, whose budget was many times larger? It must follow that the leaders are deceiving the people under the cover of secrecy.^[406]

Akhromeev also feared international repercussions:

A comparison of the budgets of the USSR and USA would prompt the thought that the claims of the USSR’s leadership to stand up to the USA “on equal terms” was a bluff, since the USSR budget was several times smaller than the budget of the USA. At the same time, the USSR’s leadership would be accused of deception.^[407]

To respond to such concerns, the Soviet government would be forced to go into the secret figures in detail and explain exactly why they too fell short:

It would be complicated to answer such questions; one would have to go into the differences in labor and material costs in the USSR compared with abroad, the social channels of redistribution of part of the income in the USSR, and to compare the purchasing power of the ruble and dollar.^[408]

Here, when Akhromeev mentioned “the social channels of redistribution of part of the income in the USSR,” he evidently had the following idea in mind. Traditionally, the Soviet government raised substantial tax revenues from the production and sale of consumer goods to households, while subsidizing production and investment in the heavy and defense industries. To the extent that taxes on consumer goods were used to subsidize other sectors of the economy, relative prices faced by buyers were distorted: civilian consumer goods were priced above their production costs while machinery and weapons were priced below costs. Thus, large parts of the country’s military activities were actually paid out of sums earmarked for government support of the economy, not of the military. This seems to be what Akhromeev had in mind.

The result was that one ruble in the hands of Soviet military agencies had greater purchasing power over state products than a civilian ruble. But there was no comparable distortion of relative prices in the US economy. Using the prices prevailing in each country, therefore, the Soviet economy would appear to support a given level of military activity from a smaller consumer sacrifice than the US economy. To correct the appearance would require detailed comparison of the purchasing power of the two currencies, as Akhromeev remarked.

An aspect of the problem that Akhromeev did not consider is that the understatement of Soviet defense costs could not be corrected just by adding in the cost of direct subsidies to the Soviet defense industries. Accounting for direct subsidies was only the start of the problem. There were also indirect subsidies that lowered the prices paid for military activities. At first sight, for example, bread and meat subsidies did not look as if they had anything to do with defense. But they reduced wage costs for all lines of production, civilian as well as military. Possibly workers in the defense industries had privileged access to subsidized bread and meat, which others would have to buy at much higher prices in the urban markets. Privileged access to food subsidies would then disproportionately lower the cost of weapons. This too would have to be taken into the account. Most likely, the final effect of all the direct-plus-indirect subsidies on the relative costs of military and civilian lines of production could only be ascertained by actually removing them all and observing the results that would ripple through the economy.

Could this be done at all? Periodically, it is true, the Soviet authorities attempted to reset industrial prices in line with costs (this was called “price reform”), with the aim of restoring the loss makers to profitability and eliminating cross-subsidies. Price reforms were costly and arduous to implement, and the cross-subsidies always soon returned.^[409] As of 1987, it was expected that the Soviet economic reforms being promoted by Gorbachev would include a comprehensive reform of state prices, but no reform had yet been undertaken.

Related to the price of Soviet weapons in the internal defense market was the question of its export price. The Soviet Union priced its exports based not on production costs but on what the foreign market would bear. For that reason alone, production costs were a carefully guarded commercial secret. Akhromeev worried that having to publicize the subsidized ruble prices of Soviet weapons in the domestic market would undercut the prices and profits of the Soviet arms trade:

By publishing information about the low production cost of military equipment in the USSR, we will spoil our [international] military-technical cooperation, [where currently] we work at world prices and get a good profit.^[410]

Finally, Akhromeev speculated that the lack of credibility of the ridiculously small figures for defense outlays published at the time gave the Soviet Union a hidden propaganda advantage:

The propagation of myths about large outlays of the USSR’s budget on military purposes and the “Cold War” was advantageous to the USSR’s leaders since it provided a justification (based on “huge” military outlays) of the low standard of living of people in the USSR.^[411]

Despite his long list of concerns arising from the prospect of disclosure, Akhromeev did not insist on secrecy forever. Whether as a tactical retreat or as a strategic repositioning, he pleaded for time:

We are not ready to enter such discussions. Let us come out onto the world level in our [international] economic relations in the 1990s, and then we will publish our military budget and have no more secrets.^[412]

As of April 1987, the situation was one of deadlock. The reformers wanted “no more secrets” now, not in several years’ time. But there was no consensus on what should take the place of secrecy. At this point, Beliakov made a remarkable suggestion:

It appears expedient not to publish the actual sum of outlays and in preparation of the documentation to proceed by way of artificial formulation of a figure for our “outlays on defense activities” that is acceptable for publication.

Given this, [we ought] to examine the actual outlays of the advanced capitalist countries on these uses and compare their outlays per head of population and in terms of percent of gross national product. The figure of our outlays on defense activities (which should include the known outlays on defense—20 billion rubles) in percentage terms should be not worse than the developed capitalist countries.^[413]

The spirit of the proposal was clear enough. Beliakov was playing the joke of the business consultant who, on being asked to solve two plus two, lowers his voice to ask: “How much do you want it to be?” It is ambiguous what Beliakov meant by “not worse.” Most likely, he meant “not less”: everyone understood that the published figure was too small, and a more expedient figure would be larger than before, one that showed the Soviet Union neither underplaying nor overplaying its hand as an equal partner of the United States.

Akhromeev was not alone in his anxieties. In the early summer of 1987, the resistance crystallized. The Kataev papers include a memorandum from July of that year, prepared for signature by seven senior figures: Prime Minister Ryzhkov, Defense Minister Yazov, Foreign Minister Shevardnadze, former ambassador to Washington Dobrynin, Central Committee secretaries Zaikov and Yakovlev, and first deputy KGB chief Bobkov.^[414] All but Bobkov and Dobrynin were Politburo members. This was a heavyweight alliance.

The Seven acknowledged that the published 20-billion-ruble figure lacked credibility. For context, they cited the CIA’s much higher estimates of the Soviet military burden at the time—15 to 17 percent of Soviet GNP.^[415] They warned, however, “Disclosure of the actual scale of overall allocations to the needs of defense of the USSR at prevailing prices will not yield a positive political outcome and can even lead to adverse consequences for the USSR.” Their supporting arguments echoed Akhromeev’s:

If we set out along the road of detailed justification of our defense outlays at current prices, then we will have to enter fruitless discussions about the comparative costs of the basic types of weapons and military equipment, design work, and other items at home and in the West. At the same time, to announce the true prices of domestic military products would create particular difficulties in the sale of our weapons to foreign countries.

For the first time, they set out a concrete plan of action:

In the situation before us, it appears possible ... to start from the proposition that the published defense budget of the USSR expresses [only] the outlays of the USSR ministry of defense on personal maintenance of the Armed Forces, their material-technical provision, military construction, the provision of pensions, and a number of other outlays.

In this secret document, the Seven chose their words with notable caution. “That the published defense budget of the USSR expresses the outlays of the USSR ministry of defense on personal maintenance of the Armed Forces,” et cetera, was termed a “proposition” from which it was “possible to start.” They did not affirm it as a fact. They went on:

Concerning the financing of scientific research and experimentation, and also the procurement of munitions and military equipment, it can be announced that we make the necessary allocations under other headings of the USSR state budget; comparison of this part of the outlays of the USSR with the outlays of other countries on analogous uses has no sense because of fundamental differences in the structure of the armed forces, prices of weaponry, and the mechanism of price formation.

Public comparison of the Soviet and American military budgets, they concluded, “will be possible in 1989 to 1990 after the Soviet Union has implemented the proposed radical reform of price formation in the course of restructuring the administration of the economy,” that is, in the next “two to three years.” To announce this position would “allow us in some degree to lift the pressure which has existed for a long time on the question of the size of the USSR military budget.”

In the short term, therefore, Akhromeev got his way. On 6 August 1987, the Central Committee adopted the position that the defense budget would be declassified, not immediately, but within the next two to three years.^[416] This would buy time for the experts to come up with a number that would be both credible and meet the foreign objectives of Soviet diplomacy.

The outcome was revealed to the world in several stages. Speaking at the Geneva disarmament conference on 15 August 1987, Gorbachev affirmed that the Soviet Union was committed to greater military openness—in principle. The following day, Deputy Foreign Minister Petrovskii told the conference that the published Soviet military budget, now 20.2 billion rubles, was accounted for largely by the maintenance of the troops and so forth, and so was only a fractional component of the true total. (This appeared to vindicate the working assumptions underlying a substantial subset of the Western estimates of the Cold War era, discussed above.)

Finally, after three more weeks, Gorbachev announced that transparent and comparable figures for the Soviet defense budget would be made available within “two to three years.” This completed the implementation of the Central Committee decision of 6 August.

If the decision is taken at face value, then the following months should have been full of activity. To expose the thousand hidden cross-subsidies that supported military production and procurement to a first or second approximation would have required a major accounting exercise, but there is no evidence that such an effort was commissioned or undertaken. Alternatively, as the Seven envisaged, it required the “radical reform of price formation” that was expected to accompany the transition to a market economy, so that the results could be observed. There is no sign of this either. On the contrary, it is a fact that no price reform took place, the reason being that all serious economic reform measures were blocked by resistance and indecision.^[417]

While nothing was being done, the balance of elite opinion was shifting. In the autumn of 1988, a party journal published a long article entitled “From the cult of secrecy to an information culture.”^[418] The author, Vladimir Rubanov, was a senior KGB officer serving as head of a research unit for “questions of the defense of state secrets and of legal regimes of assurance of state security.”^[419] Although much of it was couched in rather general and abstract terms, the article amounted to a wholesale rejection of the existing regime of secrecy. It called for public discussion of the proper limits to be placed on government information and for legal regulation of security classification based on a “presumption of nonsecrecy.”[420] It introduced the Soviet reader to the US system of information security oversight, including its acknowledged defects. It highlighted both the economic costs of excessive secrecy and the losses to Soviet diplomacy of the inability to respond with facts to Western discussion of Soviet ends and means.

That was in theory. In practice the clock was ticking. There was an external deadline: Gorbachev expected to visit London in April 1989 and wanted to know what to tell Prime Minister Margaret Thatcher about Soviet military spending and force levels in Europe.^[421] In the event, the foreign deadline did not bind. Gorbachev spoke in London about Soviet force levels but made no announcement about military spending. There was also an internal deadline. Under new constitutional arrangements, oversight of the military budget was to be delegated to a parliamentary committee of the USSR Supreme Soviet. For review of the 1990 budget, details would have to be laid before the committee in the autumn of 1989.^[422]

A new figure began to appear in the documents, although not yet in public: 77.3 billion rubles for defense in 1989, falling to 71 billion rubles in the budget for the following year.^[423] The preliminary budget account for 1989 included a figure of 20.2 billion, but this was now listed against military maintenance, construction, and pensions only. Munitions and equipment, RDTE, and “other” items accounted for the other 50 billion.

The Kataev papers place the new figures for 1989 in the context of US outlays on national defense in the same year, as shown in Table 7.6. On the new basis, Soviet defense outlays took up 8.4 percent of Soviet GDP, while the equivalent American burden was only 5.9 percent. As can be seen, the new numbers met Beliakov’s “expediency” criterion for disclosure: they suggested that the Soviet Union, with a military establishment roughly equal to that of the United States, but with fewer national resources to support them, bore a heavier military burden relative to its GDP.

To judge from the papers, there was still a residue of private defensiveness about going public. The points were made that the forecast figures

would show defense outlays falling through 1990 and 1991, and that this would underline the Soviet commitment to peace. In case anyone in the Politburo was inclined to backslide, it was emphasized that not to disclose the figures would mean reneging on the public promises made in 1987.^[424]

At all events, there was no going back. In a speech to the USSR Supreme Soviet on 30 May 1989, Gorbachev finally released new figures to the public: in 1989, Soviet defense outlays would be not 20.2 billion rubles but 77.3 billion.^[425] The long wait was over.

Or was it? As soon as they were released, the new figures evoked intense curiosity. Although many times the sums previously acknowledged, 77.3 billion rubles was still far short of most Western estimates. CIA analysts concluded privately that the true figure for 1989 remained in the range 130 to 160 billion rubles.^[426] Describing the 77.3-billion-ruble figure as “a best effort for the present,” they recorded the suspicion that the Soviet administration itself still did not know the full extent of continuing defense subsidies and was aware of its own ignorance.

Controversy was fueled by subsequent statements made by top Soviet leaders that put the Soviet military burden at 16,18, or even 20 percent of national income.^[427] These statements were typically vague about whether GNP or the net material product (at least one-third smaller) was the intended denominator. But their claims were so far above the 8 to 9 percent of GNP implied by the ruble figures announced in 1989 that imprecise definitions did not much matter.

Hidden Soviet defense industry costs were the focus of a new investigation led by Dmitri Steinberg. In Moscow in the winter of 1990/91, he consulted with statistical officials and former defense industry managers with the aim of identifying budgetary items not counted under the Ministry of Defense allocation and other costs imposed on the civilian economy by military activities and military procurement. He concluded that an upper limit on Soviet defense outlays in 1989 was 133 billion rubles.^[428] This figure was within the range of the CIA estimate for the same year, although at the lower end.

To summarize, Akhromeev correctly predicted what could go wrong. Gorbachev’s disclosure of Soviet defense costs in 1999 was an important step. But it did not build trust or allay suspicions that the Soviet leadership was still not being entirely frank. It created new puzzles and invited further enquiries. From our perspective, most important was the growing sense that what was being concealed by the Soviet leaders was not a truth that only they knew, but that no one knew the truth.

At the time this was admitted, more or less plainly, by Akhromeev. Speaking on Russian television in October 1989, he responded to an interviewer:

You are young and you want it all at once on a plate. Just think how we established this budget of 77.3 billion rubles. After all, we did not have it as such. Previously the armed forces were only interested in what they were given for upkeep, in accordance with a government decision. As far as research and development was concerned, this came under the ministries of defense industry sectors. Nor did we pay for series production deliveries. First, we had to bring this together.^[429]

Here was an important difference from the situation in the early 1930s. Then, although the public was deceived about the factual total of defense outlays, a handful of people always knew the truth. How many people knew the truth in Brezhnev’s time? According to Yurii Mashukov, the answer was “a limited circle of persons (the leadership of USSR Gosplan and not even all Politburo members).”^[430] According to a Russian military source the answer was four (the party general secretary, the prime minister, the defense minister, and the chief of the general staff).^[431] According to Gorbachev, the answer was “two or three.”^[432] But Akhromeev implied that the real answer was none. As some Western observers now concluded, maybe the decades of deception had caused even the Kremlin to lose sight of the true figure.^[433]

It is a fact that, in 1987, when Gorbachev finally admitted that the old figure was deceptive, he could not immediately find a new one to replace it. He promised that the new figure would be available in “two to three years.” That alone suggests that a substantial research project was anticipated. It does not follow that the necessary work was actually undertaken, however.

The process of disclosure awaited one more step: to trace the deception back to its origin. In 1999, a new historical series for Soviet defense outlays was published, adjusted retrospectively to the 1989 basis. This provided an opportunity to learn more about hidden Soviet defense spending and the reliability of the new figures.

The authors of the new historical series were former Soviet defense insiders Yurii Mashukov and Yevgenii Glubokov. In Soviet times, Mashukov had been head of two key bodies, the state military-industrial commission (VPK, for a few months in 1991) and the national planning board (Gosplan, from 1988 to 1991). Glubokov had been a specialist member of the VPK. Mashukov, moreover, was one of the Seven who had put together the compromise on disclosure in 1987. They wrote with authority, therefore.

The Masliukov-Glubokov findings are shown in Table 7.7. The earliest year reported was i960. In that year, as Table 7.2 showed previously, the deceptive official budget claimed 9.3 billion rubles of defense outlays. Corrected to the 1989 basis, this should have been 15.3 billion rubles. If so, hidden defense spending already stood at 6 billion rubles in 1960. After that, the table suggests, hidden spending grew rapidly—11 billion by 1970, 30 billion by 1980, and 57 billion by 1989.

But was correction to the 1989 basis sufficient? In percent of Soviet GNP, Figure 7.1 shows the tremendous gap between the deliberately deceptive figures for Soviet defense outlays before 1989 and most Western table 7.7. The new “official” history of Soviet defense outlays, 1960 to 1990 (selected years), as published by Mashukov and Glubokov on 1989 basis

Source: Masliukov and Glubokov, “Planirovanie i finansirovanie voennoi promyshlennosti,” 105. Percentages of GNP are as published in the source, except for figures in brackets, which are corrected by the author. The basis for correction is the GNP series shown below (in billion rubles):

Again, GNP figures are as published in the source, except for figures in brackets, which are corrected by the author as follows. For 1980, the figure of 661.9 billion rubles reported by Masliukov and Glubokov and used in their calculation of the defense burden in that year is most likely a typographic error; the figure of 619 billion is a better fit in the series and was published twice without revision in Goskomstat, Narodnoe khoziaistvo SSSR v 1989 g., 6, and Narodnoe khoziaistvo SSSR v 1990 g., 5. For 1988 and 1989, the figures reported by Masliukov and Glubokov appeared in Goskomstat, Narodnoe khoziaistvo SSSR v 1989 g., 6, but were revised the following year in Narodnoe khoziaistvo SSSR v 1990 g., 5. As for 1990,1 prefer the figure from Goskomstat, Narodnoe khoziaistvo SSSR v 1990 g.,5, for consistency. figure 7.1. The Soviet military burden, 1960-1990: Alternative views of defense spending in rubles, percent of GNP

Sources: CIA estimates in rubles, percent of GNP, are reported by Firth and Noren, Soviet Defense Spending, 129-30. M&G (Masliukov and Glubokov) corrected as Table 7.7. Budget pre-1989 figures are as Table 7.2, divided by GNP corrected as in the note to Table 7.7.

estimates, represented here by the CIA. Into this gap now stepped the new historical series provided by Mashukov and Glubokov. The new series tracked the CIA estimates to some extent. For example, both series now showed the Soviet military burden peaked before declining at the end of the 1980s. But still, if (and, of course, only if) the CIA set the standard of credibility, then Masliukov and Glubokov had closed less the half the gap.

In presenting these figures, Mashukov and Glubokov discussed when and why the deception began, the methodology they used to reconstruct the 1989-basis figures, and why their new series should be preferred to others. First, when and why did the deception begin? The authors wrote:

In the postwar period the USSR continually posed the issue of cutting the defense budgets of all countries so as to facilitate substantial disarmament or a lower level of confrontation. This idea was already expressed at the first session of the United Nations General Assembly and further specific proposals were put forward.

In 1963 the USSR unilaterally cut [its] allocations to the armed forces to show goodwill and the USA also made some cuts, but the West did not support the 1964 proposal to cut military budgets by 10 to 15 percent. The USSR made proposals to cut military allocations successively in 1973,1976, 1978,1980, 1982, 1983, 1984, and 1987. At the same time, the necessity of providing for strategic parity in the arms race that was imposed on us, and of improvement of the technical level of armament and military equipment, demanded further development of the defense complex and growth of the volume of finance. Therefore, defense spending in real terms rose and peaked in 1988-89. In 1989 it amounted to 77,294.2 million rubles (16.1 percent of the overall budget of the country).^[434]

On this reading, the motivation for concealment in the 1960s was the same as in the 1930s. At a time when the Soviet Union was determined to rearm, it wanted to encourage others to show restraint and, if possible, bind them into mutually agreed limitations on weaponry. To do both openly would appear inconsistent. Disclosure of the real dimensions of Soviet defense outlays would undercut the Soviet foreign-policy position on arms reductions. The only way to square these objectives was to pursue rearmament in secret, and this required deceptive budget figures.

Although they did not say it explicitly, Mashukov and Glubokov left the reader to suppose that an important moment in the emergence of the budgetary deception was 1963, the year in which the Soviet Union (with the United States and the United Kingdom) signed a treaty banning atmospheric and underwater nuclear weapon tests. Closer inspection suggests that this could not have been the start. Table 7.7 shows that that large-scale deception was already taking place in i960: it did not begin in 1963. So, it began earlier, but we still do not know exactly when.

Second, the methodology of reconstruction. Mashukov and Glubokov indicated that deception took the form of reporting one element of defense costs, the outlays on the maintenance of the troops, military construction, and pensions, as if it were the whole. To reconstruct the whole, it was necessary to look for the missing outlays. Their location in the budget accounts is described as follows:

The finance of [military] RDTE and provision for procurement of armament and military equipment was conducted, until 1989, under the national economic plans confirmed annually by decrees of the party Central Committee and Soviet government and budgetary allocations assigned to the defense ministries for RDTE, and [the finance] was entered under the item of outlays on the national economy of the country.^[435]

Taking these words at face value, Mashukov and Glubokov started from the published defense outlays and then added direct budgetary support of military research and equipment (“the finance of RDTE and provision for procurement”). As discussed above, this cannot fully account for all forms of support of the defense sector, which benefited not only from direct subsidies but also from indirect subsidies at earlier stages of the supply chain. Of course, it is possible that Masliukov and Glubokov simplified the explanation of their work for the general reader, but the lack of mention of indirect subsidies is likely to mean that this part of the work was not done or even considered.

A related problem is to understand how the government was able to phase in the start of the deception. Mashukov and Glubokov maintained that the deception mainly took the form of reassigning defense outlays on procurement and RDTE from the reported defense total to other budget headings, and that this began in the early 1960s. But these missing items were already large relative to military maintenance in the 1930s, let alone the 1950s. If there existed some end point of budgetary truthfulness in the late fifties, it is not easy to understand how the items to be concealed could have been subtracted from the published series without causing a noticeable break.

Finally, which series should be preferred? Masliukov and Glubokov noted that their historical series on the 1989 basis, like the 1989 defense budget itself, fell consistently below those of the CIA (among others), but they maintained that this was entirely defensible: the CIA was wrong.

To support their case, they offered two arguments. First, the only way to find a higher GNP share of Soviet defense was to include items from the grey zone adjacent to defense activity. For 1989 this could add another 1.4 percent of GNP to the 8.6 percent already found, making 10 percent in total.^[436] This was still far short of the CIA estimate for the same year. In any case, they pointed out, grey outlays would not be counted in the defense budget proper “in the USSR, as in other countries.” In other words, these items were not in dispute in the first place, so they could not contribute to a reconciliation.

What else could account for the gap? Masliukov and Glubokov then accused the CIA of adopting a methodology that unreasonably inflated its estimate of the Soviet military burden. This (they claimed) was the use of American salaries to value the services of Soviet military personnel and of American prices to value Soviet weapons, hugely overstating their costs in rubles.^[437] As already mentioned, dollar valuation was one stage of the CIA’s “building block” methodology. The aim of that stage was to compare Soviet and American military spending at comparable prices. But when the aim was to value the Soviet defense burden in terms of Soviet GNP, the next stage of the CIA methodology converted dollars to rubles at purchasing power parity. If done properly, the conversion eliminated the bias of which Masliukov and Glubokov complained. In short, Masliukov and Glubokov misrepresented the CIA methodology and were not able to account for the gap otherwise.

Summing up, the credibility of the historical series introduced by Masliukov and Glubokov pivots on two things: the revised defense budget total announced by Gorbachev in 1989 and the methodology that worked the defense budget totals backward from 1989 to i960. The documentary evidence suggests strongly that Gorbachev’s 1989 figure was crafted for political expediency, not accounting veracity. It did not fully close the statistical gap or the credibility gap. Working backward from 1989, Glubokov and Masliukov improved on the deceptive figures of the past, but that was a low bar. At best, they made a first approximation to a more convincing series. As for what remains to be done, most obvious is a correction for the indirect subsidies of the defense sector arising from systematically distorted Soviet prices. Is that omission large enough to explain the remaining gap? We still do not know.

Finally, through the decades of deception of the public, did the top Soviet leaders (if not the full Politburo) retain private access to a secret figure that gave them meaningful or substantial control over defense outlays? The evidence-based answer must be no. For, if such a figure existed anywhere, it is hard or impossible to explain why months and years of specialist research were required to reconstruct it, not in the disorderly 1990s that followed the Soviet collapse, but before then, in 1988 and 1989, when the Soviet regime was still in place and exactly that figure was urgently required.

A British government report (concerning intelligence on weapons of mass destruction in the Iraq war) offers a conventional distinction between intelligence secrets and intelligence mysteries:

In principle, intelligence can be expected to uncover secrets. The enemy’s order of battle may not be known, but it is knowable. The enemy’s intentions may not be known, but they too are knowable. But mysteries are essentially unknowable: what a leader truly believes, or what his reaction would be in certain circumstances, cannot be known, but can only be judged.^[438]

For much of the Cold War, Western intelligence worked on the assumption that Soviet defense costs were a secret. With hindsight, they look more like a mystery. When the time came to disclose them, no one could lay their hands on a figure. Three years went by before a figure could be identified for disclosure. At best, one may believe what was said or implied by Akhromeev and Mashukov. On that basis, the figure that was disclosed in 1989 was an incomplete approximation to the reality. At worst, placing more weight on Beliakov’s call for an “artificial formulation,” one might conclude that the 1989 figure was concocted out of thin air. Either way, even if Soviet defense costs were intrinsically knowable, for much of the Cold War they were practically unknowable, which must come close.

Introducing the words quoted above, the Butler report noted: “A hidden limitation of intelligence is its inability to transform a mystery into a secret.” Conversely, our story suggests how to turn a secret into a mystery. In the first years after World War II, Soviet defense outlays were not secret, and the published figure was not deceptive. At some point a decision was made to allow the published figure to diverge from the factual total, and the factual total became a secret. To maintain the secret, the flows of funds into the defense sector were covered by mistitled budget items and redirected by hidden subsidies. Over time, the underground river widened, and its secret tributaries multiplied. The cover was originally designed to hide the true scale of the defense sector from foreign eyes, but it worked so well that eventually the Politburo itself lost sight of it. The secret became a mystery.

Why did the Soviet leaders, who concentrated so many powers into their own hands, allow this to happen? One explanation is simple and plausible: it suited the Soviet military not to have to account for the full sum of defense costs. Akhromeev came close to admitting this in his television statement, already quoted: “Previously the armed forces were only interested in what they were given for upkeep, in accordance with a government decision. As far as research and development was concerned, this came under the ministries of defense industry sectors. Nor did we pay for series production deliveries.” In other words, in sharp contrast to the American military from the 1960s onward, the Soviet military was spared the need to count its costs, other than for the upkeep of the troops. Nor did it have to justify the costs for the simple reason that nobody knew what they were.

It is sometimes asked if the Soviet Union collapsed because of its military burden. A moment’s thought suggests that this cannot have been the single cause. The Soviet Union collapsed at the end of 1991. At this time, having already peaked, the military burden was declining. If a heavy military burden could bring about economic collapse without help from other factors, then the Soviet Union should have collapsed in 1942 or 1943 when the military burden was several times higher than in 1987 or 1988.^[439] In fact, the Soviet collapse had many causes, including economic, social, political, and moral factors. The military burden was there in the background, but it was never decisive.

What can be affirmed is that the mystery surrounding the military burden indicates the declining effectiveness of Soviet rule. Stalin, for whom divide-and-rule was a first principle of governance, would never have tolerated a situation where the military could make common cause with the leaders of the defense industries to freely spend the government’s money.^[440] During the time of Khrushchev and Brezhnev, the military acquired an unprecedented capacity to exploit secrecy in their own interest. In that sense the “golden age” of the Soviet defense complex (as the historian Julian Cooper called the Brezhnev period) was the leaden age of communist rule.^[441]

Eventually, secrecy over defense costs came at the expense of state capacity. As Gorbachev recognized, secrecy had deprived the Soviet state of the capacity to do business with its foreign adversaries—to commit to arms limitations. It also cost the Soviet state its capacity to manage government spending and limit its burden on the economy.

Stepping back, we can see that the secrecy over defense costs was one aspect of a more general phenomenon. Soviet secretiveness gave rise to an uninformed elite. In principle, one person had the right to know everything: the party leader. Below this level, even within the Politburo, all knowledge was compartmentalized by need-to-know. One agency, the KGB, had the responsibility to manage the compartments of secrecy. This was bad enough. The result of need-to-know was that most members of the late Soviet elite were brought up on myths about the history of their own country and of their own party, about how the world worked, and about the Soviet Union’s place in the world. The few that knew better had no right to share their knowledge more widely and risked punishment if they tried to do so without authorization. Most were never exposed to serious debate involving principles of logic and evidence. It is true that many products of Western scholarly research and debate were imported and translated into Russian, but their availability was strictly limited to readers whose loyalty was proven, so that they were already known to be willing to endorse the conclusions previously approved by the party.^[442]

All knowledge is one. This is why our institutions of higher learning are called universities, not multiversities. While universities generally find it convenient to compartmentalize their teaching and research into disciplines that lend their titles to departments and faculties of economics, history, or chemistry, a good deal of what passes for the advance of understanding relies on making connections across those boundaries. When knowledge is compartmentalized by secrecy, those kinds of advance are suppressed. In the early 1980s, the future of the Soviet Union had become an inherently multidisciplinary problem, involving economics, ecology, sociology, political science, and moral philosophy (at least). Few were used to thinking in all those disciplines at once, and those that were inclined to try were forced into silence or into exile. In that context it is unsurprising that, of the many plans for institutional reform put before Soviet society over the next few years, not one was realized.

Could less secrecy have saved the Soviet Union or communist rule? That might be a counterfactual too far. Before Gorbachev, all Soviet leaders saw comprehensive secrecy as fundamental to preservation of their regime. Mikhail Gorbachev, the last Soviet leader, seems to have shown that they were right: when he became a late convert to the idea of an open society, the Soviet regime lost its grip on society and fell apart.