Secret Leviathan

2. THE SECRECY/CAPACITY TRADEOFF

The Bolsheviks began to build their state from near zero. World War I had weakened Russia’s old monarchical regime and its governing institutions, and the Bolshevik insurrection completed their collapse. Now the Bolsheviks had to reestablish the minimum duties of any state—external defense, internal security, the promulgation of laws, and the policing of crime. Restoring these minimum functions received the Bolsheviks’ first attention. As the historian Lars Lih noted, the communist experiment had “many long-term destructive consequences,” but history should recognize the communists’ achievement of building a “serviceable state apparatus out of nothing” despite their inexperience.[78]

But the Bolsheviks’ goal was not to restore the state as it had been. They aimed to build a new state that was far more powerful than Russia had ever seen, with broader scope and more sweeping authority than other states of the time.

It took the Bolsheviks no more than twenty years to achieve this goal. Within that time the Soviet state became virtually the sole legal owner of land, industrial capital, and transport infrastructure, public utilities and services, urban housing, educational and health facilities, public media, news and entertainment, and virtually the sole employer of nonagricultural labor. It became the monopoly buyer of farmers’ surpluses of major foodstuffs and of nearly all nonagricultural goods and services, which it resold to the population at controlled prices. It censored the press and communications. It kept the public and private lives of the citizens under continuous surveillance. As well as strictly policing the external borders, it also controlled where broad categories of people could live and move within their own country.^[79]

As the powers of the Soviet state expanded, the rights of citizens were restricted by laws that created many novel offences. These laws extended the powers of the agencies of Soviet internal security and law enforcement. As a result, while policing is among the functions of any state, the police functions of the Soviet state became much more sweeping and intrusive than was normal elsewhere. That kind of society is sometimes called a “police state.”^[80]

The growing operational demands that the party leaders placed on the Soviet state continually challenged its capabilities. One result was continuous policy cycles in all aspects of Soviet life, from the economy to criminal justice to culture. These cycles began with aggressive plans and campaigns that designated new priorities and imposed new controls. The cycle ran from initial successes and touted achievements to overreach, disorder, and retrenchment, followed eventually by normalization. Whether the new normal involved more or less centralized regulation than the old one depended on the historical phase of the Soviet system.

The functions and powers of the Soviet police state were more comprehensive than in other countries, but this was not all. They were also more secret. The enforcement of secrecy was another challenge to the capabilities of the Soviet state. This chapter considers how secrecy could add to these capabilities, how it could also detract from them, and what was the net effect.

Wherever they took power, communist rulers pursued two objectives: to monopolize state power and to expand it. To monopolize power, they repressed opposition and silenced unauthorized criticism. To expand power in all its “hard” and “soft” dimensions, they invested in the military, economic, and cultural capabilities of the state.

This is a simplified interpretation of the goals of communist rule. Like all simplifications, it might not be useful for all purposes.^[81] At first sight it seems to exclude the influence of communist ideas (officially defined by “Marxism-Leninism”) about raising the condition of the workers and modernizing society. In fact, both Marx and Lenin are represented by the goals of monopolizing and expanding state power, although in somewhat reduced form. From Marx, the communists inherited an antipathy to private property in general and specifically to private property rights as the legitimate basis of decentralized social action. As Marx told the story of “primitive” accumulation, the capitalist class had exploited state power to help them seize the means of production from the producers. In the workers’ hands, as the Bolsheviks saw it, state power became the agent that would now expropriate the expropriators (or “loot the looters” in Lenin’s more colorful phrase).^[82] For this reason, the Bolsheviks laid tremendous weight upon seizing state power, on excluding all other influences and interests from its direction, and on developing it to the utmost.

From Lenin came the Bolsheviks’ view of global society as an arena of conflict between capitalism and socialism, where hostile classes and social systems contended for the prize of domination. Without an exclusive hold on state power, the Bolsheviks would be quickly dislodged by the internal class enemy. But to hold power within Russia was not enough. In Lenin’s words of 1917, Russia must “either perish or overtake and outstrip the advanced countries.”^[83] The workers’ state must be powerful enough to dominate its neighbors, or the Bolsheviks would be crushed by the external enemy.

The communists found secrecy useful in working toward both their twin goals. And secrecy did contribute to both the monopolization and expansion of state power—but not at the same rate, and only up to a point.

The contribution of secrecy to the ruling party’s monopoly of power does not require detailed explanation. Secrecy and censorship allowed the communist leaders to frame public discourse as they wished by excluding competing voices, values, visions, and the knowledge that might have lent credibility to the alternatives. The same secrecy and censorship also kept those with private access to alternative ideas and facts isolated from each other. Those who felt they had an alternative for Russia were prevented from finding each other and from building networks, developing strategies, and summoning others to collective action to challenge the regime. These considerations made secrecy the key to the security of the ruling party’s hold on power.

How might secrecy promote the expansion of the state’s powers—as distinct from state security? The distinction may seem like a thin one at first sight. But the experience of liberal democracies shows the difference: they have found uses for secrecy that protect the state, but without shielding the ruling party from public criticism, political competition, and frequent elections with unpredictable outcomes. Secrecy that is limited in this sense is widely regarded as legitimate in that the limits are regulated by law and can be appealed in the courts. Limited secrecy can then be thought of as something like a government prerogative: just as law-abiding citizens may close their curtains against nosy neighbors, rulers that govern by consent are entitled to conceal information that private or foreign interests might exploit to the detriment of the community.^[84]

Military secrecy and budget secrecy are examples. If the outsider is a foreign power, the government holds its war plans in secret. The gain to the community is to make the foreign adversary uncertain of its best move. Alternatively, if the outsider is a special-interest group that seeks preferential treatment, the government shields its budgetary processes by secrecy. The gain to the community comes from closing off the lobbyists’ inside track. In both cases, secrecy gives the state greater autonomy, making it more capable of choosing its own path. In both cases, it is true, secrecy also somewhat limits the rights of citizens to take part in formulating policies or to know what state officials are up to when they are being paid by the public. When there is secrecy, the voters must trust the government to do the right thing. Only afterward can they hold it accountable. For liberal democracies, therefore, some secrecy may be necessary, but it is still a necessary evil.

To summarize, secrecy has distinct uses. It can be used to enhance the capabilities of the state. It can also be used to protect the incumbent ruler. It is not hard to see why a party bent on holding and expanding state power would exploit secrecy for both reasons. Hence—Secret Leviathan.

How far can secrecy be taken? For the sake of regime security, the Bolsheviks took secrecy far beyond the limits encountered in liberal democracies. They extended secrecy to all decision making, to all information that might make them accountable, or that risked enabling or empowering civic activism independent of the Communist Party.

Why did they do this? The extension of secrecy was never an explicit goal of communist rule, nor did it have any foundation in the ideals of socialism that they inherited. Rather, they did it because they could. More precisely, the underground conspirators who seized state power in 1917 were already habituated to secret ways of working and coordinating their work. Then, their monopolization of state power and the ensuing centralization of economic, social, and cultural life under state management gave them unexpected opportunities to extend conspirative practices throughout the spheres that they controlled. By trial and error, they learned to use these opportunities to the fullest extent.

This book will show that, in doing so, the Soviet rulers encountered the increasing costs of secrecy. Ratcheting up the degree of secrecy and restricting the availability of information gave rise to both benefits and costs. The benefits were increased security of the regime and increased autonomy of the state. But the costs were manifold and were also increasing. Moreover, the costs fell exclusively on the state’s capabilities. The party’s hold on the state became ever tighter, while the state’s capabilities declined.

The rising costs were both direct and indirect. Directly, secrecy diverted the state’s resources and efforts away from other priorities to the handling and storing of secrets and the control of public discourse. Indirectly, secrecy spread fear and suspicion in the state and mistrust in society, lowered the quality of government information, and made some contribution to the eventual collapse of Soviet rule.

In short, secrecy made the ruling party more secure by helping to scatter and silence its opponents. It also added to the capacity of the Soviet state—up to a point. Beyond that point, what additional secrecy gave to the rulers with one hand it took away from the state with the other. As a result, the capacity of the state was increasingly compromised.

The concept of “state capacity” has given rise to an extensive literature at the interface of political science, economics, and history. This literature is of great interest, but a particular problem is that it has not paid close attention to the experience of twentieth-century communism. In the next two sections I will discuss what is known about state capacity. After that, I will illustrate how twentieth-century communists were exceptional state builders.

There is no single definition of state capacity. Some definitions are descriptive, while others are normative. Descriptive definitions list various attributes of the state that can help it to get things done. If the state is able to get more of a thing done, or get more things done, then it has greater capacity. For example, political scientists Hanson and Sigman define state capacity as the ability of the state to “coerce, tax, and administer.”^[85] Economic historians Johnson and Koyama propose the ability to “collect taxes, enforce law and order, and provide public goods.”^[86] As short lists these are descriptively useful, although (we will see) limited in what they cover.

Normative definitions focus on the capacity of the state to maximize an assumed objective function. Economists Besley and Persson, for example, define state capacity as “the institutional capability of the state to carry out various policies that deliver benefits and services to households and firms.”^[87] This presumes that the government’s intention is to serve society in a benign or inclusive way, which might not be the case. More neutrally the economic historian Mark Dincecco defines state capacity as “the state’s ability to accomplish its intended policy actions—economic, fiscal, and otherwise.”^[88] But this still measures capacity relative to intentions. On that basis a highly interventionist state would have no more capacity than a state that leaves nearly everything to private initiative, provided only that both succeed in the tasks they set themselves. In this book I have in mind the capacity of the state to get things done, measured on a scale that does not depend on government intentions, benevolent or otherwise. Therefore, I will focus on state capacity in its descriptive sense.

In that spirit, what is the scope of things a state can do that require capacity? The short lists provided by the literature such as “coerce, tax, and administer,” or “collect taxes, enforce law and order, and provide public goods,” are a good start. Probably what is included here does not require too much explanation. National defense and the enforcement of public laws and private contracts are textbook cases of the public goods that can be efficiently supplied by government. No government can be effective without administrative capacity, a requirement of multiple objectives that can be achieved only by subdividing tasks across a multifunctional bureaucracy and delegating them from higher to lower levels. As for fiscal capacity, it supplies the funding necessary for the other aspects of state capacity.^[89]

Is more detail necessary? In the interest of parsimony, the literature often singles out fiscal capacity not just for itself but as an indicator that can stand for state capacity generally. Fiscal capacity has the advantage of being measurable in government accounts over long periods of time and across many countries. Relying on fiscal capacity to proxy for state capacity generally is not unreasonable, given that fiscal capacity is generally complementary to some other aspects of state capacity, which, therefore, should all tend to be correlated with each other.^[90] The larger the number of diverse aspects of state capacity we need to consider, however, the less we should rely on fiscal capacity to represent them all, since the chance that at least one other significant aspect is a substitute rather than a complement is likely to increase as the scope of state capacity widens.

The scope of demands that communist rulers placed on state capacity was unusually wide, and any description of state capacity under communism demands attention to aspects that fail to make the conventional shortlists or, even if shortlisted, are rarely treated in detail. Think for example of human capacity in the sense of competent, multiskilled, knowledgeable, noncorrupt personnel, selected and motivated to pursue the state’s objectives.^[91] This might be a factor in the capacity of any state, but it was particularly important for communist states, simply because their leaders expected the state to do so much more than in other societies.

Something that communist leaders particularly demanded of state personnel was to exercise ownership and control rights over most productive capacity. Productive capacity under state ownership is completely overlooked in the state capacity literature. Indeed, it is not currently measured in any major cross-country historical dataset (as discussed in the appendix to this chapter).

A related aspect, administrative capacity, although found in the Hanson-Sigman short list, is seldom defined or investigated.^[92] Building administrative capacity was a core project of communist state builders. None of their goals could be achieved without the subdivision and delegation of tasks. In turn, to build administrative capacity posed difficult tradeoffs between the scope and specialization of executive agencies.^[93]

A feature of administration to which communist leaders gave especially close attention was the control function, as in the military sense of “command and control.” When commands were issued to officials at lower levels, how would higher officials know whether the commands were successfully executed?^[94] In economically decentralized societies, much of the economic control function is given over to markets, prices, and profits. In command economies, the control function of prices was absent or weakened, so administrative control took its place and made very significant demands on state capacity.

Arising from this, but separate from it, was the capacity of the state to diagnose the level responsible for policy failures as they occurred: a policy might fail because of lack of effort or competence at lower levels, or because the policy itself was mistaken, in which case the fault lay above. In turn, this suggests the need for capacity to learn and correct errors. In an uncertain and fast-changing world, policy makers make mistakes. Decisions have unintended consequences. The multiplying losses associated with uncorrected errors may eventually grow to represent existential risks. To avoid these, the state needs an open channel for those willing to speak truth to power. When the message gets through, the state ought to be capable of responding by adjusting its policy settings, adapting its instruments, and reordering priorities.^[95] This cannot be taken for granted, because those who seek power rarely wish to be corrected. Open political competition provides a mechanism for learning: the voters can decide whether a policy has failed and eject the authors from office. When political competition is suppressed, error correction must take place through an administrative process—if it is provided for.

A final dimension of state capacity is time. State capacity needs to be durable, because state breakdown invites civil violence and foreign intervention and is associated with persistent development setbacks.^[96] What is known about the design of durable states? One form of constitution that is relatively sustainable is a state that coevolves with civil-society activism, allowing the citizens to participate in government and restrain government by laws.^[97] Conversely, authoritarian directives that superimpose top-heavy controls on an unresponsive society have been shown to be ineffective or self-limiting.^[98] But the evidence of communist states, which have shown surprising resilience (and sometimes fragility) in peace and war, is typically ignored or mentioned only in passing in the studies that are cited here.^[99]

To summarize, state capacity can have many dimensions. Coercive, legal, fiscal, human, productive, administrative, control, and learning capacities all look as if they might have separate implications for the conduct of the state’s business—the more so, the more encompassing is the business of the state. Time is also a dimension of state capacity because an effective state must be durable.

This book investigates the relationship between secrecy and state capacity. When we look over the activities that make up state capacity, where does secrecy belong? On the face of it, secrecy belongs, first of all, to legal capacity. Legal capacity is the ability to create formal rules that allow people to resolve disputes without violence. Secrecy is the set of formal rules that restrict what people are allowed to learn, which limits the scope for people to engage in disputes in the first place. Then again, secrecy is closely related to coercive capacity: coercion restricts people’s choices by threatening them with violence, whereas secrecy restricts their choices by preventing them from learning about other possibilities. Based on examples we have already mentioned, secrecy keeps the adversary guessing. It is easier to coerce an adversary who has no idea when or where the next blow might fall, whether the adversary is a foreign enemy or a domestic opponent. In this sense, secrecy economizes on coercion.

Secrecy also turns out to affect all the other capacities. Think of administrative capacity, which is the ability to subdivide and delegate tasks in an organization. The secrecy involved in compartmentalizing information has the potential to make subordinates more compliant, because it deprives them of the information that could allow them to challenge orders or work around them, or to combine with other subordinates to resist the principal.

In the examples given so far, secrecy appears to augment state capacity. But in some other respects, state capacity might be impeded by secrecy. Think of human capacity, which is made up by the number and quality of public servants. Human capacity is influenced by secrecy because subordinates must be chosen who, in carrying out their tasks, can be trusted to keep the secrets they know and lack interest in the secrets that they do not need to know. It is a good question whether a state will be served best when officials are selected for pliability and lack of curiosity.

And it is straightforward to see how secrecy might impede the state’s capacity for error correction. Learning from mistakes and unintended consequences depends on two things—the free circulation of information and the sense that people are free to draw attention to policy failures—both inhibited by secrecy. A state that repeats mistakes because it cannot learn from them might eventually lose durability too.

To summarize, it seems that secrecy can augment several aspects of state capacity, at least up to a point. But questions arise along the way that suggest limits on the relationship. For example, lack of information that would allow officials or citizens to challenge policies and their implementation will eventually be damaging if it prevents errors from being corrected.

In this chapter I will suggest that the net effects of secrecy on state capacity are nonlinear: a certain amount of secrecy is likely to be capacity enhancing, but as secrecy becomes more intense, the net benefit to state capacities of various kinds begins to wear off and turns into a loss. The subject of this book is then how the Soviet rulers added to the state’s capabilities, what those capabilities amounted to—and how secrecy first added to them and then eroded them.

Does high state capacity promote economic development? To begin with, it can be seen that the wealthy market economies of the Organisation for Economic Co-operation and Development (OECD) all have high state capacity and that many middle-income and nearly all poor countries such as in the Middle East and sub-Saharan Africa do not.^[100] This suggests that a capable state might be productive for economic development.

Also consistent with the evidence is an alternative view, that the state is primarily extractive. Based on the possibility of reverse causation, the economist Bryan Caplan has argued that state capacity building is mostly wasteful, and richer countries acquire higher-capacity states only because they can afford the luxury of waste.^[101] On this view the state is primarily extractive, and greater wealth permits bureaucrats and politicians to extract more rents from society.

Historical data, which make it possible to compare the dynamics of economic development and state building in different countries, generally support a more nuanced conclusion. States that were capable but avoided the excesses of despotism and corruption appear to have fostered early economic development: they complemented the market-economy by suppressing internal and external predators, guaranteeing property rights, and providing public goods.^[102] Providing public goods could be called productive, whereas suppressing rent seeking and defending against external predators are protective functions. In research on these lines, however, it is often difficult to separate the productive contributions from the protective ones. The most convincing evidence relates to the protective functions of the state. Various studies have been able to link both state building and economic development to historical experiences of warfare. This suggests that the stimulus to state building was to seek protection from warfare. Notably, the evidence has been drawn from almost every region and period of history—except that of twentieth-century communism.^[103]

Geloso and Salter, economic historians, argue that protection is the other side of extraction. Richer countries might invest more in state capacity, not because state capacity is productive, but because it is extractive. In global society, countries with weak states can expect to be plundered by predators, and wealthier countries will provide higher value as targets for plunder. Therefore, wealthier countries must have invested more in state capacity, not because it brought prosperity but because it protected prosperity once achieved.^[104]

This view of state capacity would not have surprised the historian Richard Tilly, who famously wrote: “War made the state and states made war.”^[105] It would have been endorsed enthusiastically by the nineteenth-century German economist and advocate of German unification, Friedrich List, who responded to the “English” idea that a wealthy economy is more important than a powerful state as follows:

Power is of more importance than wealth because a nation, by means of power, is enabled not only to open up new productive sources, but to maintain itself in possession of former and of recently acquired wealth, and because the reverse of power—namely, feebleness—leads to the relinquishment of all that we possess, not of acquired wealth alone, but of our powers of production, of our civilisation, of our freedom, nay, even of our national independence, into the hands of those who surpass us in might.^[106]

I have argued that the experience of communist state building has been neglected by social scientists writing about state capacity. How was communist state building different? The evidence shows two things. First, on basic measures, communist states showed extraordinary capacity.^[107] And second, time mattered: communist state capacity was relatively durable.

Figures 2.1 and 2.2 compare the state capacities of the relatively wealthy market economies belonging to the OECD and the somewhat less wealthy economies under communist rule and belonging to Comecon, the Soviet-led Council for Mutual Economic Cooperation. The year is 1980, toward the end of the Cold War, at a time when communist rule was everywhere intact. The figures illustrate two measures of state capacity, the tax share and the state ownership share in the economy. These aspects of state capacity are important in their own right. In addition, at high levels they also imply something about coercive capacity. High tax rates encourage evasion, which must be policed and repressed. High levels of state ownership are likely to have required the forced seizure of private assets in the past and continued repression of private ventures that might weaken public monopolies in the present.

The figures plot these measures against the average real GDP per head of each country. As discussed earlier in this chapter, a country’s state capacity is likely to be increasing in its level of economic development (as noted, the direction of causation remains a matter of debate). Before attributing a country’s higher levels of state capacity to communist rule, therefore, we should check whether greater wealth could be the underlying correlate.

The charts answer this question clearly: communist rule was decisive. On fiscal capacity (Figure 2.1), we see that the Comecon economies had figure 2.1. Communist economies had higher tax shares in 1980, given their level of economic development

Sources and notes: See the Appendix to this chapter and Table 2A.1

generally high tax ratios (the group average is 49 percent of GDP). At first sight this does not look exceptional because a subgroup of market economies in northwestern Europe shows similar tax shares (Austria, Belgium, Denmark, Finland, France, Netherlands, Norway, and Sweden—although many other OECD countries come in below). What the figure demonstrates, however, is that the high tax shares of the Comecon group were exceptional, given their income levels. The Comecon countries were relatively poor in real GDP (6,000 to 8,000 international dollars per head). The subgroup of market economies with similar fiscal capacity were much richer (12,000 to 16,000 international dollars per head). In other words, countries under communist rule had fiscal capacity similar to market economies that were roughly twice as wealthy.

Alternatively, the same chart lets us compare fiscal capacities of the Comecon group with the OECD countries at similar levels of income at the time—Spain and Portugal. By contrast to those countries, the Comecon group show tax shares that were higher by around 20 percentage points of GDP—a very large advantage.

The contrast of the Comecon countries to the OECD countries is even more clear cut in Figure 2.2 (China, not a Comecon member state, is added figure 2.2. Communist economies had higher state ownership around 1980, regardless of their level of economic development

Sources and notes: See the Appendix to this chapter and Table 2A.2. I thank Branko Milanovic for advice on sources and discussion of comparability across countries and systems of national accounts.

to the data). Here the measure is the share of state-owned or public corporations in economic activity in or around 1980. This measure indicates the scope and variety of government activities, over and above the functions of government administration that nearly all states engage in, such as defense, policing, education, and so forth. The differences are stark, and there is no overlap. The Comecon group average for state-owned activities is 92 percent of national income, and for China nearly 80 percent. The OECD group average stands at around 9 percent. The difference cannot be explained away by relative income levels for, if high development were the statistical correlate of high state capacity, it would go the opposite way.

The evidence so far relates to communist state capacity in snapshots, taken at a single moment in time. But time itself is a dimension of state capacity. In addition to showing unusual capabilities, communist states have been capable for longer, both by comparison with other types of authoritarian regime, and also by comparison with some democracies.

When measured against the longstanding liberal democracies of northwestern Europe or North America, communist states might appear fragile or short lived. But this might be the wrong yardstick. Twenty years ago, the political scientist Barbara Geddes wrote:

Since at least the 1950s, many analyses of communist regimes have stressed their inherent dysfunctions and contradictions. When the regimes finally broke down, these dysfunctions were invoked as causes. Yet these political systems had lasted forty years in Eastern Europe and seventy in the Soviet Union.^[108]

None of the communist states that remained has broken down since then. Today they include North Korea (founded in 1948), China (1949), Vietnam (1955), Cuba (1959), and Laos (1975). North Korea matched the seventy-four years of Soviet rule in 2022, and China should do so in 2023.

Multiple datasets show that single-party dictatorships are especially long-lived forms of authoritarian rule—among them, most prominently, communist states.^[109] Single-party regimes are thought to be more durable than others (such as military and personal dictatorships) because they offer ways to institutionalize rent sharing, career advancement, and orderly succession, preventing intra-elite disputes from breaking out into violence.^[110] At the same time not all single-party regimes are the same. Particularly long lived are “revolutionary autocracies” that have their origins in civil and foreign wars. As political scientist Jean Lachapelle and coauthors write:

Revolutionary elites’ efforts to radically transform the existing social and geopolitical order trigger intense domestic and international resistance, often resulting in civil or external war. These military conflicts pose an existential threat to new revolutionary regimes and in some cases, such as Afghanistan and Cambodia, they destroy them. But where regimes survive, violent conflict produces four important legacies: (1) a cohesive ruling elite, (2) a loyal military, (3) a powerful coercive apparatus, and (4) the destruction of rival organizations and alternative centers of power in society. These legacies help to inoculate revolutionary regimes against elite defection, military coups, and mass protest—three principal sources of authoritarian breakdown.^[111]Regime longevity matters. It is not just a signal of political vitality. It is also a precondition of economic modernization. The longer a regime can persist without breaking down amid violence, the longer is the period in which the economy can grow without a setback. Communist countries were able to industrialize their economies and modernize their armies, not only because these were priorities, but because they were able to suppress political conflict for decades, giving their policies time to work.

The shared outlook of these countries’ rulers was a determination to rule indefinitely and to build economic and military power while doing so. Perhaps it can all end tomorrow. The title of Alexei Yurchak’s book about “the last Soviet generation” is a warning to communist rulers from history: Everything Was Forever, Until It Was No More.^[112] But so far, and for some of them, tomorrow has been put off.

To summarize: communist states under stable authoritarian rule have been the most formidable Leviathans of the last hundred years, perhaps of all time. They arose from top-down state building, carried out amid the ruin of interstate and civil wars. They mobilized economies against all the maxims of mainstream economics—lacking the rule of law, the protection of private property and contracts, or the enforcement of market competition. The durability and continuity of the surviving communist regimes assure the world that their plans for economic and military modernization remain deeply serious. If the economic outcomes have been dismal in some places and times, they have been striking in others. The global impact of China’s industries, exports, and military power is just the latest illustration.

Despite the capabilities and sustainability demonstrated by communist states, scholarly writing since the end of the Cold War has tended to downplay the communist experience of state building. The neglect is visible in the selection of regions and periods from which many studies of state capacity have drawn their data. Yet the communist experience shows a record of top-down state building far more capable and sustained than the consensus would allow.

Moreover, whether or not communism will be over one day, it is not over yet. Will China have time to become the first communist country to join the rich nations’ club? You, the reader, might be optimistic or skeptical.

As of now, we can’t be sure of the answer; only time will tell. Meanwhile, a better grasp of the strengths and weaknesses of communist state capacity does at least give us a chance at the question.

Communist states developed exceptional state capabilities. But how usable were the capabilities that they intended to deploy? To understand why this question comes next, think of the internet. The internet endows us with capabilities, such as instant access to global information and commerce, that would astonish a visitor from 1970. But every internet user has had the annoyance of being locked out of an account by a forgotten or misremembered password. Sometime the internet is just not very usable.

A feature of internet commerce is the “security/usability” tradeoff. An expert writes:

A colleague of mine used to quip “Got an access denied? Good, the security is working.” That means that security administration is fundamentally opposed to network administration—they are, in fact, conflicting goals. .. . Essentially, the tradeoff is between security and usability. The most secure system is one that is disconnected and locked into a safe.^[113]

How does the security/usability tradeoff work? When internet security is light and the bar is low, predators crowd in and begin to feed on the unwary. User confidence collapses. The first password is an obvious advance. It boosts my confidence in the internet and increases my willingness to use it. Soon, all the legitimate account holders have a password. An arms race develops. Thieves become proficient at hacking passwords and stealing identities. More complex passwords, memorable information, and multifactor authentication make it more difficult for criminals to break through the security surrounding my account details by stealing my identity. But these precautions don’t just make it more difficult for criminals. They also make it more difficult for me to assert my identity in my own legitimate business. The price of making my assets more secure is that they are also made less usable.

This is the security/usability tradeoff. Too little security, and the system is laid waste by thieves. More security seems like a good idea, but it also makes the system harder to use. Too much security and the thieves are locked out, but so is legitimate business. Somewhere, there is the right amount of security—but where is it?

State secrecy appears to have a similar effect on state capacity. At very low levels, secrecy is productive: it keeps out opportunists and troublemakers. And this enhances capacity. Beyond a point, however, the costs of secrecy increase so much that they begin to detract from state capacity. When the price of more secrecy is that capacity declines, there is a secrecy/ capacity tradeoff.

What, exactly, causes the tradeoff to bind? There are three factors. First, as secrecy becomes more and more intense, the government must commit increasing resources to managing the “regime of secrecy” and maintaining its integrity. These resources must be taken from elsewhere, and the diversion places an increasing burden on state capacity.

A second factor that comes into play as secrecy grows is not less important, but it is more complicated: the bundling of functions. Secrecy enables some functions and disables others in bundles that cannot be disentangled. An example is the bundling of error correction with the management of authority (or reputation). Suppose a policy has a bad outcome: the excessive rapidity of Soviet industrialization in the late 1920s and early 1930s led directly to a famine in which millions of people starved to death. To maintain the authority of the party and its leaders, this bad outcome was strictly censored. Because of this, private citizens and most public officials were deprived of information that could have been useful in preventing a repetition. Thus, in enabling the management of authority, secrecy also disabled error correction.

Evidence in support of this bundling is provided by economists Egorov, Guriev, and Sonin: using data from the period 1995-2007 and controlling for income levels, democracy scores, and natural resources, they show that bureaucratic quality across more than 130 countries was positively and significantly increased by the degree of media freedom.^[114] The authoritarian ruler’s problem is that media freedom opens a channel to the improvement of policies and their delivery, but the same channel can allow criticism to spread and opponents to coordinate action.

Another example is the bundling of concealment (from an external adversary) with cooperation (internally among managers responsible for the business of the state). Suppose the location of a sensitive facility is encrypted, with the key kept under secrecy. Outsiders who do not have the key can no longer find the facility on the map. But insiders—say, managers with legitimate business at the secret facility—who lack the key, cannot find it either. In this real-life example (from evidence to be provided in Chapter 4), secrecy switched concealment on and cooperation off at the same time.

And a third factor in the secrecy/capacity tradeoff: As secrecy becomes more intense, it provides a cover for those who choose to exploit it, not in the interests of the state, but because they are greedy or lazy. You want to inspect my work? No, it’s secret. You want to know how much I’m entitled to charge you? No, that’s secret too.

The sense of this discussion is captured in Figure 2.3. This figure shows state capacity (increasing vertically) as a function of the intensity of figure 2.3. The secrecy/capacity tradeoff: State capacity rises, peaks, and then falls away as the degree of secrecy increases

Note: In this figure the degree of secrecy (defined as increasing regulation of information and penalization of violations) is measured on the horizontal scale. State capacity is measured on the vertical scale. C_max marks the maximum attainable level of state capacity, and S* is the degree of secrecy that delivers the maximum at point X. This chapter argues that citizens of a democracy are likely to prefer combinations to the left of the maximum, such as Y, whereas an authoritarian ruler is likely to prefer a combination to the right, such as Z.

secrecy (increasing horizontally from left to right). Secrecy is made more intense by increased regulation and increased penalization of noncompliance. By assumption, secrecy has diminishing returns and increasing costs. At low intensities, the costs of secrecy are low and the benefits are high. The rising part of the curve reflects those aspects of secrecy that promote state capacity by providing minimal assurance of the power to take decisions without being manipulated by foreign and domestic adversaries. While marginal costs are low and benefits are high, state capacity rises. As costs increase and benefits diminish, a peak is reached at X, after which a decline sets in. As secrecy becomes even more intense, state capacity is driven down by the costs of assuring the growing volume and detail of secrets, the difficulty of protecting the business of the state from the growing complications of secrecy, and the growing scope for insiders to exploit secrecy for personal gain.^[115]

As this approach hints, somewhere in the figure there should be a “right” level of secrecy—but where? That very question was assigned to the journalist and former spy Mark Frankland during his training for the British Secret Intelligence Service in 1958:

We had to write an essay on “the economic level of security.” Total security led to paralysis, for only if no one said or did anything could nothing be given away. Too little security led to disaster. Where was the golden mean? It was as good, or as pointless, a subject for debate as the medieval argument about how many angels could sit on the head of a pin.^[116]

Although seemingly scholastic, the location of the golden mean, the “right” level of secrecy, is a practical issue for all states, one that is vital to get at least approximately right. But where is it to be found? It would be easy, but wrong, to identify the golden mean with the point at X, where state capacity finds its peak.

A moment’s reflection suggests the mistake. The answer depends on the constitution of the state. In a liberal democracy, where an “informed citizenry” is valued, the voters might tend to see secrecy as a necessary evil. If so, they should be willing to give away some state capacity so as to have less secrecy, preferring Y to X, for example. A degree of openness to foreign influences, whether welcome or unwelcome, or an element of domestic gridlock resulting from political competition, might be a price worth paying for the right of citizens to be fully involved in the making of decisions or to challenge the authority of the decision makers.

For the dictator, by contrast, secrecy is not a necessary evil but a good. Such a ruler should be willing to give away some state capacity so as to have more secrecy. Beyond X, where secrecy is beginning to erode state capacity, lies a point such as Z. Here the ruler willingly sacrifices some capabilities in order to conceal mistakes and wrongdoing, maintain a façade of competence, and suppress criticism and opposition.

In the case of communist parties in power, it appears that they wanted two things: to secure their own regime, and to build a powerful state. At first sight, these two objectives looked fully compatible. Looking more closely, the notion of the secrecy/capacity tradeoff drives a wedge between them. It suggests that secrecy promoted both regime security and state building up to a point. Beyond that point, increasing secrecy continued to fortify the defenses of the ruling party, but it began to work against the goals of state building. This was because the costs of secrecy began to eat away the capacity of the state to do anything more than keep its own secrets and keep the ruling party in power.

The idea that a dictator faces tradeoffs is not new. Any authoritarian ruler must spend resources on repressing opponents or on buying them off. These resources must be diverted from other uses, which must be traded off.^[117] Again, the more fearsome the dictator, the less he can trust the sycophants that surround him to share unwelcome facts: this suggests he must trade truth for compliance.^[118] Another recognized tradeoff is competence of government personnel for loyalty: autocratic regimes are likely to promote their officials based on unthinking loyalty and adherence to the leader, to the detriment of their general level of competence.^[119] The contribution of this chapter is to identify secrecy as one of the margins at which the dictator must optimize.

In claiming that communist rulers wanted to hold state power and build the state, I do not exclude that their program extended to other things as well—to build a modern, technologically progressive, well-defended society, for example. But for all the things the rulers wanted, they saw one instrument—a strong, capable, multifunctional state. Power was what made the state their instrument, and, for that and for their own survival, they wanted to hold onto power at any price. Secrecy helped to keep the rulers in power, but the same secrecy also ate into the capacity of their instrument to deliver their program. This was part of the price they were willing to pay to retain power.

The idea of the secrecy/capacity tradeoff suggests rational choice. A public-choice perspective would imply that the government can calculate and freely choose the degree of secrecy most consonant with its preferences.^[120] From this perspective, secrecy looks like a policy, similar to a tax rate or a speed limit, that can be varied in line with changing circumstances and needs.

To what extent did communist regimes consciously choose the degree of secrecy, knowing the consequences of doing so? Does the government of any country actually choose the amount of secrecy in this deliberative, calculating way? Doesn’t the amount of secrecy that the government inherited, to which everyone has become habituated, weigh more heavily than any calculation? Was the level of Soviet secrecy really a calculated choice?

There are several approaches to this issue, and public choice is not the only angle. To understand the responsiveness of the degree of secrecy to a changing world, we could range the alternative models on a spectrum, with secrecy more easily recalculated and fine tuned at one end, and more inherited from the past and inflexible at the other. To simplify, secrecy might be a policy (flexible and freely chosen), a culture (inherited and predetermined), or an institution (with elements of both).

As discussed above, the idea of a secrecy/capability tradeoff suggests that the government should freely decide the combination that is best suited to its idea of the national interest. If circumstances or needs change, the degree of secrecy should change. In this view, secrecy is just a policy that can be recalibrated and varied as flexibly as a tax rate. This idea has at least some merit, for Soviet secretiveness was not a constant. As discussed in Chapter 1 (and further illustrated in Chapter 4), Soviet secrecy was more intense than secrecy in prerevolutionary Russia, while the late 1920s, late 1930s, and late 1940s saw consecutive increases in the degree of secrecy, before a rebound in the 1950s.

Empirical support for the idea that secrecy can be varied in response to changes in the context is found in the work (already mentioned) of Egorov, Guriev, and Sonin.^[121] They note that, to retain power, authoritarian rulers need to control both society and their own bureaucracy Control of the media helps them control society, but it also impedes control of the bureaucracy because bureaucrats become the main source of information about their own performance. Media freedom gives the ruler better information about bureaucratic performance, but also empowers the opposition. But this problem is more acute for some rulers than for others. An oil-rich regime can maintain itself by spending or sharing oil rents; it need have less concern for bureaucratic performance. Egorov and coauthors show that media freedom is inversely related to the size of a country’s oil reserves: the more an incumbent ruler can expect to rely on oil rents, the less is media freedom in the country. The direction of the effect is unambiguous, and its timescale is within a few years. This provides support for the idea that the degree of censorship, at least, can be varied in response to changes in the regime’s environment.

While Soviet secretiveness was not a constant, it was clearly less flexible than, say, a tax rate. Major alterations transpired no more frequently than once a decade, and there seems to have been little fine-tuning in between. This suggests that secrecy was not so much continuously variable as a regime (the word they used themselves) with considerable persistence, adjusted rather occasionally and then quite sharply.

The idea of secrecy as a flexible policy instrument suffers from another notable weakness. It rests on the idea that costs and benefits are well defined and measurable, so that the government can balance them and find the right degree of secrecy by watching the balance change as secrecy is adjusted in real time—an implausible scenario. If optimization did take place, and if reoptimization happened when circumstances changed, it must have taken the form of time-consuming trial, error, and correction.

The idea of trial and error does not contradict optimization in general; on the contrary, much real-world optimization takes that form. But it certainly implies that finding the optimum might be much more difficult than Figure 2.1 would suggest.

A Culture of Secrecy

An opposite approach to secrecy can be found in the idea of culture— that is, the degree of Soviet secrecy might have been the outcome of a secretive culture. The word culture is often a way of capturing the sense of shared norms and practices that persist unchanged among a population or a subgroup over long periods of time. Often it is associated with the transmission of identity: look, we do things this way, not because it happens to be efficient, but because we are who we are. Cultures change, and a culture of secrecy might be made to change by external pressure, such as a competing culture or counterculture of openness, but cultures do not change of their own accord.^[122] A characteristic weakness of the “culture of secrecy” approach might be the lack of a clear sense of who are the agents of cultural persistence and what motivates them. But perhaps this is not required, because if one culture is equipped to suppress another, it is surely a culture of secrecy.

The idea of a communist culture of secrecy is clearly exemplified by the idea of “conspirative norms”: the Bolsheviks explicitly patterned their governmental practices on the habits they developed for survival in the revolutionary underground.^[123] Once the party came out into the open and seized power, it did not change its habits but rather formalized them in a written code, so that conspirativeness became a way of ruling. On this interpretation, extreme secrecy was not so much a choice as an inheritance that no one could avoid—not only a way of ruling but a way of being that was handed down from one Bolshevik generation to the next.

On this reading, Soviet leaders of the following generations had no alternative but to continue the culture of secrecy. They did not calculate how they did things; they would have been able to fine-tune it only with great difficulty, and they could not abandon it. Even Mikhail Gorbachev departed from it only under the pressure of extreme events that could not be denied, such as the Chernobyl catastrophe.^[124]

There is clearly something to the idea of a secretive culture, not only communist but also Russian. For hundreds of years, Russia has been ruled by secretive councils organized around the figure of a tsar, then a general secretary, and now a president.^[125] From time to time, Russia has become more open, but such periods have been brief, followed quickly by a return to “normal” secrecy.

At the same time, the idea of a fixed Russian culture does not explain how Soviet secretiveness went so far beyond prerevolutionary practices. The idea of a fixed communist culture does not explain the variation of secrecy over time under communist rule. It might be useful to turn to a third way of understanding secrecy, one that is neither policy nor culture.

Secrecy as an Institution

An intermediate position is available: secrecy might be an institution in Douglass North’s sense of “rules of the game.”^[126] In this view institutions are rules of interaction, which were adopted in the past and persist today, and while they persist, people make choices within them. The rules persist, not forever, but for as long as they have the support of a governing coalition. They change when the individual choices made within them give rise to new coalitions of people that have the power and will to change them.

A significant feature of Soviet history is that the “governing coalition” was often extremely narrow and, on occasion, depended entirely on the changeable will of a single person—the party leader. Thus, it is consistent with the idea of secrecy as an institution that the rules of Soviet secrecy could persist for quite long periods, and then be altered abruptly. Factors in the sudden shifts of Soviet secrecy included the break with the Left Socialist Revolutionaries and Lenin’s turn to a one-party state (in 1918); Stalin’s victories over the Left Opposition and the “Right deviation” (in 1927 and 1929); his anger at the unauthorized sharing of biomedical research (in 1947); Khrushchev’s unexpected denunciation of Stalin (in 1956), which disrupted the coalition of Stalin’s successors; and the prevailing winds that, by blowing radioactive debris from Chernobyl across half of Europe, forced Gorbachev into unprecedented disclosure (in 1986) of a domestic catastrophe.

To conclude, if we wish to understand why the Soviet Union’s leaders gave up substantial state capacity for the sake of secrecy, there are three approaches. Each of them is a simplified concept that has something to offer. I am not inclined to nominate one of them as the only correct approach to secrecy. In this book, we will see secrecy operating sometimes as a culture, sometimes as a policy, and sometimes as an institution.

At the same time, I should disclose my professional bias. As an economist, I was trained to look for costly behavior. Where a cost is incurred but could be avoided by choice, there should also be a benefit. This approach presumes rational choice, not as a matter of faith, but as a driver of investigation: if costly behavior is not explained by an apparent net benefit, then there must be some factor at work that we have not yet understood. Perhaps we should dig deeper. At all times, therefore, I will resist the temptation to describe Soviet secretiveness as “irrational.”

In the present case, if Soviet secrecy was costly, and if more transparent forms of government that avoided the costs were clearly available, then why did the Soviet rulers choose to accept these costs? The answer to which I will return again and again is that the expected benefit of secrecy to the rulers must have been commensurately large. It was not by chance that the Soviet system tended to merge security and secrecy into one. When decisions were pondered in secret, the ruler was secured against pressure from civil society. When the decisions themselves were censored, the ruler was secured against challenge. When the outcomes of bad decisions were concealed, the ruler was secured against accountability and criticism. For these reasons it was worth the rulers’ while to uphold the culture, to enforce the institution, to provide the career incentives for millions of administrators and managers to conform to the pattern, and to tolerate the very high costs to their effectiveness in other dimensions.

Finally, the idea of a secrecy/capacity tradeoff does not preclude mistakes. In the Soviet state it was possible to increase security to a point where usability became so difficult that people felt compelled to behave in “insecure” ways. I do not describe this as irrational, but it had unanticipated consequences and could reasonably be called a mistake.

The ultimate purpose of Soviet secrecy was to sustain a durable dictatorship. It does not matter much whether you ascribe the origins of Soviet secrecy to the inner drive of totalitarian single-party rule or to the external pressures of foreign encirclement.^[127] The Soviet Union’s communist rulers made no distinction between the survival of the country and the security of the regime. Whichever way you looked at it, the need for secrecy was existential. If you wanted to run a durable dictatorship, and not be overthrown by the foreign enemy or the domestic enemy, you had to pay the costs, no matter how heavy.

Moreover, secrecy worked. It did sustain a dictatorship for seven long decades—more than the lifespan of the average twentieth-century Russian male. From that perspective, Soviet secrecy was generally extreme, sometimes weird, occasionally amusing—but irrational? Over a long period of reflection, I have come to think not.

The purpose of this appendix is to set out and explain the data underlying Figures 2.1 (tax shares) and 2.2 (ownership shares). It involves discussion of the comparability of statistics across both countries and accounting systems. Comparing tax shares in national income is primarily a problem of the accounting system for national income. Comparing ownership shares presents multiple issues.

For tax shares the problem is to estimate the country share of current government revenues (the numerator) in national income (the denominator) when the denominator is measured inconsistently. The countries of the OECD shared the System of National Accounts, in which national income is measured by the gross domestic product or GDP, the aggregate value of final goods and services at market prices. The Comecon countries and China used the material product system, in which national income is measured by the net material product or NMP, the aggregate value of final goods utilized for consumption and accumulation at prevailing prices. The main difference between the two lies in the treatment of final services. Contrary to what is often asserted, the NMP does include some services— the intermediate services that enter into the production of goods. It excludes only those services that enter directly into the final consumption and accumulation decisions of households, firms, and government. Thus, education and health expenditures are not counted in material production, and retail banking and passenger transportation are also left out. But freight transportation and distribution are included.^[128]

The importance of final services in the global economy has tended to grow over time, and by 1980, the year of interest for our measures of state capacity, the difference between GDP and NMP measures was quantitatively large. In the final years of the Soviet system, official statisticians began to calculate the Soviet national income retrospectively using both measures. For 1980, they estimated the Soviet net material product utilized as 454.1 billion rubles, while GDP was 619 billion rubles, more than one-third larger.^[129]

For tax shares, what counts on the side of the numerator is approximately equivalent between the two systems, both of which acknowledge the major categories of retail sales taxes and personal income and profit taxes. The denominator is the big difference. As Table 2A.1 shows, Com-econ country tax shares in the net material product in 1980 look much larger than OECD country tax shares in GDP, and there is no overlap at the near edge of the two distributions. But NMP is much less than GDP. When we adjust, we find that Comecon tax shares are still very much on the high side, but (as noted in the chapter text) they now overlap the high-tax market economies of northwestern Europe.

Turning to ownership shares, the comparability of country statistics is compromised in several ways. First, OECD countries did not routinely report data on the value of output or assets or the size of employment by public enterprises.^[130] For the OECD countries shown in Table 2A.2,1 rely on the data for many OECD countries in the 1970s and 1980s, assembled by the economist Branko Milanovic, for output and employment shares of “state-owned enterprises in commercial activities.” In Milanovic’s data, “government services proper” are excluded; this refers to government services supplied free of charge, from defense and policing to health and education.

Second, for Comecon countries we have two sources on ownership shares. One is, again, Milanovic, as shown in the table. Alternatively, there is an official source, because such figures were also compiled and reported by Comecon statisticians, and they too are shown in the table. Their coverage table 2A.1. Measures of state capacity across countries: Fiscal ratios in 1980

Note. The countries shown in the table are ranked in declining order of their real GDP per head (in dollars and international prices of 1990) according to the Total Economy Database (January 2014) of the Conference Board at http://www.con ference-board.org/data/economydatabase/. For 1980, the unweighted mean for the OECD countries shown was $13,929, and for Comecon countries it was $6,603. The real GDP per head of the poorest OECD country, Portugal, was $8,044; that of the richest Comecon country, Czechoslovakia, was $7,982. Thus, the two groups of countries do not overlap.

Sources: (A) OECD countries: Economic Outlook, no. 108 (December 2020), https:// www.oecd-ilibrary.org/economics/data/oecd-economic-outlook-statistics-and-pro-jections_eo-data-en, gives general government current receipts as a percentage of GDP (the sum of final goods and services at market prices) (series YRGQ). (B) Comecon countries: current government revenues (taxes, levies, fees, fines, and enterprise incomes) are calculated in percent of NMP (net material product, the sum of final goods and intermediate services at prevailing prices) from United Nations, Statistical Yearbook 1981, Tables 26 and 46. Shares of net material product are then recalculated as shares of GDP, based on the observation that the Soviet GDP for 1980 reported by the government statistical agency was 36 percent greater than NMP utilized (Goskomstat, Narodnoe khoziaistvo SSSR v 1990g., 5). The same conversion factor is applied to all the Comecon countries. Unweighted averages are then calculated for this table from the figures shown.

(В) Comecon countries and China

State-owned enterprises,

Notes. The countries shown in both parts of the table are ranked in declining order of their real GDP per head in 1980, as explained under Table 2A.1, except that Cuba and Mongolia, not given in the TED dataset, are inserted on the basis of comparable estimates in the Maddison historical dataset (2009 edition), http://www.ggdc.net/ maddison/oriindex.htm.

Sources: (A) OECD countries, from Milanovic, Liberalization and Entrepreneurship, 24. Figures cover “state-owned enterprises in commercial activities,” excluding “government services proper.” Where Milanovic gives country figures for more than one year, I take the nearest to 1980. (B) Comecon countries and China, socialist sector in 1980 (col. 1), from SEV, Statisticheskii ezhegodnik, 37. Figures are for the shares of “state enterprises and institutions, social organizations, cooperatives of all kinds and types, personal sideline farms of members of cooperatives and personal sideline farms of manual and white-collar employees of state, social, and cooperative enterprises and institutions” (SEV, Statisticheskii ezhegodnik, 417-18) in the net material product in 1980, except Bulgaria where the figure for 1980 is missing, and the share of the socialist sector in productive fixed capital is used in its place. For Poland the source gives two figures. The lower figure, shown here, is at current prices, with a slightly higher figure at “comparable” prices. Shares of state-owned enterprises in the net material product in various years (cols 2 and 3) are those reported by Milanovic, Liberalization and Entrepreneurship, 21.

is subtly different from the OECD data: the “socialist sector” comprised “state enterprises and institutions, social organizations, cooperatives of all kinds and types, personal sideline farms of members of cooperatives and personal sideline farms of manual and white-collar employees of state, social, and cooperative enterprises and institutions.”

The inclusion of cooperatives and of households’ sideline farming (on state-owned land but not themselves state owned) within the “socialist sector” of Comecon countries is the major difference from the concept of the OECD figures. This is not so troubling for agricultural cooperatives (“collective farms”) which, although not legally owned by the state, were heavily regulated and monitored. The inclusion of nonagricultural cooperatives and of sideline farming is potentially misleading, however, because these were more often forms of private enterprise, either taxed and regulated or operating below the state’s radar. Still, the overstatement of state ownership in Comecon countries is far smaller than the Comecon/OECD gap.

Another source of discrepancy might be the treatment of “government services proper.” These are excluded from the OECD data, whereas the Comecon concept includes the output of “state . .. institutions.” The important thing is that whatever output is included in the Comecon data cannot refer to “government services proper” because the material product system counted these as final services, which were excluded from the value of output.

In short, the OECD data refer to the “commercial activity” of public corporations whereas the Comecon data refer to the “productive” activity of state-owned and cooperative enterprises, and this is the nearest we can get to comparable numerators.

Third, we turn to the denominator, the national income measure. Do the ownership shares of Comecon countries in national income need the same correction as the tax shares? It is a matter of judgement, but my judgement is no. For both groups of countries, we have an internally consistent measure. For OECD countries, final services are included in both numerator and denominator. For Comecon countries they are excluded. If they are to be added back to the bottom of the fraction, they must also be added back to the top. It is hard to imagine that substantial change would be the result.