The following is an entry written for the New Palgrave Dictionary of Economics, 2nd edition, to be published in May 2008. When the new edition is published, this article will no longer be available from Brookings.
Abstract: The virtual economy was the system of informal rent-distribution that arose in post-Soviet Russia in the 1990s as nonviable Soviet-era manufacturing industries sought to protect themselves from the discipline of the market. Enterprise directors and their allies throughout the economy (including government officials) colluded to use nonmarket prices and various forms of nonmonetary exchange such as barter to transfer value from resource sectors to manufacturing industry. The article discusses the system’s historical roots, describes some of its characteristic phenomena, and outlines a model for behavior of enterprises.
The virtual economy was the name given to the system of informal rent-sharing or value distribution that prevailed in Russia in the 1990s. Featuring widespread use of nonmonetary exchange and nonmarket prices to conceal transfers of value from especially resource sectors to manufacturing industry, the virtual economy reached a peak in the run-up to the country’s financial crisis in August 1998.
The strategies used by enterprise directors to participate in this nonmonetary economy fundamentally changed the behavior of hundreds and thousands of noncompetitive manufacturing enterprises in Russia during the transition process. The behavioral adaptation permitted enterprises to survive in the transition environment where they ought to have failed. The expectation had been that when the old Soviet industrial structure was shocked by the sudden collapse of central planning and the subsequent launching of radical market reforms — including mass privatization and elimination of overt subsidies — economic agents would be forced to change their behavior to become competitive in a market economy. The transition was thus intended as a Darwinian process whereby only those enterprises that could transform themselves into competitive operations would survive. But in the case of Russia, the dinosaurs survived — without restructuring. They did change. Only, instead of adapting to the market, they changed to protect themselves from the market.
In essence the virtual economy was a peculiar system of rent distribution in which the primary vehicle through which agents laid claim to rents was production. The virtual economy was the set of informal institutions that facilitated the production of goods that were value subtracting, that is, worth less than the value of the inputs used to produce them. Enterprises were able to engage in such production because they had recipients who were willing to accept fictitious (nonmarket) pricing of the goods at levels that masked their lack of profitability. Buyers and sellers colluded to hide the fictitious nature of the pricing. In the classic form of the virtual economy, they did so by avoiding money, instead using barter and other forms of nonmonetary exchange, as well as even more intricate subterfuges.
Since value was being destroyed as the system operated, there had to be a source of value. The ultimate “value pump” in Russia was the fuel and energy sector, above all one single company, Gazprom — Russia’s natural gas monopoly. In exchange for the rights to keep what it earned from exports, Gazprom pumped value into the system by supplying gas without being paid for it (or, more generally, at a cost that was low enough to keep enterprises operating). Gazprom subsidies, which then led to arrears to the government, were the primary way in which unprofitable activity was supported in Russia.
The virtual economy evolved and persisted because it met the needs of so many actors in the economy. Workers and managers at industrial dinosaurs benefited because the virtual economy postponed the ultimate reckoning for loss-making firms. Government, especially at the subnational level, where much of the important action took place, benefited because the virtual economy system maintained employment and the provision of social services. Gazprom also benefited, since the value transfers it made to the virtual economy earned it the right to appropriate the massive rents from exports.
The roots of the virtual economy mechanisms lay in the Soviet system, especially the production relationships that had developed under the Soviet command economy. These relationships represented a peculiar type of asset, “relational capital,” which supplemented the enterprise’s conventional physical and human capital. Thanks to relational capital, market reform policies did not necessarily compel the enterprise to restructure to be able to compete in the market environment. Enterprises chose between whether to become more competitive in the market, by investing in physical and human capital, or to be better protected from the market, by investing in relational capital.
The term “virtual economy” was coined in 1998 by Gaddy and Ickes, building on terminology in a Russian government report from 1997. In early 1996, alarmed by the extent of tax delinquency in the country, President Boris Yeltsin appointed a special blue-ribbon panel to investigate the low rate of collection of taxes in Russia. Presenting its findings after an 18-month investigation, the panel reported that the country’s largest companies conducted 73 percent of all their business in the form of barter and other nonmonetary forms of settlement. Especially alarming was the extent of nonmonetary payments of taxes. During the period of review, these large enterprises paid less than 8 percent of their tax bills in actual cash. They simply failed to meet 29 percent of their obligations at all, while “paying” the remaining 63 percent in the form of offsets and barter goods. The market value of the goods delivered was far below the nominal price used in the offsets, leaving the government with substantially less in real revenues than officially accounted for. In summing up their own conclusions about the contemporary Russian economy, the investigatory commission wrote:
An economy is emerging where prices are charged which no one pays in cash; where no one pays anything on time; where huge mutual debts are created that also can’t be paid off in reasonable periods of time; where wages are declared and not paid; and so on. [...] [This creates] illusory, or virtual earnings, which in turn lead to unpaid, or virtual fiscal obligations, [with business conducted at] nonmarket, or virtual prices. (Karpov, 1997).
Gaddy and Ickes (1998) suggested that the entire system be called a virtual economy “because it [was] based on illusion, or pretense, about almost every important parameter of the economy: prices, sales, wages, taxes, and budgets.” The pretense that had become the norm was as characteristic of the virtual economy as were the colorful forms of nonmonetary exchange.
The Nonmonetary Economy
The nonmonetary means of payment that characterized the virtual economy spanned a wide range. They included direct exchanges of goods (true barter), either bilaterally or through “chains” with multiple participants, offsets (where debts accrued by one party were later paid off not in money but in goods), and promissory notes called veksels. Veksels — the name for which is derived from the German Wechsel — were a widespread non-monetary payment mechanism that ranged from being a substitute for money to essentially a form of barter.
There were several key nodes in the barter chains, above all the major natural monopolies known popularly as the “Three Fat Boys” (tri tolstyaka) — Gazprom (the natural gas monopoly), RAO UES (the electricity monopoly), and MPS (the state railways). All three frequently complained that they collected as little as 10 percent of their revenues in cash. Almost all enterprises in Russia were consumers of the output of these three companies: rail freight transport, gas, and electricity. The three monopolies also accounted for about 25 percent of taxes due to the federal budget. The fact that everyone needed to purchase services from the “fat boys” meant that there was a ready demand for the veksels (IOU’s) of these companies. It was this special position that put them at the core of the non-payments system in Russia.
The other key player in the barter economy was the government, or rather, governments at all levels. Here again was an agent to whom nearly everyone had an obligation. The volume of accrued unpaid taxes, plus the huge fines and penalties levied for nonpayment, presented governments with an almost inexhaustible supply of debts. And, in turn, governments themselves owed many others. They were, like the natural monopolies, a key node for barter.
One particularly important phenomenon was tax offsets. An enterprise owed taxes to the government, and concluded an agreement whereby those tax obligations were settled by delivering goods or performing services for the government. Of all the forms of nonmonetary transactions observed in Russia in the 1990s, the mechanism of tax offsets was the most characteristic of the virtual economy. Russian governments at all levels grew increasingly willing to offset enterprises’ tax obligations against goods or services delivered to the government. By the end of 1997, the accumulated tax debt was enormous. Industrial enterprises were particularly egregious delinquents. The sum owed by the enterprises at the end of 1997 was equal to 46 percent of the amount they actually remitted in taxes for all of 1997. These enormous debts gave impetus to the practice of tax offsets.
Consider, for example, an enterprise that was able to supply the local government with services in lieu of taxes. The enterprise could have paid its tax liability in money, but that would have required selling its output for cash. Alternatively, the enterprise could negotiate with the government to supply some service as an offset for taxes. If the enterprise had resources that were not fully utilized, the latter alternative was likely to reduce the effective tax burden on the enterprise. Moreover, once the government showed itself to be willing to engage in tax offsets, the options open to enterprises expanded. Now the enterprise could potentially pay its taxes not only with its own products but also with products it received in barter deals from other enterprises. This greatly reduced the cost to the seller of accepting goods rather than cash.
The motivation for governments to join in the barter economy was simple. They reasoned that if they could not get cash, it was better to reach some sort of settlement than receive nothing at all. In some cases, especially at the local level, an enterprise could offer to deliver goods or services to the city or regional government in lieu of taxes. At the federal level, it was more common for the government to cancel tax arrears or taxes due by writing off the government’s own debt to the enterprise in question for state orders. Once the practice was established with respect to past arrears, there was an anticipatory factor: enterprises began to feel confident that they could henceforth ship off products to the government, knowing that later they would be allowed to offset their taxes in an equivalent amount. Less than 60 percent of all federal taxes collected in 1997 were paid in cash; the rest were in the form of offsets.
The federal government was particularly victimized by these schemes. Enterprises frequently colluded with regional and local officials to hide income and hence keep revenues away from the federal government for taxes whose revenues were split between local and national authorities. In other cases, local governments demanded that enterprises pay their taxes in the form of goods and services that could only be used locally and not be shared with the federal government (for instance, by providing road construction or repairs of buildings). Often, if the federal government received anything at all in these schemes, it was only what the regional governments did not want.
In one notorious case reported in the Russian press in the spring of 1998, the oblast (province) government of Samara had permitted enterprises to pay their regional taxes in the form of goods. One of the items offered turned out to be ten tons of toxic chemicals from a local chemical plant. Although the plant claimed (and was given) credit for 400 million rubles [$80,000] in taxes, auditors later determined that the chemicals were worthless (and indeed dangerous). The Samara government never suffered from this curious deal, however, since it had previously sought and received permission from the federal ministry of labor to fulfill its obligations to the federal unemployment compensation fund by delivering goods instead of money. Among the goods it offered were ... the ten tons of toxic chemicals. (Gaddy and Ickes, 2002, p. 176)
As a result of these practices, the Russian budget ran massive deficits. Even using the inflated prices used in the offset deals, federal revenues plummeted — from 16.2 percent of GDP in 1995 to 12.4 percent in 1998. To finance its deficits, the government had resorted to extensive borrowing outside and inside Russia at increasing and unsustainably high costs, thus digging itself even deeper in debt. Finally, on August 17, 1998, the government defaulted on about $40 billion worth of its own ruble-denominated debt instruments (so-called GKOs), some $17 billion of which were held by foreigners.
The Soviet Roots
The roots of the virtual economy lay in the structure and institutions bequeathed to Russia by its Soviet predecessor. Some parts of the economy, notably the resource industries, were value-adding. But most of the vast manufacturing sector that Russia had inherited could not compete in a market setting. In fact, by the final years of the Soviet era, the manufacturing sector was in poor condition even on the terms of the planned economy. By official Soviet standards, more than one-third of equipment in Russian industry was physically obsolete. Soviet planning practice, which emphasized output over costs, set physical, rather than economic, obsolescence as the criterion for removing a machine from the factory. As long as the machine could produce anything at all, it was kept in production. The result was very low replacement rates for capital equipment.
The location of industry in the Soviet economy was another problem. Not only did Soviet location policy ignore transportation costs but it also failed to take into account the costs associated with the cold Russian climate — in terms of energy use, health maintenance, and many other factors. By being placed in some of Russia’s coldest and most remote regions, the manufacturing enterprises were rendered even less competitive and less attractive for foreign investment.
Equally important as the structure of the Soviet economy and its lack of competitiveness was the fact that this reality was hidden. As the market transition began, past history and performance gave no information about which sectors, or enterprises, were value-adding and which were value-destroying. The culprit was distorted Soviet pricing.
Soviet Pricing and the “Circus Mirror” Effect
Soviet prices were not based on opportunity cost, or value; rather they were simply an accounting instrument to measure plan fulfillment. Although Soviet prices were set arbitrarily, they were not set randomly. They were determined by specific rules of the system, and this produced some systematic biases. First, the planners underpriced raw material inputs, especially energy. They based raw materials prices only on the operating costs of extraction, while ignoring rent. In so doing, they disregarded the opportunity cost of using the resources now rather than in the future. The planners’ overriding goal was to increase today’s output. Scarcity pricing might have induced more conservation, but it would have militated against maximizing current production. This bias in raw materials prices fed into the system of industrial prices. Heavy consumers of energy were, in effect, subsidized. So, too, were heavy users of capital, thanks to the absence of interest charges. In short, costs of production were calculated on the basis of an incomplete enumeration of costs. This led to lower prices for inputs, especially resource inputs, than for final uses and thus an understatement of the share of gross output used in production and, hence, an overstatement of net output.
In addition to incomplete cost-based pricing, the Soviet system was explicitly biased toward certain users. The Soviet leadership assigned priority in the economy to heavy industry, especially defense industry, and it was important that it appear that these sectors were producing value. This nonscarcity–based pricing was like a distorting mirror at the carnival. It created the illusion that many enterprises were value producing when in fact they were value destroying.
The “Loot Chain”
A further factor contributing to the opaqueness of the Soviet economy and its post-Soviet successor was the way in which income from control of assets was passed down as payoffs through what Gregory Grossman (1998) referred to as the loot chain. In the USSR, wealth diverted from the official state economy into private hands was shared among networks of individuals in the form of payoffs, bribes, and other schemes. Over time an ever greater proportion of people’s incomes depended on the chain of corruption and side payments.
The virtual economy perpetuated the loot chain in post-Soviet Russia. The living standards of a huge number of people depended on the chain of production and distribution of goods and services in the virtual economy system. In the virtual economy, value redistribution, in contrast to looting pure and simple, occurred in a form that paralleled and was intertwined with actual productive economic activity. This made it especially difficult for agents to discern what their own value and the value of their assets would have been in a well-developed and transparent economy. Basic ideas of a market economy, such as the relationship between individual effort and reward, became almost impossibly obscure. One’s static position in the production process — for instance, membership in the work force of a particular enterprise — was more important for success than individual skills and abilities. The Soviet system separated “what you get” from “what you do.” The reality was that the effort-reward nexus was random. Instead of “from each according to his ability, to each according to his needs (or ability),” it was “to each according to some unknowable, random criterion.” The durability of the misperception depended on its opaqueness. There was no alternative, competing information about the real relationship. This meant that the loot chain was also a constraint on the future evolution of the economy. Individuals were dependent on the prevailing system at the same time that they could not know what an alternative system would offer. The uncertainty caused them to resist abandoning the prevailing system.
Impermissibility of True Reform
While there was no accurate information about the economic importance of the large Soviet manufacturing sector, its social and political importance was unavoidable. Many of the least competitive enterprises — the so-called dinosaurs of Soviet industry — were socially the most important. They employed millions of workers and provided for tens of millions of their family members. Entire cities depended on them. The sheer size of this sector — as shown by employment — operated to maintain its social and political importance, and the illusion of its economic performance. In a sense, then, the importance of the manufacturing sector in Russia was an illusion economically but continued to be a political and social reality.
This latter reality constrained serious market reform policies. Russia did not formally reject the policies themselves; instead, it continued with a pretense of market reform. Policymakers launched one measure after another in their attempt to transform Russia into a market economy. But very few of those measures were allowed to play themselves out to their full extent. The consequences of complete and proper implementation would have been politically intolerable. Thus, while the nation’s leadership proclaimed reform policies, enterprises and other agents continued to behave in ways that rendered the policies ineffective.
The Behavioral Implications
The range of behavioral options in the virtual economy was broad. The ability to use nonmonetary mechanisms to pay taxes to governments and bills to the natural monopolies fundamentally changed the range of opportunities for action available to Russian enterprise directors. By allowing enterprises to settle their obligations by delivering goods for which there was no effective demand, the governments and the monopolies offered an incentive to avoid restructuring. For many enterprises it was easier to produce such goods than to restructure and earn additional monetary income to pay bills in cash. Producing those goods allowed for use of idle capital and labor. In short, offsets and barter permitted some enterprises to survive without restructuring. To represent the full range of choice, not only market-oriented activity but also behavior characteristic of the virtual economy, Gaddy and Ickes (2002) employed the notion of a two-dimensional space, called r-d space. The following sections outline their model.
Market Distance, d
The impact of liberalization on the Soviet economy can be expressed with a spatial metaphor: liberalization revealed the distance that a Russian enterprise would have to travel to compete in the world economy. Let d designate the enterprise’s “distance to the market” at the start of transition. Clearly, d depends on the enterprise’s initial endowments of the things that matter for market viability: physical and human capital, as well as the enterprise’s marketing structure and organizational behavior, but also the characteristics of the good that the enterprise produces (its quality and cost of production). Formally, define an enterprise’s d as the amount of capital expenditure needed to enable the enterprise to produce a product that is competitive in the market. The fundamental reason for measuring d in terms of the investment cost is that transition causes a divergence between the value of existing (inherited) capital and that of newly installed capital.
One may begin to grasp this point by recalling what happened to traditional models of investment in market economies during the energy crisis of the 1970s. Those models predicted that investment would decline, given the tremendous increase in the price of energy. In fact, however, spending on new equipment and buildings soared. The reason for this discrepancy between model and reality was the divergence between the value of installed capital that was energy intensive and new capital that was energy saving. The conventional model ignored the sharp decline in the economic value of the existing capital stock as a result of the 1973 energy crisis. Installed capital had been the result of investment decisions based on low energy prices; hence, its value fell dramatically once energy prices quadrupled. This in turn only increased the demand for new investment in energy-saving equipment. The result was a divergence between the value of installed capital (which lost value) and that of new capital (which had full economic value). In the Russian context, measuring market distance d by the need for new capital investment is a way of capturing the cost of filling the gap between the value of inherited (Soviet) capital and new (market) capital.
Distribution of d
The level of d differs widely among enterprises in the economy. An enterprise that already produces a product it can sell in world markets at a price above cost will have a value of d equal to 0. A completely noncompetitive enterprise will have an enormously large d. Everyone else will be somewhere between. For example, an oil-producing enterprise will have a very low d. Its product is already right for the market. It may need only relatively small investments in marketing, and so on. A Soviet-style machine tool producer, in contrast, is likely to have a long distance to travel.
The distribution of d’s in transition economies differs in two respects from that in market economies. In transition economies the range of d’s is greater and the distribution is more skewed. Both differences stem from the dissimilarity in the process of entry and exit in market and planned economies. In a market economy, whether or not a new firm attempts to enter an industry depends on the founders’ expectations about the new firm’s competitiveness. They will enter if they expect the firm’s potential costs to be lower (its productivity to be higher) than those of existing firms. No firm enters an industry in which it expects it will be noncompetitive. Over time the competitiveness of some firms declines, so d increases. But if a firm in a market economy has too high a level of d, it will be forced to close. Competition and hard budget constraints cause high-d enterprises to shut down.
In a transition economy, in contrast, some enterprises have very high d's that would not be observed in a market economy. There are several reasons for these high-d enterprises. First, in socialist economies entry was not determined by expectations of profitability or competitiveness but rather by the need to fulfill plan targets. Second, insulation from the world economy meant that enterprises were created that produced goods for which the country might not have had a comparative advantage. Third, especially in the case of Russia, the priority given to defense production led to a proliferation of enterprises that produced goods whose market collapsed with the end of the Soviet Union. Fourth, since the geographic location of industry in the Soviet period was based on ignoring transportation costs (as well as the costs associated with extraordinarily cold temperatures), the location of enterprises was also a factor in increasing the d in many cases. For all of these reasons, the distribution of the d’s in Russia at the onset of the transition had a much higher mean and was more skewed to the right than in a mature market economy. This extra mass of high-d enterprises was the burden of the Soviet legacy. And it was this burden that was the essence of the restructuring problem: so many enterprises had to all radically reduce their distance to the market at the same time.
One way to think of the purpose of economic reform is to reduce the average distance in the economy. This occurs through three means: (1) exit of high-d enterprises; (2) entry of new low-d enterprises; (3) and reduction of the d of surviving enterprises. In an ideal market world, market distance would be the only condition that characterized the state of an enterprise. If the only important difference in enterprises were their initial level of d, then policies that put pressure on existing high-d enterprises and encourage creation of new low-d enterprises would have the effect of pushing the distribution in the direction of the market.
The conventional view of restructuring, that reform means reducing d, assumes that each enterprise has one set of resources — its physical and human capital — that it must use ever more efficiently in order to survive. The virtual economy view, in contrast, posits that some enterprises have another resource, relational capital, which they can draw on to enhance their chances for survival. Relational capital is the stock of goodwill that an enterprise can use to avoid the strictures of the budget constraint. An enterprise that has high relational capital can undertake transactions (bartering, using tax offsets, delaying payment) that other enterprises, with low amounts of relational capital, cannot get away with. To put it another way, relational capital is goodwill that can be translated into the ability to continue to engage in production and exchange without reducing the distance to the market. It is therefore the existence of this second dimension that can explain the persistent survival of high-d enterprises in the Russia of the 1990s.
At the onset of transition enterprises differed in their inherited relational capital — call it r — just as they differed in their d. Some enterprises (or their directors) had very good relations with local and/or federal officials. Relations with other enterprises also varied.
Origins of Relational Capital
The relational capital of Russian enterprises was initially accumulated in the Soviet system. Enterprise directors relied heavily on the accumulation and use of personal connections. Relational capital was passed forward to the post-Soviet system in a deceptively simple manner: it was spontaneously privatized. And here lies an important aspect of economic transition in Russia. As Hewett (1988) described, plan fulfillment in the Soviet economy required enterprise directors to use informal skills. Their ability to accomplish this, and their position in the economic hierarchy, was critical to their incomes. While directors earned income from these positions, they did not legally own the source of these incomes. The demise of the planning system, which had already begun with Mikhail Gorbachev’s reforms in the late perestroika period, had the effect of increasing the autonomy of enterprise directors. With the start of economic reform and privatization, the role of the enterprise director increased; other mediating actors (planners, party officials) played less and less of a formal role in economic allocation. Directors used this opportunity to appropriate the returns to the relationships they had developed and cultivated under the previous system. However, in order for directors to appropriate these returns, the enterprises had to continue to operate. Much of the relational capital was both enterprise specific and person specific. To the extent that it was enterprise specific, the director could not cash out the relational capital. The primary form of these connections was relationships with directors of other enterprises, often in related lines of activity, and with ministerial officials and local government officials. The relational capital was worthless to the incumbent director unless he remained in that particular enterprise. He could not leave the enterprise and take the relational capital with him. Furthermore, because it was person specific, he could not sell it to someone else. Instead, in order to appropriate the rents accruing to his relational capital, he had to remain in the enterprise and keep it operating. The privatization of relational capital is thus an important part of the explanation of why directors fought to keep open enterprises that had few prospects in the market economy.
The concept of relational capital can be used to revise the spatial representation of the Russian transition economy. There are now two state variables that describe the nature of an enterprise. In addition to the dimension of market distance, enterprises can be arrayed in terms of their level of relational capital. The initial conditions of an enterprise can thus be described by a two-dimensional space, r-d space, in which each enterprise has its own location.
Whether one views the enterprise sector in a single (d) dimension or in the two dimensions of r-d space is critical for how reform policy is understood. The conventional, unidimensional, view assumes that economic reform measures will have the greatest impact on those enterprises that have the highest level of d. According to this assumption, for example, if budget constraints are tightened, enterprises that are farthest from the market will be under greatest competitive pressure. Similarly, it is assumed that if the economy is opened to international competition, the greatest impact will be on those enterprises that are most in need of restructuring. In the two-dimensional r-d space environment, the effects of market-type reforms need not have this property at all. Tightening the budget constraint will not necessarily put the most pressure on the enterprise that is most inefficient (has the highest d). If the enterprise has been endowed with high r, it may be insulated against the impact of this policy; it can use relations to evade the budget constraint. And if tight budget constraints are enforced against enterprises that are lower in r, then the policy may, in fact, have greater impact on low-d enterprises than high-d enterprises.
It is not just the initial levels of either r or d that matter, of course. An enterprise’s location in r-d space is not the immutable relic of its past. It depends on the path of enterprise investment decisions. If the enterprise has invested in r, it will improve its resistance to policies of tight budget constraints. The enterprise director’s problem is to decide how much to invest in reducing distance and how much to invest in relational capital.
Gaddy, C. and B. Ickes. 1998. “Russia’s Virtual Economy.” Foreign Affairs, 77, 53-67.
Gaddy, C. and B. Ickes. 2002. Russia’s Virtual Economy. Washington, DC: Brookings Institution Press.
Grossman, G. 1998. “Subverted Sovereignty: Historic Role of the Soviet Underground.” In The Tunnel at the End of the Light: Privatization, Business Networks, and Economic Transformation in Russia, ed. S. S. Cohen, A. Schwartz and J. Zysman. Berkeley: University of California Press.
Hewett, E. 1988. Reforming the Soviet Economy. Washington, DC: Brookings Institution Press.
Karpov, P. 1997. “On the Causes of the Low Rate of Tax Collection (Nonpayments in the Fiscal System), General Causes of the ‘Payments Crisis’, and the Possibility of Restoring the Solvency of Russian Enterprises.” Report of the Inter-Agency Balance-Sheet Commission, chaired by P. A. Karpov. Moscow (December). [In Russian].