BARRIERS TO COLLECTIVE ACTION

Steelworkers and Mutual Dependence in the Former Soviet Union

STEPHEN CROWLEY

Why, despite tremendous economic hardship, have industrial workers in almost every sector of the former Soviet Union remained quiescent? Why, despite the disintegration of the Soviet state, do the same trade unions, formerly dominated by the Communist Party and largely despised by their constituency, still hold a virtual monopoly on the representation of workers? As the former Soviet republics confront the daunting task of economic reform, few questions can be as significant as those concerning industrial workers, the predominant social group in this still very industrial society. What effect will privatization, and with it unemployment and the very probable deindustrialization of the economy, have on the former Soviet working class? How will workers respond? 

This study began as a look at a small question within a much larger event. The larger event was the 1989 Soviet coal miners’ strike, the first mass industrial strike in the Soviet Union in over sixty-five years. Over four hundred thousand miners in several regions and republics seized control of their mines, occupied city squares, and advanced broad economic demands. The small question was this: why, in such centers of strike activity as the cities of Donetsk and Novokuznetsk, did steelworkers remain at the mills? Or put in broader terms, not why was there so much labor unrest in the Soviet Union, but why was there so little? 

If the miners' actions had not continued beyond the first strike, the problem would not seem so intractable. But the miners formed a self-described workers' movement, and within two years, in March and April 1991, they renewed their strike with much more radical goals: they renounced their past economic demands and sought to lead a general strike aimed at forcing Gorbachev to resign, giving independence to the republics, and removing the Communist Party from power. Moreover, this radicalization occurred against a backdrop of political fragmentation and increasing economic hardship. The miners have since continued their militancy and strike activity, while steelworkers and almost every other category of industrial worker have hardly made a sound. 

This question of workers' collective action is addressed through a "most similar case study" of coal miners and steelworkers. Since these two groups are involved in heavy industry and are often located in the very same communities, many variables can be held constant [1]. Moreover, this contrast in behavior holds in two very different settings-Ukraine and Siberian Russia, located in different republics and ultimately different states, with sharply contrasting market positions-allowing for still greater explanatory power. 

The present study briefly enumerates alternative explanations for worker mobilization (which will be examined in detail following the case studies) and then examines how previous explanations of the relationship of workers to the Soviet state might account for the miners strikes. An alter-native perspective is advanced, that of the "mutual dependence" of the worker and the Soviet state enterprise. It is argued that due to the short-age economy the worker has been dependent on the state enterprise for the distribution of goods and services in short supply, while the enterprise has been dependent on workers in a labor-short economy. The distribution of goods and services through the workplace in the former Soviet Union ties workers to the firm and prevents collective action. Moreover, because the level of these benefits varies from industry to industry and from firm to firm, it can account for differences in strike activity. This thesis is examined in case studies of two steel plants, located in the center of the miners strikes. 

EXPLAINING STRIKES 

There has been no shortage of attempts to explain the presence or lack of working-class activism in certain groups [2]. The present case of mobilization of one group but not another within the same community leads one first to explanations rooted in industry. Kerr and Siegel's isolated community thesis has been the classic explanation for the higher level of strike activity in certain industries. The fact is explained by the "isolated communities" in which they live, where "the miners, the sailors, the longshoremen, the loggers ... form isolated masses, almost a ‘race apart’," and fail to form the cross cutting cleavages" which characterize the multi-industry town" [3]. A s we shall see, however, the opposite was the case: miners in multi-industrial towns were more radical and strike prone than miners in isolated mining communities. 

Another potential  explanation for collective action in certain groups is relative deprivation theory [4]. For this explanation, once the predominant interpretation, to be valid, we would expect miners' conditions to have objectively deteriorated relative to other workers and for miners to have experienced a greater subjective sense of deprivation. However, while their conditions are indeed quite bad, it will be shown that during this period of dramatic economic decline, steelworkers suffered significant deprivations and injustices, and the steel plants themselves were centers of serious conflict. 

Resource mobilization theory particularly as employed by Tilly and his collaborators, is more useful here [5]. This perspective holds that what is necessary is not simply the desire but the ability to carry out strikes and other collective action with some chance of success. A related approach has focused on technology and the resulting organization of work as a source of obstacles or advantages to workers' collective action. Thus, that "coal miners around the world are typically the 'aristocracy' of militant labor" is said to be due in no small part to the fact that in mining, the very nature of the labor process-such as the extreme danger and the close trust this engenders among work partners-promotes unity [6]. 

Conversely, the wide variety of skills required in a modern integrated steel mill appears to create obstacles to unity. These differences in skill extend both vertically and horizontally : job ladders create a greater hierarchy within the basic professions of steel manufacturing than in coal extraction, and each shop engages in different tasks, which means that workers often face very different production problems [7]. 

One further significant technological difference between coal mines and steel mills is size. While the workforce of a mine ranges in number from one to several thousand, the giant steel plants in the former Soviet Union can range from several thousand to over seventy thousand employees. Beginning with Marx, some have hypothesized that "the larger the factory, the more workers interact with one another, and the less they interact with their superiors" [8]. Others, to the contrary, cite the collective action problem in arguing that large plants deter solidarity [9].

We will return to each of these potential explanations in the final section of the paper. Here, however, one needs first to address the institution-al context of the Soviet political economy to understand why other workers in the now former Soviet Union have not joined the coal miners in the past four years. Western scholars have advanced several theoretical models to explain the labor peace in the post-Stalin era; each suggests a dissimilar explanation for the presence or lack of strikes in the former Soviet Union [10]. 

In the totalitarian view, workers have been deeply alienated, above all by the fear of repression that kept them from organizing to defend their rights. Consequently one would expect greater strike activity, since in the absence of state repression, workers can give voice to dissatisfaction held in for decades [11].  Yet strike activity and worker organization has been extremely low in most sectors, well after the downfall of the Communist Party and the Soviet state, suggesting the need for another explanation. 

The most popular explanation for labor peace in the post-Stalinist Soviet Union has been that of the "social contract" between the regime and society, of which industrial workers were among the main beneficiaries [12]. In contrast to the totalitarian model which has difficulty reconciling continued social peace with the decline in overt repression, the social contract model emphasizes voluntary compliance based on the state's ability to "deliver the goods." Political elites favored blue-collar workers, above all through policies of frill employment, roughly egalitarian wages, and the toleration of slack work rules; in return, workers provided the social basis of support for the state. 

The problems with such a conception soon become clear, however [13], in particular, with the often implicit assumption that Soviet industrial workers were conservative and supported the status quo. This conception does not accord well with the demands made by the striking coal miners [14]; nor with the demands of steelworkers who did not strike. Moreover, this assumption is grounded in the behaviorist premise that, with the lessening of overt repression, "social peace" must be explained by voluntary compliance and legitimacy; it misses hidden sources of conflict and the obstacles to translating consciousness into collective action.

Others have suggested an alternative approach, one that might be labeled “workers' control”. Burawoy and Stark, for example, using a lower level of analysis, both found sources of conflict despite the lack of strikes or other collective action [15]. They also built on the insights of Janos Kornai, who argued that the economic mechanism in state socialist societies that was able to create stunning rates of growth eventually produced widespread shortages that put a brake on further growth [16]. The continual shortage of the factors of production, including labor, created uncertainty in the labor process and in turn led management to cede control to core production workers. The labor shortage also removed the sanction of firing workers for breaking work rules or lowering productivity, furthering workers’ control  over the production process. 

Accordingly, it is not elite preferences but the economic system that creates labor and other shortages and thereby gives workers autonomy. If workers in a capitalist economy are atomized through labor market com-petition and the continual threat of unemployment, the situation in state socialist societies has been quite different. In Hirschman's terms, while workers in these societies have traditionally been denied "voice," they have used the possibilities of individual "exit," leaving jobs in a labor-short economy to find a better deal for themselves elsewhere [17]. 

MUTUAL DEPENDENCE 

Yet if workers hold such power resources, why have there not been more examples of collective action among workers in these societies, especially when voice became a real possibility? It is argued here that the worker in a state socialist society has been in a position of dependence, not simply as an individual dependent on the state, but as a working person dependent on the place of work-in particular the industrial enterprise - as the direct provider of one's basic life needs [18]. If a worker in capitalist society received a wage packet and health benefits, a worker in state socialist society received that and also housing, access to the enterprise hospital, to day care and other forms of education for one's children, often employment for one's spouse, trips to rest homes and vacation centers, and consumer goods ranging from automobiles to perishable food items. Such a set of ties can be a powerful disincentive to collective action, particularly when the distribution of these goods and services takes place largely at the discretion of management and when alternatives are few. 

This depiction of enterprise paternalism is not entirely new. Andrew Walder's account of "Communist neo-traditionalism" speaks of social and .economic dependence on the enterprise, political dependence on management, and personal dependence on one's superiors. Nevertheless, although his book has received well-deserved praise as a case study of industrial relations in contemporary China, it does not successfully make the case that his “type-concept” of Chinese industrial relations is among the "generic features of modern communism," [19] since unlike the Soviet Union and Eastern Europe, China remains an overwhelmingly peasant society. The implication is that workers have not faced excess demand for their labor; rather, they have experienced considerable unemployment. This may explain Walder's rigid model of domination, which offers little chance for subordinates to escape their plight: accommodating themselves to the sys-tem is the only option. More recently Walder has revised his position along the lines suggested here, arguing that because of the lifetime tenure of Chinese workers in state enterprises, management is dependent on them and hence must provide a high level of benefits in order to maintain productivity [20]. 

Indeed, in the Soviet Union this dependence has been not simply top-down, but also two-way, or "mutual": the enterprise is dependent on the worker in an economy that has created shortages of all inputs, including labor, and workers are dependent on the enterprise in that virtually all of their life needs are met through the workplace [21]. Workers have often used the taut labor market to obtain a better packet of enterprise benefits elsewhere [22], leading managers to acquire better goods and services in order to retain current workers and attract others. Not surprisingly, turnover occurs most frequently among young (and single) workers, in part because the spouse and children of married workers are also dependent on the enterprise, many seniority rules discourage turnover, and the wait for housing is several years at the least. Moreover, the goods and services at the disposal of the enterprise are distributed very unevenly, in order to retain those skilled workers most needed for production. In this way the provision of goods and services by the enterprise forms a distinct version of the internal labor markets found in capitalist firms [23].

Workers, I argue, are caught in a collective action problem: the same selective incentives that enterprise managers have used to prevent workers from acting individually and seeking work elsewhere can also be used to prevent workers from acting collectively [24]. While workers everywhere face the problem of collective action [25], workers in Soviet enterprises have run the risk that an unsuccessful strike would likely deprive the initiators not only of wages, but also of housing, day care, summer vacations, and the rest [26]. 

How were the miners able to overcome this dilemma? The distribution of these goods and services is also highly uneven between industries, depending on the preferences of planners and the ability of management to barter its product. In some industries and enterprises workers have been better provided for, while in others, such as coal mining, workers have had less to lose. Hence, the variation in the level of this enterprise dependence between industries and even firms can account for much of the variations in strike activity. 

STEELWORKERS 

The two steel plants examined here are located in the two epicenters of the miners strikes. The Soviet economy has been characterized as "a coal and steel economy" [27], and the two industries remain closely linked in the former Soviet republics. As a research strategy, it was hoped that the plants' locations, so close to the miners strikes, would more readily reveal the fault lines of conflict within the plants. The plants are also located in two very different regions - the Donbass and the Kuzbass  - which during the course of events examined here became part of different sovereign states. 

The plants themselves are quite different. The Donetsk Metallurgical Factory (or Donetskii) was established in the late 1860s by James Hughes, a Scottish entrepreneur.28Th e plant, like the other old plants in Ukraine, had been rebuilt and enlarged during the Soviet era but still relies on outdated open-hearth technology. Its current workforce numbers more than seventeen thousand, quite large by world standards though not so in the Soviet Union.29 The history of the West Siberian Metallurgical Complex could not be more different. Whereas blast furnaces have been fired at Donetskii for over 120 years, West Siberian was constructed in the 1960s and 1970s. This placed it on the cutting edge of the Soviet steel industry.30 Steel at the complex is manufactured mainly in oxygen converters rather than through open-hearth casting. The West Siberian Metallurgical Complex employs over thirty-two thousand workers. Because workers in both plants acted similarly despite these strong contrasts in size, age, technology, market position, region, and ultimately different states, the conclusions about the differences between steelworkers and coal miners are strengthened. 

The following case studies are based in part on in-depth interviews with steelworkers in both regions, both worker-activists and workers met at random, and in part, indeed largely, on a close reading of plant newspapers (mnogotirazhki) during the two-year period surrounding the miners strikes. The strikes had a strong effect on these plants, and along with the greater political changes developing throughout society, helped turn the papers into real forums for debate and even sources of conflict between management and the workforce. 

From the central press, one got the impression that steelworkers actively opposed the coal strikes, which directly affected steel production. In an interesting twist on working-class rhetoric, telegrams sent to striking miners from steel plants in 1989 and in 1991 appealed to "working-class solidarity" in asking them to go back to work.31 Closer inspection reveals a more complex picture, however. These plants were indeed conflict ridden, and the miners strikes shook them greatly. Strikes occurred at both plants, though they never grew beyond the level of a single shop. 

In fact the West Siberian complex experienced strikes several months before the July 1989 miners strike. The first strike was led by an independent "contract brigade, "which would have been paid handsomely for fulfilling its plan but was unable to do so because of the lack of materials. The demands quickly went beyond norms and supplies to questioning why certain categories of workers had rights to additional holiday pay, while others did not. This strike was followed a month later by another in a different shop with almost identical demands.

Still one month before the miners’ strike, workers in the rail transport shop presented demands to the administration that were very similar to those the miners would present, including additional pay for evening and night shifts, an increase in the regional wage coefficient, and reductions in the number of managerial personnel. And "to accelerate the resolution of the demands and to not allow them to be shelved," workers in the shop formed an "initiative group," an act taken as an implicit strike threat. 

When the first coal strike broke out in fill in Ukraine and Siberia, both plants held meetings and passed resolutions stating, "We support the demands, but reject the strike as a method." "Workers committees" were set up by managers, with the rather clear goal of extracting concessions from the Ministry of Ferrous Metallurgy. Workers tried to form genuine workers committees and to lead strikes, but their efforts were thwarted. 

Although neither West Siberian nor Donetskii (nor any other major industrial plant, for that matter) struck along with the miners in July 1989, these plants were hardly without conflict. Clearly, in contrast to the coal mines, the administration retained the upper hand in steel and other plants; the question remains how it did so. The explanation given most often by the participants themselves-whether managers, trade union officials, or worker-activists in favor of a strike-for the lack of strike activity has to do with the role of these enterprises not only in producing steel but also in reproducing their labor force, that is, in providing goods and ser-vices to satisfy virtually all the life needs of the plant's employees. Moreover, the distribution of these goods and services created the greatest source of conflict within these enterprises. 

ENTERPRISE DEPENDENCE 

Soviet industrial enterprises, in the face of shortages of consumer goods and services, as well as labor, developed a unique system for providing their workforces with services, from housing to food [32].  At the Donetsk plant, as with many others, the factory provided vacations at its centers on the Azov Sea, the Black Sea, and elsewhere, had a close relationship with a state farm, as well as a department for subsidiary agriculture. The enterprise also distributed such scarce durable goods as automobiles. By providing nurse-care for workers' children and payments for workers' funerals, the plant's services went quite literally from cradle to grave [33]. Indeed, the plant directly provides half of the services to its borough, including housing and transportation to and from work, so much so that it provided a base for the plant director's successful campaign for the borough seat in the Ukrainian Congress of Peoples' Deputies [34].

At West Siberian, the workforce was even more dependent on the enterprise for goods and services; located in harsh Siberia and so far from Moscow, the alternatives were that much starker. At the same time, these conditions increased the difficulty of attracting and retaining workers. Moreover, with its thirty-two thousand employees, the plant was twice the size of Donetskii, and its newness and relative profitability left greater resources at its disposal. If the Donetsk plant, the biggest in the city, had dominated the borough in which it was situated, West Siberian was itself the entire borough, unimaginatively named "Factory Borough."

The plant controlled several vacation centers and eight pioneer camps, some as far away as Central Asia and the Crimea. It had its own state farm, to which various shops sent gas and other materials, as well as workers at harvest time. As this was apparently insufficient to feed the workforce, the plant set up a rabbit farm and an aquaculture program. The entire operation was supervised by the steel plant's "deputy director for agriculture." Food production was further expanded when the plant contracted with a Moscow engineering cooperative to build a shop for producing sausage and other meat products at the plant. In addition to the plant cafeterias, which provided workers with their main meal of the day at subsidized prices, the plant had twelve stores selling foodstuffs inside the production shops and two more selling consumer goods. 

This description of the social infrastructure of these enterprises creates the impression that all the workers under the wings of the enterprise have been quite well provided for. But given the enormous size of the workforces involved, these benefits are rather less impressive when considered on a per capita basis. More importantly, they were not distributed equally, and therein lies a major source of conflict within the enterprise, one that workers challenged when the political opportunity to do so presented itself. This enterprise paternalism has been closely related to the labor shortage, with privileges distributed above all to retain the skilled workers most in demand. 

Thus, certain occupations were given privileges in the distribution of housing, consumer goods, such as cars and TVs, and vacation trips. Machine toolers and pipe fitters might be given apartments in five years, whereas steel founders and rollers had to wait ten to fifteen. This sort of uneven distribution was illegal; what was allowed was privileged distribution to exemplary workers as defined by management. Other workers greatly resented both forms of privileged distribution. 

The divisions were not simply between skilled and unskilled workers, but were embedded in a hierarchy that pervaded the life of the plant. If skilled workers were privileged because of the need to retain them, service workers peripheral to production were practically ignored. These latter were primarily women, who serviced the plant's enormous social infra-structure, as cooks, teachers, and janitors. Thus, in order to get housing, skilled cooks might leave the plant cafeterias to work in steel production, where shop bosses, themselves short of workers, were reluctant to let them return to their former professions [35]. 

Divisions also occurred between shops. The economic reform program under Gorbachev not only devolved some decision-making power from the ministry to the enterprise but also made each subunit "economically accountable."A s a result, pay and conditions began to vary widely between shops [36]. 

By far the shop in the worst position at Donetskii was the open-hearth shop, which was over a hundred years old. During the 1989 miners strike management only narrowly averted a strike in that shop. Because of outmoded technology, the plan was not being met, bonuses were withheld, pay declined, and workers left the shop for jobs elsewhere, creating a vicious cycle. The situation in the shop became so critical that workers in other shops were asked by the administration to "render aid" to the open-hearth workers to increase their pay, as if they were a Third World country [37]. 

In 1990, almost a year after the proposal to render aid, the plant's labor council (STK) met to discuss the tremendous problems of the open-hearth shop. The discussion revealed the interrelation of production, turnover, and "social" problems. Rather than production or turnover, "the discussion in the hall began with a no less important problem today, and one, unfortunately, very interconnected, as open-hearth workers more than once underlined, with the worsening of production in the shop. This is the housing problem." According to the steelworkers, "people are leaving production [work] because of the absence of housing, [since] the line at the factory stands for fifteen to seventeen years." After heated discussion, the council decided to grant additional privileges to steel founders and assistant steel founders (decreasing the wait to five years), since while "housing, undoubtedly, isn't the only cause [of the shop's problem], it's one of the main problems" [38]. 

Many workers opposed this method of solving the labor shortage through the privileged distribution of housing, and fought for a single line for housing. For if the distribution of housing was connected in the eyes of management with turnover, it was connected in the eyes of the workforce with social justice, which meant distribution according to a single line, rather than according to one's position in the labor market or to one's status as an exemplary worker (peredoviki). 

At West Siberian it was asked, "Why does belonging to a certain profession (machine tool operator, wire drawer, metal worker) become the basis for receiving an apartment at the front of the line? Why are rollers, steel founders, and members of other professions worse?"[39]. At West Siberian the fight for a single line for housing led to a protracted struggle within the enterprise, which divided workers-the privileged and not-as well as workers and management, but eventually the supporters of the single line prevailed. At Donetskii, the conflict remained one over which categories of workers should be admitted to the privileged housing line [40].  

Oddly enough, as market pressures increased, the steel plants began to devote more and more attention to activities outside of steel production. This was especially so at West Siberian, one of the newest and most profitable in the industry. In 1990 the plant contracted with a Turkish firm to construct a surgical wing for the borough hospital. With the help of a German firm and a Yugoslavian construction team, the plant began building a large furniture factory inside the steel complex, with the first output earmarked for workers. In July 1990 it was said that "the production of consumer goods is occupying an ever larger place in our plant's activities." The plant opened a shop for assembly of electronic devices, including VCRs. "This year the shop will produce 10,000 VCRs, and the next year 20,000" [41]. Since the local housing construction agencies had fallen apart for lack of materials and especially workers, the plant, with the help of foreign expertise, created its own construction firm [42]. 

Even in 1993, after the downfall of the Soviet Union, these steel plants continued to increase the provision of goods and services. Donetskii increasingly used barter deals to provide consumer goods to employees who could not afford the spiraling prices on the market, while West Siberian built a savings bank, a shoe factory, and a brewery at the plant, and also assured its employees that no one would be left without a job [43]. In all this, talk of investment to increase steel production-the plant's raison d'etre - was played down, though this would seem all the more compelling under market conditions. 

THE TRADE UNION 

In theory, workers had an array of organizations charged with defending their rights. Besides the Communist Party, they were represented by the trade union and the Council of the Labor Collective (STK). In reality, however, most workers saw these groups as part of the problem rather than as its solution. The striking feature of the Communist Party organizations in the enterprise during this period is how little they seemed to matter as they withered and disappeared. Though party activists continued to participate in plant meetings, the party committee itself began to withdraw quietly into the background at Donetskii, while at West Siberian it found itself increasingly on the defensive, as the Kuzbass miners began calling for removal of the Communist Party from all enterprises. When both were removed after the August 1992 putsch, their absence was not readily apparent. The Communist Party was not the only organization forced to justify itself in the face of the miners strikes and wider political change, the STKs-intended to increase workers' self-management-were also called into question. They spent most of their time handling the distribution of the plants' goods and services. As such, they failed to define a role for themselves distinct from the trade union. 

Indeed, it was the official trade union, still dominant at these and most other plants, that was most fully integrated into the system of enterprise dependence. For instance, at Donetskii, less than three weeks after the region's coal miners had first filled city squares and seized control of mines, the trade union committee met to discuss fulfilling the plan. The trade union chair announced, "In July there were twenty absences from work, twenty-eight violations of public order, and twenty-two cases of drunkenness.' The results of the latest round of socialist competition were announced: the winning brigades received red banners, a diploma, and a small monetary prize" [44]. 

But rather than dealing with production questions, the trade unions spent most of their time handling the distribution of the plant's goods and services. Even before the miners’ strike, one trade union official at West Siberian criticized the union in the following fashion:

 It's no secret that some trade union representative see their basic function as distributive, letting the main problems get out of control, which immediately leads to conflicts between workers and management.... In the course of many years, people began to relate to the trade union in a purely consumer sense. The distribution of this or that good ... became the chief indicator of the work of the trade union. The most important questions, such as workers' safety, pay and rate setting, fell to second place. 

After the miners strikes, the trade unions began to speak increasingly of defending the rights of workers. Yet, in practice this meant increasing the amount of goods handed out rather than worrying about safety, wages, and norms. For example, two months after the miners first struck, at a plant wide conference at Donetskii to elect a new trade union chair (the old one was reelected), a delegate from the casting shop raised some serious issues: due to the lack of steel, his shop had not fulfilled the plan for the second month, thereby cutting workers' wages drastically. This apparently did not concern the trade union, however. "As for the trade union committee," he added,

it has solved the problem of the family vacation center on the Azov Sea. I consider this a social victory for the trade union committee. And our children vacation at the pioneer camp "Metallurgist.". . . Now many are saying that the trade union should not be occupied with the problems of providing workers with potatoes, meat, soap. But who will take care of these problems? I consider this the trade union committee's job [45]. 

Others were not so sure [46]. As workers learned more about how goods and services were distributed within the enterprise, the trade unions became mired in scandal and workers came increasingly to resent those who were supposed to represent them. 

Over a year after the miners first struck, little had changed. One dele-gate to a congress of the steelworkers union, which had promised big changes, reported back: "It became clear, very clear, what trade unions do. Now, we must look truth in the eye, we devote ourselves to handing out tobacco and candy, distributing consumer goods in short supply and so forth-anything but defending the interests of workers." Another delegate said, "With such 'defenders,' life will be very difficult under the market" [47]. 

ANALYSIS 

Steelworkers did not join the miners in striking, forming independent trade unions, and pressing political demands on the government, I argue, because most industrial workers in the Soviet Union, and now in the successor states, have found themselves in a collective action problem. If all workers in a plant struck, they would all be better off, since they would be able to control the distribution of goods and services within the enterprise, much as the miners had done [48]. But if one or more workers tried to lead a strike and failed, they would likely lose access to the privileges distributed by management. 

As a steelworker from Magnitogorsk told an American historian: "We're completely dependent on them. Food, clothes, apartments, furniture, day care, summer camps, vacations-everything is allocated by them according to their lists, with which they rule over our lives. Everyone has something to lose " [49]. 

Explaining why the miners were the first to break out of this dilemma is not so hard. In many respects, strikes by the coal miners are over-determined. There is indeed something unique about the labor process in coal mining, with its dangerous underground work; and this very likely fosters professional pride and solidarity. But more important are the specifics of the Soviet coal industry. 

As far back as the late 1950s, when planners gave priority to oil and gas, “a brake was applied to the development of the coal industry” [50]. From this there emerged a whole series of problems, not least of which were the extremely hazardous conditions under which Soviet miners were forced to work. While mining everywhere is dangerous, the problem became particularly acute in the Soviet Union. Accident rates there were extremely high-much higher than in the next most dangerous industry, steelmaking; or as the miners put it, for every million tons of coal extracted, one miner paid with his life [51]. 

Declining investment created difficulties not only with working conditions, but also with "social problems," like housing, day care, and the rest. Miners' living conditions were often appalling: in the Kuzbass and Vorkuta, for instance, miners and their families often lived in barracks, some of which were left over from Stalin's gulags. 

Moreover, the problems in the Soviet coal industry only deepened with the first attempts at economic reform. The introduction of full khozraschet, or self-financing, was intended to increase management's concern with profits and losses; but since it was not accompanied by price reform, it created a perverse effect, particularly in the coal industry. In order for the administrative prices in the Soviet system to balance, state purchase prices for raw materials had to be set extremely low, so that the prices of producer and consumer goods could also remain low, even after the prices reflected some of the added value. This meant that the price a mine received for one ton of coal-the official wholesale price-was roughly one-half the cost of extracting it, making the coal industry the only "planned-loss" branch in the country [52]. Since under khozraschet bonuses and social expenditures were intended to come out of plant profits, the already austere living conditions of the Soviet coal miners were strained even further.

Although this seems to be a case of relative deprivation, there are problems with such an argument in this context. First, the miners themselves felt the changes in investment patterns only very gradually. Second, in the case studies of the steel plants and in interviews with steelworkers met randomly, the sense of anger, frustration, and deprivation is quite clear. The very same issues raised by coal miners are raised forcefully by steelworkers again and again [53]. 

More importantly, miners have been the highest paid of any category of industrial worker; at the time of their first strike they were making almost twice as much as the average steelworker. Due to decreased investment in social infrastructure, however, coal mines had fewer goods and services to distribute to their workforce. While industrial-level data on social infra-structure does not exist, regional-level data shows this to be the case, as in the following graphic statistic: the Kemerovo region of Siberia (Kuzbass), Russia's largest coal basin, ranked thirteenth in 1989 in industrial production in rubles, even at artificially low prices for coal, but was forty-third in the provision of housing, fifty-eighth in children's establishments, and eighty-ninth for social clubs.54I n an examination of housing units built in 1979 in twenty-eight large Soviet cities, Donetsk, the coal capital of Ukraine, was twenty-third; it was greatly outpaced not only by the Ukrainian cities of Kiev and Kharkov but also by the nearby metallurgical center of Dnepropetrovsk, which came in first out of all twenty-eight cities [55]. 

Thus, steelworkers were compensated with low wages and high in-kind benefits, making them more dependent, while coal miners were compensated with a low level of goods and services but high wages, giving them greater autonomy. The strategy some workers pursued was to begin their career as a highly skilled steelworker, get an apartment quickly on a privileged basis, and then to leave for the coal mines, getting both the apartment and the high wage [56].

How did compensation with high levels of in-kind benefits create greater dependence than high wages in rubles? First, when workers are tied to a job through a paycheck alone, this "cash nexus" becomes a fragile connection because it is a single strand: a disruption in wage levels can quick-ly transform quiescent workers into militant ones [57]. Second, unlike many of the benefits distributed through the enterprise, wages can be saved to be spent later, for example, when scarce goods became available or during a strike. Third, wages have been easier to replace, even in the event of losing one's job, than in-kind benefits. Due to the labor shortage, workers fired for disciplinary reasons were able to find new employment soon thereafter, in many cases at higher wages than in their previous jobs.58 But for a work-er who had invested years waiting for an automobile or an apartment and was nearing the top of the list, loss of the position in line would be a terrible blow [59]. 

Alternative arguments remain, in particular, those pertaining to the technology of production and the resulting organization of work. Arguably, if the labor process in mining fosters solidarity, the workforce in a modern integrated steel mill is more stratified-along job ladders and across shops-lessening the potential for solidarity. While this argument appears forceful, there are reasons to question several of its assumptions. First, the steel industry is highly strike prone in some countries but not in others [60], suggesting that such factors as unions, the state, and cultural and other institutions are more important than technology. Moreover, as Katherine Stone has argued with reference to the origins of job structures in the American steel industry, there was actually a "lack of important skill differentials between the jobs in steel-making"; and furthermore, those job ladders and classification schemes that did arise were largely artificial constructs of management, rather than the result of technological imperatives [61]. 

In the present case, moreover, the divisions that occurred among workers were most salient for the workers themselves when the privileged distribution of goods and services was involved. The in-kind benefits distributed within the enterprise greatly facilitated the creation of an internal labor market in several ways. This privileged distribution allowed managers to respond to a taut labor market, for skilled labor especially, despite the often rigid central determination of wage structures. Further, while workers often knew what other workers were receiving in their monthly paycheck, the privileges entailed in the distribution of benefits according to lists were a mystery. In fact, the sharpest conflicts at these two steel plants during this time occurred when workers first tried to apply" openness" (glasnost)t o the system of privileged distribution and then tried to remove those privileges. Hence, many of the demands raised by steelworkers, as well as miners, concerned not higher wages and benefits but information on the economics of their enterprise and the system of intra-enterprise distribution. Even as management attempted to use the goods and services at its disposal to respond to the taut labor market, it tried to hide that fact from a workforce whose sense of justice includes the notion of pay according to labor.

One further significant difference between the coal mines and the steel mills examined here is size. The lack of full-scale strike activity at these huge plants would seem to be a confirmation of the thesis that large plant size is an obstacle to collective action. Indeed, in terms of its workforce a mine is more comparable to a single shop than to an entire steel complex. Stated this way, however, the explanation is insufficient. While the difficulties in communicating and organizing activity between the shops of these large plants would certainly be significant, this was no less an obstacle than that originally faced by miners in separate mines. Yet these miners combined: while only several individual mines struck in the spring of 1989, by July the Mezhdurechensk miners had set off a ripple effect throughout the Soviet Union [62]. By contrast, individual shops struck in the two steel plants studied, both before and after the miners, yet they failed to spark similar actions in other shops. 

Typically missing from discussions of plant size and collective action is the corresponding size of management in these large plants. According to one steelworker:

We're all in one group, in one place, with one boss [kulak] ... if some kind of revolt begins in one shop, then the administration tries to somehow pacify it, and somehow frighten workers with its actions, so there is no solidarity. Here I'll give one example from my own shop: when the shop announced that there would be a strike, then the administration literally immediately got together and there was a leaflet sent around, that participants in the strike will suffer this, and this, and this punishment: lose their wages, lose subsidies for their children, then all the privileges that workers get [63]. 

As we have seen in the cases of the two steel enterprises, management - at both the shop level and the enterprise level-intervened often to maintain control over the plant and its workforce. And yet, as the last part of the above quotation indicates, the issue was a question not simply of management domination but also of the levers that it wielded: the privileges extended to workers to prevent them from acting individually (to take advantage of the taut labor market to exit) can also be used to prevent them from acting collectively. Once again, we return to the importance of mutual dependence. 

One might argue that we are left with partial explanations. Even the "most similar cases" of coal mining and steelmaking still leave many differences in the work experiences of the two groups, which may account for differences in strike activity. How can one untangle these alternative explanations? Fortunately, for present purposes, the most similar comparative case method can be pursued even further to include counterexamples - coal mines that did not strike.

Participation during the 1991 strike was spotty: some mines joined in the middle of the strike, others dropped out, and still others produced coal for local use with the blessing of the strike committee.t5 There were also mines that did not strike at all. Though the mines that did not strike were often quite different from one another, they all shared a common characteristic: the enterprises were all linked to a trading partner in such a way that a strike would cut off the provision of vital goods or services to the workforce. 

Three types of mines fall in this category. One group of mines less prone to strike, as described above, consisted of those that made up the very "isolated communities" that were expected to be most strike prone. These mines often had close links with nearby state or collective farms that pro-vided foodstuffs to the workforce. Had the mines struck, they would have violated their contracts with these farms and in turn would have lost their source of food supplies, especially hard to obtain in industrialized coal basins. 

A second group of mines that did not strike were those, more often located in urban areas, which, either through rich geological endowments or through managerial skill, had created links not only with state farms but with producers of consumer goods. These arrangements typically involved barter deals that provided scarce and valuable goods, including VCRs and foreign cars. 

One such mine, Zasiad’ko, is something of a local legend in the city of Donetsk.65 During the 1991 strike, in an article entitled "Why the Zasiad'ko Mine Is Not Striking," a coal face worker explained how the mine, through the work of its director, had developed "an entire trade and industrial complex" that included contracts with trade organizations for clothes, shoes, furniture, and automobiles, while barter deals provided the miners with cassette players, VCRs, and televisions at subsidized prices. To transport all these goods, the mine had obtained its own truck depot, and to improve export prospects, it had joined with other enterprises in purchasing a freight ship.66A s for striking, the face worker explained: 

We ... well understand that if we shut down the mine, we will worsen our material position, since no one will pay our wages.... There will be no means to maintain the sanitoria, meat and vegetables will disappear, housing construction will stop, the day care centers will close. What's more, our trading partners will demand our forfeiture. We will suffer enormous losses [67].

A third category of mine that did not participate in the 1991 strike included those controlled by the miners themselves, as a result of their earlier strike. Some of these miners had created joint-stock companies with foreign partners, which could promise a relatively high standard of living for all employees. One miner-activist, Yurii Gerol'd, managed to achieve this for his mine. Gerol'd was elected chair of his mine's STK after the July 1989 strike, although he had only just started working there as a foreman. Young and forceful, he was immediately elected deputy chair of the Kuzbass regional strike committee and helped prepare the protocol, signed by the miners and the government commission, that ended the strike. He was later elected co-chair of the Confederation of Labor, as well as a people's deputy of the Russian Federation, the latter on a program of "enterprise independence" and end to "the alienation of the working person ... from the product of his labor" [68]. 

By March 1991, when the Kuzbass miners were striking to bring down the Soviet government, Gerol'd resigned from the city and regional strike committees and the Confederation of Labor (although he remained a people's deputy).H is mine, Polosukhinskaya, continued to operate. Remaining true to his principles of creating independence and ending labor's "alienation," at least in one mine, Gerol'd had sought and helped find a foreign partner for a joint-stock company, which promised a relatively high standard of living for all of Polosukhinskaya's employees. Participation in the strike would have meant an end to the joint-stock arrangement. Thus, Gerol'd and his colleagues were faced with a choice between maintaining solidarity and remaining part of the workers' movement, or keeping their relative prosperity and independence-the final goal, in the miners' eyes, of their political strike against the Soviet government. 

The common denominator for each of the three categories of mines was a contract arrangement with horizontally linked trading partners, which provided the mines' employees with goods and services at levels significantly higher than those of the average mine. Of course, miners who did strike sacrificed a good deal. But while striking miners were provided with a strike fund, however meager, these contract arrangements provided miners with goods and services that a strike fund could not buy [69]. 

Thus, even across firms in the same industry, dependence on the enterprise as the distributor of vital goods and services has remained a strong inhibitor of strike activity and workers' collective action. Can the lack of a high level of enterprise dependence account for strikes in industries other than coal mining? While a definitive answer lies outside the scope of the present study, a look at the industries and sectors where strike activity has occurred suggests that this may indeed be the case. As we have noted, the provision of goods and services at the disposal of an enterprise varies greatly across enterprises as well as across industries. While some strike activity has occurred in sectors that control a valuable product which can be sold for high prices on the market-for example, among oil and gas workers and gold and silver miners-most strike activity, outside of coal mining, has occurred among transportation workers (railroad and public transport workers, dock workers, merchant marines) and white collar employees (doctors, teachers, and air-traffic controllers) [70]. While there may be alter-native explanations for strikes among these particular workers, all lie out-side the traditional heavy industrial sector, where worker collective action could be expected to occur and where enterprise goods and services have been concentrated. 

Just how generalizable is the concept of "mutual dependence" to other former communist societies? The dependence of the worker and the state enterprise is a product of the shortage economy, itself a result of economic centralization. Thus, where the economy has been more centralized, such as in Bulgaria or Romania as well as the Soviet Union, one would expect this mutual dependence to be stronger than in a relatively less centralized economy, such as in Hungary or Poland [71]. Moreover, shortages, especially in the consumer sector, were alleviated to a greater extent in some Eastern European societies through private farming, toleration of petty entrepreneurship, and the shorter legacy of Communist Party rule. This independent economic activity also extended to the social realm, where independent institutions created greater social space. Hence, workers in Poland were able to overcome dependence on the workplace and act collectively due to greater alternatives and the space provided by such institutions as the Catholic church [72]. 

What are the implications of this system of enterprise distribution? First, it is time to reconsider some of the conceptions used by Sovietologists to explain the relative labor peace and the compliance of workers and other groups in Soviet society. The totalitarian perspective emphasized the domination of the Communist Party, but as we have seen, many of the institutions within state enterprises have survived the down-fall of the Communist Party and, indeed, of the Soviet state. On the other hand, the antithesis of the totalitarian perspective, the social contract approach, assumed that the quiescence of labor, in the absence of terror, signified voluntary compliance, largely because the distribution of goods and services throughout Soviet society appeared to favor blue-collar laborers. Yet we have also seen that the distribution of these goods and services, especially through the industrial enterprise, was itself the source of great conflict. 

The concept of "mutual dependence" has been proposed here to explain the relationship of the industrial worker to the Soviet enterprise. Although we have emphasized the dependence of the worker on the enterprise in order to explain why steelworkers and others did not strike, the other side of this equation should not be underestimated. It was the shortage of workers that compelled managers to increase the supply of housing and other benefits at their disposal. And while these emoluments may have reduced turnover and prevented strikes, managers, faced with such a short-age and the need to meet the plan, were much less able to control workers on the shop floor. Conversely, as the labor shortage disappears and is replaced by a labor surplus, the relative position of workers will decline further. It is far from certain that workers will be compensated for the loss of the exit option with a gain in their ability to use their voice. 

Thus, the Soviet institution of distributing social services through the workplace will very likely affect the course of reform efforts, and the concept of mutual dependence suggests some hypotheses concerning the economic and political transformation in the former Soviet Union. One of the major obstacles to the privatization of the state sector remains the absence of viable substitutes for the enterprises now supplying vital services to a large part, perhaps a majority, of the population. To cut credits to these enterprises, forcing many to close, would deprive their workers not simply of wages and benefits (which could be replaced temporarily at least by unemployment payments), but also of access to housing, health care, child care and important sources of food and consumer goods as well. Other alternatives are not currently feasible: private providers of such services would have to charge prices beyond the reach of their potential customers; and local governments are reluctant to take on more burdens, as they are already dependent on taxes from these very enterprises for even the paltry services they now provide. Thus one hypothesis would be that privatization (and the closure of unprofitable enterprises) will be most difficult in the Soviet-style company towns, where one or more enterprises provide virtually all social services to the urban population [73]. 

Moreover, the system of enterprise distribution also explains the continued domination throughout the former Soviet Union of the once official trade unions. Because the trade unions are the distributors of housing, automobiles, and televisions, as well as pensions and sick pay, workers have been afraid to sign a form stating they wish to leave the old trade union [74]. Thus, these trade unions have survived the collapse of the Communist Party and the state, even though they are generally abhorred by their constituency. They hold a virtual monopoly on workers' representation and are attempting to form corporatist bargains in their name with the state [75].  Therefore, a further hypothesis would propose that the former state trade unions will continue to flourish until the state(s) finds an alternative mechanism for performing the social-welfare functions currently handled by the trade unions (with the corollary that alternative trade unions, in order to compete for members, will have to perform these same functions) [76].

The dissolution of the Soviet Union and the collapse of the central allocation economy has affected the paternalism of state enterprises in contradictory ways. Some enterprises, deep in debt and unable to obtain sufficient credits, have simply abandoned many of the former services (such as vacation centers), leaving workers to fend for themselves. Other enterprises, such as those in the coal and steel industries examined here, have actually increased their provision of goods and services in the face of continued shortage, spiraling prices, and the inability of local governments to provide additional services [77]. 

Why might managers continue to expend valuable resources on procuring goods and services for the workforce, especially when market pressures are increasing and the threat of unemployment is replacing the labor shortage? Walder has recently argued that in China, despite considerable market reform and high unemployment, the position of state enterprise manager remains as much a political position as an economic one. The enterprise itself comprises a "socio-political community' that places considerable pressure on managers from below. Thus, managers must be concerned not only with maximizing profits, but also with "enhancing employee income and . .. delivering a wide range of other benefits and services to their employees" [78]. 

In the former Soviet Union such positions have become even more directly political. As noted above, the director of the Donetskii steel plant was elected to the Ukrainian Supreme Soviet, joining many other industrialists who have formed powerful blocs in the Ukrainian and Russian parliaments [79]. While striking miners in most cases removed their old directors and elected new ones, Donetskii's Sledenev used his plant's largesse to get himself elected political representative from the borough. He has since used his new political position to seek contacts, both foreign and domestic, for expanding the economic base on which his power rests [80].

This ability of managers to maintain political control of the enterprise also helps explain why most directors prefer privatization plans that give large numbers of shares to the labor collective. Managers clearly believe that far from leading to direct worker ownership, this is the best way for them to remain in command [80]. This leads to a further hypothesis: current plans for "employee ownership" through the purchase of majority shares in joint-stock companies will give workers de jure control of the enterprise but will remain a formality, just as did the Gorbachev-era law on electing enterprise managers. (To this should be added the corollary that, just as the coal strikes transformed the enterprise election law into a source of real power for the miners, enterprises experiencing strikes under such an ownership system could result in de facto workers' control) [82]. 

In the longer term, the future of enterprise dependence is limited, even if it is difficult to determine when the longer term might arrive [83]. If credits to state enterprises are eventually cut off, leading to bankruptcies and full-scale unemployment, it would eliminate the main reason for enterprises to supply workers with goods and services in the first place-to attract and retain skilled workers in a taut labor market. At the same time, unemployment and the uncertainties of a market economy would likely create new obstacles to, as well as opportunities for, workers' mobilization. 

If the concept of mutual dependence is historically limited, it is theoretically limited as well. Structural and rational-actor perspectives, on which the concept of mutual dependence has drawn, have been helpful in explaining how individuals are or are not able to overcome obstacles to collective action. Yet the rational actor approach-assuming preferences as static and indifferent to ideology - is rather less helpful in explaining what occurs once individuals do act collectively. For example, such individual-level, rational choice analysis does not account for how the coal miners, once mobilized, transformed their objectives from supporting "perestroika from below" to trying to bring down the Soviet state. But that is a topic for another paper.

* Research for this article was made possible by an SSRC/IREX Fellowship for Soviet Sociological Research. The author would also like to thank the following for their suggestions and comments: William Bianco, Michael Burawoy, Paul Christensen, Zvi Gitelman, Jerry Hough, Michael Kennedy, Herbert Kitschelt, Anna Temkina, and William Zimmerman.

NOTES

[1] Why choose steelworkers even though their work regime is not entirely similar to that of miners? As mentioned, they are located in the same regions and communities. Second, the two production processes are linked, since coking coal is a major component of steelmaking, steelworkers were thus directly affected by the coal strikes, as well as more generally by the economic downturn. More importantly, steelworkers, like miners, were at the core of the once privileged Soviet proletariat, comprising a major part of the country's coal and steel economy. If the miners were to form a "workers' movement," as they professed to be doing, they understood that steelworkers had to play an integral role.

[2] For a particularly useful review of the literature, see Gordon Marshall, "Some Remarks on the Study of Working Class Consciousness," Politics and Society 12, no. 3 (1983).

[3] Clark Kerr and Abraham Siegel “The inner-industry propensity to strike”, in Kerr, ed., Labor and Management in Industrial Society (Garden City, N.Y.: Anchor Books, 1964), 109, 111; Seymour Martin Lipset, Political Man: The Social Bases of Politics, enlarged ed. (Baltimore: Johns Hopkins University Press, 1981); David Lockwood," Sources  f Variation in Working-Class Images of Society," Sociological Review  14 (November 1966). This perspective has met with a good deal of criticism, for its highly structural nature, for its assumption that radicalism develops in isolation rather than in interaction with other perspectives and ideas, and for the fact that the inter-industry propensity to strike has varied greatly over time. However, the Soviet case would seem to provide another example supporting the thesis of the isolated community. Paul K. Edwards, "A Critique of the Kerr-Siegel Hypothesis of Strikes and the Isolated Mass: A Study in the Falsification of Sociological Knowledge," Sociological Review 25 (August 1977); Edward Shorter and Charles Tilly, Strikes in France, 1830-1968 (Cambridge: Cambridge University Press, 1974), esp. 287-95; James E. Cronin, "Theories of Strikes: Why Can't They Explain the British Experience?" Journal of Social History 12, no. 2 (1978-79); Howard Kimmeldorf, Reds or Rackets? The Making of Radical and Conservative Unions on the Waterfront (Berkeley: University of California Press, 1988), 13-15. For recent support of the isolated community thesis, see the discussion in Albert Szymanski, The Capitalist State and the Politics of Class (Cambridge, Mass.: Winthrop, 1977), chap. 3.

[4] Ted Robert Gurr, Why Men Rebel (Princeton: Princeton University Press, 1979).

[5] Shorter and Tilly (fn. 3); Tilly, From Mobilization to Revolution (Reading, Mass.: Addison- Wesley, 1978).

[6] Michael Yarrow, "The Labor Process in Coal Mining: The Struggle for Control," in Andrew Zimbalist, ed., Case Studies in the Labor Process (New York: Monthly Review Press, 1979), 187; see also Shorter and Tilly (fn. 3), 13.

[7] Kerr and Siegel (fn. 3), 110.

[8] Szymanski (fn. 3), 63; see also Lipset (fn. 3), 252. For Marx, see "Revolution and Counter- Revolution in Germany," in Selected Works (Moscow: Progress Publishers, 1969). For some empirical support, see Richard F. Hamilton, Affluence and the French Worker in the Fourth Republic (Princeton: Princeton University Press, 1967), 205-28.

[9] Shorter and Tilly (fn. 3), 15. Russell Hardin, Collective Action (Baltimore: Johns Hopkins University Press, 1982), 222-23.

[10] See the discussion of alternative views on Soviet industrial relations in Peter Rutland, "Labor Unrest and Movements in 1989 and 1990," Soviet Economy (December 1990), 193-95.

[11] Joseph Godson and Leonard Shapiro, eds., The Soviet Worker (London: MacMillan, 1981).

[12] Walter Connor,  The Accidental Proletariat (Princeton: Princeton University Press, 1991); Linda J. Cook, "Brezhnev's' Social Contract' and Gorbachev's Reforms”, Soviet Studies  44, no. 1 (1992); Janine Ludlam," Reform and the Redefinition of the Social Contract under Gorbachev," World Politics 43 January 1991); Peter Hauslohner,"  Gorbachev's Social Contract," Soviet Economy 3 , no. 1 (1987); George Breslauer, "On the Adaptability of Soviet Welfare-State Authoritarianism," in Erik Hoffman and Robin F. Laird, eds., The Soviet polity in the modern era (New York: Aldine, 1984), Stephen White, "Economic Performance and Communist Legitimacy," World Politics 38 (April 1986).

[13] Beyond the fact that this was a contract that was never negotiated, let alone ratified, by the two parties, there is a deeper problem. Within this conception, there is no theory of the power of the social groups, such as workers, that could explain why they were the "Winners" in this bargain, other than "the regime's de facto preference for certain groups' interests over others." Hauslohner (fn. 12), 59.

[14] Rutland (fn. 10); Theodore Friedgut and Lewis Siegelbaum, "Perestroika from Below: The Soviet Miners' Strike and Its Aftermath," New Left Review, no. 181 (Summer1 990); Stephen Crowley, "From Coal to Steel: The Formation of an Independent Workers' Movement in the Soviet Union, 1989-1991" (Ph.D. diss., University of Michigan, 1993).

[15] See Michael Burawoy and Pavel Krotov, "The Soviet Transition from Socialism to Capitalism: Worker Control and Economic Bargaining in the Wood Industry," American Sociological Review 57 (February1 992); Burawoy and Janos Lukacs, The RadiantP ast: Ideology and Reality in Hungary's Road to Capitalism (Chicago: University of Chicago Press, 1992); Burawoy, The Politics of Production (London: Verso Press, 1985); David Stark, "Organizational Innovation in Hungary's Emerging Mixed Economy,"  in Stark and Victor Nee, eds., Remaking the Economic Institutions (Palo Alto, Calif.: Stanford University Press, 1989); idem, "Rethinking Internal Labor Markets: New Insights from a Comparative Perspective," American Sociological Review  51, no. 4 (1986), 492-504. 16 Kornai, The Economics of Shortage( Amsterdam: North Holland, 1980). See the summary of his work in David Stark and Victor Nee, "Toward an Institutional Analysis of State Socialism," in Stark and Nee (fn. 15), 16-20.

[17] Albert Hirschman, Exit, Voice and Loyalty: Responses to Decline in Firms, Organizations and States (Cambridge: Harvard University Press, 1970). For a slightly different application of this concept, see Stark and Nee (fn. 16); and George Bergsten and Russell Bova, "Worker Power under Communism: The Interplay of Exit and Voice," Comparative Economic Studies 32 (Spring 1990).

[18] Such paternalism is not unique to this particular time and place. The company town of American history and the current Japanese enterprise are but two other examples. There are important differences, however. In the Japanese context, for instance, the employee has access, depending on the wage level, to a range of alternatives through consumer markets. Given the shortages in state socialist economies, such alternatives have been greatly limited. See also Andrew Walder, Communist Neo-Traditionalism: Work and Authority in Chinese Industry (Berkeley: University of California Press, 1986), 15-17.

[19] Walder (fn. 18).

[20] See Andrew Walder, "Factory and Manager in an Era of Reform," China Quarterly, no. 118 (June 1989), esp. 249-53. It should be noted, however, that because of continued unemployment rather than a labor shortage, Chinese workers have not been able to leave one job for another, a major strength of workers in the enterprises examined here. Thus, Chinese workers have had neither exit nor voice, though they have used what Scott calls "the weapons of the weak."James C . Scott, Weapons of the Weak. Everyday Forms of Peasant Resistance (New Haven: Yale University Press, 1985).

[21] This mutual dependence is also suggested by the term "plan-fulfillment pact." See Ulrich Voskamp and Volker Witkke, "Industrial Restructuring in the Former German Democratic Republic," Politics and Society 19 (September 1991); Burawoy and Krotov (fn. 15).

[22] V. Kalmyk and T. Sil'chenko, "Sotsial'no-ekonomicheskaia obuslovennost' otnosheniie k mestu raboty," in E. Antosenkov and V. Kalmyk, eds., Otnoshenie k Trudu i Tekuchest' Kadrov (Work attitudes and labor turnover) (Novosibirsk. Institut Ekonomiki i Organizatsii Promyshlennoyo Proizvodstva, 1970).

[23] See Paul Osterman, ed., Internal Labor Markets (Cambridge: MIT Press, 1984); Robert Althauser and Arne Kalleberg,  "Firms, Occupations and the Structure of Labor Markets," in Ivar Berg, ed., Sociological Perspectives on Labor Markets( New York. Academic Press, 1981); Peter Doeringer and Michael J. Piroe, Internal Labor Markets and Manpower Analysis (Lexington, Mass.: Heath, 1971).

[24] Hardin (fn. 9); Michael Taylor, The Possibility of Cooperation (Cambridge: Cambridge University Press, 1987).

[25] Regarding workers' collective action specifically, see Adam Przeworski, "Marxism and Rational Choice," Politics and Society 14, no. 4 (1985); Claus Offe and Helmut Wiesenthal, "Two Logics of Collective Action: Theoretical Notes on Social Class and Organizational Forms," in Maurice Zeitlin, ed., Political Power and Social Theory, vol. 1 (Greenwich, Conn.:  JAI Press, 1980); James Johnson, "Symbolic Action and the Limits of Strategic Rationality: On the Logic of Working-Class Collective Action," in Maurice Zeitlin, ed., Political Power and Social Theory, vol. 7 (Greenwich, Conn.: JAI Press, 1986);J on Elster, Ulysses and the Sirens (Cambridge:  Cambridge University Press, 1984); Colin Crouch, Trade Unions: The Logic of Collective Action (London: Fontana, 1982).

[26] Indeed, the stakes are quite high on both sides. Even as the miners clearly gained through their strikes, managers did not lose profits; they did lose their jobs, however, as the miners often removed the old bosses and elected their own.

[27] Connor (fn. 12), 137.

[28] Theodore Friedgut, Iuzovka and Revolution, vol. 1 (Princeton: Princeton University Press, 1989).

[29] While open-hearth furnaces produced only 7 percent of the steel in the United States in 1985 and none were being used in Japan and West Germany, the Soviet Union continued to rely on them for more than half of its steel production. Boris Rumer, Soviet Steel: The Challenge of Industrial Modernization in the USSR (Ithaca, N.Y.: Cornell University Press, 1989), 64.

[30] Rumer (fn. 29), 54.

[31] For instance, the workers' council of the Donetskii plant wrote to ask the miners to return to work because" the working class of steel workers" would otherwise lose their pay and privileges. Kuzbass, July 18, 1989.

[32] In a chapter entitled "New Cities: The Politics of Company Towns," William Taubman refers to the more than one thousand cities built in the Soviet Union since the 1917 revolution, most of which "have been born and raised as Soviet-style company towns, in the shadow of one industrial establishment or with several establishments dividing responsibility or competing for control." These enterprises provide "housing and whatever meager services" there are. See Taubman, Governing Soviet Cities: Bureaucratic Politics and Urban Development in the USSR (New York: Praeger,1 973), 54.

[33] Metallurg, August 19, 1989.

[34] Director  Sledenev was subsequently elected to the higher legislative body, the Ukrainian Supreme Soviet.

[35]  Metallurg Zapsiba, July 11, 1989.

[36] Burawoy and Krotov (fn. 15) provide evidence of shops being pitted against each other in the wood industry (pp. 25-28).

[37] Metallurg, July 26, 1989.

[38] Ibid.

[39] Metallurg Zapsiba, October 11, 1989.

[40] Ibid., November 23, 1989; author interview with AR and BM, deputy chairs of the trade union committee at Donetskii, Donetsk, July 2, 1992. Why did workers at West Siberian, where management had more resources, succeed in establishing a single line while those at Donetskii failed? Managers may have had more resources, but workers had greater countervailing resources as well. First, the labor shortage was more severe in Siberia than in Ukraine. And second, West Siberian was potentially profitable on the market, giving workers a positive incentive to organize, whereas at Donetskii everyone realized that the plant was going to remain dependent on state subsidies.

[41] Metallurg Zapsiba, July 13, 1990.

[42] Ibid., February 10, 1990.

[43] Interview with AR and BM (fn. 40); Metallurg Zapsiba, September 4, 1993. Walder (fn. 20) notes that managers in China have behaved in the same fashion, increasing rather than decreasing their efforts to provide benefits to their workforce in the face of market forces (pp. 249-53).

[44] Metallurg, August 5, 1989.

[45] Ibid., October 25, 1989.

[46] Ibid., for how another delegate to the same conference put it: Frankly speaking, it pains me that you have organized this conference as if it were ten to fifteen years ago. We don't talk about the sore points here. We won't revolt, if we find out that apartments are being given to Afghan war invalids ahead of the line, that a vacation trip in short supply [defitsitnaya] is being given to a steel founder, a furnace worker or a rolling mill operator. But we demand the just distribution of social goods. We need glasnost here, so that all will know, who obtained what.... I think we've ruined this conference.

[47] Metallurg Zapsiba, October 10, 1990. The Central Council of the once official steelworkers union has since declared itself independent from the central trade union confederation (FNPR) and has allied itself with the independent miners union. Galina Koval'skaya, Svobodnie Profsoyuzi Rossii (Free trade unions of Russia) (Moscow: Allegro Press, 1993), 8-12. But this was a decision made by the top union leadership in Moscow, on the level of the enterprise, little has changed.

[48] Were the miners actually better off after striking, or were steelworkers actually smart to avoid striking, with all its attendant problems? While the miners' successes were certainly uneven, they clearly gained from their strikes in several fundamental ways. First, they gained greater control over their workplace. Second, they gained greater political and economic changes, first in the Soviet Union and later in the newly independent states. Other workers have been represented by the former state trade unions or management. Finally, the material position of miners, while hardly ideal, has improved, at least relative to that of other workers. In addition to greater control over intra-enterprise benefits, wage increases have outpaced even the astronomical rates of inflation in Russia and Ukraine, so much so that subsidies to the coal industry alone (before recent efforts to raise coal prices in Russia) were absorbing 20 percent of Russian state revenues and accounted for nearly 33 percent of the state budget in Ukraine. Izvestiia, May 8, 1993, cited in RFE/RL Daily Report, May 11, 1993; Moscow News, June 18, 1993.

[49] He continued, "Perestroika, you say?I have a family, children, grandchildren. They want to go away for the summer. What can you say? Daddy told the boss he didn't like how he was being treated, so this year and for the next couple of years no one is going anywhere." Stephen Kotkin, Steeltown, USSR (Berkeley: University of California Press, 1991), 28-30.

[50] "Travma," Ekonomicheskaya Gazeta, no. 7 (1989), 17.

[51] Kuzbass, July 13, 1989; Argumenty i Fakty, no. 30 (1989).

[52] Viktor Kostyukovskii, Kuzbass: Zharkoe leto 89-ogo (The Kuzbass: The hot summer of 1989) (Moscow- Sovremmenik, 1990), 26. Subsidies to the branch in 1988 were reportedly 5.4 billion rubles and in 1989 a billion more.

[53] Reflecting on this very question, Burawoy produces a complex argument to explain why steel-workers did not follow the example of coal miners. First, as basic goods producers in a supply-con-strained economy, miners had been privileged by the state but lost their privileged position with perestroika. Second, steelworkers were able to avoid a similar decline by shifting their product profile, thus getting higher state purchase prices for their goods. Third, miners, owing to their unique work regime, experienced an exaggerated form of "workers' control" over production, which contrasted more sharply with their exploitation by the state. Burawoy, "The End of Sovietology and the Renaissance of Modernization Theory," Contemporary Sociology 21 (November 1992), 780-81. This argument breaks down on each point, however: the steel industry in the former Soviet Union has suffered dramatically during the overall decline in production; changing one's product profile is no longer valid given the lack of state orders and the glut on the steel market; and the miners' political militancy toward the state was a product of their strike activity and subsequent organization rather than a precursor to it. On this last point, see Crowley (fn. 14).

[54] Kostyukovskii (fn. 51), 62. Mines were doubly disadvantaged in terms of social infrastructure. First, a mine has a definite life span. Unlike an industrial factory that theoretically can be continually modernized, there is less incentive to sink capital into permanent infrastructure such as housing or cultural and educational facilities in a mining settlement. Second, many mining communities are indeed isolated, and while this in itself does not breed radicalism as the isolated community thesis maintains, it does mean there is less social infrastructure than in a larger city. Indeed, the differences in the living conditions for miners and steelworkers in the Donbass were evident from its initial industrialization over a hundred years ago. See Friedgut (fn. 28).

[55] The numbers are even worse for Donetsk when the number of marriages and thus presumably the number of new housing applicants is compared with the number of new housing units built; then Donetsk was twenty-fifth. Henry Morton, "The Contemporary Soviet City," in Morton and Robert Stuart, eds., The Contemporary Soviet City (Armonk, N.Y.: M. E. Sharpe, 1984).

[56] Metallurg Zapsiba, November 7, 1989.

[57] John H. Westergaard, "The Rediscovery of the Cash Nexus," The Socialist Register 1970 (London: Merlin Press, 1970), esp. 120-21. Thus miners in Vorkuta, of Russia's Far North, were drawn from other parts of the Soviet Union in the hope that savings from high wages would allow them to buy a home and retire in Russia's temperate South. Inflation immediately wiped out those plans. See "In Russia's Far North, Inflation Destroys a Dream," Moscow Times, March 3, 1993.

[58] Connor (fn. 12), 172.

[59] Thus, although important, wages represented only one component of an overall compensation package. Both steelworkers and miners demanded wage increases. The Hungarian experience suggests that the preference for wages versus in-kind benefits varies with the scope of the consumer market. In Hungary many years of economic reform created a more advanced market for consumer goods and ser-vices, and consequently wages were valued above enterprise emoluments. See Stark (fn. 15, 1986); Burawoy and Lukacs (fn. 15). Given prices out of the reach of many consumers, this has not yet occurred in Russia.

[60] John E. T. Eldridge, Industrial Disputes: Essays in the Sociology of Industrial Relations (London:  Routledge and Kegan Paul, 1968).

[61] Stone, "The Origin of Job Structures in the Steel Industry," Review of Radical Political Economics 6 (Summer 1974), 156.

[62] On the creation of a ripple effect as a way out of the collective action problem, see John Chamberlin," Provisions of Collective Goods as a Function of Group Size," American Political Science Review 68 (June 1974), 707-16.

[63] Author interview with Oleg Semyenov, Novokuznetsk, May 3, 1991.

[64] In this last category, some mines provided coal to local metallurgical and power plants. The phrase that some mines were working "with the blessing of the strike committee" may have been a way for both the particular mines and the strike committee to save face given the less than full participation in the strike.

[65] For more on the Zasiad'ko mine, see Stephen Crowley and Lewis Siegelbaum, "Survival Strategies: The Miners of Donetsk in the Post-Soviet Era"( Manuscript,1 993). The following paragraph is largely drawn from that paper.

[66] Verchernyi Donetsk, April 17, 1991.

[67] Ibid. A Siberian mine with a similar history and paternalistic scope is described by Petr Biziukov and Simon Clarke, "Privatization in Russia: The Road to a People's Capitalism?" Monthly Review (November 1992).

[68] Author interview with Yurii Gerol'd, Moscow, February 27, 1991; Nasha Gazeta, February 20, 1990.

[69] There were no cases that I found, in the central, regional, or local press, or during fieldwork, where miners broke such a significant contract arrangement to strike.

[70] Linda Cook, "Labor's Response to the Soviet Post-Communist Transition" (Paper presented at the annual meeting of the American Political Science Association, Chicago, September 1992).

[71] Regarding a different social consequence of the shortage economy, see Katherine Verdery, "Nationalism and Nationalist Sentiment in Post-socialist Romania," Slavic Review 52 (Summer 1993), 182-83. Trade unions in Bulgaria and Romania have been transformed from above, without wide-spread collective action (outside mining and a few other sectors) from below.

[72] A full accounting of the rise of Solidarity is dearly beyond the scope of this paper. Among the other factors present in Poland and not in the Soviet Union were nationalism and the experience of previous worker uprisings; all contributed not only to workers uniting but also to workers uniting with other classes. See Roman Laba, The Roots of Solidarity: A Political Sociology of Poland's Working-Class Democracy (Princeton: Princeton University Press, 1991); Michael Kennedy, Professionals, Power and Solidarity in Poland (Cambridge: Cambridge University Press, 1991). In Hungary, where little collective worker activity occurred even as shortages decreased as a result of market reform, workers continued to pursue individual strategies in seeking to take advantage of the consumer market-a reminder that the market also presents obstacles to workers' collective action.

[73] One clear analogy with the steel enterprises discussed here are the company towns of early capitalism. Ironically, however, whereas enterprise dependence in the former Soviet Union appears to be a case of a bloated welfare state, it was the establishment of the welfare state in capitalist societies that finally ended the worker's near total dependence on the employer. By guaranteeing a minimum standard of living regardless of one's work performance, the welfare state ended the employer's direct control over the reproduction of the labor force. See Burawoy (fn. 15), 125-26.

[74] To call such items "selective incentives" is misleading in at least two respects: they were originally intended not to retain members in the trade union, but rather to keep workers in the enterprise, and they have become the primary service these organizations provide. On the term "selective incentives" as applied to trade union membership, see Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups (Cambridge: Harvard University Press, 1977); and the critical discussion in Crouch (fn. 25).

[75] On the corporatist bent of these trade unions, see Cook (fn. 70).

[76] According to the official trade union chair at one Donbass mine, "At first the NPG ... pushed aside health issues, daily life concerns, and all the rest. But now the NPG takes care of everything up to trade and the distribution of foodstuffs, that is, those things for which the NPG leaders always cursed us." Pozitsiia, May 6-12, 1992, p. 1. The same appears to be true in Vorkuta. See Michael Burawoy and Pavel Krotov, "The Economic Basis of Russia's Political Crisis," New Left Review, no. 198 (March-April 1993), 60-64.

[77] This was certainly the case in Donetsk in the summer of 1992 and Novokuznetsk in the fall of 1993. This is especially so since the barter economy provides workers in privileged sectors with scarce consumer goods rather than high wages. See Crowley and Siegelbaum (fn. 65). 78 Walder (fn. 20), esp. 249-53. 79 As of this writing it is too early to tell whether the latest Russian elections have sent fewer enterprise managers to parliament. Preliminary analysis by the author suggests that managers have indeed used their resources to elect, if not themselves, then at least like-minded representatives.

[80] Interview with AR and BM (fn. 40). The director of the Zasiad'ko mine mentioned above first used his economic position to be elected a member of the Ukrainian parliament, was later elected mayor of Donetsk, and from there was appointed prime minister of Ukraine.

[81] Simon Clarke," Privatization and the Development of Capitalism in Russia," New Left Review, no. 196 (November-December 1992).

[82] Indeed, as the miners' experience shows, the ability of managers to control such institutions is not absolute. Thus managers must carefully balance a paternalism that on the one hand ensures their continued control and on the other hand does not provoke concerted action by workers that could remove them from office. 83 The work of economic historian Douglass North makes one cautious about predicting the rapid demise of such institutions, no matter how revolutionary the change in economic structures or even property rights. See North, Institutions, Institutional Change, and Economic Performance (Cambridge: Cambridge University Press, 1990).

This paper was published in: World Politics, Vol. 46, No. 4 (Jul., 1994), pp. 589-615.