Workers Control: Its Structure Under Allende

This essay is reproduced here as it appeared in the print edition of the original Science for the People magazine. These web-formatted archives are preserved complete with typographical errors and available for reference and educational and activist use. Scanned PDFs of the back issues can be browsed by headline at the website for the 2014 SftP conference held at UMass-Amherst. For more information or to support the project, email

Workers Control: Its Structure Under Allende

by Andrew Zimbalist 

‘Science for the People’ Vol. 5, No. 6, November 1973, p. 26-28

The following articles, solicited on short notice, come from two observers, holding academic positions in the U.S., who spent part or all of the past year in Chile. Both articles deal with an important component of the then advancing Chilean Revolution: workers’ control in the factories. This phenomenon resulted from two parallel and frequently joint processes: 1) moves by the government against private owners or managements which were uncooperative toward, or active saboteurs of, the government’s economic policies, and, 2) the actions of rank-and-file workers who seized factories because for them that was the concrete meaning of collective ownership of the means of production.

“Workers’ Control: Its Structure Under Allende” is based on a survey of 40 enterprises in the socialized sector. It describes the organizational changes and kinds of worker participation that developed within enterprises, often in spite of the government guidelines proposed for these situations. The article points up the difference between industrial democracy from below as was beginning to happen in Chile, and job enrichment from above as is being introduced ln the U.S. by some managements to increase output.

The second article, “At The Side of The Workers”, is a personal account by a physicist who worked alongside Chilean workers in the metallurgical factory they had seized. It presents (somewhat romantically) concrete examples of the problems that arise in developing relations between production workers and technical or professional workers. In the course of dealing with real production problems, such as the maintenance of deteriorating machinery, or interpreting design drawings, the vast potential of worker participation and control is illuminated and the technical professional discovers how much he must also learn.

These articles do not deal with the difficult problem of maintaining and defending such cooperative gains in a country where the working class is not in full control, and the embittered, money-dispossessed are actively sabotaging the revolutionary gains. The political reality is that on September 11 the global oppressors could smash, unretarded, the fruits of the people’s struggle. That setback demands that the political and organizational lessons of Chile’s struggle also be learned.

Also unanswered in these articles are questions as to the nature of the gains that were realized in the workplace. The “technical proficiency campaign” to which Maurice Bazin addresses himself is in itself a consciousness raising experience. But it is not clear from his article how the workers perceived their role in a larger struggle and whether this was a subject of discussion and political education in the factory. Clearly socialization of the factories was not complete, but to what extent was this due to the unpreparedness of the Chilean working class, or the power of foreign imperialism, or remnants of the administrative bureaucracy, and how did the workers approach these problems? Although we know that priorities of production were modified, it is not clear how or to what extent workers determined what they would produce.

We must also ask ourselves how we can contribute to this struggle? How does the scientific and technical work-force employ its revolutionary potential? Bazin engaged his commitment to the revolutionary process in the factory. Working at the side of the Chilean laborer he attempted to translate his technical learning into pragmatic tools which might advance the possibilities for real industrial democracy. By what other processes did professionals in Chile channel their skills and energy into the revolutionary tide? And the problem that presses us at home: what can be the role of the professional in factories which are not socialized, amongst a work force which is not so politically astute and hardly so organized as the forces which militated for their own liberation in Chile?

The author of the following article spent the past year in Chile working on a survey of 40 randomly selected socialized factories, in collaboration with the Chilean Notional Labor Organization (now abolished). The article sketches the emergence and experience of worker’s management in these enterprises.

Some 275 enterprises passed into the socialized sector of Chile’s economy during Allende’s 34 months as president. These firms were managed by an administrative council theoretically composed of 5 worker representatives (3 blue-collar, 1 technical person and 1 professional person), 5 state representatives and 1 state appointed administrator. In practice, only a handful of the largest enterprises actually followed this scheme. In the great majority of these “socialized” firms the worker representatives had more than 50% of the seats on the administrative council. Where the worker representatives didn’t control the council outright, it was the case that several of the state representatives were chosen amongst the workers of the factory in question. Indeed, in many instances the workers themselves chose their administrator.

At the shop floor level worker run production committees were formed. These committees dealt with production issues in their sections. Suggestions and information flowed from the shop floor to the administrative council through the coordinating committee. The latter body was run by the president of the largest union. (Typically, the Chilean firm had 2 unions, one white-collar and one blue-collar. During Allende’s period there was a tendency for these two groups to merge and form a single class-unified union.) The heads of the production committees and the worker representatives on the administrative council also attended the meetings of the coordination committee.

Aside from the task of presiding over the coordinating committee, the union leaders were not allowed to serve as worker representatives on the production committees or on the administrative councils. This decision to separate union and worker management functions was based on a dual proposition. One, so long as the State itself was not completely controlled by the working class, the working class needed to maintain an independent basis for its self-defense and for advancing the class struggle, viz., the union. If the union assumed management functions, then it would cease to effectively lead the workers in the class struggle. Two, the Chilean union, although rarely as co-opted and undemocratic as the U. S. union, has historically concerned itself with economistic demands and has tended to develop a paternalistic and bureaucratic style. Such an organization was clearly not optimal to develop industrial democracy.

The theory behind the separation of union and management powers was sound, but it created contradictions in practice. Specifically, the more bureaucratic unions (those controlled by the Communist Party and many of those controlled by the Christian Democratic Party) opposed the formation of parallel institutions to represent the workers, i.e., the organisms of workers’ control at the firm level. In these instances, the participatory apparatus was either not set in motion or it failed to effectively develop.

Workers tended to comprehend and participate immediately in union related issues, e.g., safety, security, wage levels, work rules, etc. Other administrative matters such as hiring, firing, scheduling and training courses were also quickly grasped and dominated.

Soon after participation blossomed in the above areas, workers would actively make recommendations and decide upon the physical ordering and reorganization of production — job transfers, job rotation or expansion, maintenance of machinery, quality control, provision of raw materials and spare parts, the selection and modification of technology, etc. In the past, foremen were discouraged by their employers from accepting suggestions from the workers. This gave the worker a sense of ignorance and impotence regarding his or her work environment. With socialization, either the foreman was eliminated or he would assume a different role.

Workers and former foremen generally agreed that the production worker knew more about the machinery than either his or her immediate or remote boss. Participation on this level, thus, liberated an enormous creative energy. It is not surprising then, that in 32 of the 40 socialized factories that we surveyed worker productivity either stayed the same or increased (it increased at a rate of over 6% per annum in 14 firms.) This record is all the more impressive given the raw material shortages and generalized macro-economic problems of the Chilean economy during this period. Without diverting resources from productive uses, socialized factories rapidly expanded social services available to the worker: plant medical facilities, day care centers, cafeterias and consumer cooperatives, athletics fields, libraries and so on. In several factories cultural departments were created. These departments sponsored factory folk groups and theater troups that visited other factories singing and acting the story of the expropriation of their factory and their social liberation.

In virtually all the socialized factories wage differentials were drastically reduced. The ratio of the highest to the lowest wage would typically have been 30 or 40 to 1 beforehand. Now, it varied from 5 to 1 to 11 to 1. This result along with increased production constitutes an anomaly to bourgeois economic theory which maintains that differentiated incentives are necessary to motivate efficient economic performance. To the contrary, the reduced wage disparities in Chile’s political context produced less invidious division and more solidarity within the working class. The resulting cooperative spirit motivated the productivity gains.

Matters of financial and economic management — investment financing, production plans, cash flow strategy, pricing policy, etc. — were the most difficult to decentralize. Only in the most advanced factories did the worker representatives knowledgeably and energetically participate in related decision making. More often, the technical and professional staff were relied on for these matters. In most cases, middle and upper management remained on the job after socialization of the firm and cooperated more or less enthusiastically depending on their political position. Production problems tended to emerge more frequently where the middle and upper management were in opposition to Allende.

Consistent with the above line of argument, we found production to be positively correlated with participation. The extent of participation, in turn, was a function of the system of communication and information flow within the factory, the style and political ideology of middle and upper management, the degree of labor mobilization (for instance, whether the firm was socialized by decree or by a workers’ take-over) and the union and political party structure in the enterprise. The most important of these variables, we found, was the level of political consciousness and activity amongst the factory’s workers. Variables such as technological intensity (capital-labor ratio), technological complexity (percent of personnel preferentially dedicated to maintenance), technological type (artisan, machine-tending, mounting or assembly, assembly line and continuous process), size of firm, vertical or horizontal integration, average educational attainment were poor predictors of the extent of participation.

The Chilean experience reconfirms the desirability of industrial democracy. Capitalists in the United States are discovering the profitability of “job enrichment” and “team” production. While the latter, at least in the short run, tends to enhance the workers’ identification and satisfaction with the job, it should not be confused with industrial democracy. Ultimately, employer introduced “job enrichment” programs will only co-opt the pressure and undermine the perspectives for true democracy. If, however, issues as job rotation or expansion, speed rates, working conditions are raised from below and the “job enrichment” is implanted and directed by the workers, then workers’ control begins to be a real alternative.


>>  Back to Vol. 5, No. 6  <<