PRODUCTION

It is an abuse of words to apply the same term "discipline" to such unrelated concepts as the mindless reflex motions of a body with a thousand hands and a thousand legs, and the spontaneous coordination of the conscious political acts of a group of men.

-Rosa Luxemburg

Now, as to occupations, we shall clearly not be able to have the same division of labor as now: vicarious servanting, seweremptying, butchering, letter carrying, boot-blacking, hairdressing, and the rest of it, will have come to an end ... we shan't put a pattern on a cloth or a twiddle on a jughandle to sell it, but to make it prettier and to amuse ourselves and others.

-William Morris

Workers' Councils

The first step toward establishing nonhierarchical work is to establish workers' councils. Every workplace is governed by a workers' council in which each worker has one vote. Smaller councils are organized for divisions, units, and work teams as circumstances dictate. All issues are ultimately subject to majority rule in the workers' council, but this does not preclude establishing more refined procedures where different degrees of consensus make sense. In fact, forging a workplace decision making structure that allows people to influence decisions in proportion to the degree they are affected would almost always require more refined decision making than simple voting since not all workers are equally affected by all workplace decisions. By leaving decisions that only affect a subgroup of workers to those workers and their councils, by assigning initiative to those most affected by matters, and by weighing voting to reflect differential impacts, democratic workers' councils would fashion their own best approximation to self-management.

We claim neither that conceiving and agreeing on the most appropriate arrangement will be free of dispute within actual workers' councils, nor that any particular arrangement will be universally applicable. The first point is simply that meetings of workers' councils where each worker has one vote would be the final arbiter just as stockholders' meetings where each owner votes as many times as shares she or he owns are the final arbiter in privately owned enterprises. And the second point is that in a situation where each worker has an interest in selfmanagement and none has disproportionate power, it is not unreasonable to hope that workers' councils will strive for decision making structures and ways to delegate authority that accord with selfmanagement rather than establish hierarchies.¹

But even procedures that poll council members in proportion as they are affected may not ensure collective self-management. If some people hold authoritative jobs while others only obey, even in formally democratic meetings people who hold more authoritative and information-rich jobs will exert more influence. Beneath formal democracy, a hierarchy of managers can still dominate deliberations.

Democratic councils, therefore, do not by themselves ensure full participation. Additionally, the organization of work must guarantee that all members of councils are equipped to express their desires and opinions.

Work Organization

We know that not all work is equally desirable and that even in formally democratic councils rote workers will lack the information, skills, and energies to participate equally with conceptual workers. Democratic councils help foster participation and equity but something must equalize daily work assignments so the impact of people's work experience does not destroy equity and self-management. If some have greater information and responsibility to use to their advantage to dominate policy making, they can become a ruling "coordinator class," both thwarting workers' participation and monopolizing desirable work roles.

Balanced Job Complexes

The remedy is conceptually simple: If you work at a particularly unpleasant or disempowering task, you should spend some time working at other more pleasant or empowering tasks as well. Pleasant and unpleasant work, rote work and conceptual or administrative work must be balanced. People should not do one type all the time. To foster participation and equity people must be assigned to a balanced mix of tasks. Which is not to say every person must perform every task in every workplace. This is impossible and unnecessary. The same person need not work as a doctor, an engineer, and a literary critic, and those who assemble cars today need not assemble computers tomorrow. Nor should everyone who works in a hospital perform brain surgery. The point is simply that people should rotate in some reasonable time period through some sequence of tasks for which they are adequately trained so that no one enjoys consistent advantages over others.

However, we are not merely suggesting that doctors occasionally clean bed pans or that secretaries attend an occasional seminar. Parading through the ghetto does not yield scars and slinking through the country club does not confer status. Short-term stints in alternative circumstances do not rectify inequalities in basic responsibilities. Likewise, for those doing elite work in one workplace to do rote work in another does not challenge mystification, deference, and authoritarianism in either workplace. Only primary job complexes balanced for desirability and empowerment can give all workers an equal chance of participating in and benefiting from workplace decisions.

Since disparate empowerment destroys participatory potentials and creates class differences that in turn lead to inequity, we focus on empowerment without implying that balancing to rectify differential desirabilities is not also important. But balancing for empowerment is more complicated and solutions are also applicable to balancing for equity. Likewise, for the reasons noted above, our initial focus is each individual's primary responsibility. Is it a number of different rote tasks, a number of different skilled tasks, or a combination of diverse tasks in a single package balanced with others in the unit for empowerment?

Whereas capitalist and coordinator jobs combine tasks with the same qualitative characteristics so that each worker has a homogeneous job complex and most people do one level of task, in contrast, participatory jobs combine tasks into balanced job complexes where everyone does many levels of tasks. Everyone works at a particular bundle of diverse tasks, but all bundles prepare people to participate as equals in democratic workplace decision making.

While one might question the efficiency of organizing workplaces and industries this way, it is certainly theoretically possible. Formally, for any economy E:

1. We construct the set of all possible single types of productive activities, {PA}.

2. We construct the subsets of "practically identical production activities" of {PA}, {PIPA}ⁱ where each member of each {PIPA}ⁱ has essentially the same qualitative effect on workers as any other member of that subset, and a different effect than any. member of any other practically identical subset,{PIPA}^j so that:

i. The set of all practically identical activities subsets of {PA} constitutes a partition of {PA}.

ii. The subsets{PIPA}ⁱ can be ordered along a spectrum of qualitative effects on workers ranging from empowering to debilitating.

The critical practical issue in creating this spectrum of subsets of activities is how finely to measure "qualitative effects." If we measure too precisely, each {PIPA}ⁱ will have only one element. If -we measure too flexibly, each {PIPA}ⁱ will be huge. Both partitions would be useless. In practice, experience and special evaluative boards would settle this matter. In theory, there is no problem.

3. We construct the menu of available job complexes by combining different productive activities from different subsets of practically identical production activities to form balanced job complexes so that the menu of available job complexes includes only offerings that incorporate a mix of tasks which together have comparably empowering effects.

Balancing Across Workplaces

In balancing job complexes within each workplace for equal empowerment, the goal was to prevent the organization and assign ment of tasks within a workplace from preparing some workers better than others to participate in decision making. But balancing job complexes for empowerment within workplaces does not guarantee that work life will be equally empowering across workplaces, any more than balancing job complexes for desirability within workplaces guarantees that working in different industries will be equally desirable. Establishing conditions for a participatory, equitable economy requires overall balance in addition to balance within each workplace. Clearly, to guarantee equal opportunity to participate both within units and in the economy as a whole requires internal as well as overall balance. Strictly speaking, overall balance for desirability is sufficient for work equity. But since, in practice, this will usually be easiest to achieve by arranging internal balance regarding desirability and then correcting for differences between workplaces, the conclusion that a participatory, equitable economy requires approximate internal and external balance for both empowerment and desirability is a useful rule of thumb.

The only way to balance for desirability and empowerment across workplaces is to have people spend time outside their primary workplace. How can an economy accomplish this? How can it calibrate balance? For that matter, how do people come to work in one or another workplace in the first place?

In a participatory economy, everyone will have the right to apply to work wherever they choose, and every workers' council will have the right to add any members they choose. We must wait until after explaining participatory allocation to analyze when and why workers' councils would wish to add or release members, but for now it is sufficient to know that once the plan is finalized, each workers' council will have a list of openings for which all can freely apply. So any worker can apply for any opening, and move to a new workers' council that wants them should they prefer it to their present council.

While this resembles traditional labor markets, much is different. Under traditional labor markets people generally change employment to win higher pay or enjoy more desirable working conditions. But if a participatory, equitable economy balances job complexes across as well as within workplaces, and if equitable consumption requires "equal pay for equal effort," people will be unable to attain these traditional goals by changing workplaces. This is not to say people will never change workplaces in a participatory, equitable economy. If a person would prefer a different group of workmates, or working at a different combination of tasks, she or he would have good reason to apply for membership in a different workers' council. But to the extent job complexes are balanced, these personal preferences will be the only motives to move, and, conversely, the freedom to move will provide a check on the effectivneess of efforts to balance for equity across workplaces. Higher pay will not be available by changing jobs, nor will objectively better work conditions, since pay and conditions will be balanced.

Just as workers must balance complexes internally through a rating process, so will delegates of workers from different councils and industries develop a rating process to balance across workplaces. There will thus be "Job Complex Committees" both within each workplace and for the economy as a whole. The internal committees will combine tasks and assign work times to achieve balanced work complexes within workplaces. The economywide committee will assign workers in less desirable and less empowering primary workplaces offsetting time in more desirable and empowering environments, and, vice versa, it will also assign workers in more desirable and empowering primary workplaces offsetting time in less desirable and empowering environments. The check on internal calibration will be whether more members of a workers' council want one assignment or another. The check on overall balance will be excessive applications to change workplaces. Economywide committees will adjust rotation times to eliminate significant discrepancies in people's applications to work in different workplaces.

Balance in Practice

Creating balanced job complexes is theoretically possible. And if we combine balanced job complexes with democratic councils, we would have nonhierarchical production relations that promote equity and participation. But in practical situations, could workers really rate and combine tasks to define balanced complexes within and across plants?

Provided we are talking about a social process that never attains perfection, but does fulfill workers' own sense of balance, the answer is yes. The idea is that workers within a plant would engage in a collective evaluation of their own circumstances. As a participatory economy emerged from a capitalist or coordinator past, naturally there would be a lengthy discussion and debate about the characteristics of different tasks. But once the first approximation of balanced complexes within a workplace had been established, yearly adjustments would be relatively simple. If the introduction of a new technology changed the human impact of some task, thereby throwing old complexes out of balance, workers would simply move some tasks among affected complexes or change the time spent at different tasks in affected complexes to attain a new balance.

The new balance need not be perfect, nor the adjustments instantaneous, nor would everyone need to agree completely with the result of a democratic determination. As a matter of fact, individual preferences that deviate from one's workmates would determine who would choose to apply for which balanced complexes. If I am less bothered by noise but more bothered by dust than most, I will apply for a complex whose rote component is attending noisy machinery rather than a complex with a sweeping detail.

Participation Versus Efficiency

We have argued that balancing work complexes for empowerment is a necessary condition for guaranteeing all actors equal opportunity to participate in economic decision making. But many believe that this would destroy efficiency. In our own view, the relation between participation and efficiency will more likely be positive than negative, but in any case it is more complicated than contemporary "common wisdom" asserts.

One prevalent view has it that some people make decisions better than others, so that the only efficient arrangement is for them to make all important decisions. In other words, participatory decision making is inherently inefficient since it gives everyone decision making input proportional to the degree she or he is affected even though some people make decisions better than others. Of course, this was the same argument used to support the rule of kings against demands for popular representation, the rule of "wise" and "benevolent" dictators against demands for democracy, the rule of men against demands for women's suffrage, and the rule of a "vanguard" party against demands for political pluralism. Do those who accept the wisdom of universal suffrage, democracy, and political pluralism do so at the expense of efficiency? Does some distinction between political and economic decisions warrant popular participation in the former but not the latter?

Human beings have the potential to make conscious choices in light of our understanding of the consequences of alternatives . Particular social environments can subvert or suppress this potential in the majority as when political dictatorship, patriarchy, dogmatic religions, slavery, and authoritarian education stifle people's decision making potentials. But one important goal of desirable social institutions is that they develop rather than thwart our most creative potentials. If we assume that noneconomic institutions develop these potentials, and if we choose economic institutions that do likewise, it is reasonable to assume participants would be capable of making economic choices in light of predicted consequences.

This is not to assert that everyone is equally knowledgable about every decision, or to deny a role for "expertise" in economic decision making. We often need "experts" to interpret the consequences of choices and to explain the likely implications of possible decisions. If experience shows a greater role for expertise in economic decisions than in political or other decisions, so be it. But once consequences are known because ordinary people have had the opportunity to hear diverse expert opinions, what remains is for the affected people to register their choices. While those with expertise in a particular matter may well predict consequences more accurately, those affected know best whether they prefer one outcome or another. In other words, making economic choices entails both determining consequences and evaluating them. While efficiency requires an important role for experts in determining complicated consequences, efficiency also requires that those who will be affected decide which consequences they prefer. Therefore, it is just as inefficient to keep those affected by decisions from making them
after experts have analyzed and debated consequences, as it is to prevent experts from explaining and debating consequences of complicated choices before those affected register their desires. In sum, for informed self-management, there is no conflict between participation and efficiency.

Still, there is a sense in which guaranteeing the conditions necessary for establishing a fully participatory economy might be at the expense of efficiency. If a specific skill is scarce-either because an essential innate talent is only present in a fraction of the work force, or because training required to develop the skill is time consuming for trainees or trainer and less pleasurable than alternative human activities-then balancing work complexes for empowerment by assigning someone with this talent or training to work at an activity that requires a less scarce skill would be inefficient. Even so, assignments of this type might be less inefficient than any humanly feasible alternative. For example,

1. If always working at the same task proved so monotonous and boring that people's concentration, effort, and performance deteriorated, balancing might be more efficient than not balancing.

2. If failure to balance was deemed unfair, and resentment led to deteriorating performance on others' part, it is possible for balancing to be more efficient than nonbalancing.

3. If working at a complex including a number of tasks provides an overview of how different tasks depend on one another thereby enhancing the intelligence of people's efforts in their primary responsibilities, then even for the most skilled balancing might be more efficient than not.

4. If performance is positively correlated with participation, and participation is positively correlated with balancing, balanced complexes might prove more efficient than unbalanced complexes.

Besides a remarkable propensity not to consider these possibilities, the traditional view of the relation between job-mixing and efficiency rests on questionable assumptions regarding talent and training. In the traditional view, many productive talents are assumed present in only tiny fractions of the population, while most training is assumed terribly burdensome. Since doctors changing bed pans is an example those opposed to job balancing like to cite, it is not inappropriate to point out:

1. Ample experience demonstrates that it is possible to train only moderately good students to become physicians, so the "talent" necessary to become a doctor is apparently distributed among a rather large fraction of the population, and the inefficiency of having doctors change bed pans could be greatly reduced by training more doctors.

2. Protests of medical students aside, a substantial portion of medical training could be highly illuminating and have a positive "consumption" component so that the time occupied in the training process need not be as burdensome as the AMA. would have us believe.

3. There is an implicit assumption that while it is inefficient for doctors to change bed pans, it is not inefficient for others to change them. However, while it is true that in societies where a large number of perfectly talented members of the work force receive no training, the opportunity costs of having them change bed pans are obviously less than the opportunity costs of doctors doing so, this would not hold if everyone received ample training in some area for which they had relevant talent. If all have their particular talents trained, there would be significant opportunity costs no matter who changed bed pans. As to whether it is more efficient for an economy to train all in accord with their talents, or more efficient to train the productive talents of only a few, the answer is obvious.

Choice of Technology

A new technology is superior if it yields greater net social benefits than existing ones. The simplest kinds of superior technologies use less of some scarce input without using more of any other input, or yield more of some desirable output without yielding less of any other output. These "pure input saving" and "pure output enhancing" techniques are probably not the most common superior technologies, although interestingly enough, new organizations of production that increase opportunities for workers' participation can often be counted among them. However, most new technologies use less of some inputs, but more of others, or generate more of some outputs but less of others. In these cases, whether the new technology is superior depends on whether the social cost of the inputs saved outweighs the social cost of the inputs used, or whether the social benefit of the outputs increased outweighs the social benefit of the outputs lost.

Whether the information available and incentives for those who choose technologies in different kinds of economies can be relied on to lead them always and only to implement superior techniques is, of course, of great concern. John Roemer presents an interesting model in which capitalist economies fail to achieve these goals whenever the rate of profit exceeds the growth rate.³ Moreover, numerous proponents of the "conflict theory of the firm" have suggested that capitalists will not always choose superior techniques for reasons additional to the one Roemer analyzed. ⁴ And we have clarified the logic of why capitalists, even under competitive conditions, will reject superior techniques if they are sufficiently worker empowering. ⁵ On the other hand, our analysis of coordinator economies concludes that choice of technique in those systems will also be biased against worker empowering technologies. ⁶ While the full explanation of why we believe our participatory economy will implement superior techniques must wait until we present our allocation procedure, it should be clear that there is no reason for workers' councils to be biased against superior technologies because they are worker empowering! Moreover, since the criteria of choice are different, there is every reason to expect technologies developed in participatory economies to diverge from those that have developed in capitalist and coordinator systems. Over time this could dramatically improve the nature and quality of work life.

It is important to note that nothing we have said implies that workers in different workplaces must make decisions or organize job complexes entirely alike. There will be a variety of ways workers can pursue the principles discussed in this chapter, so that each participatory workplace will optimize within constraints imposed by the allocation system, but according to the tastes of their own council members. Differences in workers' preferences and leeway in the constraints on council decision making will permit technologies and social relations to vary from workplace to workplace.

Notes to chapter 2

1. For those familiar with John Rawls' idea of an "Original Position" and his argument that those in such a position would logically agree to the "liberty principle"--maximum liberty for each consistent with equal liberty for all-we might point out that the argument here is similar.

2. Those interested in the details of how all this might be accomplished and what the result might look like in different workplaces and industries should see Albert and Hahnel, Looking Forward: Participatory Economics for the Twenty First Century (Boston: South End Press, 1991).

3. See theorem 4.9, John Roemer, Analytical Foundations of Marxian Economic Theory (Cambridge: Cambridge University Press, 1981), 102-3. While Roemer's model assumes one primary input, homogeneous labor, we see no reason the problem he illustrates should not obtain in more realistic settings.

4. For example, see Herb Gintis, "The Nature of the Labor Exchange and the Theory of Capitalist Production," Review of Radical Political Economics 8, no. 2 (Summer 1976).

5. Theorem 8.1, Welfare Economics.

6. Theorem 9.1, Welfare Economics.