The Political Economy of Participatory Economics - by Michael Albert and Robin Hahnel



[I]t is clear that someone (some institution) has to tell the producer about what the users require. If that 'someone' is not the impersonal market mechanism it can only be a hierarchical superior. There are horizontal links (market), there are vertical links (hierarchy). What other dimension is there? .. In a complex industrial economy the interrelation between its parts can be based in principle either on freely chosen negotiated contracts (which means autonomy and a species of commodity production) or on a system of binding instructions from planning offices. There is no third way.

-Alec Nove

Necessity is the argument of tyrants. It is the creed of slaves.

-William Pitt Jr.

In this chapter we describe an alternative system of allocation, which we call decentralized participatory planning. The system permits consumers' and workers' councils to participate directly in formulating a plan and has strong egalitarian properties. Because workers' and consumers' councils propose and revise their own activities prior to initiating those activities, the planning process is a decentralized, social, iterative procedure.

We consider specifying this procedure and analyzing its welfare theoretic properties our most important contribution to developing a libertarian egalitarian economic vision. The idea of "associated producers" democratically determining their own plan is no more original to us than the vision of workers' and consumers' councils. But whereas many before us have contributed to the theory of the internal workings of democratic councils, few have attempted to explain, in detail, how those councils might jointly settle on a plan. 1 In fact, most economists agree that no third procedure qualitatively different from markets and central planning exists, or, if there is another alternative, that it has not been articulated at a level permitting meaningful comparison with markets and central planning.

Alec Nove, for example, threw down the gauntlet in unequivocal terms: "I feel increasingly ill-disposed towards those who ... substitute for hard thinking an image of a world in which there would be no economic problems at all (or where any problems that might arise would be handled smoothly by the 'associated producers'.... In a complex industrial economy the interrelation between its parts can be based in principle either on freely chosen negotiated contracts [i.e., markets], or on a system of binding instructions from planning offices [i.e., central planning]. There is no third way."2

Allen Buchanan posed the challenge in a somewhat more agnostic vein: "It is impossible to show that a feasible nonmarket system at least approaches the productivity of the market unless (1) a rather welldeveloped theoretical model of the nonmarket system is available, and (2) it is demonstrated that a sufficiently productive approximation of the ideal ... system described in the theoretical model is practically possible. Unfortunately, [no one] has achieved even the first step-that of providing a theoretical model for a nonmarket system."3

In this chapter and the next two we present a rebuttal to Nove and a direct answer to Buchanan by providing a "rather well-developed" theoretical model of a decentralized planning procedure and offering a preliminary analysis of its efficiency properties as well as arguing that a "sufficiently productive approximation of the ideal ... system described in the theoretical model is practically possible," including description of a number of experiments through which other economists might sensibly test this claim.

Participatory Information and Communication

Our description of participatory workers' and consumers' councils assumed that the necessary information about their relations with others would be available. But what precisely do workers in a council need to know to regulate their production activity cognizant of the effects on themselves, other workers, and consumers? And what must consumers know to formulate their consumption requests in light of their own needs as well as the needs of other consumers and workers?

Participatory workers must be able to weigh the gains from working less or employing less productive though more fulfilling techniques, against the consequent loss of consumer well-being. Participatory consumers need to be able to weigh the gains of a consumption request against the sacrifices required to produce it. Participatory workers must be able to distinguish an equitable work load from one that is too light or too heavy. And participatory consumers need to be able to distinguish reasonable consumption demands from ones that are unreasonable or overly modest. Finally, all actors must know the true social costs and benefits of things they demand or supply, that is, all the nonhuman and human, quantifiable and nonquantifiable consequences of their choices, if they are going to participate in informed collective self-management.

First Communicative Tool: Prices

Prices providing accurate estimates of the full social costs and benefits of inputs and outputs are the most important quantitative communicative tools we use. They arise in the process of participatory planning and serve as guides to proposals and evaluations. And this is an important point. All too often theoretical economists view "efficiency" prices or "shadow" prices as quantitative measures that can be found via technical procedures. In the literature on central planning, for example, shadow prices arise as the solution to the dual of the primal planning problem that central planners "solve." And in neoclassical literature on market systems, an equilibrium price vector is studied as something implied by preferences and technologies taken as givens. While these conceptions are useful in some regards, they are misleading as well. Real people's preferences arise in social communicative processes. Not only do results depend on what those processes are like, but the very preferences that lie at the basis of the results depend on the processes as well. So, without engaging in undue mystification, we should remember that estimates of social costs and benefits with any claim to accuracy must arise from social, communicative processes. The trick is to organize these processes so people have no incentives to dissimulate regarding their true desires, and all have equal opportunity to manifest their feelings. It is precisely because our participatory planning process is different from the flawed communicative processes of market and centrally planned allocation that the prices to which it gives rise will be different as well.

In any case, prices are "indicative" during the planning process in the sense of indicating the best current estimates of final valuations. They are not binding but flexible in the sense that qualitative information provides important additional guidance. And they are not the product of competition or authoritarian determinations, but of social consultation and compromise. The idea is that qualitative information is necessary if quantitative indicators are to be kept as accurate as possible. But qualitative information is also necessary to develop workers' sensitivity to fellow workers' situations and everyone's understanding of the intricate tapestry of human relations that determines what we can and cannot consume or produce. So both to assure accuracy and to foster solidarity we need a continual, social resetting of prices in light of updated qualitative information about work lives and consumption activity. Thus, the cybernetic burden of a participatory allocation procedure is considerably greater than for nonparticipatory economies. Not only must a participatory economy generate and revise accurate quantitative measures of social costs and benefits in light of changing conditions, but it must communicate substantial qualitative information about others' conditions as well.


Second Communicative Tool: Measures of Work

As we explained in chapter 2, job complexes are balanced in each workplace, and in plants with above average work conditions workers spend time doing more menial tasks elsewhere, while in plants with below average work conditions, people put time into more interesting pursuits elsewhere.

For an individual to work more or less than the social average in a given period and not disrupt a humane balance of work, she or he need only diminish or increase her or his hours worked at all tasks in the same proportion. Then, each individual could receive from her or his workplace an indicator of average labor hours expended as an accurate indicator of work contributed. Over a sufficient period, whenever a person's indicator was high (or low) compared to the social average, the individual would have sacrificed more (or less) for the social good, and would be entitled to ask for proportionately more (or less) consumption in return. Unlike what emerges from the marxist labor theory of value: (1) In our system job complexes are balanced by a real social evaluation, and (2) our "hour counts" serve only as guidelines for decisions since councils can grant exceptions for higher (or lower) consumption requests as conditions and needs warrant.

In short, participatory planning can obtain a reasonable first estimate of effort expended by counting labor hours because people's job complexes have been balanced. These estimates can then be revised in light of effort intensity ratings by one's work mates. In attempting to gain consumption flexibility, only unbalancing job complexes is prohibited.

Third Communicative Tool: Qualitative Activity

To guard against "reductionist accounting" each actor needs access to a list of everything that goes into producing goods directly and indirectly, and a description of what will be gained from consuming them. This means those who produce and consume particular goods must try to communicate the qualitative human effects that cannot be captured in quantitative indicators. This does not entail everyone writing Upton Sinclair length novels about their work and living conditions. It does mean generating concise accounts that substitute for the fact that not everyone can personally experience every circumstance. Of course, not every worker and consumer will use all this qualitative information in all calculations. But over time people will become familiar with the "congealed" material, human, and social components of products they use just as people are now familiar with the products themselves. In this way, everyone can more accurately assess the full effects of others' requests in a way that enhances solidarity. Both producers and consumers must therefore receive not only quantitative summaries of overall social costs and benefits, but detailed qualitative accounts as well. Only this will ensure that the human/social dimension of economic decision making is not lost and guarantee that summary price data remains as accurate as possible.

Allocation Organization

Every workplace and neighborhood consumers' council participates in the social, iterative procedure we call participatory planning. But besides workplace councils, we also have industry councils and regional federations of workers' councils. And besides neighborhood consumers' councils, we also have ward, city, county, and state federations of consumers' councils as well as a national consumers' council. Moreover, in addition to all these councils and federations of councils, various facilitation boards assess collective proposals and large-scale investment projects, regional and industry boards assist workers changing places of employment, and household boards help individuals and families find membership in living units and neighborhoods. Finally, at every level of the economy facilitation boards help units revise proposals and search out the least disruptive ways of modifying plans in response to unforeseen circumstances.

In our companion volume, Looking Forward, we provide comprehensive descriptions of planning institutions and procedures, including hypothetical case studies for particular kinds of workplaces and consumers' councils intended to illustrate the texture of participatory planning. Here we present a summary of results sufficient for theoretical purposes.


Participatory Planning

Each consumption "actor" proposes a consumption plan. Individuals make proposals for private goods. Neighborhood councils make proposals that include approved requests for private goods as well as the neighborhood's collective consumption request. Higher level federations make proposals that include approved requests from member councils as well as the federation's collective consumption request.

Similarly, each production "actor" proposes a production plan. Workplaces enumerate the inputs they want and outputs they will make available. Regional and industry federations aggregate proposals and keep track of excess supply and demand. Every actor at every level proposes its own plan, and, after receiving information regarding other actors' proposals and the response of other actors to its proposal, each actor makes a new proposal. As every actor "bargains" through successive "iterations," the process converges.


Preparing First Proposals

The real world always has a "just completed year." If production and consumption of the just completed year was recorded, we would have information about last year's plan for each actor. If the prices used to calculate social costs, benefits, and income last year were recorded, we have a set of "indicative prices" that could be used to begin this year's estimates as well. By storing last year's full plan in a central computer, access to relevant parts including indicative prices could be made available to all actors in the planning process. Additionally, each unit knows what its own proposals were in each iteration last year.

So, how do workers' and consumers' councils plan?

1. They access relevant data from last year.

2. They receive information from facilitation boards estimating this year's probable changes in prices and income in light of existing knowledge of past investment decisions and changes in the labor force.

3. They receive information from higher level production and consumption councils regarding long-term investment projects or collective consumption proposals already agreed to in previous plans that imply commitments for this year.

4. By reviewing changes in their own proposals made during last year's planning they assess how much they had to scale down their consumption desires or their plans to improve the quality of work life, and look to see what increases in average income and improvements in the quality of average work complexes are projected this year over last.

5. Using last year's final prices as indicators of social costs and benefits they develop a proposal for the coming year, not only enumerating what they want to consume or produceand implicitly what they think society's total output should be-but also providing qualitative information about their reasons.

This does not mean units must specify how many units of every single good they need down to size, style, and color. Goods and services are grouped into classes accordingly as they are roughly interchangeable regarding the resources, intermediate goods, and labor required to make them. For planning purposes we need only request types, even though later everyone will pick an exact size, style, and color. Individuals present consumption requests to neighborhood councils where they are approved or disapproved. Once approved, individual consumption requests are summed and added to the neighborhood collective consumption request to become the neighborhood consumption proposal. These in turn are summed with consumption requests from other neighborhoods into ward proposals, which are summed along with consumption requests from other wards into city proposals. Having the next higher level council review, approve, or contest lower level requests until they are ready to be passed on saves a great deal of planning time.

In the same way, a firm's iteration board provides all its workers with summaries of last year's production schedule, including what was initially proposed, changes made during planning iterations, and what was (finally) approved, as well as a prediction of this year's requests based on extrapolations from new demographic data and the trajectory of last year's iterations. Individual workers consider this information, discuss ideas for improving the quality of work life, and enter proposals which are averaged into the firm's first proposal for "inputs" and "outputs." After some number of iterations, firm proposals are discussed, negotiated, and decided as a unit rather than each individual making his or her own proposal and these being averaged.

Besides quantitative proposals for each production and consumption unit, a qualitative addenda including descriptions of changes in circumstances and conditions is also entered into the computerized planning system. At any point any council can access the data banks of any facilitation board and any other council.

Proceeding from One Proposal to Another

The first proposals are in. We have all answered how much we want to work and consume in light of our own presumably over-optimistic assessments of possibilities. Do the proposals constitute a plan, or must we have another round? To decide, it is only necessary to sum all proposals and compare total demand and total supply for every class of final good and service, intermediate good, and primary input. In a first iteration, where consumers propose in part a "wish list" and workers propose substantial improvements in their work lives, while some goods may be in excess supply, for most goods initial proposals should generate excess demand. In other words, initial proposals are not supposed to sum to a feasible plan. As the next step, every council receives new information indicating which goods are in excess supply or demand by what percentage, and how its proposal compares to those of other relevant units. Most important, iteration boards provide new estimates of indicative prices projected to equilibrate supply and demand.

At this point, consumers reassess their requests in light of the new prices and most often "shift" their requests for goods in excess demand toward goods whose relative prices have fallen because they were in excess supply or less in excess demand than others. Consumers' councils whose overall requests were higher than average would also be under pressure to "whittle down" their requests in hopes of winning approval. Equity and efficiency emerge simultaneously. The need to win approval from other similar councils forces councils whose per capita consumption request is significantly above the social average to reduce their overall requests. But the need to reduce can be alleviated by substituting goods whose indicative prices have fallen for those whose prices have risen. Attention focuses on the degree to which units diverge from current and projected averages, and on whether their reasons for doing so are compelling.

Similarly workers' councils whose ratios of social benefits of outputs to social costs of inputs were lower than average would come under pressure to increase either efficiency or effort, or to explain why the quantitative indicators are misleading in their particular case. Before increasing their work commitment, workers would try to substitute inputs whose indicative prices had fallen for inputs whose indicative prices had risen, and substitute outputs whose indicative prices had risen for outputs whose indicative prices had fallen.

Each iteration yields a new set of proposed activities for all economic actors. Once summed, these yield new data regarding the status of each good and the average consumption per person and production "benefit cost ratio" per firm. All this allows calculation of new price projections and new predictions for average income and work, which in turn lead to modifications in proposals until excess demands are eliminated and a feasible plan is reached.

Flexible Updating

Converging and updating are related because both can benefit from algorithms that take advantage of the large scale of the planning process. Assume we have settled on a plan for the year. Why might we need to update it during the year, and how might this be done with least disruption?

Consumers begin the year with a working plan including how much of different kinds of food, clothing, meals at restaurants, trips, books, records, and tickets to performances they will consume. What if someone wants to substitute one item for a slightly different one? Or what if she wants to delete or add entries? Or what if she changes her mind and wants to save or borrow more than planned? She belongs to a neighborhood consumers'council which in turn belongs to a ward council, a city federation, and so on. Some changes will cancel out among all the consumers within the neighborhood, others will cancel out at the ward level, and so on. As long as adjustments by many consumers cancel at some consumption federation level, production plans need not change. Indeed, making adjustments without disrupting production plans is one function of consumer federation boards.

But what happens if aggregate demand rises for some good? Suppose individuals record their consumption on "credit card" computers that automatically compare the percentage of annual requests "drawn down" with the fraction of the year that has passed, taking account of predictable irregularities such as birth dates and holidays. This data can be processed by planning terminals which communicate projected changes to relevant industry councils which in turn communicate changes to particular firms. The "technology" involved is little different from the now common system of computerized store inventories where cash register sales are automatically subtracted from inventory stocks. In any case, what would then happen is that consumer federations, industry councils, and individual work units would engage in a dialog to negotiate adjustments. Such dialogues may lead to work diminishing in some industries and increasing in others, including possible transfers of employees, but there need be no more moving about than in other types of economies. In any case, the need for workers to change jobs or increase or diminish work loads would be a factor considered in the dialogue over whether to meet changed demands.

However, since each firm's activities have implications for other firms, if planned matches between supply and demand are calculated too closely, any change in demand could disrupt the whole economy. For this reason a "taut" plan would prove unnecessarily inconvenient since it would require excessive debating and moving. To avoid this and to simplify updating, the plan agreed to should include some excess supply for most goods. A practical knowledge of those industries most likely to be affected by nonaveraging alterations would facilitate this type of "slack planning."

Converging to a Plan

A little thought reveals that convergence can be a complicated matter. Adjusting indicative prices to reduce excess supplies and demands is more complicated in practice than in economists theoretical models with all their convenient assumptions. For example, a product in excess demand in one iteration could overshoot equilibrium and be in excess supply in the next iteration as workers offer to produce more and consumers offer to request less in response to a price increase. Worse still, considering that each product's status affects the status of many others, progress in one industry could disrupt equilibrium in another. Theoreticians' solutions to these headaches always assume away the troublesome phenomena. Whether the issue is market equilibrium and stability or convergence of iterative planning procedures, it is well known that convexity and gross substitutability assumptions are good aspirin for these theoretical headaches. But simplifying assumptions are no aspirin at all for practitioners operating in the real world.

To make our participatory planning procedure more efficient, specific economies will incorporate flexible rules that facilitate convergence in a reasonable time but do not unduly bias outcomes or subvert equity. Procedures can range from rote algorithms carried out by computer that take short cuts toward equilibrium, to rules that
prohibit actors' responses that would yield time consuming loops, to adjustments fashioned and implemented by specialized workers experienced in facilitating convergence when particular situations arise. Devising and choosing from among these and other possibilities is a practical issue in implementing any actual participatory economy. Assuming the procedures chosen do not violate principles essential to participatory planning, considerations include (1) the extent to which iteration workers could bias outcomes, (2) the extent of reductions in the number of iterations required to reach a plan, (3) the amount of planning time saved through compartmentalizing subsets of iterations, and (4) how much less onerous to producers and consumers their calculations could be made.


A Typical Planning Process

Since the procedure we have described is dramatically different from traditional market and central planning allocation, it is useful to summarize by describing what a typical planning process might look and feel like to its participants.

The first step is for each individual to think about her or his plan for the year. Individuals know they will end up working in a balanced job complex, and can expect to consume an average consumption bundle unless their work effort is above or below normal or special needs dictate otherwise. The first decision is whether they want to "save" by working longer or consuming less than average, or "borrow" by working less or consuming more than average. Facilitation boards provide an initial estimate of what average consumption and average work loads will be for the year based on last year's levels, last year's investments in equipment and training, and adjustments that occurred during last year's iterations. When you make your first proposal you are not only proposing to do specific work and consume specific items, but you are proposing a level of work contribution and consumption request for yourself, and, implicitly, at least on average, for everyone else as well. To be realistic you must coordinate your work and consumption proposals, though you need not agree with facilitation board growth estimates.

In other words, what you propose is: "I would like to work so much at my job complex and to consume so much broken down in the following way." And this proposal is based on last year's experience, your prediction of economic growth, and your individual decision about saving and borrowing. Everyone makes such a choice, trying to optimize given their particular preferences and within the constraint that the overall amount consumed must be produced and that responsibilities and rewards in this endeavor will be distributed equitably.

After first proposals are summed, new indicative prices are calculated and new projections of social averages estimated. Note that it would not even be possible to implement most initial production proposals since in most firms one person in a team may have proposed working more hours than another person in the same team, even though they can only work together. Moreover, most goods will be in excess demand so the initial plan is infeasible as well.

Again every individual would formulate a new response. You compare your proposed work load and proposed consumption to the average proposals of others. You might also consider more localized averages, for example in your firm or industry, and in your council or neighborhood. You certainly consider the status of each item you ordered or proposed since excess demands and supplies will be reflected in changes in indicative prices. That is, you will be faced with summaries of the statuses of goods as well as new estimates of social opportunity costs and benefits. After you consult descriptive explanations for what seems odd to you, like large changes in worker productivity or consumer choice, and after you consult with whomever you like and whatever data you are interested in, you then make any desired changes before entering your second proposals.

And, once again, all these new proposals are summed and the new information made available for the third iteration. So far there have been no rules or limits on workers' or consumers' responses. Now, however, there could be a change. Instead of being able to change proposals in any direction by any amount, limits might be imposed. For example, consumers might be prohibited from increasing their demand for certain goods beyond some maximum percentage above projected averages for the economy. Or producers might be prevented from lowering output proposals by more than some percent in this and subsequent rounds.

The point is simply that it is possible to impose rules limiting changes to specific ranges to keep the status of goods from varying excessively from round to round. Any particular implementation of participatory planning settles on socially desirable and mechanically efficient rules to guide the behavior of producers and consumers in different iterations.

In the third or fourth iteration, proposals might be limited to councils instead of individuals. Consumers meet in their local neighborhood councils and workers in their workplace councils to settle on councilwide proposals so that work proposals are no longer abstract unimplementable averages but consistent work plans that could be enacted if inputs requested were made available.

Note that nothing about our procedures pushes different actors to consume the same amounts of different goods. Individual consumers and producers can hold pat on proposals that are far from average. On the other hand, workplaces do feel pressure to measure up to average "benefit cost ratios," and consumers will be pressured to keep their overall requests from exceeding average income. Indeed, at this stage production councils that persist, after allowance for acknowledged different circumstances, in proposals with benefit cost ratios below their industry's average, might have to petition their industry for permission not to be disbanded. And, similarly, although again with sensible allowances, local consumers' councils with above average proposals might have to petition higher federations explaining special circumstances to warrant their requests.

The fifth iteration in our hypothetical procedure might deploy still another rule to accelerate planning. This time facilitation boards extrapolate from the previous iterations to provide five different final plans that could be reached by the iterative process. What distinguishes the five plans is that each entails slightly different total product, work expended, average consumption, and average investment. All actors then vote, as units, for one of these five feasible plans. Each plan is a consistent whole and implementable. Once one of the five is chosen as the base operating plan, units adjust requests in subsequent iterations in conformity with the base plan until individual agreements are also reached.


While we must still address important aspects of participatory allocation in chapter 5, it is useful to summarize here where our argument stands. In chapter I we argued that hierarchical production and consumption, markets, and central planning were individually and in combination incompatible with efficient, egalitarian, economies in which people control their own lives and enjoy solidarity. In chapter 2 we presented a description of participatory, nonhierarchical production. In chapter 3 we added a description of participatory, equitable consumption. In this chapter we have described a planning procedure that promotes participation, equity, solidarity, and variety and supports rather than undermines participatory production and consumption within units.

What remains is to demonstrate that a participatory economy can yield desirable outcomes efficiently. To do that we examine the convergence and efficiency properties of a mathematical model of participatory planning in chapter 5, and propose various simulation and other social experiments regarding its "practical feasibility" in chapter 6.


Notes to Chapter 4

1. One recent exception is Pat Devine who presents a model of "democratic planning" in part IV of Democracy and Economic Planning (Boulder, Colo.: Westview Press, 1988).

2. Alec Nove, The Economics of Feasible Socialism (London: George Allen and Unwin, 1983), ix-x, 44.

3. Allen Buchanan, Ethics, E fficiency, and the Market (Totowa, N.J.: Rowman and Littlefield, 1985), 29.


Next Chapter