On Theories of a Democratic Planned Economy and the Coevolution of “Pro-democratic Planning” Preferences

Abstract

The motivation problem is a common component of discussions about democratic planning. This paper focuses in particular on individuals' motivation to work hard and to engage in the scrutiny of public affairs, both necessary traits for the success of a planned economy. The analysis draws on tools from the evolutionary game theory literature to consider whether the models of Pat Devine (Negotiated Coordination Economy, or NCE) and Michael Albert and Robin Hahnel (Participatory Economics, or Parecon) offer convincing solutions. I argue the proliferation of those traits would not naturally occur at the outset of democratic planning. I discuss selection mechanisms that increase the likelihood of successful proliferation of individually costly traits: segmentation, conformism, and group-level interactions. However, I argue that, under the impact of the defining institutions of NCE/Parecon, segmentation or conformism would not achieve the desired proliferation. The central conclusion of this paper is, then, that more sanctions and incentives at the group-level should be incorporated to the institutional design of these models.

Keywords:

• democratic planning
• Pat Devine
• Michael Albert and Robin Hahnel
• institutions and preferences

Acknowledgments

I am grateful to Peter Skott and Leila Davis for thoughtful comments on the whole text, and to Zoe Sherman for her comments and for English language editing. Anders Fremstad helped me with English editing as well. I also want to thank Samuel Bowles and David Kotz, whose classes in the Ph.D. program in Economics at the University of Massachusetts Amherst inspired this article. The usual disclaimers apply.

Notes on Contributor

Leopoldo Gómez-Ramírez is a doctoral student in economics at the University of Massachusetts at Amherst, USA. He holds a Bachelor's degree in philosophy from the National Autonomous University of Mexico, a Master's degree in economics from University of Guadalajara, and a Master's degree in economics from UMass-Amherst. He also spent two years teaching kids and teenagers in rural Mexican towns classified as being under extreme poverty. His research interests are in theories of distributive justice, new theories of a democratic planned economy, and the impact of inequality on growth in contemporary Mexico.

Notes

¹For brevity, NCE sometimes refers to Devine's model and sometimes to the society itself he is describing, and likewise Parecon. In each case the meaning should be clear by the context. Also for brevity, I use the terms “democratic planned economy,” “NCE,” “Parecon,” and “socialism” interchangeably. In all the paper, “Devine” denotes “Devine (1988).” Other important contributions to NCE model are Devine (2002, 2012). Core contributions of the Parecon model were written by both Michael Albert (2000, 2003) and Robin Hahnel (Albert and Hahnel 1991, 1992, 2002), but the most developed version was written by Albert (2003). For this reason, and for brevity, I often refer to Parecon as originating with Albert alone. “Albert” denotes “Albert (2003).”

²Likewise, evolutionary game theory literature has ignored the topic of socialism.

³However, different from this paper, Laibman's crucial “collective morale function” seems to be “operational” not at the outset of socialism but at a later stage (Laibman 2011, 43–44).

⁴For the interested reader, a more mathematical working paper version is available by request.

⁵It could be that average people of a democratic planned economy will have greater rational capabilities than under capitalism. But this is the kind of personal characteristic that cannot be expected to be the case at its outset.

⁶Two notes on terminology. By “proliferation” I mean “a significant fraction of the population (say more than one half) adopt such behavior.” By “protective mechanisms” I mean “segmentation and/or conformism and/or group-level emulation” (to be explained in section 3).

⁷Of course, here I just present an overview of the models. By definition, it has to leave out many important topics. For example, I don't mention the objectives of a democratic planned economy (Devine 1988, 113; Albert 2003, 9).

⁸On the centrality of investment ex ante decisions, Devine says he follows Dobb (1955, 1960), whom he calls “the most insistent and persuasive advocate of planning to date” (Devine 1988, 16, 190). By the way, Devine (1988, 23, 278) also argues Dobb was not an advocate of market socialism.

⁹Indeed, Albert (2003, 277) criticizes allowing some scope to markets on the grounds that “The logic of markets invalidates the logic of participatory planning and of the whole Parecon, and it is also imperial, once it exists trying to spread as far as wide as it can.” This in an interesting claim but its proper evaluation is out of the scope of the paper.

¹⁰In NCE, the governing bodies of production units include a wider range of interests than just the workers' ones.

¹¹Interestingly, a branch of the market socialism proposal, called “Labor Management Market Socialism,” has also defended workers' self-management; see Weisskopf (1993), Schweickhart (1992). And indeed, they have as well defended the point not only on moral grounds but on efficiency ones. Schweickhart (1992, 31), for example, argues that “the interaction here of workplace democracy and the market gives each worker a powerful incentive to work hard and efficiently.” The similarities of democratic planning, especially as proposed by Devine (1988), and market socialism, are interesting for future research.

¹²First, Devine envisions it as happening across your lifetime, in the sense that you work, say three years, in an empowering job and three doing rote work, while Albert envisions it as going on during any moment of life: you work, say three days per week, in some empowering workplace and three doing rote work. But there is probably a more important difference, which is that for Devine “abolition” does not imply every person has to do as many different things as Albert suggests. That is why Devine (1988, 168) distinguishes between “functional” and “social division of labor” and says that only the second is realistically abolishable. I'm aware of Albert's claims that Parecon's “balanced job complexes” do not imply “everybody does everything” (Albert 2003, 103–4), but it seems that they do imply that everyone must do a greater number of different work activities than under Devine's proposals.

¹³A hypothetical clarifying example could be needed here. If we rank job desirability with numbers, everyone should have the same average, say 20 points per week. This should be done not only within the workplace but also across workplaces. A miner, whose workplace average desirability is 10 should therefore also work in a TV station with workplace average of 30. And if the desirability of the TV station jumps to 40, then the TV actors/actresses will have to work more at the mine and/or less at the TV station. Albert (2003, 222) believes the operation of this same principle (“everybody should enjoy comparable job responsibilities”) will prevent the workers of FBs to “bias” the plan towards something they would prefer, because workers of FBs are like any other so they should also work in other places with lower average desirability.

¹⁴On one hand, it could imply that consumers don't demand much more than necessary of some labor-intensive good, even if they don't work at the workplace producing it, because they would end up bearing the labor costs of its production as well (Albert 2003, 218). But, on the other hand, it could imply weak incentives for the producers to improve their workplaces, because if they made such improvements and these improvements were not simultaneously replicated in all workplaces, the result would be that they would work less time in desirable places and/or more time in less desirable ones. Albert (186–87) addresses the issue. In my opinion, however, more than offering a proper answer to the issue for Parecon, he offers good criticisms of beliefs that are widespread among economists about how those benefits are reaped under capitalism.

¹⁵I could think of at least three other “pro-democratic planning preferences.” (3) “Not ask for more than I really need to consume.” (4) More generally, “sacrifice luxury desires so as to have balance between supply and demand.” (5) “Do the fair share of each category of social activity during life” as opposed to “live under basic income grant always,” or cooperate at first and then defect and live under basic grant. One reason not to discuss (3) (“consumerism”) is that I broadly agree with Devine and Albert when they say it is to a large extent created by capitalist advertising, and that it is also partially a fake “escape” (see Devine's discussion [1988, 164–67], about Bahro's [1978] speculations in this respect). Thus I think getting rid of consumerism under socialism is at least easier than spreading “working hard” or “engaging in scrutiny of public affairs” behaviors. With respect to the more encompassing behavior of sacrificing personal “greedy” desires so as to get balance between supply and demand, Albert (2003, 131–32, 190, 218) mentions peer pressure, that everyone would deal with the costs of producing (because of the principle that everybody should enjoy comparable job responsibilities), and “the need to get approval from other similar councils” (because of the principle that everyone affected by a decision should have a say on it) as inbuilt solutions to the issue. Devine, for his part, usually mentions the fact that people's “sectional interests” are all brought to the table and agreement has to be reached (Devine [1988, 185] is one of many examples). Again, one reason not to discuss it further is that, however difficult it could be, it seems still easier to solve than the problems of working hard and engaging in the scrutiny of public affairs. With respect to (5), one reason not to discuss it further is that it could be considered a special case of the problem of “working hard when you have to do rote work.”

¹⁶Note that in a complex economy there would be dozens of goods involved in this decision. But in this paper the “information problem” of socialism is disregarded, and thus we assume the information is publicly available. Assume also the worker has enough education to be able to participate. Under NCE/Parecon this opinion on inputs would actually not be done “after hours.” Thus, assume as well the other job responsibilities can actually be done fast enough so as to have time for this one.

¹⁷It is worth mentioning that, in my opinion, this example is more favorable for the NCE/Parecon case than other work situations we could think of, both because in it Cooperate implies less effort than in many other work situations and because in it identifying a defector is relatively easier. A different example, like “working in team production in which individual contributions are hard to identify” would be less favorable. I deliberately choose the inputs example to highlight even favorable imagined situations are problematic.

¹⁸Assume in the short run such appraisal does not fundamentally affect his/her consumption basket or his/her political rights. Note as well, again, that in a complex economy this assessment would entail dozens (hundreds?) of production units. But in this paper we just assume the information is publicly available. And assume also C has enough education and free time to participate.

¹⁹Interestingly, Albert says at least once something related with public participation, if only in relationship with the criticism raised by Nancy Folbre (1991) in “A Roundtable Discussion on Participatory Economics,” which implies there could be an unnoticed issue (Albert 2003, 253). In response, Albert (2003, 253) says “Parecon has every intention of leading people to participate, but no doubt, some will drink more deeply from the well of participation than others, and those who do, will . . . probably influence decisions disproportionately.” Then he adds in any case the critics should agree that “it would be better to have a dictatorship of the sociable with no material privileges accruing to them, than a dictatorship of the propertied . . . with great material privileges accruing” (255). However, no matter how much better it would be to have “a dictatorship of the sociable with no material privileges” compared with current capitalism, if some “drink more deeply from the well of participation” but they are too few then Parecon could not succeed.

²⁰Player W or C is the Row player. As the game is symmetric, only his/her payoffs are shown. It is not necessary to interpret the Column player as a single individual, and thus to interpret the population dynamics as coming out of “pairwise” interactions. Column could be a whole range of different agents: from the big “rest of the regional consumers’ council” (as in example two above) to the smaller “rest of the production unit” (example one), passing by the single partner in an individual household interaction.

²¹A reviewer critically commented that in a modern technological environment is intuitive. This specification would imply Cooperate would proliferate in the whole population if only the payoff from Cooperate were more than twice the payoff of Defect: (plugging payoffs in equation 2 to be presented shortly). In my view, however, it is not clear at the outset of socialism under any interpretation of the payoffs, because of the difficult situation such society would most likely be in. But in any case, under the interpretation of payoffs of this paper, does not seem to be the case.

²²In addition to the condition of a ``Prisoners' Dilemma,'' is sometimes added, ``so there is no gain from taking turns'' playing Cooperate and Defect (Gintis 2009, 50).

²³A similar equation, easier to work with, , could have been derived. However, it would have lost an important further message of equation 2, which is that a very homogenous population changes less than an heterogeneous one: is smaller for very large or very small p.

²⁴Note that the last three channels in fact presuppose the institutions of NCE/Parecon are already successfully operating since its outset, which I have pointed as faulty in the absence of the more basic proliferation of pro-democratic planning preferences. However, for the sake of the argument, assume the prospects of effective participation, democratic decision-making, and public information are in many people's minds, and that such prospects are enough to foster Cooperate.

²⁵Eric Olin Wright (2010, chapter 9) offers an interesting analysis of “ruptural transformations.”

²⁶Bowles and Gintis (2011, chapter 8) accommodate the possibility that group level emulation does not come from conflictual interactions, but they think it was not the case for ancestral humans (147). See Bowles, Choi, and Hopfenzits (2003) too.

²⁷This is more likely to be the case when Cooperate is more prevalent than Defect: .

²⁸It is worth mentioning that conformism has been posited as explaining, ``counterintuitively,'' ``the punctuated equilibria that characterizes rapid cultural change, as well as the long-term survival of individually costly norms, whether group-beneficial or not'' (Bowles 2004, 376). Formally, it can be shown that the conformist mixed equilibrium is unstable/stable if , from which it can be seen that the higher the degree of conformism the more likely the mixed equilibrium will be unstable. A high degree of conformism therefore can result in rapid social change. Now, in this paper the explanation of the movement towards a socialist society is assumed. But in any case the idea that conformism is a two-edge knife for social change is quite interesting for future research for the literature on the transition to socialism.

²⁹By definition, that average is .

³⁰In the degenerate case of all groups having the same number of cooperators () there is no group selection effect. Likewise, in the degenerate case of groups being internally homogenous (either all cooperators or all defectors, but in any case every ) there is no individual-level effect.

³¹It can be shown that .

³²Bowles and Gintis (2011, chapter 7) offer an extension of this model in which there is egalitarian redistribution within groups (through taxation). As result, the assortment necessary to sustain the costly trait is reduced. From an egalitarian (but not socialist) point of view, it arises as interesting for future research.

³³To the risk of being repetitive, I want to stress I am thinking about the outset of NCE/Parecon, in which the overwhelming majority of the people would be experiencing this gigantic social change for the first time. In earlier presentations of this paper, it has been pointed out to me a socialist society would elicit more segmentation and not less than capitalism because in the former people would be interacting somehow with more freedom to choose, probably closer at an emotional level, and with more information. All this could be the case in “mature” socialism, but I don't find it reasonable at its outset.

³⁴In terms of the formal model presented in subsection 3.2, assuming inequality i holds, my point of view can be captured by specifying so that . Alternatively, assuming inequality ii holds, it can be captured by specifying so that : there is an enlargement of the basin of attraction of the nice equilibrium but it is not enough to achieve the proliferation of Cooperate.

³⁵Here it is worth clarifying the following. As conformism may help to explain cultural change (Bowles 2004, see footnote 28), it could explain the social change that leads to socialism. But then we could mistake the possible high degree of conformism that could have helped such social change (at ) with the degree of conformism after it (at ). Now, this paper is only concerned with the coevolution of institutions and preferences after the social change, and in any case maybe it was not related with too much conformism. I thus only ask how conformism could be after the social change, under the influence of the new institutions.

³⁶One reason Bowles (2001, 169) offers to argue conformism is important is that, “under quite general conditions where learning is costly,” conformism learning is itself (taken as a trait) capable of proliferation. But the openness of information reduces the costs of learning.

³⁷In earlier presentations of this paper, it was pointed out to me that a socialist society could yield more and not less conformism but in a somehow “conscious” way. The argument goes as it follows: in a socialist society people would gather to decide what to do and as a result of such discussion they would cooperate, even if that were costly from a payoffs point of view, precisely because they discussed and chose what to do. This is indeed an interesting point, but so far I believe at the outset of socialism such “conscious conformism” would not be significant enough. I find it fair to recognize, however, that the topic of how would conformism be under socialism is open for future research and debate.

³⁸In terms of the formal model presented above, assuming inequality i holds, my point of view can be captured by specifying so that . Alternatively, assuming inequality ii holds, it can be captured by specifying so that the enlargement of the basin of attraction of the nice equilibrium is not enough to achieve it.

³⁹There is a vast literature related with the issue of unintended effects of policy-making. Gneezy and Rustichini (2000) is a classic. Bowles and Polania-Reyes (2012) is a recent contribution.

⁴⁰As mentioned in the Introduction, Robin Hahnel (2012, 202–3) has recently claimed that “recent research in evolutionary biology and evolutionary game theory suggest that not only have successful societies developed social norms to induce such behavior, but there is every reason to believe natural selection would have favored genetic dispositions toward behavior that helped the group, not the individual, to survive.” I disagree with this claim, however, because of the following reasons. First, what is the proper unit of analysis when dealing with the motivation problem of socialism: the nation or something smaller? I think something quite smaller. Second, and more importantly, the claim overlooks one of the main messages of the invoked literature: “Group competition and culturally transmitted group differences in institutional structure [reproductive leveling and segmentation] are central to our explanation of the evolution of cooperative behaviors among humans” (Bowles and Gintis 2011, 111). In my opinion, an adherent to a socialist project must at least think about how to deal with the inconvenient (for socialism) thesis such that high levels of cooperation among some groups occurred because they helped those groups to win conflicts (literally wars) against other groups.

⁴¹To capture this formally I think the right way is to specify the success of NCE requires a smaller mixed equilibrium than the success of Parecon, and the temptation of making “engage” more costly under the latter than under the first (in table 1 making b larger and/or d smaller under NCE, if referring to the behavior of “participating in public life”) should be avoided. The latter would imply saying that market exchanges make “engage” less costly, which is very different (much stronger and hard to defend) from saying that fewer engaged people are needed when there are market exchanges.

⁴²Other important frequently discussed topic, not approached here, is innovation under socialism. See Kotz (2002).

⁴³The “information problem” discussion goes back, at least, to the 1930s' “socialist calculation debate”; Hayek (1935), Von Mises (1935), and Lange (1938) are classic references. Adaman and Devine (1996) offers an evaluation of the “revival” of the debate that occurred in the 1980s–1990s after the new arguments of the “Austrian school” based on the “tacit information” problem.

⁴⁴Albert (2003, 268) speaks about the foreign non-pareconist world once. There, he tries to deny the internal logic of Parecon could allow a foreigner businessman to create a firm inside Parecon, by offering higher wages to attract workers. But my concern is that, if willing to destroy Parecon, foreign businessmen would not only offer higher wages to Parecon workers but would bring their armies. Devine for its part, seems to leave open the possibility of international NCE. He says that even though he does not discuss “the international dimension,” he believes “the principles underlying the model of negotiated coordination could be applied to international economic transactions” (Devine 1988, 253).

⁴⁵Such a confined society would certainly forego international benefits, like technological improvements or commercial links, and it would very likely face the hostility of foreign non-socialist countries. It is my impression that through these two channels the costs of Cooperate and the benefits of Defect would in turn increase; so that, to put it formally, inequalities i or ii would hold. I am aware that these are mere impressions about issues that have been lengthy disputed among socialists.