Notes

Chapter 1

1. Whether or not one believes that free enterprise (capitalist) systems can in fact realize socialist goals and values is perhaps the decisive difference between socialists and social democrats.

2. For discussion OfYugoslavia, see Pejovich (1966), Milenkovitch (1971), Vanek (1977a), Estrin (1983), and Lydall (1984; 1989). For Mondragon, see Thomas and Logan (1982). For the kibbutz experience, see Leviatan (1991) and the references cited therein. The cases OfYugoslavia and Mondragon will be discussed in more detail in chapter 6.

3. For a fuller articulation of this position, see Buchanan (1982, 60-81). The crit­icisms that follow in the text are also due to Buchanan.

4. See, for example, Nozick (1974, 233).

5. For a useful historical discussion of utopian socialist thought, see Kolakowski (1978, 182-233).

6. Actually, it is easy to prove that the concept of a good society is not the same as the concept of an ideal society. A good society has an economic system. Any society with an economic system has lawyers (to interpret the rules of the system). No ideal society has lawyers. Therefore, no good society is an ideal society.

7. In principle, the condition in question could be a sufficient condition for a soci­ety to be a good society. However, it would be an impoverished theory that held that a society’s having a certain type of economic system is sufficient for it to be a good society. On the face of it, it seems self-evident that a society could have the right type of economic system and still fail to be a good society for other reasons.

8. Unfortunately, a good deal of contemporary economics is a purely formal, mathematical enterprise. The economics that is relevant to this dispute must have a fairly substantial empirical component. Notice also that a full-blown economic theory (of prices, distribution, etc.) may not be necessary to substantiate some central claim or claims that a defender of one or the other type of system makes.

9. John Rawls says, ‘Justice is the first virtue of social institutions just as truth is the first virtue of systems of thought” (1971, 3). Thoughjustice may be the first virtue of social institutions, it is certainly not the only one. All good societies are just, but the converse is highly doubtful; some contemporary liberals notwithstanding, there is more to the good society than satisfying the demands of justice.

10. This view derives from a certain normative conception of human nature, that is, what makes a life a good life for human beings. For discussions of this normative element in Marx’s writings about human nature, see Elster (1985, chap. 2.2), Warten- berg (1982, 77-78), Buchanan (1982, 18-19), Miller (1981, 324-27), Nasser (1975, 486-88), Gregor (1968, 377).

11. For some representative recent socialist criticisms of the free enterprise system, see Schweickart (1980), Horvat (1982), Nove (1983), Nielsen (1985), Miller (1989b), and Le Grand and Estrin (1989a). For a sample of criticisms of socialism, see Hayek (1944), Kolakowski (1974), Rutland (1985), Lavoie (1985; 1986), and Evers (1989).

12. This objection to John Roemer’s account of exploitation under capitalism has been raised by Allen Buchanan (1985, 121-22). I have argued elsewhere that the problem that Buchanan calls attention to is legitimate but that Roemer’s definition of exploitation can be repaired so as to avoid it (Arnold 1990, 126-28).

13. Because of the way the terms ‘free enterprise system’ and ‘socialist economic system’ have been defined, there might be some minor exceptions to this general principle. Suppose that a critic of free enterprise systems believes that widespread pri­vate ownership of the means of production is malum in se. Since a prohibition on this is part of the very definition of a socialist economic system, this critic of the free enter­prise system is committed to the proposition that no socialist economic system has that problem.

14. In a similar vein, when proponents of a free enterprise system criticize exist­ing socialist systems for the shortages, generalized poverty, and tyrannical political systems that they associate with those economic systems, they explicitly or tacitly hold out the prospect of escaping or avoiding these social problems by embracing a free enterprise system. This was one of the messages of Hayek (1944).

15. I have reconstructed and critically evaluated these arguments elsewhere (Arnold 1990, 193-201, 268-81).

16. See note 13.

17. This is especially true of socialists, both those opposed to the existing capital­ist order in the West and those opposed to the bureaucratic socialism that character­ized Eastern Europe. For some examples of the former, see Levine (1984), Schweick- art (1980), and Miller (1989a). For an example of the latter, see Horvat (1982).

18. Exceptions are claims that display a logical relation between type-defining fea­tures and social virtues, for example, some libertarians’ claims linking free enterprise systems to respect for basic human rights.

19. This paragraph touches on some important and difficult issues in the philoso­phy of social science, such as the correct understanding of ceteris paribus clauses and the role Ofbackground assumptions in social scientific explanation. For a nuanced discussion of these and related issues, see Kincaid (1990).

20. This view is so widespread among socialists, especially contemporary social­ists, that documentation is scarcely needed. For representative views, see Nielsen (1985) and the articles in Le Grand and Estrin (1989a).

21. Indeed, it used to be thought to be the only task. Once the concepts had been clarified, it was time to clock out and head for the local tavern. Almost no analytic philosophers believe this anymore.

22. For a recent example, see DeJasay (1990, 9-10).

23. This tacitly assumes that the best society that one can reasonably hope for is not beset by widespread and pervasive alienation.

24. As a point of logic and dialectics, this requires more than criticizing one’s oppo­nents’ arguments, since their premises might be false and yet their conclusions true. It is surprising how many critics of free enterprise systems rest content with criticiz­ing arguments offered by those who favor such systems. To criticize the free enterprise system, it is insufficient to discover deficiencies in the reasoning of, for example, Robert Nozick or Milton Friedman.

Chapter 2

1. The best representations of mainstream opinion are to be found in those enor­mous introductory economics textbooks that college students are forced to buy. The successive editions of Paul Samuelson’s Economics provide a good chronicle of the changes in mainstream opinion among economists over the past two decades. An examination of the relevant sections of this text over a number of editions reveals that Samuelson’s overall assessment of central planning has become increasingly less favor­able over the years. Some truly embarrassing quotations on the Soviet economy from Samuelson and other economists are collected in Arnold Beichman’s column in the Washington Times of December 18, 1989.

2. The consensus was not unanimous. The passage quoted from Heilbroner and Thurow has been retained in later editions of The Economic Problem. Heilbroner and Thurow, like many economists in the public eye, maintain a remarkable serenity and equanimity in the face of events that reveal their predictions and other public pro­nouncements to be contrary to the plainest facts. Events such as these would produce an intellectual, if not a personal, crisis for most individuals. The only historical par­allel to this equanimity that I am aware of is the ancient Pyrrhonian skeptic’s attitude of detachment (ataraxia) from the world of sensory experience, an attitude cultivated by an intensive and careful study of skeptical arguments.

3. See, for example, Wlodzimierz Brus’s review essay in the New Left Review (Brus 1985).

4. See, for example, Brus and Laski (1989, 132-52) and the writings of Janos Kor- nai over the past four decades. The most comprehensive statement of Kornai’s views can be found in Kornai (1992).

5. The best and most comprehensive discussion of the debate is to be found in Lavoie (1986), which explains why Mises’s critics missed the simple point alluded to in the text. This book also contains the most complete citation of sources on the debate up to about 1985. For more recent discussions of the Mises-Hayek argument and the socialist calculation debate, see Rutland (1985, 24-48), Shapiro (1989), and Arnold (1989, 177-91; 1990, 246-63). Much of Nove (1983) restates and illustrates in con­siderable detail some of the general points first made by Mises and Hayek.

6. Lavoie has argued that some of this vital information takes the form of tacit knowledge, which by its very nature cannot be articulated and thus cannot be trans­mitted to the planners (1986, 103).

7. See the quotation from Heilbroner and Thurow. One can be reasonably sure that those who believed this had no detailed empirical knowledge about what life was like under such a system. What has changed in recent years is not the performance of centrally planned economies, which has always been dismal, but the perception of those systems among some segments of the elites in both the West and the East.

8. It is ironic that most formerly socialist countries now have a ministry of priva­tization instead of a ministry of socialization. Though statues of Mises have not yet been erected, many statues of Lenin and Marx have been toppled.

9. See, especially, Schweickart (1980, chap. 3) and Le Grand and Estrin (1989b, 4). This point is discussed in greater detail in the second subsection of the last section of this chapter.

10. There are some difficult and complex issues here in the philosophy of eco­nomics, notably, the usefulness of the Pareto concepts to assess the efficiency of eco­nomic systems.

11. A question that lurks around the edges of this discussion is whether or not there could be some third alternative to central planning and the market. The only contemporary theorists of whom I am aware who advocate neither central planning nor markets are Michael Albert and Robin HahneL See Albert and Hahnel (1978; 1987; 1991). For penetrating criticisms of their model, see Prychitko (1988).

12. As was pointed out in chapter 1 (p. 7), the state in capitalist societies often lim­its the various rights, terms, and conditions associated with ownership of the means of production. However, as indicated, so long as residual rights of control are in pri­vate hands, ownership is private. This is not to deny that there will be borderline cases in which there is no determinate answer to the question of whether or not there is pri­vate ownership, but in general, there is a determinate answer to the question of whether residual rights of control are in private hands or not.

13. The qualifier about a fully functioning market economy is important. At least until recently, socialist thought has been notorious for its vagueness about the insti­tutions for future socialist society. Because of this, the claim in the text has more ana­lytical than historical reach. For example, syndicalism fits neither of the subsequent models discussed in the text; under syndicalism, entire industries are controlled (owned?) by those who work in them. However, syndicalism does not constitute a counterexample to the claim to which this note is appended, since as a matter of empirical fact, there could be no fully functioning market economy in which each industry is effectively constituted by one firm. If for no other reason, problems about monopoly pricing would soon produce institutional paralysis.

14. Nationalization is particularly attractive if all one really cares about is getting rid of capitalism. The simplest and most direct way to achieve that end is to seize state power and nationalize most enterprises. This thought must have occurred to Lenin, probably around the age of six.

15. What is true in principle may not be true in practice. Janos Kornai argues that there is an affinity between markets and private ownership on the one hand and cen­tral planning and state ownership on the other hand (Kornai 1992, 447-50, 497-500). The point at issue here, however, concerns what is theoretically possible.

16. The root of this idea can be found in Marx’s later theory of the state. See Marx ([1852] 1963, 122; [1871] 1975, 329-30). For an illuminating discussion of Marx’s the­ories of the state, see Aveneri (1968, 51-52).

17.1 do not mean to suggest that all of these authors unanimously agree about all of the details that follow in the text. Rather, what all of these writers share, or seem to share, is a belief that the economic system should be changed so that some type of worker cooperative would be the main form of economic organization. Some of these authors are too vague in their descriptions for one to ascribe to them anything more definite than this. What follows is what I believe to be the best motivated type of mar­ket socialist economic system in which some of the more important details about prop­erty rights have been spelled out. The motivations for this type of system will be dis­cussed, primarily in the next section. Chapter 8 discusses some other alternatives.

18. The account of the capital usage fee that follows in the text echoes Schweickart (1980, 52-53).

19. See note 16.

20. Three authors who favor market socialism but do not straightforwardly endorse state control of new investment are David Winter, Saul Estrin, and David Miller (Estrin and Winter 1989, 115-19; Estrin 1989, 183-91; Miller 1989a, 309-11). They see potential dangers to democracy and efficiency in this idea. However, their positive proposals do not make it clear whether they believe that new investment should be determined largely outside of the political process or not. This question is of considerable importance, since political control of society’s economic destiny seems to be a common thread in nearly all nonanarchist socialist thought. The political con­trol of society’s economic destiny is discussed in greater detail in the next section, and alternative forms of market socialism are discussed in chapter 8.

21. Of course, there is some state planning of new investment in existing free enter­prise systems under the guise of industrial policy, but no one seriously maintains that the state controls most new investment in, for example, Germany or Japan. In other words, there is a categorical difference between what passes for industrial policy in existing free enterprise systems and the kind of planning that market socialists envi­sion for the system they favor. This will become clearer in the next section, which exam­ines the motivations for investment planning in this type of market socialist system.

22. For ease of reference, the term ‘market socialism’ will be reserved for this fam­ily of types of economic systems from this point up until chapter 8, where some alter­native forms of market socialism will be discussed.

23. An important caveat emptor: the reader should not infer that I endorse the arguments and explanations that follow in the text. To the extent that anything later in this book depends on accepting or rejecting any of them, I will state and defend my own views at the appropriate time. Also, to keep the length of this chapter within reasonable bounds, I shall attribute various views to Marx without much in the way of supporting arguments and documentation. These interpretive arguments and doc­umentation can be found in Arnold (1990), chaps. 2 (alienation) and 3 (exploitation).

24. This account of alienated labor combines objective features (e.g., lack of worker control) and subjective features (e.g., various forms of dissatisfaction). For more on subjective versus objective alienation, especially as it has shaped research in sociology, see Seeman (1991) and the other articles in Oldenquist and Rosner (1991).

25. Some discussion of these problems is fairly common in the writings of most market socialists, not to mention standard economic textbooks. For a clear, concise statement from a market socialist perspective, see Estrin and Winter (1989, 107-15). For a more elaborate discussion (not in the context of the debate between capitalism and socialism), see Buchanan (1985, chap. 2).

26. For a sensitive discussion of the tension in market socialism between commit­ment to the market and commitment to certain end-states as part of the socialist vision of the good society, see Plant (1989). Not all socialists see the contrast between free enterprise systems and market socialism in this way. Some see it as a choice between the undemocratic procedures that are found in the former and the democratic pro­cedures that would be found in the latter. I owe this observation to Justin Schwartz.

27. I have attempted a comprehensive discussion and critical evaluation of these reformulated arguments in Arnold (1990, chaps. 4 and 5). The reader is directed there for a critical discussion of these arguments. My purpose here is to represent without criticizing central elements of a socialist critique of free enterprise economic systems as a way of motivating the elements of market socialism identified in the last section. In addition, in what follows I will make no attempt to argue for my representation of Marx’s argument as the correct interpretation of what he had in mind. I have argued for this interpretation in detail in Arnold (1990, chap. 3).

28. See Arnold (1990, chap. 3, esp. pp. 74-86) for an argument in support of this harsh judgment on the labor theory.

29. For a discussion of this point, see Becker (1986, 143).

30. This view, or something close to it, is advocated or discussed in Holmstrom (1977, 359); Elster (1985, 167); Miller (1981, 337); Reiman (1987, 4); DiQuattro (1984, 70-71); Young (1978, 441-44).

31. Writers such as Holmstrom (1977) and Reiman (1987) have maintained that this forcing is a logically independent evil (and, in some sense, more important than the failure of reciprocity). In other words, even if the capitalists were entitled to their profits, the unfreedom the worker is subjected to by being forced to work for the cap­italist is independently objectionable. This view is worked out in detail by Justin Schwartz in an unpublished manuscript, “What’s Wrong with Exploitation?” See also Brenkert (1985).

32. Roemer (1982a; 1982b) has a very different conception of exploitation that does not seem to be most naturally understood in terms of a failure of reciprocity. For a critical evaluation of Roemer’s conception of exploitation and the related charge against capitalism, see Arnold (1990, chap. 5).

33. The literature on this is enormous. For some examples, see Nielsen (1985), Dworkin (1981a; 1981b), Griffin (1986), G. A. Cohen (1989), Sen (1987).

34. This latter view of the relationship between the state and the ruling class is argued for by Jon Elster (1988, 213-15).

35. It is on this point that socialists and social democrats have a fundamental dif­ference of opinion. Unlike socialists, social democrats believe that significant redis­tribution by the state can take place in the context of a free enterprise system. The dif­ference between socialists and social democrats on this point may be as much about the meaning of ‘significant’ as it is about the empirical question of how much of the right kind of redistribution the state can accomplish in a society with a free enterprise system.

36. This, of course, is the mirror image of one of the main social vices that has been attributed to state socialist economic systems, namely, their failure to meet people’s basic material needs satisfactorily.

37. Two especially clear examples of this are Levine (1984) and Miller (1989a).

Chapter 3

1. For a comprehensive discussion and critical evaluation of the literature on this charge, SeeArnold (1990, chaps. 3-5).

2. One notable exception is the theory of exploitation developed by Miller (1989a, chap. 7). His theory, like the one presented in this chapter, is a perfectly general account of exploitative exchange.

3. Cohen’s argument does not make exactly this supposition. His claim is only that the capitalist is not a producer, but he seems to believe that this is sufficient for a fail­ure of reciprocity. For a full discussion of this wrinkle in Cohen’s argument, see Arnold (1990, 99).

4. The following is a very compressed statement of criticisms of Cohen and Schweickart that I have made in more detail elsewhere (Arnold 1990, 102-8); see also Arnold (1985).

5. The account Ofexploitation being developed here makes exploitation a prop­erty Oftransactions or exchanges. Though that is the dominant view in the literature on economic exploitation, it is not the only one. Over the past decade, John Roemer has developed a theory of exploitation in which classes are the primary subjects and objects of exploitation (Roemer, 1982a; 1982b; 1985). Individuals are said to be exploited only derivatively insofar as they are members of the exploited class. I have discussed Roemer’s views in detail elsewhere (Arnold 1990, chap. 5).

6. Notice that this distinction all but disappears in the case of the contribution of the laborer, if the labor theory of value were in fact true. According to the labor the­ory, value just is (socially necessary) labor.

7. There may be other forms of exploitation that take place in and through the economic system. For example, sexual harassment and some other forms of degra­dation in the workplace may be forms of exploitation. But however intimately they are tied up with the workings of the economic system, these forms of exploitation are not instances of economic exploitation, since they are not directly concerned with the distribution of economic value (wealth or income). For a discussion of how the work­ers might be noneconomically exploited in a capitalist system, see Arnold (1990, 112-15).

8. For an accessible discussion of ideal markets and perfect competition, see Buchanan (1985, chap. 2).

9. This conception of pure profits and entrepreneurship is due to the so-called Austrian economists (which these days include many Americans and other anglo­phones). For relatively recent and complete statements of this conception, see Kirzner (1973; 1979).

10. It is perhaps best to think of F-exchanges as those that take place within a somewhat narrow range of values, instead of at a particular value. F-exchanges are going to turn out to be a subset of the nonexploitative ones, and it would be implau­sible to say that what makes an exchange exploitative is that its terms differ only slightly from the terms of an exchange that is nonexploitative. For instance, one would not want to say that someone who paid $1,000 for something that was “really” worth only $999 was being exploited.

11. Though ‘fair’ seems to be the most natural term to describe these exchanges, the foregoing is not meant to be an analysis of the meaning of the terms ‘fair’ or ‘fair exchange.’ The choice of the term ‘fair’ is meant to be evocative only. The main sub­stantive claim being advanced here is not about fairness but about the value of what someone has to offer in an exchange.

12. This claim presupposes one restriction and one simplifying assumption. The restriction is that ‘well-being’ is to be understood as referring to well-being insofar as it can be affected by exchangeable goods and services. Some elements of well-being cannot be secured by exchangeable goods and services, which just means that there is more to the good life than what the economic system can provide. This restriction is implicitly recognized in that the topic of this chapter is economic exploitation. The simplifying assumption is that given all the relevant factual information, individuals are the best judges of what enhances their own (material) well-being. In this context, this means that they are better judges than the most likely alternative judge, the state. Though this assumption does not hold in certain cases (which is why it is a simplify­ing assumption), it is difficult to see how one could deny it as a general principle and still favor the market as a way of organizing production.

13. For the sake of simplicity, this illustration abstracts from other determinants of the new equilibrium price, including the government’s monetary, fiscal, and tax poli­cies. These policies can affect the equilibrium price of a good or service in a variety of ways (e.g., through income effects). This simplification does not affect the general point, which is that whatever the determinants of the new equilibrium price, the lead­ing-edge price of a market in transition at any given time is simply whatever price is closest to what will emerge as the new equilibrium price.

14. It is important to note that these so-called unfair exchanges need not be exploitative; unfair terms of exchange are a necessary but not a sufficient condition for exploitation. As the next section argues, if buyers have a “real alternative” to an unfair exchange, then they are not being exploited.

15. For more on the ethical and political implications of this conception of the role of pure profits in a market economy, see Arnold (1987c).

16. This account of the failure of reciprocity is similar to one of the conditions in David Miller’s theory of exploitation (1989a, 186-89). One important difference is that for Miller, the benchmark equilibrium situation is defined not in terms of exist­ing entitlements to resources and the like (as it is in the present account) but instead relative to a set of entitlements that are regarded as morally defensible on other grounds (p. 187). This may lead to a problem in identifying exploitative exchanges insofar as it is difficult to determine what exchange rates would emerge under the preferred set of entitlements. Another difference is that Miller’s account does not countenance the contribution of the entrepreneur.

17. John Roemer (1985) has taken a similar position, though on different grounds.

18. Another possibility is that she does not know that she is giving up more than she is getting. There is some suggestion of this in Marx’s claim that the wage contract helps to mystify the extraction of surplus value from the worker. Though this might be part of his explanation for the medium-term stability of capitalism, the principal Marxist explanation for why the workers put up with the alleged failure of reciproc­ity is that the workers are, in some manner, forced to work for the capitalists.

19. This account is officially agnostic about the question whether or not she is forced to work for the company. It does not really matter. Under the circumstances, she has no feasible alternative to working for the company; if she is getting less than the value of what she is contributing, then she is being exploited.

20. For criticisms of many extant alternative accounts of exploitative exchange, see Arnold (1990, chaps. 4 and 5). The main alternative account of exploitation in which exploitation is not a property of exchanges is Roemer’s. For criticisms of Roe­mer, see Arnold (1990, chap. 5).

Chapter 4

1. As stipulated in chapter 2, for ease of exposition, the term ‘market socialism’ will be reserved for the type of economic system described and motivated there. Chap­ter 8 discusses some alternative forms of market socialism.

2. In some free enterprise systems (e.g., Germany), some members of the boards of directors are representatives of the workers, but these board members are not so numerous or powerful that they (or the workers whom they represent) have residual rights of control over the firm’s assets. In other words, these firms are still essentially privately owned even if workers have some say at the highest levels. Although the sys­tem is called a social market economy (SozialmarkttDirtschaft), it is not a form of market socialism, as the term is used both in this book and, indeed, in most of the theoretical literature on socialism.

3. Indeed, one of the heroines of the Right, Margaret Thatcher, is revered for her privatization drives against the bureaucratic kulaks in Great Britain. There remains, however, an unresolved question about the role of the state in the right-wing con­ception of the good society. This is as it should be, since there is significant disagree­ment on the Right about this question. A more complete defense of the free enter­prise system would have to address this issue.

4. Elsewhere (Arnold 1987a; 1987d) I have argued in detail that cooperatives would have a strong tendency to degenerate into capitalist organizations. Gilboa (1991) shows that this argument is not plausible in the case of cooperatives compara­ble in size to the open corporation. The argument should be restricted to smaller cooperatives, which would be likely to degenerate into classical capitalist firms. A dif­ferent argument would be needed to show that large cooperatives would evolve into open corporations.

5. For more on these Ulysses-and-the-sirens type problems, see Elster (1979).

6. Socialists are likely to object to this presumption as it applies to market social­ism even before they hear the reasons for it. The first section of chapter 6 contains a thorough discussion and defense of this assumption.

7. It is worth pointing out that this book does not aim at a comprehensive evalu­ation of worker cooperatives. Cooperatives can take many different forms, especially as it pertains to the ownership of capital. Moreover, cooperatives can exist not only in market socialist systems but also in free enterprise systems and in systems that are nei­ther socialist nor free enterprise. The focus of this book is narrow in the sense that it is concerned only with cooperatives as specified (and motivated) in chapter 2. The appropriateness of these restrictions is dictated by the larger framework within which this book is located, namely, the capitalism/socialism dispute.

8. This assumption about some stability in the environment is crucial to the evo­lutionary approach. In a rapidly changing environment, advantageous traits will not have the opportunity to take root and spread. This means that the evolutionary hypothesis must be restricted in scope or otherwise qualified to take account of fac­tors that would disrupt the selection process. More qualifications and cautionary observations about the evolutionary strategy are added at the beginning of chapter 5.

9. An unstated assumption in this article is that the central contracting agent has ultimate decision-making authority. Although the authors do not provide it, it is easy enough to give a transactions cost explanation for joining these two roles. The basic idea is that separating them creates a potential for conflict of interest and associated inefficiencies that do not exist if they are joined.

10. Those who provide raw materials and semifinished products may have on­going contractual relations that are difficult to distinguish from the relations that employees have with employers. For this reason, the firm-market boundary on the input side is sometimes blurred.

11. Opportunistic behaviors are necessarily deceptive and unanticipated. If it were widely known up front how much input would be withheld as a result of a person’s opportunism, that could be factored into his remuneration. See also chapter 5, note

6.

12. The notion of a zone of acceptance comes from March and Simon (1958, 90). See also Williamson (1985, 218, 220). It is arguable that these residual rights of con­trol (management rights), together with residual claimancy status (income rights), con­stitute the essence, in some important sense, of ownership. For an illuminating dis­cussion of residual rights of control, see Grossman and Hart (1986).

13. See note 9.

14. Proponents of market socialism undoubtedly believe that the cooperative would be afflicted with much less shirking than the classical firm. The second section of chapter 6 explains why that is probably not true.

15. Since Simon’s earlier writings (e.g., Simon 1961), the psychological picture of these limitations and defects has been filled out in some detail. See Nisbett and Ross (1980) and Kahneman, Slovic, and Tversky (1982) for a more detailed and up-to-date picture of the relevant psychology.

16. Implicit in these different perspectives are, I believe, some important philo­sophical issues for normative economics—issues that warrant a fuller treatment than can be provided here.

17. This is closely related to what Milgrom and Roberts (1990) call “influence costs.” Airplane hijackers usually do not bill the airlines for the expenses they incur in taking over a plane. Organizational hijackers almost always do.

18. What is at work in these examples are bounded rationality and uncertainty in the form of private information. The general problem, of which these examples are illustrations, is that complete, costlessly enforceable contracts governing every con­tingency cannot be written.

Chapter 5

1. This particular comparative claim will be made explicit and argued for in detail in the next section of this chapter.

2. It is not surprising that some of the critical treatments of transactions cost analy­sis have come from sociologists, especially sociologists of organizations. By inclination and training, they are disposed to see the hand of non-economic forces at work. See, for example, Perrow (1986, chaps. 6 and 7), Zald (1988), and Granovetter (1986). For a more sympathetic treatment of transactions cost analysis by a political scientist, see Moe (1984). This article is an excellent comprehensive summary of recent work in transactions cost analysis. Despite his sympathy for much of this work, Moe argues elsewhere (1989; 1990a; 1990b) that the explanatory principles invoked in transac­tions cost analysis have little application to political organizations. More on this in chapter 7.

3. For another good illustration, see Williamson's transactions cost explanation of the common promotion and pay policies (“internal labor markets,” as the phenome­non is sometimes called) found in modern corporations (1975, 57-81).

4. As the term is defined here, in the open corporation most of the capital that the firm uses is owned or supplied by shareholders and thus does not take the form of debt. Of course, there are large corporations in free enterprise systems that do raise much of their capital through debt financing. The transactions cost advantages and disadvantages of debt versus equity financing in large corporations will be discussed at the end of the subsection on equity ownership in the third section of this chapter. See also Jensen and Meckling (1976).

5. Sometimes this type of organization is nominally owned by a family instead of an individual—as, for example, when all and only family members are on a board of directors of a corporation. Incorporation is often done for tax purposes, however. Whatever their legal status, these organizations are essentially classical capitalist firms since one person is in charge. They are to be distinguished from the closed corpora­tion (the term comes from Fama and Jensen 1983b) in which a small number of peo­ple, who may or may not be related to the chief executive officer, have a real and direct equity stake in the firm and have some say over how it is managed. More on the closed corporation at the end of the next section.

6. Notice that shirking and other forms of opportunism are necessarily deceptive and unanticipated. If the behaviors in question (e.g., loafing on the job) are wholly anticipated, their value is reflected in the market price for the factor in question. Everyone—including bosses—expects janitors, nightwatchmen, maintenance men, state highway workers, assistant deans, and so on to spend much of their time out of sight doing nothing productive. Although these people are loafing, they are not shirk­ing. To shirk in jobs like these requires an unusual penchant for opportunism and a really strong antipathy toward productive labor.

7. This is true of the typical case, which is the object of the present investigation. One can imagine unusual circumstances under which other, more egalitarian moni­toring arrangements would be superior. Putterman discusses some of these, but makes no claim that these circumstances are common (1984, 173-74).

8. This is not to be confused with ultimate decision-making authority. Someone may have ultimate decision-making authority without having any operational control, including residual rights of control. Indeed, this is precisely the situation of stock­holders in the open corporation. They have ultimate decision-making authority in the sense that the firm’s assets are managed in their interests. However, they cede resid­ual rights of control—-effective control—to management. More on ultimate decision­making authority in the classical capitalist firm shortly.

9. There are, of course, exceptions. Professional athletes under multi-year con­tracts who have had an especially good year do exactly this (pull a “holdup”) when they threaten not to report to work or when they warn that they will “not be able to concentrate on the game” unless and until their contracts are renegotiated. Unlike other inputs, the central contracting agent does not have physical control over human labor or its owner.

10. Perhaps the most frequently cited article is this literature is Stephen Marglin’s (1974) “What Do Bosses Do?” See also Karen Stone’s (1974) “The Origins of Job Structures in the Steel Industry,” and Charles Perrow’s (1986) Complex Organizations, chaps. 7 and 8. For a general account of economic power that is consistent with trans­actions cost analysis on this point, see Bartlett (1989, chap. 7).

11. This contrasts with the standard Marxist explanation for this phenomenon, which presumes that the capitalist holds onto this power with white-knuckled fear lest the workers discover that they could carry on production just fine without him.

12. See note 4, this chapter. Also, the term ‘equity owner’ seems to have more than one meaning, depending on the context and who is using it. Sometimes, it just means ‘residual claimant,’ but in the law, the equity owner is also liable for the execution of debt. The problem is that economists, lawyers, and accountants have systematically different concerns, so they conceive of equity ownership in systematically different ways. The sentence to which this note is appended is a Stipulative definition adopted for the purposes of this book.

13. Williamson was instrumental in changing the attitude of the U.S. Justice Department toward mergers and acquisitions. The previous policy saw attempts at vertical integration simply as attempts to amass market power with no efficiency advantages. The current policy recognizes that there can be transactions cost effi­ciencies in mergers and acquisitions. (See U.S. Department of Justice 1984, sec. 3.5.)

14. Notice that at the time of its breakup, AT&T, even in its long-distance opera­tions, might have reached its size not because of any transactions cost efficiencies but instead because of its protected status as a monopoly. However (up to a point any­way), large size seems to have been an advantage in the ensuing struggle for survival. Its primary competitors are also quite large, which suggests (but does not prove) that there are transactions cost advantages to large size in this particular market.

15. Economies of scale are usually conceived of as production cost efficiencies instead of transactions cost efficiencies. However, as has been pointed out elsewhere in this book, technology does not directly and uniquely determine an ownership structure. For this reason, it is most appropriate to treat economies of scale as a trans­actions cost efficiency.

16. These forms of opportunism are practiced at all levels in a bureaucratic hier­archy—whether in the public or the private sector—which is not to say that the prob­lems are everywhere the same. In particular, it matters quite a bit who monitors the monitors (more on this in chapters 6 and 7). Nevertheless, the problem is ubiquitous, and because of imperfect monitoring, those who bear the wealth losses of these bureaucratic inefficiencies have nowhere else to go; they are, therefore, exploited. For an entree into the relevant sociological literature, see Perrow (1986); for more spe­cific discussions of the economic inefficiencies of hierarchy, see Williamson (1985, chap. 6), Milgrom and Roberts (1990) and Liebenstein (1987).

17. For a useful discussion of decision making and monitoring in true nonprofit organizations such as charities, cultural organizations, and churches, see Fama and Jensen (1983b, 318-20). Unlike the open corporation, these organizations are char­acterized by the absence of alienable claims on the residuals (however residuals are understood in such an organization) and the active involvement in decision making by people who donate money and time.

18. In practice, this type of organization is fraught with risks for the limited part­ner. For a discussion of some of these risks, see Wolfson (1985). The exact nature of these risks and how they are dealt with need not detain us here, since the concern of this section is the open corporation. Suffice it to say that limited partnerships are rel­atively uncommon and, not coincidentally, they are suited only to unusual economic microenvironments. They are the marsupials of the economic kingdom.

19. The term comes from Jensen and Meckling (1979). They argue that the pure rental firm is an impossibility because certain intangible capital (e.g., product design and engineering, distribution systems, good will) must be owned by the firm and so cannot be rented (1979, 476-78, 480-81). Perhaps, then, one should speak of the ‘nearly pure rental firm,’ though I shall not. Nevertheless, the points about to be made in the text must be slightly qualified or moderated to take into account the possibil­ity that firms must own at least some of the capital that they use. The discussion of this subsection touches on some interesting questions in financial economics about the explanation of the ratio of debt to equity in the open corporation. The seminal arti­cle in this literature is another paper by Jensen and Meckling (1976). More on this issue shortly.

20. The organizational alternatives discussed in this and surrounding subsections seem exceedingly odd and counterintuitive in the abstract. However, they are worth discussing not only because of the comparative nature of the transactions cost effi­ciency claims being advanced on behalf of the distinctive features of the open corpo­ration but also because some of these exotic organizational formations will be found on the wall of Plato’s cave, market socialism. To anticipate chapter 6, under market socialism, the worker cooperative will turn out to have some of the features of the pure rental firm with the taxpayers as capital providers and workers as ultimate decision­makers and residual claimants. Some of the discussion in chapters 6 and 7 details some of the ways in which the taxpayers can be exploited by the workers through this orga­nizational structure.

21. Another reason debt is attractive, at least in the United States, is that it is sub­ject to favorable tax treatment. This, of course, is a non-transactions-cost considera­tion in favor of debt financing.

22. They are also relatively rare. The average debt-equity ratio in nonfinancial institutions in the United States in recent years has hovered below.8. Even in Japan, where firms are more highly leveraged, debt-equity ratios have been below 1.4 (Econ­omist, August 10, 1991, p. 69).

23. See, e.g., Schweickart (1987a; 1987b). The theoretical foundations for this pic­ture were laid by Berle and Means (1932). For an interesting collection of articles on Berle and Means and the accountability of corporate managers, see the June 1983 issue of the Journal of Law & Economics, especially the article by Demsetz (1983).

24. The information in this paragraph comes from personal communication with certified public accountants with experience in major public accounting firms.

25. For a fuller discussion of the ways in which boards can monitor management, see MacAvoy et al. (1983).

26. Demsetz and Lehn (1985) provide some empirical evidence of this point by examining how far ownership is concentrated in the five hundred largest corpora­tions. They find, among other things, that the five largest ownership interests control, on average, 25 percent of the shares issued.

27. Another option is the proxy fight whereby dissident shareholders try to oust managers and boards they believe are incompetent. For a discussion of the relative merits of the takeover as compared to the proxy fight, see Williamson (1983, 361-65).

28. This organizational innovation was invented by Pierre DuPont at the DuPont Nemours Corporation and Alfred P. Sloan at General Motors. The definitive history of the development of the multidivisional corporation is Alfred D. Chandler’s (1966) Strategy and Structure.

29. For references to additional and more elaborate critical discussions of these arguments, see chapter 3, notes 1 and 3-5.

30. For a clear and concise statement of this view, see Hayek (1979). For a more elaborate statement, see Hayek (1960). See also Rothbard (1972).

31. See chapter 3, note 13.

32. I hasten to emphasize the hypothetical nature of the claims made in this exam­ple. The current sorry state of education in the United States has led many thoughtful people to the conclude that teachers are getting far more than the value of their contri­butions and through the power of their unions and their monopoly on public financing of education, are leaving those who employ them with no real alternative but to pay inflated wages. As a general point, it is worth remarking that sometimes the reason that asset owners have nowhere else to go is that their assets are highly overvalued in com­parison to what they would get from other uses or from other users of their assets.

33. What follows is highly speculative. Moreover, even if there is something to it, it is probably a good illustration of how transactions cost efficiencies may be only one factor in explaining the persistence of a phenomenon.

34. Over the past twenty-five years union power has declined in the industrialized West, in part because of the growth of the world economy; that growth has provided both consumers and employers with real alternatives and has resulted in a decline in exploitation by these unions.

35. See note 24.

36. The only other alternative is that the market is highly volatile, that is, prices are in a state of flux and are not heading toward a new equilibrium. Recall from chap­ter 3 that the value of something in such circumstances is undefined. It follows that no such exchange could he exploitative. This may have counterintuitive consequences in the case of an exchange that takes place at terms that are very far out of line with what occurs elsewhere in a volatile market. Perhaps the account of exploitation and the associated definition of value given in chapter 3 should be amended to take this into account, though that is not obvious. This problem has no bearing on the issue at hand, however, since the systematically exploitative nature of the labor contract in a free enterprise system could not be established by appeal to this type of exchange.

37. Of course, another reason that workers might be facing about the same prospects everywhere is that the relevant labor markets are competitively efficient, and workers are, in fact, getting the value of their respective contributions (and thus are not exploited). Notice that if one confuses necessary with sufficient conditions, free enterprise systems would appear to be massively exploitative. Very often, work­ers have no real alternative, in the sense defined, to their current situation. And, very often, workers are not being paid the value of their contributions because the market for their labor is in transition, and they are not on its leading edge. If either or both of these conditions were a sufficient condition (and not merely a necessary condition) for exploitation, then exploitation of workers would indeed be rampant in existing free enterprise systems.

38. In the employment relation, members of these latter two groups often have nowhere else to go, precisely because they are receiving more than the value of their contributions!

Chapter 6

1. The economic literature on the Yugoslav system is voluminous. Information in this and subsequent paragraphs is based on Pejovich (1969), Milenkovitch (1971), Sirc (1979), Estrin and Bardett (1982), Estrin (1983), Lydall (1989) and Kornai (1992, chap. 20).

2. For relevant discussions of Mondragon, see Thomas and Logan (1982), Thomas (1982), and especially Bradley and Gelb (1982).

3. It is important nevertheless to distinguish what people say they believe from what they actually do believe. As Hume points out, “Many eminent theologians... affirm that tho’ the vulgar have no formal principles of infidelity, yet they are really infidels in their hearts, and have nothing like what we call a belief of the eternal dura­tion of their souls. For let us consider on the one hand what divines have display’d with such eloquence concerning the importance of eternity.... And after this let us view on the other hand the prodigious security of men in this particular: I ask if these people really believe what... they pretend to affirm; and the answer is obviously in the negative” ([1739] 1978, 113-14).

4. Though the exact proportion of internally financed new investment cannot be known, it will doubtless be low for the following reason. If a cooperative pays out residuals to its workers, they get IOO percent of those funds. By contrast, if those funds are reinvested, the assets purchased with those funds become part of the firm’s capi­tal; effective ownership of those assets would then revert to the state. Sometimes, the new residuals they can earn from reinvestment will justify the decision to do it, but usually it will make more sense to take the money and run. Also, if the business oppor­tunity they are looking at is sufficiently attractive, it is likely they can fund it through normal channels. Finally, the state would be reluctant to permit or encourage too much new investment to be beyond its purview, since that could easily frustrate its own plans. For more elaborate discussions of internal versus external financing of new investment, see Pejovich (1969; 1973), Furubotn and Pejovich (1970), and Bonin and Putterman (1987, 68-72).

5. For an entree into the substantial empirical literature on this, see Deutsch (1985, chap. 15).

6. As was pointed out earlier, even if some of the workers in Mondragon are not getting the value of their contributions, they are nevertheless not exploited because they do have somewhere else to go, namely, to noncooperative firms elsewhere in the area.

7. The supposition that the unskilled workers are in the majority may not hold in all firms. Firms that sell professional services come to mind as possible exceptions, though they often have substantial supporting staff. However, most firms will have more unskilled (or less skilled) workers if only because the talents and training required of more highly skilled labor are relatively scarce.

8. See Robert Nozick’s discussion Ofjudgments of self-esteem (1974, 239-43). See also Frank (1985).

9. See the second and third sections of chapter 3 for a discussion of why margin- alist thinking about value is the appropriate way to think about the economic value of someone’s contribution.

10. This discussion abstracts from the racial, ethnic, regional, and sexual biases found in every culture. If those biases are factored in, one could imagine different coalitions forming, different distributions of income emerging, and different groups of people being exploited. I owe this point to Richard Arneson.

11. It used to be thought that this phenomenon could be explained solely by ref­erence to the fact that cooperatives aim at maximizing net income per worker, rather than net income (as capitalist firms are assumed to do). This has sometimes been called the “Ward effect” (see Ward 1958). However, Jacques Dreze (1989) has constructed a model of a market economy consisting of labor-managed firms in which workers receive income shares in the firm; these income shares adjust in such a way that labor markets clear as readily in his model as they do in corresponding models of free enter­prise systems. This allows him to prove that his model and a model in which all firms are profit maximizers (i.e., a model of a free enterprise system) are equivalent in equi­librium. Nevertheless, these results have no direct bearing on any of the claims in this book, since Drezes model of the worker-managed enterprise abstracts from many of the details about organizational structure—as well as other nonorganizational factors— that are crucial to the discussion in this and subsequent sections.

12. Estrin’s own proposal for market socialism, which differs from the one under consideration here, does not face these problems in such a stark form. His proposal will be considered in some detail in chapter 8.

13. It does not follow, however, that there would be less exploitation, all things considered, in the very small cooperative. As explained later in this section, there are opportunities for exploitation of the capital providers that more than offset whatever modest advantage very small cooperatives might have in monitoring workers.

14. One of the best discussions of the relationship between base and superstruc­ture in Marx’s thought is to be found in Cohen (1978, chaps. 8-10).

15. Actually, this assumption is overly optimistic on behalf of market socialism in two respects. First, as explained in the last section, cooperatives are less inclined to expand and contract in response to changes in the market. This means that it is pos­sible that some transactions cost inefficiencies (and the opportunities for exploitation they represent) would persist in a market socialist system that would be eliminated in a free enterprise system. Second, in the next chapter, it is argued that mergers and acquisitions would be less frequent in a market socialist system than in a free enter­prise system, which also implies that some transactions cost inefficiencies might per­sist that would otherwise be eliminated.

16. What this assumption means is that if the small cooperative has advantages, as far as exploitation is concerned, over the large, open corporation (or vice versa), that will make no difference in any final overall assessment of the two types of sys­tems. Similar considerations apply to comparisons between the classical capitalist firm and the large cooperative on the potential for exploitation. I believe that restricting the terms of comparison to the small cooperative versus the classical cap­italist firm and the large cooperative versus the open corporation is not conse­quential in a final comparative evaluation of the two types of systems. However, the discussion that follows would be significantly more complicated if the small-to- medium-sized cooperative had to be compared to both the classical capitalist firm and the open corporation and if the large cooperative also had to be compared to both of these types of organizations.

17. This is one reason why it would not be a good idea for governments in former communist countries to insist on profit-sharing cooperatives in their current attempt to join the world market economy. Reports from Eastern Europe, the former Soviet Union, and China suggest that shirking has been and is pandemic under communism. To turn organizations around in these countries, someone is going to have to grapple with deeply entrenched habits and attitudes. To insist that this person share equally all the gains to be realized if he succeeds in these battles would be the height of folly.

18. See the discussion in the second section of chapter 5 of these roles in the clas­sical capitalist firm. In what follows in the text, the assumption that one and the same person is monitor, central contracting agent, and director of the firm’s product is not essential. Though it makes it easier to tell stories about what is likely happen, the main points of those stories remain unchanged if these roles are separated. Indeed, I sus­pect that separating these roles would create additional opportunities for exploita­tion, though I shall not press that point here.

19. Notice that once again, in the very small cooperative, these problems may effec­tively vanish. If everyone is involved in setting (relatively narrow) parameters for rou­tine decision making and if everyone is involved in entrepreneurial decisions, then the opportunities for exploitation under discussion here may well be insignificant. This puts the very small cooperative on virtually the same footing as the very small classical capitalist firm in this respect.

20. To foreshadow some of the results of the second section of chapter 7, this exploitation is likely to go on longer in a market socialist system than in a free enter­prise system because the state will be slower to pull the plug on losing operations than the market is in a free enterprise system.

21. Might not the firm offer him all of the entrepreneurial gains as an inducement to join the firm? Possibly, but that might well put the cooperative on the road to being a classical capitalist firm. For more on this, see Arnold (1987a; 1987d, section 2).

22. The main argument of Arnold (1987c) links entrepreneurial contribution to desert. In other words, entrepreneurs deserve the pure profits (positive or negative) that their firms earn because of their entrepreneurial contribution. It also explains why the fact that the successful entrepreneur does not deserve his special talents and abilities (if indeed that is a fact) does not attenuate his deserts. Independently of this observation, that argument is not fully consistent with the account of the entrepre­neur’s contribution developed in chapter 3 and the observation here about the sig­nificance of luck. It needs to be amended to take these points into account.

23. The points in this paragraph are explained in detail in any basic accounting textbook. See, for example, Kieso and Weygandt (1983, 524-27).

24. Because there are no tradable shares of equity ownership in the cooperative, assessing the quasi-rent value of firms would be very difficult, to say the least. This creates real problems if the cooperatives are expected to pay the state a capital usage fee on this quasi-rent value. More on this in the second section of chapter 7.

25. Some of the points in the remainder of this section are based on Jensen and Meckling’s (1979) discussion of the inefficiencies of the “pure rental firm” and the Yugoslav cooperative.

26. Both types of firms, of course, would want to prevent individual workers from that form of opportunism that consists in misusing equipment to make the job go eas­ier, but the point in the text is about those forms of misuse, abuse, and failure to per­form maintenance that improve the bottom line in the near term.

27. These methods of appropriating quasi-rents were thought up by two econo­mists (Jensen and Meckling) and a philosopher (this author). It is worth point out that all three, especially the latter, are rank amateurs when it comes to these matters. In a market economy that has the structure of ownership rights under discussion, indi­viduals of consummate cleverness and unparalleled rapacity (in other words, the real professionals) would crawl out from under their respective rocks and sell their ser­vices to the highest bidders. It is at least possible that a self-reinforcing process would get underway resulting in the virtual destruction of the economy. Though there is no way to predict this with moral certainty, what can be predicted with confidence is that these opportunities for exploitation would exist and would not go unnoticed.

28. One way to deal with all of these problems would be for the capital providers— the state—to have a direct hand in managing the firm or at least in choosing the man­agers. This, of course, is inconsistent with the fundamental right of worker self-man­agement in the cooperative, which is why it has not been considered here. This possibility will be discussed in chapter 8, which systematically considers various alter­natives to the form of market socialism under discussion here.

Chapter 7

1. In cooperatives of sufficient size, the workers’ council might be so large that it would make sense for them to elect or appoint a smaller body, a kind of politburo, to hire and monitor management. This is the case in Mondragon (Thomas 1982, 135). In what follows, I ignore this possible complication and assume that the workers’ council exercises the same functions that a board of directors does in the open cor­poration. As will become apparent, the interposition of another group between the workers and the monitors of management would not alter the general conclusions of this chapter in a way that is favorable to market socialism. At most, this two-tiered supervisory structure would create additional distinctive opportunities for exploita­tion in a market socialist system.

2. However, the problem of opportunistic on-the-job consumption may not go away; it may instead just go underground. Recent revelations indicate that commu­nist dictators were outstanding practitioners of the art Ofinconspicuous consumption (subject to occasional indiscretions), an art that cooperative managers might revive and master.

3. For the evidence, see Estrin and Bartlett (1982, 86) and the references cited therein.

4. For a variety of reasons, it is also generally true that state officials are not well monitored by the public. This point is discussed in more detail in the next section.

5. This is not universally true. Individuals or institutions that own relatively large blocks of a company’s stock have a more difficult time getting out with their quasi­rents more or less intact.

6. Clearly, in a free enterprise system, managers of target firms have a strong incentive to block acquisitions and sometimes mergers. One of the devices that has evolved over the years to neutralize that incentive is the “golden parachute.” This is a severance payment to senior managers that they themselves can trigger by resign­ing after an unfriendly takeover. Of course, these can be misused. For a useful dis­cussion, see Williamson (1985, 314-16). Imagine the cost of providing “golden para­chutes” for all the ultimate decision makers in a large cooperative who might lose their jobs as a result of a merger or acquisition!

7. For more on the inefficiencies of hierarchy, see Leibenstein (1987) and Williamson (1985, chap. 6).

8. For a useful account of collective versus individual preferences and collective versus individual decision making in this context, see Schweickart (1980, 108) and the references he cites.

9. See Lavoie (1985) for a good discussion of these problems.

10. In free enterprise systems, the public accounting firm reports to the board of directors, which represents the interests of the equity owners, that is, the suppliers of capital. It is likely the state would hire public accounting firms in a market social­ist system, or they might do the audits themselves, since it is their assets that are at risk.

11. Firms in existing free enterprise systems do have other ways of exploiting customers and taxpayers through the state, such as tariffs on imports, regulations that differentially harm competitors, and so on. These forms of exploitation are not discussed here for two reasons. First, they are not features of free enterprise sys­tems per se but are artifacts of the political system of existing free enterprise sys­tems. Generally, defenders of the free enterprise system favor abolishing forms of government intervention designed to allow firms to exploit their customers in these and myriad other ways. Second, even if tariffs and the like were endogenous to free enterprise systems, it is virtually certain that they would be endogenous to a mar­ket socialist systems as well. Indeed, because the economic and political systems are more intimately connected, problems of this sort are likely to be even worse in a market socialist system.

12. The assumption throughout this section is that the organizations that handle new investment (i.e., the banks) are state organizations. They could not be regular profit-making cooperatives because they must be answerable, in some fairly direct way, to political authorities if they are to implement whatever economic plans issue from the planning agencies and, ultimately, the political process. If they were regular cooperatives and their only criterion for making loans was profitability, planning would have been effectively abandoned. Given the assumption of planning, then, it is reasonable to assume that the actual dispensers of investments funds would be state organizations.

13. For a clear, accessible statement of this issue, see Gwartney and Wagner (1988, 11-17). For a useful introduction to public choice theory, see Wagner (1989).

14. The problems with this assumption are addressed in the so-called social choice literature. For overviews of this literature, see Coleman (1988) and Seabright (1989). In this book, I have avoided a discussion of the social choice and public choice litera­tures because both are relatively independent of the details of the organizational struc­ture of public institutions, whereas the focus of this book has been on organizational structures. Though I cannot argue for it here, I believe that an application of the more theoretically secure areas of social choice and public choice would be very damaging to socialists’ faith in the efficacy of political decision making.

15. See the subsection on debt financing in the open corporation in chapter 5 for a discussion of the transactions cost attributes of debt financing.

16. This seems to have been the case in the government bailout of the U.S. savings and loan industry. It has proven difficult to distinguish loans that went bad despite due diligence on the part of lenders from foreseeably bad risks, which in turn have been difficult to distinguish from outright fraud.

17. For surveys of that literature, see Furubotn and Pejovich (1972) and De Alessi (1980).

Chapter 8

1. Some terminological innovation is called for in this chapter, since different ver­sions of market socialism will be identified and discussed. Let us call the version that has been the main subject of this book the worker control-state ownership model. I use the term ‘model’ because it is much less cumbersome than ‘version of market socialism.’ However, it is worth pointing out that the term ‘model’ is misleading because models in economics are typically formal, whereas the objects of discussion in this book are systems of property rights that are specified without the benefit—or hindrance—of formal machinery.

2. The reasons why those who occupy these two roles in a free enterprise system cannot regularly and systematically exploit laborers were discussed in the last section of chapter 5. A crucial part of that story is that, in general, there are adequately func­tioning labor markets. Notice that there are no corresponding capital markets in the worker control-state ownership model, since the state owns all the capital. Another important difference between labor assets in a free enterprise system and (nonhu­man) capital assets in the worker control-state ownership model is that the value of labor assets (i.e., the value of human capital) cannot ordinarily be dissipated or sold off without the owner’s knowledge.

3. A similar point is made by Putterman (1988b, 334). For a survey of portfolio the­ory, seejensen (1972) or Fama (1976).

4. See the account of the amelioration of alienation in the final section of chapter 2 for the main reasons why socialists do and should favor self-management as a means to important elements of the socialist conception of the good society (i.e., social virtues).

5. Ellerman has a conceptual schema for thinking about ownership rights that sys­tematically differs from the more conventional one employed in this book (1990, chap. 3). In accordance with his schema, he denies that the internal capital accounts repre­sent equity ownership (p. 79); instead, he refers to them as internal debt. I have tried to ensure that the differences between his exposition and my exposition of his views are merely terminological.

6. Of course, state bureaucracies in any society may be thought of as economic organizations, since they produce exchangeable goods and services. However, the Ellerman model does not envision widespread state ownership (or other forms of state control) of the means of production.

7. Ellerman’s model also allows for this type of financing, though he does not envi­sion it as the predominant form (1990, 87); see also Vanek (1977a, chap. 11) and McCain (1977, 358-59). This system does not rule out—and, as a practical matter, may require — some worker ownership of shares in the firm. Outside investors may not be willing to invest in a firm in which the workers put up no capital and have no status as residual claimants (Putterman 1988a, 258; Estrin 1989, 181).

8. To be fair to Jay, he seems relatively unconcerned with establishing the social­ist credentials of his system, although he occasionally refers to it as a market socialist system. Instead, his main arguments are addressed to a British audience at the dawn of the Thatcher era. His proposal was intended to solve some problems with that par­ticular system at that time. Jay does not make a concerted effort to link his proposal to the larger socialist tradition. One of my purposes in the text is to show that those linkages are, in fact, quite weak.

9 To buttress his socialist credentials, Estrin might call attention to the fact that he favors a form of noncoercive investment planning called Indicativeplanning (Estrin and Winter 1989, 115-17). Indicative planning involves the state as a provider of infor­mation and a coordinator of some investment plans for the cooperatives. In effect, it is a somewhat more ambitious version of what these days is called “industrial policy.” However, the state’s power to direct resources to particular projects or even sectors of the economy is no greater than the power of the state in free enterprise systems. (It “indicates” rather than “commands” the direction new investment should take.) This has led some observers to question whether or not this process really should be called “planning.” For more on this problem, see Lutz (1969, 99) and Pejovich (1966, chap. 2).

10. For more on these opportunities, see the discussion of equity ownership in the second section of chapter 5. Of course, there are highly leveraged firms in free enter­prise systems, but for reasons explained in the second subsection of the second sec­tion of chapter 5, few firms approach the debt-equity ratios envisioned in Estrin’s model. Moreover, highly leveraged enterprises often cede substantial decision-mak­ing authority to the debt holders, and the latter also receive a risk premium. Under these circumstances, the line between debt and equity becomes blurred, and the nom­inal debt holders begin to resemble equity owners.

11. As noted earlier (see note 9), Estrin favors indicative planning, but it is at least arguable that this more modest role for the state really involves giving up on the idea of collective control of the rate and direction of economic growth.

12. This follows from Marx’s theories of the state. See Marx ([1852] 1963, 122; [1871] 1975, 329-30). See also Elster (1988, 213-15).

13. For a discussion of this see, for example, Barry (1979).

14. I owe this observation about the connection between payment in accordance with contribution and self-determination to James Rachels’s discussion of desert (1978, 159).