Chapter 5 The Logic of Group Behavior In Business and Elsewhere 9 nonmembers and which would (through, for example, the wearing of uniforms) distinguish members from nonmembers; (4) collective sharing of all property and all communal work; (5) submission to public confession and criticism; and (6) expressed commitment to an identifiable power structure and tradition. Needless to say, the cost implied in these “commitment mechanisms” would tend to discourage many free riders from joining the society. By identifying the boundaries to societies, these mechanisms made exclusion possible. As Kanter points out, the importance of these commitment mechanisms is illustrated by the fact that their breakdown foreshadowed the end of the community. Other means of bringing about collective behavior on the part of group members are suggested by the cattlemen’s associations formed during the nineteenth century. During the nineteenth century, cattle were allowed to run free over the ranges of the West. The cattlemen had a common interest in ensuring that the ranges were not overstocked and overgrazed and in securing cooperation in rounding up the cattle. To provide for these common interests, cattlemen formed associations which sent out patrols to keep out intruders and which were responsible for the roundups. Any cattleman who failed to contribute his share toward these ends could be excluded from the association, which generally meant that his cattle were excluded from the roundup or were confiscated by the association if they were rounded up. 13 The family is a small group, which by its very nature is designed to promote the common interest of its members. That common interest may be something called “a happy family life,” which is, admittedly, difficult to define. The family does not escape difficulties. At present its validity as a viable institution is being challenged by many sources; however, it does have several redeeming features that we think will cause it to endure, imperfect though it may be, as a basic component of social fabric. Because of the smallness of the group, contributions made toward the common interest of the family can be shared and appreciated directly. Parents usually know when their children are failing to take the interest of the family into account, and children can easily ascertain similar behavior in their parents. Family members are able, at least in most cases, to know personally what others in the group like and dislike; they can set up an interpersonal cost- and-benefit structure among themselves that can guide all members toward the common interest. Most collective decisions are also made with relative ease. 14 However, even with all the advantages of close personal contact, the family as a small group often fails to achieve the common interest. Although all family members may be encouraged to “go their own way” up to a point, some individuals may take this too far. They may fail to contribute their share to the common goal and may cause bitterness and, perhaps, the demise of the family. Given the frequent failure of the family as a viable organization 13 For a very interesting historical investigation of the cattle business during the late nineteenth century, see Rodgers Taylor Dennen, “From Common to Private Property: The Enclosure of the Open Range,” Ph.D. dissertation, University of Washington, 1975. 14 See, for more discussion on the economics of the family, Richard B. McKenzie and Gordon Tullock, “Marriage, Divorce, and the Family,” in The New World of Economics (Homewood, Ill.: Richard D. Irwin, Inc. 1978), chap. 8 Chapter 5 The Logic of Group Behavior In Business and Elsewhere 10 with a common interest, 15 the failure of much larger groups to achieve their expressed common objectives is not difficult to understand. Large Groups In a large-group setting, the problems of having individual members contribute toward the development of the common interest are potentially much greater. The direct, personal interface which is present in small groups is usually lacking in larger groups; and, by the nature of large groups and the public good they produce, the benefits generated by any one person are generally spread over a large number of people, so much so that their actions have a significant effect on anyone, even themselves. As a result, they may perceive neither direct benefits in terms of what their behavior does for themselves, personally, nor indirect benefits in terms of what their behavior contributes to the welfare of others. On the other hand, an individual may be able to detect benefits from his actions, but he must weigh these benefits against the costs he may have to incur to achieve them. For a large group the costs of providing detectable benefits can be substantial—or they can escalate with the size of the group. This is not only because there are more people to be served by the public good, 16 but also because large groups are normally organized to provide public goods that are rather expensive to begin with. Police protection, national defense, and schools are examples of very costly public goods provided by large groups. If all people contribute to the public good, the cost to any one person can be slight; but the question confronting the individual is how much he will have to contribute to make his actions detectable, given what all the others do. In the context of a very large group, suppose there are certain common national objectives to which we can all subscribe, such as a specific charitable program. It is, in other words, in our “common interest” to promote this program. Will people be willing to voluntarily contribute to the federal treasury for the purpose of achieving this goal? Certainly some people will (as Harry does in Figure 5.1 with a marginal cost of MC 2 ) , but many people may not. As they do each April 15 (the deadline for filing tax returns), most will contribute as little income tax as possible. Under a system of voluntary contributions, some people will contribute nothing. A person may reason that although he agrees with the national objective, or common interest, his contribution—that which he can justify—will do little to achieve it. He can also reason that withholding his contribution will have no detectable effect on the scope and effectiveness of the program. (If you or your parents did not pay taxes, would the level of public goods that benefit you 15 Approximately one-third of all families based on the institution of marriage end in divorce. Many others fail, in terms of the presence of intense hostility, even though there is no legal recognition of that fact. 16 For a pure public good, the costs, by definition, do not rise with a few additional members. However, most groups provide services that are less than a pure public good. Education is an example of an impure public good; all education does not benefit all members of society simultaneously and to the same degree. Under these circumstances, the costs can rise, as we have suggested, with the membership, although by a lower percentage. Chapter 5 The Logic of Group Behavior In Business and Elsewhere 11 be materially affected?) It is for this reason that compulsory taxes are necessary. Olson writes: Almost any government is economically beneficial to its citizens, in that the law and order it provides is a prerequisite to all civilized economic activity. But despite the force of patriotism, the appeal of the national ideology, the bond of a common culture, and the indispensability of the system of law and order no major state in modern history has been able to support itself through voluntary dues or contributions. Philanthropic contributions are not even a significant source of revenues for most countries. Taxes, compulsory payments by definition, are needed. Indeed, as the old saying indicates, their necessity is as certain as death itself. 17 The general tenor of the argument also applies to contributions that go to CARE, a voluntary charitable organization interested mainly in improving the diets of impoverished people around the world. Many of the students reading these pages have been disturbed by scenes of undernourished and malnourished children shown in television commercials for CARE. All those who are disturbed would probably like to see something done for these children. They have had an opportunity to make a contribution, but how many people ever actually contribute so much has a dollar? Needless to say, many do give. They are like Harry in Figure 5.2, who is willing to dig, voluntarily, some of the weeds from his yard. On the other hand, we emphasize the point that a large number of people who have been concerned never make a contribution. (It would be an interesting classroom experiment to see how many students are disturbed by the CARE commercials and how many have ever given to the organization.) There are many reasons for people not giving, and we do not mean to understate the importance of these reasons; we mean only to emphasize that the large-group problem is one significant reason. True, if all members of a large group make a small contribution toward the common interest, whatever it is, there may be sizable benefits to all within the group. But, again, the problem that must be overcome is the potential lack of individual incentives form which he collective behavior must emerge. Through appropriate organization of group members, the common interest may be achieved, even if the membership is large. This, however, merely shifts our attention to the problem of developing that organization. The organization of a large group can be construed as a public good, and there are likely to be costs to making the organization workable. This is likely for two reasons: first, there are a large number of people to organize, which means that even if there is no resistance on the part of the people to be organized, there will be costs associated with getting them together or having them work at the same time for the same objectives. Second, some individuals may try to “free-ride” on the efforts of others, which means it will cost more to get people to become members of the group. Further, each free rider implies a greater burden on the active members of the group. If everyone waits for “the other guy to take the initiative,” the group may never be organized. It is because of the organization costs 17 Olson, Logic of Collective Action, p. 13 Chapter 5 The Logic of Group Behavior In Business and Elsewhere 12 that students complain so often about the instructional quality of the faculty or some other aspect of university life without doing anything about it. This is also why most people who are disgruntled with the two major political parties do not form a party with those who share their views. The probability of getting sufficient support is frequently very low, which is another way of saying the expected costs are high. Because an organization may appear to be an obvious way to promote the public good, individuals who try to organize people for that purpose may go through a learning experience before they conclude that it is too costly a venture for them. Even if the organization is successful, the success may be temporary. Eventually, the free-rider problem emerges and the group may fall apart. During the winter of 1973-74, the United States was in the midst of an “energy crisis.” Prices of gasoline and other fuels were being held down in spite of the limited imports of fuel coming into the country from the Middle Eeast. Truckers were having a difficult item obtaining adequate supplies of diesel fuel and of passing their higher operating costs through to the buyers of truck services. Independent truckers sensed that it was in their common interest (not the public’s, of course,) to halt their deliveries of goods and services and, in that way, put pressure on the authorities to increase rates and to allocate more fuel supplies for the use of truckers. The call for cooperation met with some success; some truckers did terminate operations and some caught headlines by blocking traffic on major highways. However, there were many unwilling to go along with the work stoppage—something that was in their common interest. Consequently, the supporters of the work stoppage resorted to violence, and it was the threat of violence, and not the common interest, which kept many truckers off the road. If it had not been for the violence and the initial willingness of state police departments to allow truckers to flaunt the law by stopping traffic, including other truckers, it is very doubtful that the truckers would have had as much success as they did. Qualifications to the Economic Theory Obviously, there are many cases in which people acting in what may appear to be rather large groups try to accomplish things that are in the common interest of the membership. The League of Women Voters during the mid-1970s pushed hard for passage of the Equal Rights Amendment. To the Constitution; labor unions work for wage increases; and the American Medial Association does lobby for legislation that is in the common interest of a large number of doctors. Churches, the Blood Mobile, and other charitable groups are able to work fairly effectively for the “public interest,” and several of the possible explanations for this observed behavior force us to step outside the scope of the public goods theory. Why may people work for the “public interest”? First, as Immanuel Kant, an eighteenth century philosopher, said they should, people can place value on the act itself as distinguished from the results or consequences of the act. The act of making a charitable contribution, which can be broadly defined to include picking up trash in public areas or holding the door for someone with an armful of packages, may have a value in and of itself. This is true whether the effects of the act are detectable to the individual making he charitable contribution or not. The personal satisfaction (or value) Chapter 5 The Logic of Group Behavior In Business and Elsewhere 13 that comes form the act itself is probably the dominant reason why some people do give to CARE. To the extent people behave in this way, the public good theory loses force. Notice, however, that Olson, in formulating his argument, focused on rational economic man as opposed to moral man, envisioned by Kant. We expect that as the group becomes larger, greater effort will be made to instill people with the belief that the act itself is important. Second, the contribution that a person has to make in group settings is often so slight that even though the private benefits are small, the contribution to the common interest is also small and can be a rational policy course. This may explain, for example, student membership in groups like the National Association of Student Teachers. All one has to do in many situations like this one is show up at an occasional meeting and make a small dues payment. Further, the private benefits of being with others at the meetings and finding out what the plans are for the association can be sufficient incentive to motivate limited action that is in the common interest. Third, all may not equally share the benefits received by group members from promotion of the common interest. One or more persons may receive a sizable portion of the total benefits and, accordingly, be willing to provide the public good, at least up to some limit. Many businessmen are willing to participate in local politics or to support advertising campaigns to promote their community as a recreational area. Although a restaurant owner may believe the entire community will benefit economically from an influx of tourists, he is surely aware that a share of these benefits will accrue to himself. Businessmen may also support such community efforts because of implied threats of being socially ostracized. Fourth, large organizations can be broken down into smaller groups. Because of the personal contact with the smaller units, the common interest of the unit can be realized. In promoting the interest of the small unit to which they belong, people can promote the common interest of the large group. The League of Women Voters is broken down into small community clubs that promote interests common to other League clubs around the country. The Lions Club collectively promotes programs to prevent blindness and to help the blind; they do this through a highly decentralized organizational structure. Political parties are structured in such a way that the local precinct units “get out the votes.” The surest way for a presidential contender to lose an election is to fail to have a “grass-roots” (meaning small-group) organization. Churches are organized into congregations, and each congregation is decentralized further into circles and fellowship groups. Most of the work in the Congress is done in committees and subcommittees. Quiet often a multiplicity of small groups is actually responsible for what may appear to be the activity of a large groups. The decentralization that is so prevalent among voluntary groups tends to support the economic view of groups 18 18 Admittedly, other explanations for decentralization can be made, one of which relates to diseconomies of scale. That is, the organization just becomes technically less efficient as its size is expanded. The economic theory of groups rests on the motivational aspect of large organizations, rather than on the technical capabilities of the organization. Chapter 5 The Logic of Group Behavior In Business and Elsewhere 14 Fifth, large groups may be viable because the group organizers sell their members a service and use the profits from sales to promote projects that are in the common interest of the group. The Sierra Club, which is in the forefront of the environmental movement, is a rather large group that has members in every part of North America. The group receives voluntary contributions from members and nonmembers alike to research and lobby for environmental issues. However, it also sells a number of publications and offers a variety of environmentally related tours for its members. From these activities, it secures substantial resources to promote the common interest of its membership. The American Economics Association has several thousand members. However, most economists do not belong to the AEA for what they can do for it. They join primarily to receive its journal and to be able to tell others that they belong—both, private benefits. (The AEA also provides economists with information on employment opportunities.) Sixth, the basic argument for any group is that people can accomplish more through groups than they can through independent action. This means that there are potential benefits to be reaped (or, some may say, “skimmed off”) by anyone who is willing to bear the cost of developing and maintaining the organization. A business firm is fundamentally a group of workers and stockholders interested in producing a good (a public good, to them). They have a common interest in seeing a good produced which will sell. The entrepreneur is essentially a person who organizes a group of people into a production unit; he overcomes all the problems associated with trying to get a large number of people to work in their common interest by providing workers with private benefits -- that is, he pays them for their contribution to the production of the good. The entrepreneur-manager can be viewed as a person who is responsible for reducing any tendency of workers to avoid their responsibilities to the large-group firm. Because it is in their interest to eliminate shirking, the workers may be just as interested as stockholders in having and paying someone to perform this task. 19 An individual worker may be delighted if he is allowed to remain idle while no one else is, but he will want to avoid the risks of all workers shirking. If all shirk, nothing will be sold, the firm will collapse, and workers will lose their wages. We may, therefore, expect that even in communist societies, managers will be paid handsomely (relatively speaking) for the tasks they perform. It is interesting to note that the wage differential between workers and managers is greater in the Soviet Union than it is in the United States. 20 MANAGER’S CORNER I: The Value of Tough Bosses What does the “logic of group behavior” have to do with the direct interest of MBA students who seek to run businesses and direct the work of others? In a word, “plenty,” as we will see throughout the rest of the book. We will show how the “logic” is central to 19 These points have been made in a much more complete and technical manner by Armenia A. Alton and Harold Demotes, “Production, Information Costs, and Economic Organization,” American Economic Review, vol. 62, pp. 777-795, December 1972 20 Some managers in the Soviet Union are paid less than industrial workers in the United States; however, the ratio of a manager’s salary to a worker’s salary is typically greater in the Soviet Union. Chapter 5 The Logic of Group Behavior In Business and Elsewhere 15 how competitive markets (and cartels) work and will discuss a multitude of ways to apply the “logic” directly to management problems. For now, we can stress a maxim that emerges from the economic view of group behavior: Being (or having) a tough boss is tough, but a boss who isn’t tough isn’t worth much. And because tough bosses are valuable, and lenient bosses are not, there is a reason for believing that existing organizational arrangements serve to impose the discipline on bosses necessary to ensure that they do a good job imposing discipline on the workforce. Competition will press firms to hire tough bosses, and, as we will show in this chapter, the owners of the firm, or their manager-agents, not workers, will tend to the bosses. That is to say, owners or their agents will tend to boss workers, not the other way around, for the simple reason that worker-bosses will not likely survive in competitive markets. Workers may not like tough bosses, but we will explain that, if given the option, workers would choose to hire tough bosses. 21 Everyone recognizes that firms compete with each other by providing better products at lower prices in a constant effort to capture the consumer dollar. This competition takes place on a number of fronts, including innovative new products, cost cutting production techniques, clever and informative advertising, and the right pricing policy. But a continuing theme of this and other management books is that none of these competitive efforts can be successful unless a firm backs them up with an organizational structure that is competitive -- one that motivates its employees to work diligently and cooperatively. Before addressing the issue of organization, however, let’s first examine why workers value tough bosses. Those firms that do the best job in this organizational competition are the most likely to survive and thrive. The organizational arrangements used by the most successful firms are most likely to be adopted by other firms, because of the force of profit maximization and market competition. So we should expect business firms to be organized in ways that motivate bosses to work diligently at motivating workers to work diligently and at the least cost. We should expect that the choice between workers and owners of capital as to which group will market the better bosses will depend on which group can be expected to press the other to work the most diligently or at the least cost. We have already given away the answer: Owners (or their manager-agents) will tend to boss the workers, a perfectly acceptable outcome for the owners, of course, but also for the workers, which might not be expected. To understand that point, we must first appreciate why workers would want tough bosses. Take this Job and . . . Though probably overstated, common wisdom has it that workers do not like their bosses, much less tough bosses. The sentiment expressed in the well-known country song “Take This Job and Shove It” could only be directed at a boss. Bosses are also the butts of much humor. There is the old quip that boss spelled backward is “Double SOB.” 21 As we will see, even when workers own the firm and could be their own bosses, they invariably hire a boss, typically a tough one at that. Chapter 5 The Logic of Group Behavior In Business and Elsewhere 16 And there is the story about the fellow who went to the president of a major university and offered his services as a full professor. Noticing that the fellow had no advanced degree, the president informed him that he was unqualified. The fellow then offered his services as an associate professor and received the same response. After offering his services as an assistant professor and hearing that he was still unqualified, the fellow muttered. “I’ll be a Son-of-a-Bitch,” at which point the president said, “Why didn’t you tell me earlier? I’m looking for someone to be dean of the business school.” If it were not for an element of truth contained in them, such jokes would be hopelessly unfunny. Bosses are often unpopular with those they boss. But tough bosses are much like foul tasting medicines are for the sick; you don’t like them, but you want them anyway because they are good for you. Workers may not like tough bosses, but they willingly put up with them because tough bosses mean higher productivity, more job security, and better wages. The productivity of workers is an important factor in determining their wages. 22 More productive workers receive higher wages than less productive workers. Firms would soon go bankrupt if they paid workers more than their productivity indicated they should be paid, but firms would soon lose their workers if they paid them less than their productivity. Many things, of course, determine how productive workers are. The amount of physical capital they work with, and the amount of experience and education (human capital) the workers bring to their jobs are two extremely important, and commonly discussed, factors in worker productivity. But how well the workers in a firm work together as a team is also important (a point that will become more apparent in the “Manager’s Corner” on “The Value of Teams” later in this chapter). An individual worker can have all the training, capital and diligence needed to be highly productive, but productivity will suffer unless other workers pull their weight by properly performing their duties. The productivity of each worker is crucially dependent upon the efforts of all workers in the vast majority of firms. So all workers are better off if they all work conscientiously on their individual tasks and as part of a team. In other words, it is collectively rational for everyone to work responsibly. But there is little individual motivation to work hard to promote the collective interest of the group, or firm. 23 While each worker wants other workers to work hard to maintain the general productivity of the firm, each worker recognizes that her contribution to the general productivity is small. By shirking some responsibilities, she receives all of the benefits from the extra leisure but suffers from only a very small portion of the resulting productivity loss, which is spread over everyone in the firm. She suffers, of course, from some of the productivity loss when other workers choose to loaf on the job, but she 22 It is also true, as we will see in a later chapter, that how wages are paid can be an important factor in determining how productive workers are. 23 This line of analysis has been developed at length by Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups (Cambridge, Mass.: Harvard University Press, 1965). Chapter 5 The Logic of Group Behavior In Business and Elsewhere 17 knows that the decisions others make are independent of whether she shirks or not. And if everyone else shirks, little good will result for her, or for the firm, from diligent effort on her part. So no matter what she believes other workers will do, the rational thing for her to do is to capture the private benefits from shirking at practically every opportunity. With all other workers facing the same incentives, the strong tendency is for shirking on the job to reduce the productivity, and the wages, of all workers in the firm, and quite possibly to threaten their jobs by threatening the firm’s viability. The situation just described is another example of the general problem of the logic of group behavior, or more precisely a form of the prisoners’ dilemma that is endemic to that logic. This involves a classic police interrogation technique in which officers separate two suspects, indicating to each that if she confesses, then she will get off with light charges and penalties. Collectively, they might both be better off if neither confesses (which implies that the two suspects work together for their common objective, a lighter sentence), but each can be even better off if she confesses while her cohort doesn’t. More formally, a prisoners’ dilemma is a situation in which each individual is better off by acting independently of other parties in the group, no matter what the other parties do, but all parties in the group are better off by working together. Consider a slightly different form of the prisoner’s dilemma that is described in the matrix in Table 5.1, which shows the payoff to Jane for different combinations of shirking on her part and shirking on the part of her fellow workers. 24 No matter what Jane believes others will do, the biggest payoff to her (in terms of the value of her expected financial compensation and leisure time) comes from shirking. Clearly, she hopes everyone else works responsibly so that general labor productivity and the firm’s profits are high despite her lack of effort, in which case she receives the highest possible payoff that any one individual can receive of 125. 25 Unfortunately for Jane, all workers face payoff possibilities similar to the ones she faces (and to simplify the discussion, we assume everyone faces the same payoffs). So everyone will shirk which means that everyone will end up with a payoff of 50, which is the lowest possible collective payoff for workers. 26 Workers are faced with self-destructive incentives when their work environment is described by the shirking version of the prisoners’ dilemma (which we have discussed now in several other contexts). It is clearly desirable for workers to extricate themselves from this prisoners’ dilemma. They can double their gain. But how? 24 The payoff can be in dollars, utility, or any other unit of measure. The only important consideration is that higher numbers represent higher payoffs. This is in contrast to the original prisoners’ dilemma example in which the number in the payoff matrix represented the length of prison sentences, so the higher number represented lower payoffs. 25 Of course, not everyone can receive this payoff. 26 Jane would receive a lower payoff of 25 if she were the only one who did not shirk, but because of her effort the collective payoff would be higher than if she did shirk, as her effort would raise the payoff to the shirkers to something slightly higher than 50. Chapter 5 The Logic of Group Behavior In Business and Elsewhere 18 Table 5.1 The Inclination to shirk on the Job Other Workers None shirk Some shirk All shirk Don’t shirk 100 75 25 Jane Shirk 125 100 50 In an abstract sense, the only way to escape this prisoners’ dilemma is to somehow alter the payoffs for shirking. More concretely, this requires workers to agree to collectively subject themselves to tough penalties that no one individual would unilaterally be willing to accept. While no one will like being subjected to tough penalties, everyone will be willing to accept the discipline those penalties impose in return for having that discipline applied to everyone else. The situation here is analogous to many other situations we find ourselves in. For example, consider the problem of controlling pollution that was briefly mentioned in an earlier chapter. While each person would find it convenient to be able to freely pollute the environment, when everyone is free to do so we each lose more from the pollution of others than we gain from our own freedom to pollute. So we accept restrictions on our own polluting behavior in return for having restrictions imposed on the polluting behavior of others. Littering and shirking may not often be thought of as analogous, but they are. One pollutes the outside environment and the other pollutes the work environment. An even better analogy is that between workers and college students. The “productivity” of a college from the student’s perspective depends on its reputation for turning out well-educated graduates with high grade a reliable indication that a student has worked hard and learned a lot. But students are tempted to take courses from professors who let them spend more time at parties than in the library and still give high grades. But if all professors curried favor with their students with lax grading policies, all students would be harmed as the value of their degrees decreased. While students may not like the discipline imposed on them by tough professors, they want tough professors to help them maintain the reputation of their college and the value of their diplomas. (The ideal situation for each student is for the professor to go easy on him or her alone and to be demanding of all other students. 27 ) Similarly, workers may not like bosses who carefully monitor their behavior, spot the shirkers and ruthlessly penalize them, but they want such bosses. We mean penalties sufficiently harsh to change the payoffs in Table5.1 and eliminate the prisoners’ dilemma. As shown in Table 5.1, the representative worker Jane captures 25 units of benefits from shirking no matter what other workers do. If she had a boss tough enough to impose more than 25 units of suffering, say 35 units, on Jane if she engaged in shirking, her relevant payoff matrix would be transformed into the one shown in Table 5.2. Jane may not like her new boss, but she would cease to find advantages in shirking. And with a 27 See Dwight Lee, “Why It Pays to Have Tough Profs,” The Margin (September/October 1990): 28-29. . will result for her, or for the firm, from diligent effort on her part. So no matter what she believes other workers will do, the rational thing for her to. try to organize people for that purpose may go through a learning experience before they conclude that it is too costly a venture for them. Even if the