Comments on << Principles of Group Solidarity>> by Michael Hechter
Xiangyang Li
The problem of group solidarity is very important almost for every group.
Group solidarity measures the group’s capacity to influence members’ behavior. The more solitary a group, the greater the influence it casts upon its members. Groups influence their members by subjecting them to a variety of obligations to act in the corporate interest and by ensuring that these obligations will be fulfilled. Sociologists provided two solutions to the problem what accounts for the varying solidarity of groups. The first is normativist. The normativists argue that some groups, such as families, are more solitary than others, such as choruses, because the members of families have internalized more extensive norms. Whereas internalization is a perfectly plausible mechanism of solidarity, it suffers from two liabilities. Firstly, we only observe actual behavior and it is difficult to measure the effects of internalization.
In the second place, this explanation begs a fundamental question: why are some groups more effective than others? And normativists do not offer answers. The structuralist solution to the problem is quite different. Individuals are seen coalesce into solitary groups not because they carry group norms within them, but because they share common individual interests. When we are employed by Yale University, we have common interests. In this way, the greater the commonality of individual interest, the greater the resulting solidarity of the group. However the author think that this solution predicts far more group solidarity than the historical record reveals. I guess maybe that can be solved by anal sizing the difference between the representative agent’s expectation and what he gets in the group. If the difference is very small and
1

even what he gets from the group is larger than what he expected, then the solidarity will be larger.
The author develops a rational choice theory of group solidarity. Actors initially form groups, or join existing ones, are to consume various excludable jointly produced goods---goods whose attainment involves the cooperation of at least two individual producers. Those goods are public goods. The characteristic of the public good is that rational actors will choose to free ride that is to consume the good without contributing to its production. How can free riding problem be solved? Three kinds of answers have been proposed. The first answer is coercion: the state enforces social order by threatening to imprison anyone who violates it. The second answer is selective incentives: a professional association sustains itself by supplying various goods selectively, to members alone. The third answer is repeated exchange. To solve the problem, the group requires the establishment of several different kinds of rules.
Once these rules are agreed upon, members actually abide by the rules out of compensation and/or a sense of obligation. Compliance entails solidarity, for these obligations an only be fulfilled by expending some of the members’ private resources on the group’s behalf. In fact, the author thinks that solidarity can be best understood as compliance in the absence of compensation. I think maybe the author emphasizes obligation too much. Many members join in the group for something, which they cannot get by themselves, just as in the economics analysis. We first assume that we are selfish and just care ourselves. I agree with the idea that economists omit many colorful aspects of human characteristics. But I think the most stable (or solidarity) system or groups are those, which can satisfy what the members need.
While dependence increases the extensiveness of obligations, it doesn’t insure compliance with obligations. Compliance requires formal controls because only then
2

can free riding and/or deviance be precluded. Some agency of the group must have the ability to monitor the members’ behavior and to provide sanctions to reward the compliant and punish the noncompliant. Dependence and control capacity are each necessary but insufficient conditions for the production of solidarity. Individual compliance and group solidarity can be attained only by the combined effects of dependence and control. The author thinks about the role of dependence by considering the causes of differential solidarity among legislative political parties. He concludes that party solidarity is enhanced when these representatives are dependent on party leaders for reelection and career advancement. However the quality of the evidence is far from ideal. The author points out that the case of legislative party solidarity is unusual in that the roll call voting procedure enables party leaders to monitor their members costless. Few groups are blessed with such an ample, readymade control capacity.
The necessity of formal controls for the production of joint goods and the attainment of solidarity is challenged by two different kinds of argument about prosocial behavior. The first argument is that prosocial behavior—such as altruism and helping behavior—is likely to emerge in the absence of any controls at all. As I mention above, there is nothing in rational choice that denies that individuals can pursue altruistic or prosocial ends. However we cannot explain why there are so many formal controls in the society. The second thought comes from selection theories or game theory which argues that prosocial behavior can spontaneously emerge either on the basis of selection mechanism or repeated exchange. But we know that the common knowledge base needed to sustain the cooperation is unlikely to be available in large groups. So formal controls are necessary for the attainment of solidarity in large groups. I cannot agree with it more. The most big obstacle confronted with
3

every economist, politics, or other scholars is that we donot know full information.
Every agent does not have full information in the world. Asymmetric and incomplete information is the basic characteristic of interrelation. We only know that we don’t know each other. Just as Knight said, there is something (uncertainty) in the world, which even cannot be measured by probability. Although there are some papers on
AER and other journals, which discussed Knight’s idea, I have not seen operatable models.
Since the formal controls necessary to produce solidarity are themselves a collective good, why would rational members ever consent to establish them in the first place? The author attacked the question by considering the evolution of formal controls in two nonhierarchical groups, rotating credit and insurance groups. It argues that members institute these formal controls when they are the only means of gaining access to some joint good. However the mere existence of these controls offers no guarantee that rational members will abide by them. Whereas the enforcement of production and allocation rules is generally inexpensive in small groups, the costs of enforcement rise exponentially in large ones. These costs are the major obstacle to the attainment of compliance in large groups and in society as a whole.
The author particularly studies the problem of group solidarity and examines the cost of controls in the firms. One of necessities of formal controls is low labor efficiency. Although the firm compensates its workers to comply with production rules, the workers’ interest may lie in shirking rather than in full compliance. In order to reduce the risk of shirking and to attain full compliance with production rules, firms must resort to costly formal controls. There are many different types of controls, such as direct supervision and scientific management. Of course, each has its merits and weakness. The most important factor which determines which of these types is
4

adopted is cost-benefit analysis. Direct supervision typically leaves employers vulnerable to high agency costs. Yet scientific management and other piece-rate schemes are ineffective within it comes to compliance with nonroutine and complex tasks that are difficult to monitor. In the face of tasks that involve costly monitoring, compensation offers a limited basis for compliance. Sometimes firms even establish internal labor markets to solve the problems. The classic example of the internal labor market in industry is the large Japanese corporation with its lifetime employment, seniority-plus-merit wage system, and high level of firm welfare benefits. Firms will always invest in control, but how much and what kind depends on the nature of the good being produced and the opportunities for monitoring and sanctioning. Whether monitoring is based on input or output assessment depends to a significant degree on the nature of the relevant talks. Costs involved in measuring compliance and mobilizing resources for sanctioning are the major constraints on the adoption of any control system. Thus there is no single solution to the problem of control; the most effective strategy in one situation may prove to be relatively costly in another.
I think those ideas are not very new. And you have detailed with those ideas better. The firm is a set of contracts, implicit or explicit, between managers and shareholders, managers and workers, etc. As you have written in your book, accounting is very important in the control system and is useful for group solidarity.
In the macro-level, conventions, standard, and law control firms and its employers’ actions. Those controlling systems can enhance the group solidarity.
I want to emphasize the importance of company’s culture in the control and group solidarity. I think the most efficient way of group solidarity is to set up the compensation and standards. But maybe we can develop some firms’ culture environment and feelings with the employees to enhance group solidarity. It will
5

become more and more important when human capital became the most important resources in the world. I will expound this point in the other paper.
6