ABSTRACT
Why do people make collective decisions by one procedure rather than another? A collectivity’s first decision is necessarily made by a mode from the agreement-onoutcome family. In most cases, this first decision involves voluntary participation, and the procedures for making further collective decisions are then chosen deliberately and collectively, from a range of possibilities. On the other hand, it is possible for a collectivity’s first decision to involve coercive participation, subjugating one individual or group to another, in which case subsequent collective decisions can represent the exercise of authority by the subjugator over the subjugated. Then the choice of a mode expresses the exercise of power rather than pursuit of the common good. But such cases are exceptional. Thus in most cases collective decision procedures are chosen because they satisfy criteria that make them more attractive than their alternatives. This chapter explores a taxonomy of these sometimes conflicting criteria. The taxonomy is shown in Table 4.1. Table 4.1 A taxonomy of criteria for evaluating collective decision procedures Efficiency Equity Stability
It might be argued that the selection of a procedure determines the outcomes that will be selected through its use, and therefore collectivities do not apply criteria to procedures, but rather to outcomes implied by procedures. While there is an element of validity to this argument, it is usually not possible to foresee precisely what outcomes are implied by the choice of a procedure. Thus procedures are evaluated on the basis of generalizations about their likely consequences, with the outcomes that will be chosen remaining conjectural. The explanation of preferences for collective decision procedures therefore involves entering the imagination of a person confronted with the uncertain prospects of different
procedures and asking what criteria he or she would apply. This method has much in common with that of Buchanan and Tullock in The Calculus of Consent. Efficiency The appeal of efficiency is the appeal of achieving more of whatever people want. More can be achieved either by increasing outcome efficiency, that is, by increasing the overall attractiveness of selected outcomes, or by increasing procedural efficiency, that is, by using a less costly procedure for selecting outcomes. Ideally, the efficiency of an outcome would be measured by summing the amounts of utility that individuals gain and lose by having that outcome instead of a specified alternative. However, since there is no way of measuring individual utility that yields incontrovertible results, economists generally measure the efficiency of an outcome by summing the amounts of money that the members of a collectivity would be willing to give up, or would need to be paid, to have that outcome instead of a specified alternative. For example, if in a household of seven persons there are three who would value a one-hour postponement of the time when dinner is served at $5 each, while the other four would experience a cost of $3 each for the postponement, then an economist would say that it is efficient to postpone the time when dinner is served, because the benefits ($15) exceed the costs ($12). This calculation makes no allowance for the possibility that money is more valuable to some persons than to others. This accommodation, in the measurement of efficiency, to what is feasible means that measured outcome efficiency is a less pertinent concept than might be desired, but it still has a significant bearing on the attractiveness of collective decision procedures. Procedural efficiency is measured, inversely, by the magnitude of the resource and psycho-social costs of making decisions. One important resource cost of making collective decisions is the time of members of the collectivity. Time spent making decisions generally has valuable alternative uses. There may also be transportation costs in bringing the members of the collectivity together physically, or communication costs in bringing them together by mail, telephone or other means. There may be costs of preparing materials and exhibits for the members of the collectivity, and of paying speakers for them to hear. In addition to resource costs, there are psycho-social costs of making collective decisions. For example, if an academic department tries to reach a consensus on how money available for raises ought to be allocated among them, they may experience a deterioration in goodwill among themselves that is a cost of that way of making collective decisions. When collective decisions are made by extortion, additional procedural costs are encountered. These are the costs from deliberate destruction of resources,
injury to persons, pain, loss of freedom and death. There are also psychological costs from the threat of these events. A final and very significant element of procedural cost from extortion is the expense that individuals incur to avoid being subjected to the direct costs of extortion. The combination of outcome efficiency and procedural efficiency would fully account for the attractiveness of collective decision procedures only if individuals were entirely selfish, and if they believed either that money always had the same value to all persons or that it was not possible to identify factors that are correlated with the value of money. But these conditions are not met. People are not entirely selfish. And it is generally unreasonable to suppose either that money always has the same value to all persons or that it is not possible to identify factors that are correlated with its value. Thus the distributional considerations of equity and stability, as well as efficiency, enter into the evaluation of collective decision procedures. Equity Equity is a distribution of goods that conforms to shared understandings regarding what people ought to receive. The taxonomy identifies three aspects of equity: symbolic equality, material equality, and productivity-based inequality. Symbolic equality is the equality of one-person-one-vote, of equality before the law, of equality of opportunity. Symbolic equality possesses the attraction of the idea that no one of us is inherently more worthy than any other. Material equality, on the other hand, is equality in the control of physical resources. Material equality is measured, inversely, by an index of inequality like the Gini coefficient,1 applied to income or wealth. The attraction of material equality comes substantially from the utilitarian idea that additional resources are more valuable to those who have little than to those who have much. Productivity-based inequality is allocation according to the value of what each person adds to the output of a collectivity by his or her efforts (technically, the person’s marginal product). The attraction of productivity-based inequality is illustrated by the widespread agreement that those who gratify their inclinations toward laziness ought to receive less than otherwise similar persons who overcome such inclinations. There are four interrelated but distinct reasons why it is reasonable to expect concern for the two aspects of equality to influence preferences for collective decision procedures. First, many persons find symbolic equality an attractive ideal, something they strive to express in their private and public lives. Promoting material equality is a way of giving expression to a commitment to symbolic equality. Second, the implementation of collective decisions often requires individuals to act against their short-term material interests. One way of inducing individuals to
do so is to inspire loyalty in them, and commitment to symbolic equality among the members of a collectivity is instrumental in generating loyalty. Third, the level of one’s material resources is the most obvious factor that is likely to be correlated (negatively) with the value of money. By choosing a collective decision procedure that sacrifices some efficiency for equality, one is, in effect, purchasing wealth insurance: One’s expected wealth is lower, but in the event that one is among those who do not fare so well, more assistance is provided. Fourth and finally, people tend to care differentially about others and to have more concern for those with less wealth than for those with more. Therefore many people value changes that increase the degree of equality among others than themselves. By choosing collective decision procedures that generate more equality, individuals advance their interest in securing greater equality among others than themselves. While much of our concern for equity is captured by the concepts of symbolic and material equality, productivity-based inequality attracts us even though it diminishes equality. One of the most widely-quoted statements of the virtue of productivity-based inequality occurs in John Locke’s Second Treatise of Government:
That productivity-based inequality is attractive is easiest to see at the extremes of productivity. Consider people who are bad-tempered and compulsive liars. It is possible that if we were convinced that these qualities were inherited, we would want to compensate individuals who possess them for the misfortune of their genes. However, we believe that there is at least an element of free will in individual behavior. And we want individuals to receive less when, through their free will, they diminish the well-being of others by expressing their tendencies toward bad temper and compulsive lying than when they control these tendencies. Thus complete equality is not what we seek. At the other end of the productivity spectrum, consider individuals like Shakespeare, Mozart, Jefferson and Einstein. We acknowledge our great debt to their efforts and are pleased when they achieve more than average material rewards. Even if their contributions rolled from them without great effort, it seems unlikely that our impetus toward equality would be so unwavering that we would seek to limit these highly productive individuals to shares no larger than those of
others. Thus our understandings of what individuals ought to receive incorporate productivity-based inequality. While the attraction of productivity-based inequality is easiest to discern at the extremes of productivity, the idea is attractive throughout the range of productivity. Thus whenever a person’s freely-chosen actions result in more being produced, it is attractive that the producer have discretion over how the additional value is allocated. Stability Any pattern of social coordination necessarily includes elements of stability. At a minimum, it is essential that almost all words mean today what they meant yesterday. From an economic perspective, the desire for stability in collective decisions can be characterized as a desire to avoid outcomes that reduce the wellbeing of any member of the collectivity. If complete stability were attained, then every collective decision would be Pareto improvement, a change that made at least one person better off and no one worse off. One important attraction of stability comes from its contribution to procedural efficiency. The issue of who is to control what can be debated and fought over endlessly. Much of this conflict takes the form of some persons striving to seize control over things from others, while those who would be dispossessed by these proposals resist. (There are also instances of individuals seeking redistribution from which they would receive no direct personal benefits, but these are less prevalent.) The resources devoted to striving to secure what others have and to striving to retain things currently possessed usually represent waste from the perspective of the collectivity as a whole. This is particularly true when the efforts of the two groups exactly offset each other, so that no net change in allocation occurs. Thus it is attractive to have collective decision procedures that foreclose such wastes of resources. A related attraction of stability arises from the understanding that it should be possible, in principle, to introduce every improvement in efficiency as a Pareto improvement. That is, for every option that makes some persons better off and others worse off but is on net advantageous, there are in principle modified options differing only in financial compensation from those who would benefit to those who would lose, making everyone better off. The identification of such modifications of proposals that appear to be on net advantageous is highly problematic in practice, but the conviction that they must exist in principle adds to the attraction of distributional stability. Another attraction of stability arises from the “ratchet effect” with respect to wealth changes. People generally regard the increment in their well-being that comes from a given increase in their wealth, say from W1 to W2, as smaller in magnitude than the decrement in their well-being that results from a decrease in
their wealth from W2 to W1. Thus a rearrangement of the identities of the individuals who receive different incomes, leaving the distribution of incomes unchanged, generally produces an overall reduction in well-being. While stability is thus attractive, in many situations there are conflicting conceptions of what stability entails. This is reflected in the second level of the taxonomy, where stability is divided into “restoration” and “preservation.” Restoration is the aspect of stability entailed in a commitment to undo any illegitimate departure from some initial status quo. Preservation is the aspect of stability entailed in a commitment to resist any departures from the current status quo. Preservation can conflict with restoration because preservation incorporates a commitment to resist even those departures from the current status quo that would restore a previous status quo. The appeal of restoration is clearest in the shared understanding that things that have been stolen must be returned to those from whom they were stolen. The conflicting appeal of preservation is revealed by the diminution, as time passes, in the impulse to undo the consequences of theft and other illegitimate changes. Thus there is a statute of limitations for prosecuting thieves. The passage of time also tends to diminish the appeal of restoration by making it difficult to identify what restoration entails. If my father stole a painting from your father and passed it on to me, I could be required to restore the painting to you, but if my father sold the painting and used the money to buy me an education that I might not have chosen for myself, what does restoration then entail? That the passage of time does not invariably extinguish the salience of restoration may be seen in riots that occurred in India in 1990. Hindus claimed that 300 years earlier, when Moslems controlled India, a mosque was built on a site where a Hindu temple had previously stood. They wanted the mosque torn down so that the temple could be restored. While it may seem extreme to attempt to rectify a 300-year-old injustice, it is also troubling to allow the mere passage of time to legitimate a new status quo. In the rhetoric surrounding Iraq’s take-over of Kuwait in 1990, one saw efforts by Iraqi leaders to give the new situation the imprimatur of the status quo, while other world leaders strove to perpetuate a perception of the impropriety of Iraq’s actions and the need to restore the previous status quo. World opinion reflects an awareness that if the imprimatur of the status quo is easily attained after a coercive change, then aggression will be all the more attractive and prevalent. And so the tension between preservation and restoration persists. Conclusion Each of the general criteria of efficiency, equity and stability has a rationale that seems to allow no compromise with the other criteria. If efficiency is attractive because collective decisions ought to provide as much as possible of whatever
people want, then what place is there for any other criterion? But similarly, if equality and productivity-based inequality are attractive because we have shared understandings about what individuals ought to receive, how can any other criterion be relevant? And finally, if our concern for stability reflects a determination not to allow change to make any individuals worse off, how can we compromise on that position without making it meaningless? It appears that the attempt to identify criteria for evaluating collective decision procedures has the unintended consequence of creating slogans that tend to exclude the attraction of competing criteria. But the fact that the competing criteria are attractive when we focus on them reveals that each slogan overstates the attraction of the general criterion for which it speaks. To understand what we really want from collective decision procedures we must not succumb to the monistic tendencies suggested by the rationales for individual criteria. We must be alert to the attraction of criteria in all categories as we ask how the set of consequences of each collective decision mode compares with those of all other modes. The next chapter provides a discussion of the relative advantages of the various modes in terms of the criteria identified in this chapter. Notes
