The Calculus of Consent: Logical Foundations of Constitutional Democracy

Compared with the more standard works in political science, our analysis may seem to involve a “pessimistic” view of human nature. For scientific progress, however, it is essential that all conceivable assumptions about human behavior be tested. If our models provide some explanations of real-world events, and we believe that they do, our assumptions must have some empirical validity, quite apart from the “attractiveness” of the human characters that inhabit our hypothetical model world.*59

In one sense our approach taken as a whole is more “optimistic” than that taken by standard writers in political theory. Our assumptions about human nature may be judged “pessimistic,” but our conception of the political process, as such, is surely more congenial to those seeking “sweetness and light,” “peace,” and all such good things than the conception usually implicit in political discourse. We view collective decision-making (collective action) as a form of human activity through which mutual gains are made possible. Thus, in our conception, collective activity, like market activity, is a genuinely co-operative endeavor in which all parties, conceptually, stand to gain. By contrast, much of orthodox political thought seems to be based on the view that the collective-choice process reflects a partisan struggle in which the beneficiaries secure gains solely at the expense of the losers. If the political “game” should be, in fact, similar to that conception which seems to be implicit in much discussion, especially that concerned with the doctrine of majority rule, the maintenance of political order must depend, in fact, on the strength of moral restraints placed on human actors. If, by contrast, a broader and, we think, a more “correct” conception of political choice is adopted, there need be less reliance on moral restraints of individuals.

The Ethics of Trade

In our analysis we have assumed that individuals are motivated by utility-maximizing considerations and that, when an opportunity for mutual gains exists, “trade” will take place. This assumption is one of the foundations on which economic theory is constructed. Let us examine some of the ethical issues that may arise in the operation of ordinary market exchange before considering the much more complex problems inherent in this approach to the political-choice process.

Initially it seems useful to distinguish two separate stages in the organization of economic exchange, although in practice these two stages are simultaneous and the two decisions made by the individual participant are interdependent. First, the individual must decide to enter into an exchange relationship, and, secondly, he must agree to the specific terms at which exchange shall take place. The point to be made here is that, in normal discussion, ethical issues are considered to arise in the second decision, but not in the first. In Chapter 8 we discussed the simple two-person bargaining model in some detail. Where a bargaining range exists, the terms of trade will determine the division of the total benefits among the participating parties. Moreover, since this division is essentially a distributional question, the whole problem of “fair shares” arises, a problem that can only be discussed in terms of ethical norms.

As we suggested in our earlier discussion, this admittedly ethical problem is reduced to a minimum in the operation of competitive markets because the proper functioning of a market organization will insure that the single buyer or the single seller has little control over the terms of trade. If the terms of trade (the conditions of exchange) are set independently of the individual participant’s own behavior, no ethical question can arise concerning his “fairness” in dealing with other parties to exchange, ruling out fraudulent behavior. Thus we find that ethical issues about market behavior present themselves only when individuals or groups are in noncompetitive positions, when they possess some power to influence the terms of trade in their favor.

In ordinary exchange no ethical question is presumed to arise concerning the decision of the individual to engage in trade, regardless of whether or not he possesses independent power to influence the terms of trade. Moreover, a moment’s reflection suggests that there could hardly be an ethical issue posed regarding this sort of behavior. Not only is the individual presumed to secure some benefit by entering into trade, but he must also be providing benefit to the other parties in the contract. On almost any set of ethical norms, trade would seem to be an activity that would be accepted as fully consistent with the moral standards of the community.

It is difficult for the modern student of social progress to keep in mind the fact that this apparently obvious interpretation of trading activity has only been dominant since the eighteenth-century Enlightenment. Before that time, “trade,” as an activity, was suspect; and, implicitly, individuals engaged in trade were somehow supposed to be following less moral pursuits than other members of society. This suspicion of trade, as an activity, still dominates the non-Western world and has not yet entirely disappeared from Western thought. Some of the elements of such suspicion can perhaps explain the neglect of the study of the political process in terms of an “exchange” relationship.

Under what conditions is “trade” or “exchange,” even in the modern world, considered to be immoral in and of itself and quite apart from the terms of trade? To begin with, we must recognize that each person has certain moral standards, and these normally will include certain criteria for human behavior. An individual may consider it perfectly moral behavior to sell his own labor services to a business firm, but he may think that it is grossly immoral for a woman to sell the services of her body to a man. An economist may consider it morally acceptable to sell his educational services to a university, but morally unacceptable to sell his professional services to a political party. Each person will have a set of such moral values, and these may include attitudes toward certain commodities or services that other people “ought” not to sell for “money.” There is nothing inconsistent between the existence of such moral standards and an individualistic ethic until and unless the individual desires to constrain others to conform to his own moral standard of behavior. It is quite consistent for the individual to hold a set of values which dictates that a woman ought not to sell her body on the market, and at the same time to include within this set of values the attitude that he should not attempt to constrain the prostitute and her client from exercising their own free choices. In evaluating behavior in others which he thinks morally wrong, the individual, in effect, says: “I think that they are doing themselves harm by such actions; but, since I value freedom of individual choice and since there is no harm imposed on me by their actions, I do not wish to interfere by placing constraints on their behavior.”

Note that this attitude is to be distinguished from a second one in which the individual thinks that other individuals, through behaving in a way that he considers immoral or unethical, are actually reducing his own utility. If an individual interprets the prostitute plying her trade in this way, her activity is, in a real sense, imposing external costs on him. However, even when external costs are imposed on third parties, we must distinguish two separate reactions. The individual may recognize that the activity is imposing external costs on him, but he may also recognize that constraints on this activity may open up the way for other collectively imposed constraints, some of which may affect him directly and adversely. Looking at the problem in this way, the individual may rationally choose to accept the external costs (the reductions in his own utility) which the free play of individual choice in the activity under consideration introduces.

We must distinguish this attitude from that of another individual who holds somewhat stronger views on the immorality of the activity in question, say, prostitution. This individual may rationally seek collective action aimed at preventing the activity from taking place. In reaching a decision of this kind, the individual evaluates the external costs that the activity imposes on him and estimates that these are sufficiently great to offset all possible adverse collective decisions that might be taken against some of his own accepted practices were the State allowed to legislate on moral issues. In this case the individual must consider the external costs to be high enough that he is willing to pay some positive sum in order to secure elimination of the activity. In this sense he must be willing to “trade” something in return for the elimination of the activity under consideration. In many cases, of course, moral standards would become significantly weaker than they initially appear if those who hold them were asked to contribute positive sums toward the elimination of that behavior in others which they condemn as immoral.

Normally, of course, there is sufficient standardization of moral values over the population of a community to prevent serious issues of the sort posed from arising. Most trade falls within the accepted moral schemata of the great majority of the population. Nevertheless, it should be acknowledged that there is never a sharp dividing line between the many trading or exchange activities that are generally accepted and the relatively few genuine trading activities that may be suspect.

Exchange of Political Votes

This discussion of the morality of exchange is helpful because it points directly toward one interpretation of prevailing attitudes on the exchange of votes in Western societies. Individual votes on political issues seem to be among the scarce commodities or services that many members of the community consider inappropriate for open buying and selling. The free marketing of votes, either by an individual or by a member of a legislative assembly, is considered to be an activity in which individuals “ought not” to engage. This attitude toward the marketability of political votes, interpreted in the sense of outright vote-buying and vote-selling, seems to be an empirical fact. The attitude toward vote-trading through indirect methods is considerably different.*60 Our task, therefore, is to examine the logical basis for this combination of attitudes, if indeed one exists.

Why should the rational individual consider the sale and purchase of votes among his fellow citizens to impose external costs on him, that is, to reduce his own utility? Suppose that A observes B selling his political vote on an issue to C. Why should A’s utility be affected by this transaction? B and C mutually gain from the exchange, or else it would not take place. One approach would suggest that A’s utility is reduced (that he bears external costs) because the transaction gives to C political power that C would not otherwise possess. If open buying and selling were to be permitted, A could have an equal opportunity with C to purchase the vote of B. However, what is meant by “equal opportunity” in this case? If the distribution of economic power among the citizens is unequal, open buying and selling of political votes might be said to give “unfair” advantages to the richer members of the group. To be sure, both the poor (exemplified by B), who would find their over-all economic position improved by selling their political decision-making power, and the rich (exemplified by C) would gain from the vote exchange. However, if majority voting prevails, A can be more readily exploited by the votes of B and C in a coalition “owned” and organized by C.

Figure 25

This argument, which is probably characteristic of much orthodox thinking, would seem to contradict some of the conclusions reached earlier to the effect that full side payments (that is, open vote-buying and vote-selling) would tend to reduce, not to increase, expected external costs from the operation of decision-making rules. We are obliged, therefore, to examine the argument quite carefully. Again we may use a simple illustrative example. Figure 25 shows the location of three families, A, B, and C, in a community. The sizes of the squares indicate the economic position of the three families; for simplicity, assume these to be houses. Suppose now that the community is granted sufficient outside funds to construct one road to be run horizontally from the western to the eastern boundary of the territory. If open vote-buying is allowed, C may purchase B’s vote and, by majority rule, choose the road shown as II in Figure 25. On the other hand, if all vote-buying and vote-selling should be prohibited, A and B might form the majority and construct the road shown at JJ. This road, being closer to both A and B, would seem to be a more “desirable” choice on the grounds that “political equality” is more nearly satisfied by this decision than by the alternative one.

This line of reasoning is quite convincing, up to a point, and it does tend to contradict some of our earlier conclusions. It does so, however, only because the market is assumed to be imperfect. If, instead, the vote market is assumed to be perfect in all respects, A and B might well form the majority coalition, as in the no-trade case, but they would still construct the road at II. They could, by acting as a coalition, force C to purchase both of their votes (or to pay as much for one as if two were purchased) and to pay an amount sufficient to reduce his own net gain to zero (a negative sum if the road is to be tax financed). A, acting as a “political entrepreneur,” could offer B just as much for his vote as does C under these circumstances, because he would be aware that he will have the opportunity to sell both votes (as one) to C. One additional transaction or “bargain” would be required in this solution, but with perfect markets this will be no barrier.

It seems evident, however, that some imperfection in the vote market might arise, and, in this case, bargains or trades between C and B would seem much more likely to emerge. Expecting this, the rational individual may consider the open buying and selling of votes to impose an external cost on him.

If we consider the question of vote-marketing at the time of constitutional choice, differences in economic position are not predictable. Therefore, to generalize our discussion we need to allow, not for predictable differences in economic position, but for differences in interest on particular issues, which may or may not be based on differences in economic status. The individual, considering organizational rules, may well think that vote-marketing, if it could operate perfectly, would reduce expected external costs. However, he may also predict imperfections in this market which may more than offset this advantage. With expected market imperfections of a certain type, the individual may choose rationally to try to prohibit the open buying and selling of political votes.*61

If the market imperfections are expected to take the form of the exclusive exchange of votes between the most interested and the least interested groups, with the absence of “political entrepreneurs” or “vote brokers” in the mildly interested groups, the individual may expect interest coalitions to solidify and to become permanent. The basis on which his constitutional decisions rest may be changed if he does, in fact, expect permanent coalitions to form.

Closely analogous to this is the operation of competitive markets. If, in fact, markets could be expected to work perfectly, there would never be any need for the State to intervene with antimonopoly legislation. The firm securing a monopoly position temporarily would tend to be restrained in its efforts by the emergence of other firms producing closely related goods and services. Any restriction on the freedom of firms to merge, to enter into pricing agreements, etc., would, under these conditions, amount to a denial of “gains from trade.” However, when it is recognized that certain types of agreement may lead to the establishment of market-power positions that are not readily displaced due to the imperfect operation of the mechanism of adjustment, it becomes reasonable to seek prohibitions on such agreements.

There are two separate reasons why such agreements should be prohibited under these circumstances. First, once attained, the firms may be able to exploit their bargaining advantage; they may be able to secure an “unfair” share of the total gains from trade by manipulating the terms of trade in their favor. This is not, however, the relevant part of the antimonopoly analogy for our purposes. Here the aim of intervention is not that of prohibiting trade, but rather that of insuring more acceptable terms of trade. The second reason for trying to prevent the attainment of positions of dominant market power lies in the expected ability of firms, once having attained this power, to prevent the emergence of other rival groups (competitors). This reason, which is the central theme in the legal if not in the economic history of the antimonopoly laws, seems closely analogous to the argument that we have developed above regarding the open buying and selling of votes in the market. The individual may not have sufficient confidence in the perfection of the vote-market’s operation; he may fear that open buying and selling will quickly lead to the emergence of specific interest-group coalitions, which will tend to become permanent and which will possess the power to prevent the emergence of alternative patterns of coalition formation.

The whole institution of vote-buying and -selling is exceedingly difficult to analyze because of the unique nature of the items traded. A vote in the collective-choice process, operating under less-than-unanimity rules, represents potential power to impose external costs on other individuals. There are few fully acceptable analogies in the operation of ordinary markets. The potential power exists, of course, whether or not the individual holder places it on the market. Thus it is relatively easy to see why moral and ethical questions of major import tend to arise when any consideration of vote-trading is introduced.

The Imperfect Ideal

We recognize, however, that some forms of vote-trading are accepted as being consistent with the prevailing moral standards in Western democracies. The individual calculus in this respect—if prevailing attitudes can be taken to reflect rationally reached conclusions—suggests that, in reference to the whole issue of vote-trading, the “ideal” is neither “none” nor “all” but somewhere in between. As the analysis above indicates, if market imperfection is expected to be present and if the results of this imperfection can be predicted in advance, the placing of prohibitions on open buying and selling of political votes may be quite rational. If a full and open vote-trading market, where transactions take place in money, could be predicted to result in the most interested individuals and groups purchasing votes from the least interested on all or a substantial number of issues, then the rational utility-maximizing individual might expect such an institution to result in unbearable external costs or even in the overthrow of the constitutional system, the “social contract.” On the other hand, the individual might also recognize the advantages to be secured from vote-trading under certain circumstances. Indeed, if all vote-trading were prohibited, he would probably be unwilling or at least quite reluctant to agree to any less-than-unanimous decision-making rules for collective choice.*62 He may consider, therefore, that the “optimal” amount of vote-trading is provided by that system which prohibits open markets in political votes as such but which sanctions indirect methods of accomplishing roughly the same purposes. The opportunity to trade votes on separate issues through logrolling, explicit and implicit, provides an essential protection to interested minorities against discriminatory legislation. The value of this protection may be widely acknowledged, and at the same time the “open” sale of votes may be condemned as immoral. We have shown that this attitudinal pattern need not be internally inconsistent, even within the limited framework of the individualistic ethic.

The conflict between democratic ethics and economic efficiency need not, therefore, exist in so distinct a form as it might have appeared at earlier stages of our construction. Economists recognize that unrestricted trade can be guaranteed to lead to greater “efficiency” in resource usage only if markets are expected to operate perfectly. If imperfections are predicted and the characteristics of these imperfections can be identified, specific restrictions on trade may, under certain conditions, actually increase “efficiency.” These restrictions will rarely, however, extend to the prohibition of all trade.

This is not to imply that the existing set of legal prohibitions and restrictions (along with the existing moral attitudes toward the exchange of votes) is necessarily that set which the rational utility-maximizing individual “should” support. Our purpose has been that of indicating that this set is not necessarily inconsistent in itself, and that the possible conflict between ethical standards and economic efficiency is not demonstrated.

We are aware, of course, that other arguments can be developed to justify the moral attitudes on vote-trading that seem to exist. We neither wish to deny the value of these arguments nor to compare them with those we have presented. For our purposes, which are those of developing the implications of rational individual behavior in political choice-making, the other arguments are irrelevant. Much of the orthodox discussion has been based, as we have suggested, on different assumptions about the behavior of the human actor in the political process. If individuals are assumed not to try to further their own interests but instead to seek some “public interest” or “common good” when they participate in collective choice, the sale of a vote becomes clearly immoral since the receipt of a money payment provides definite proof that the individual is receiving “private” gains from his power to participate in political action. Much of the standard attitude toward vote-trading probably stems from this approach to the governmental process. We note only that the immorality of vote-trading in this context is wholly different from that which we have considered in some detail above, and much behavior which seems to be accepted as standard practice in modern democratic institutions must also be held to be immoral on this alternative approach. It would be interesting to examine the full implications of the behavioral assumption which holds that the individual always seeks the “public interest,” but, as Frank Knight has often observed, no one has yet provided us with an analysis of the organization of a society of angels.

Vote-Trading and the Rule of Unanimity

The analysis of vote-trading above applies only when collective decisions are made under less-than-unanimity voting rules. If the unanimity rule is required for collective action, the political vote of an individual no longer represents the potential power to impose external costs on other individuals. Here the vote represents only the “right” or the “permit” to participate in the division of the mutual gains that collective organization and action can secure. This major change in the very meaning and significance of the political vote of the individual modifies the analysis of vote-trading.

If, for all collective decisions, all members of the group are required to agree, there would seem to be no rational basis for imposing any prohibition on the purchase and the sale of political votes of individuals. The dangers discussed above, those of permanent power blocs being formed, no longer can exist since the effective coalition on all issues must always consist of all members of the group. Prohibitions on vote-trading under the unanimity rule serve only to create inefficiencies in the use of collective resources. An illustrative example may be helpful. Suppose that all vote-trading were to be strictly prohibited and that the rule of unanimity is operative for all collective action. Any proposal that stands a chance of adoption must include within a single “package” elements that provide net benefits to each individual and group in the community. In order to organize such a “package” proposal, many rather wasteful and inefficient projects may have to be included. If open vote-trading were to be allowed, there would be no need for any genuinely inefficient projects to be undertaken. True “pork-barrel” legislation would never be observed under this institutional scheme. With vote-trading prohibited, this sort of “pork-barrel” legislation would be quite prevalent under the operation of the unanimity rule, although it would be present under other voting rules also. However, under these other voting rules this “pork-barrel” inefficiency must be compared with the greater danger of permanent coalition formation that open vote-buying and vote-selling might encourage. This second danger, or cost, is wholly absent when the rule of unanimity is operative.

The Rent-Control Analogy

We may construct, without difficulty, an economic analogue to the prohibition of vote-trading. This analogue serves perhaps to bring out clearly the nature of the questions raised by our whole approach to the political process.

Assume that, in a ten-man (family) community, a new apartment unit is to be made available by the municipal government. There are six units in this apartment building, and the rentals are strictly controlled, being established at a predetermined level. Suppose further that this rental price is below the demand price of each of the ten families in the group. In other words, every one of the families will desire to move into the new subsidized housing if possible. We should also expect, however, that demand prices for this opportunity will vary over the ten individuals in the group. Some families will be relatively satisfied with their current accommodations and thus would secure relatively little net advantage from the new opportunity. Other families will find that the net advantages from residence in the community housing project would be substantial.

Figure 26

Now assume that no plan for rationing the six available units is adopted. The municipal authority simply announces that the first six families to sign up on a particular morning will be assigned the units. It is clear that a queue will form on the designated day, and also that no predictable configuration can be placed on this queue. Let us suppose that a queue forms which is like that shown in Figure 26. Individuals 5 through 10 will be successful in securing the desired housing, and Individuals 1 through 4 will be left to live in their current residences. This result seems to be almost precisely equivalent to the operation of simple majority voting when no vote-trading of any kind is allowed to take place. Individual 8, for example, who receives only a slight net advantage, secures, nevertheless, one of the municipal units. Similarly, in the analogous political process the individual who cares relatively little about the outcome on a particular issue counts for as much as any other individual.

Let us now modify our example. Suppose that housing permits are to be issued. Since only six units are present, permits totaling to six are to be made available. However, assume also that the municipal authority does not choose to discriminate among the various citizens. Following the principle of “housing equality,” the authority therefore issues a permit of 6/10 to each family, but it encourages the buying and selling of these permits among families. No one can secure an apartment unit unless he presents to the authority one full permit; no family can secure an apartment under these circumstances unless it purchases additional permits from others. We assume that fractional permits are marketed in units of 1/10. The result is readily predictable; queuing will no longer determine the outcome. Individuals 2, 3, 5, 6, 7, and 9 will secure the apartments. Individuals 1, 4, 8, and 10 will sell all of their permits. All members of the group are better off as a result, and a Pareto-optimal solution is attained Pareto optimally (if we neglect the costs of organizing the exchanges). This example seems precisely analogous to that political voting rule which requires unanimity but which allows for full side payments (full vote-buying and vote-selling). Note especially that “political equality” is maintained in the sense that each man is given an equal “vote” at the outset.

Let us now introduce a third version of our rent-control illustration which will be analogous to simple majority voting with full side payments (open vote-trading). As before, assume that housing permits are issued equally to all families. However, assume that, instead of 6/10 being issued to each family, each family is now given one full permit. Thus, the authority issues more permits than there are apartments available for disposition. Full purchase and sale of permits is encouraged during a period prior to the announcement of a date for distributing units among permit holders. As in the first case, units will be allocated to the first six individuals presenting permits on one designated day. However, prior to this time, those persons desiring the community housing most strongly will be able to “purchase” additional permits to prevent their losing out in an allocation solely by queue. It is easy to see that this “market” will not work nearly so smoothly as in the previous example. While the market in permits will tend to insure that the six families desiring the municipal housing most strongly will end up with the available units, there is nothing in the operation of this market that will insure that each of the ten families gains in the over-all operation. Only six of the families will be certain to secure net gains (although more could do so), and these six need not be the same ones who finally secure the housing units. They may be fully “squeezed” by other families who are better “vote brokers.” The reason for this difference in result is that, finally, four of the permits originally issued will prove to be of zero value at the time of assignment. Those families caught holding the worthless permits will gain nothing at all.

These rent-control analogies seem helpful in pointing up the issues of democratic ethics. In the second example, full vote-trading would seem to be desirable, and little ethical argument could be advanced against it. This is because the institution of trading here not only insures greater efficiency in the allocation of housing space but it also insures that every family in the group secures some positive benefits. This last result is due to the equivalence with the unanimity rule of choice. If we move to the third example, the analogue with simple majority voting, the case for allowing open vote-trading is considerably less strong. While the open marketing of votes may insure increased measured “efficiency” by guaranteeing that those families securing the apartments will be those desiring them most strongly, this advantage may be more than offset by the secondary inefficiencies stemming from the free operation of the vote market.

Conclusions

The interesting conclusion is reached that, under our behavioral assumptions, vote-trading per se cannot be condemned on the basis of a rational individualistic ethic but that vote-trading under rules for collective choice requiring less than full agreement among all members of the group may be condemned. The fact that the political vote of the individual is wholly different in these two cases makes for an extremely important difference in the attitude of the rational individual toward vote-trading. In the one case, the vote represents the potential power to impose external costs on other individuals in the group, and it is because of the fear that market imperfections may cause this power to become solidified into permanent or quasi-permanent coalitions that the individual may choose to restrict in some way the institution of vote-trading. In the other case, when unanimity is required for action, the vote does not represent the potential power to impose costs on others. No offsetting reason arises to oppose the efficiency reason for allowing full and free marketing of political votes.

This distinction is very similar to that made in discussing ordinary economic transactions. For the most part, these transactions directly affect the parties participating in exchange to the exclusion of third parties. Such transactions are, therefore, fully accepted as falling within the standard behavior patterns of democratic society. Trade is not suspect under these conditions. This is equivalent to saying that the group unanimously approves trade of this sort. Ordinary market exchange is, in a real sense, equivalent to the political rule of unanimity. On the other hand, trade does become subject to question when the services exchanged (produced) come to represent the power to affect third parties adversely, that is, to impose spillover or external costs on individuals outside the contractual relationship. Somewhat interestingly, however, the form of the suspicion is rather different in the two cases of political and economic decision-making. When economic or market activity is observed to result in the imposition of costs on parties outside the exchange relationship, economists have tended to call attention to the “inefficiency” in over-all resource usage that this organizational arrangement generates. They seem rarely to have brought into question the morality or ethics of the individuals participating in such activity. Individuals are assumed to seek to maximize their own utility within the limits of the effective constraints imposed on their action. Not bringing the underlying motivational assumptions into question, the economist tends, therefore, more or less automatically to think in terms of modifying the set of constraints on individual action (the redefining of property rights, the changes in the legal structure, etc.) with a view toward eliminating the inefficiencies, if possible.

By contrast, the student of political processes, observing what is essentially the same phenomenon in another form (that is, the imposition of external costs on third parties), has not considered the inefficiency aspects seriously. Instead he has—through his emphasis on moral restraints on self-interest, his concept of the “public interest,” etc.—sought to accomplish reform through a regeneration of individual motives. Ethical and not structural reforms tend to be emphasized. Breakdowns and failures in the operation of the system are attributed to “bad” men, not to the rules that constrain them.