The Calculus of Consent: Logical Foundations of Constitutional Democracy

In recent years the role of the pressure or special-interest group in democratic political process has come to be more widely accepted as inevitable, if not “desirable.” In 1951 David B. Truman, building on the earlier work of Arthur Bentley and to some extent on that of E. Pendleton Herring, made a bold attempt to construct a positive theory of American politics on the basis of an acknowledged interplay of group interests.*63 In 1958 an interesting, if abortive, effort to examine the role of the pressure group was undertaken in a round-table discussion at the University of Chicago.*64

Special Interest and the “Public Interest”

Most attempts to examine the role of pressure groups have bogged down in their efforts to define the “public interest.” If this cannot, in fact, be defined, it becomes impossible to determine, even conceptually, the extent to which the activity of special-interest groups either advances or retards progress toward the “general welfare.” Analysis becomes impossible without a well-defined criterion. Our essentially economic approach to the political process is useful in that it allows us to escape from the ambiguities surrounding the concept of the public interest. The literature of modern welfare economics is especially helpful in this respect. The discussion in this field has clarified some of the more troublesome issues that seem to arise. One approach recognizes that definitive meaning can be attached to “social welfare” or the “public interest” only if a social-welfare function is fully described. This function conceptually orders all possible states of society, and quite unambiguously allows for the selection of the “best” or from a restricted set of available alternatives, the relatively “best.” However, in order to describe this function, some individual must make quite explicit his own value judgments. There is no escape from the responsibility of individual ethical decision. In this construction the “public interest” is what the individual says it is. Moreover, each individual will have a meaningful conception of what he conceives to be the public interest; there will be as many social-welfare functions as there are individuals in the group. “Social welfare” or the “public interest” does exist, for the individual, as something apart from and independent of special group interests, but the usefulness of this approach disappears when we come to those issues on which individual evaluations of alternatives differ.

We have rejected this approach. Instead of initially developing a social-welfare function which unambiguously orders all social situations, we start from the presumption that we are quite ignorant as to what is “better” or “worse” for the group. Falling back on the Pareto criterion for assessing changes, we admit as “better” only those changes that are observed to be approved unanimously by all members of the group. Any change that secures unanimous support is clearly “desirable,” and we can say that such a change is “in the public interest.” Few would, we suspect, dispute this half of our criterion for evaluating social changes. However, we go further and state that, for any change in the public interest, unanimous support can be achieved. This half is perhaps less acceptable until the economic meaning of “improvement” is fully understood. If the political process is conceived as one means through which individuals co-operate to attain mutual advantage, it is clear that, conceptually, all persons can be made “better off” by any change that does, in fact, produce sufficient “improvement” for mutual advantage to be possible.

Our unanimity criterion does not, however, seem to get us very far toward a definition of the public interest in any practical applications of this term. Actually we observe day-to-day decisions being made on other-than-unanimity rules for choice. Is there no criterion by which we may judge whether or not specific changes are or are not “desirable”?

At this point our construction becomes equivalent to that conception of the public interest raised by Mr. Roy Larsen in the Chicago round-table discussion and cited at the beginning of this chapter. There is clearly no way of determining the degree to which the public interest is advanced by the operation of ordinary rules for legislative decision-making on any single issue. Here we should expect only particular or group interests. We expect that some of these will be advanced and that others will be thwarted. The “public interest” becomes meaningful only in terms of the operation of the rules for decision-making, and these rules can be evaluated only over a long and continuing series of separate issues. Our conceptual and analytical separation of the constitutional and the operational level of collective decision allows us to discuss the unanimous choice of rules and at the same time to recognize the arbitrary results that will be produced by the operation of any given rule on specific issues.

At the ultimate constitutional level of decision, the implied requirement of consensus prevents the partisan struggle among group interests that characterizes operational decisions. If identifiable and permanent coalitions are expected, genuine constitutional process, as we have defined this term, is not possible. We do not, of course, deny that conditions may be present in which separate class or group interests are so solidified that no democratic constitution can be chosen for the community. However, we should emphasize that at the ordinary operational level of decision, within defined constitutional rules, pressure-group conflicts are fully consistent with the democratic process. Indeed it is precisely because the individual anticipates that economic interest will manifest itself in the operation of ordinary collective choice-making rules that he is willing to choose processes that involve considerable investment of resources in strategic endeavor. Our analysis is not, therefore, inconsistent with a structure of political institutions closely approximating those found in Western democracies. If, in fact, the individual could be “trusted” not to follow economic interest, and if all pressure groups could be assumed away, there might be, on some grounds, considerably less strength in the argument for many of the checks and balances that characterize modern democratic process.

Pressure Groups and Big Government

The activities and the importance of special-interest groups in the political process are not independent of either the over-all size or the composition of the governmental budget. A hypothesis explaining the increasing importance of the pressure group over the last half century need not rest on the presumption of a decline in the public morality. A far simpler and much more acceptable hypothesis is that interest-group activity, measured in terms of organizational costs, is a direct function of the “profits” expected from the political process by functional groups. In an era when the whole of governmental activity was sharply limited and when the activities that were collectivized exerted a general impact over substantially all individuals and groups, the relative absence of organized special interests is readily explainable. However, as the importance of the public sector has increased relative to the private sector, and as this expansion has taken the form of an increasingly differential or discriminatory impact on the separate and identifiable groups of the population, the increased investment in organization aimed at securing differential gains by political means is a predictable result.*65

This relationship is not, however, one-sided. While the profitability of investment in organization is a direct function of the size of the total public sector and an inverse function of the “generality” of the government budget, both the size and the composition of the budget depend, in turn, on the amount of investment in political organization. The organized pressure group thus arises because differential advantages are expected to be secured through the political process, and, in turn, differential advantages for particular groups are produced because of the existence of organized activity. A spiral effect comes into play here, the results of which may be observed in the federal income-tax structure, federal tariff legislation, federal resource-development projects, and many other important areas of economic legislation in particular. This spiral effect has an important bearing on the individual constitutional calculus, and it is therefore worth discussing in some detail.

Conjecturally, and certainly not without considerable historical validity, we may imagine a government that undertakes only those activities which provide general benefits to all individuals and groups and which are financed from general tax revenues. Under these conditions there would be relatively little incentive for particular groups of individuals to organize themselves into associations designed specifically to secure special advantages through governmental action. Suppose now that this institutional “equilibrium” is disturbed through the efforts of one particular interest group, which organizes in an attempt to secure the adoption of favorable legislation. Assume that, through some means of side payments, this group is successful in its activity. It secures the passage of legislation which provides the group represented with special benefits that are not applied generally to the whole population. The measure adopted protects a specific industry, exempts a particular form of association from the antimonopoly laws, grants differential tax privileges, or any of the many other commonly accepted current practices.

The results will be that total collective action is increased and, secondly, that the door is opened for differential class-, group-, or sectional-interest legislation. Other functional or interest groups, observing the success of the first, will now find it profitable to invest resources (funds) in political organization. The pressure group, as such, will rapidly become a part of the political decision-making process. Moreover, because of the activities of such groups, the range and the extent of collective action will tend to be increased. As more and more groups come to recognize the advantages to be secured by special political dispensation, this organizational process will continue. The ultimate “equilibrium” will be reached only when all groups have become fully organized.*66

Pluralistic Equilibrium

Many modern students of pressure-group phenomena seem to rely on this “equilibrium” and expect it to produce, if not “optimal,” at least “satisfactory” results. It is often noted that the individual will simultaneously be a member of several organized interest groups: his trade union, his church, his local political unit, etc. Moreover, because of this multiple membership he will restrain the self-seeking activities of any particular group to which he belongs. Some such restraint cannot be overlooked, but it must also be acknowledged that few, if any, single individuals will be members of all groups simultaneously, and, even disregarding this, membership in separate groups will generate different degrees of individual interest. The fact that the member of the trade association or the trade union is also a consumer will not effectively restrain his activities in seeking differential advantage for his particular producer group because of the predominant importance of his producer role with respect to any single decision that the government might confront relating to the specific industry.

The difficulty is not removed if we postulate that each functional group has “equal power.” In this case mutual “exploitation” will proceed to take place under ordinary democratic processes. Discriminatory legislation will continue to be adopted. The only difference between this situation and that in which “power” is distributed unequally among organized groups, and between organized groups and the unorganized members of the community, is that the costs will tend, over time, to be distributed over the whole population somewhat more “equitably.” In this “equal-group-power” model, all groups will, over a whole sequence of issues, bear roughly the same share of the total costs of pluralistic organization. This conclusion is readily demonstrated by referring to the simple logrolling model developed in Chapter 10. In that model we may substitute a single group interest for each individual farmer, thus guaranteeing “equal power” to each group, and then examine the results. Given any collective decision-making rule other than that of unanimity, external costs will tend to be imposed by collective action. Differential or group legislation is precisely equivalent to the special road-repair projects financed out of general-tax revenues which were introduced in the model of Chapter 10.

External Costs and “Optimal” Organization

As we have repeatedly emphasized, the existence of external costs imposed by the operation of the rules for making collective decisions is neither a necessary nor a sufficient condition for “nonoptimality” in an organizational sense. The advantage of our construction lies in the fact that we are not required to explain away the effects of the special-interest groups in describing the “optimal” organization of collective decisions. Pressure- or interest-group activity is one institutional manifestation of external costs, and external costs are expected to be present even in the “ideal” organization. The question remains, however, as to whether or not the existing organization reduces the over-all interdependence costs (external costs plus decision-making costs) to the lowest possible level. Saying that external costs will be present in the “ideal” organization is not equivalent to saying that any organization embodying pressure-group activity is, in any sense, “ideal.”

No direct measurement of the total interdependence costs under existing or alternative decision-making rules is readily available. Certain conclusions can be drawn, however, on the basis of the facts of history. We may observe a notable expansion in the range and extent of collective activity over the last half century—especially in that category of activity appropriately classified as differential or discriminatory legislation. During the same period we have witnessed also a great increase in investment in organized interest-group efforts designed specifically to secure political advantage. These facts allow us to reach the conclusion that the constitutional rules that were “optimal” in 1900 are probably not “optimal” in 1960. If we may assume that the fundamental rules for organizing collective decisions were more closely in accordance with the “ideal” in 1900 than in 1960, these same rules will tend to produce a higher level of interdependence costs than necessary. This suggests that some shifting in the direction of more inclusive decision-making rules for collective choice and some more restrictive limits on the range of collective activity might now be “rational” to the individual considering constitutional changes. The contrary possibility, of course, also exists. If the operation of existing constitutional rules produces roughly “optimal” results today, clearly these same rules were overly restrictive in earlier stages of development marked by relatively less organized pressure for differential legislation.

We express an explicit value judgment here, but we consider the first alternative interpretation to be more applicable to American society. Moreover, because of this judgment we consider the external costs imposed by the operation of existing rules to be excessive. Nevertheless, we can also be somewhat optimistic, over the long run, regarding the prospects for securing some genuine improvements in political organization. If, in fact, the organization of special interests has advanced to the point at which no one interest can expect, in the long run, to secure differential advantage, the way may be open for some changes in the organizational rules themselves. Each interest group will, of course, turn every effort toward improving its own position, within the limits of the prevailing rules; but if, in fact, all interests come to recognize that the external costs involved in this continuous struggle of interests are excessive, all might agree on some changes in the rules that allow such behavior to take place. It seems doubtful whether American democracy has as yet reached this point of mutual recognition of the advantages to be secured from the requisite constitutional changes. However, as more and more groups organize to secure political support, and as more and more discriminatory action does come to characterize separate political decisions, reaction will surely set in at some point. We begin to see, perhaps, the beginnings of such reaction today with respect to income-tax legislation. More and more criticism is being raised against the maze of special exemptions and deductions that has come to characterize income-tax laws. Although the brief experience of late 1959 showed that, when actual changes of a more general sort are proposed, the special beneficiary groups are still sufficiently strong to retain the currently existing structure, the criticism is still likely to mount. While the excessive external costs involved in discriminatory tax legislation are perhaps more likely to be recognized than those involved in other legislation, the current discussion of tax policy does seem to bear out the prediction that could be made on the basis of our construction.

Ultimately the hope for some “improvement” must lie in the mutual consent of the special interests themselves for constitutional changes which will act so as to reduce the excessive costs that discriminatory legislation imposes on all groups over time. It is in seeking such changes in the organizational rules themselves that genuinely enlightened self-interests of these groups may be expressed. It seems sheer folly to expect that the interest groups will, unilaterally and independently, exercise sufficient self-restraint, given existing rules. To expect them to do so amounts to expecting them to act contrary to their raison d’être.

General and Special Legislation

If all collective action should be of such a nature that the benefits and costs could be spread equally over the whole population of the community, no problem of the interest group, and indeed few of the problems of government, would arise. If each individual, in his capacity as choice-maker for the whole group, could, in his calculus, balance off a pro-rata share of the total benefits against a pro-rata share of the total costs, we could expect almost any collective decision-making rule to produce reasonably acceptable results. Under these relatively “ideal” circumstances, individuals and groups would have relatively little incentive (because there would not exist much genuine possibility) to utilize the political process to secure advantage over their fellows. However, few collective decisions, if any, can be reduced to such general dimensions. Almost any conceivable collective action will provide more benefits to some citizens than to others, and almost any conceivable distribution of a given cost sum will bear more heavily on some individuals and groups than on others. As the analyses of Chapters 10 through 15 have shown, it is the opportunity to secure differential benefits from collective activity that attracts the political “profit-seeking” group. Moreover, these differential benefits may be secured in either of two ways. First, activities may be approved which cause benefits to accrue to selected individuals and groups but which impose costs generally on all members of the community. This was illustrated by our initial road-repair examples. Secondly, activities may be approved which provide general benefits to all members of the community but which impose costs on certain selected individuals and groups. The necessary condition for the presence of external costs, as we have used this term, is some difference in the distribution of the benefits and costs of collective action among members of the community.

One means of modifying the organizational rules so as to produce results akin to those that would be produced under truly “general” legislation would be to require that those individuals and groups securing differential benefits also bear the differential costs. This legislative generalization of the benefit principle of taxation would, in effect, produce results similar to those that would take place under “general” legislation. Note that this change in the rules need not be equivalent to requiring a larger majority or unanimity, although the results need not be significantly different from those produced by such changes. While the requirement of unanimity would tend to insure that all collective action is based on a “benefit principle” of sorts, the requirement that the benefit principle be followed need not insure that all proposals receive unanimous support. The reason for this difference is that presumably in the second case “benefits” would be measured or estimated in some manner that would be independent of the individual’s own evaluations. Therefore, a practical equivalent to the unanimity rule might be, say, majority voting under reasonably strict constitutional requirements about the matching of special benefits and special costs, as measured in some reasonably objective manner. This inversion of the Wicksell scheme, in which he proposed the rule of relative unanimity in order to insure the matching of benefits and costs, would, in any case, reduce the external costs imposed by the operation of any given rule for collective decisions other than the unanimity rule. Moreover, for all issues of collective choice other than those in which redistributive objectives are of primary importance, some improvement could, conceptually, be achieved along these lines.

A practical example may be helpful here. Suppose that a constitutional requirement is adopted to the effect that all irrigation projects, all river-valley-development and flood-control projects, all harbor and inland waterway developments, and the like must be financed, at least in part, by the levy of a special income tax on residents of those areas directly benefited by the projects in question. The number of such projects approved, even under unchanged voting procedures, could either be reduced or increased. It would be clear that those projects failing to win support would be “inefficient” and should therefore be eliminated, provided only that the differential benefits and differential costs are measured with some degree of accuracy. If all areas of the country should become sufficiently “organized” in support of such localized federal resource-development projects, and if all units were in some proximate equality as to power, it would be in the genuine interest of all groups to implement constitutional changes of the sort illustrated. The fact that the interest or pressure group as such tends to develop an interest in continuing to exist will, of course, be a real barrier to such reform.

Analogous but different constitutional changes could be instituted which would reduce the excessive external costs imposed by the operation of special-interest groups in those cases where over-all redistribution objectives cannot be put aside. Many collective projects are undertaken in whole or in part primarily because they do provide benefits to one group of the people at the expense of other groups. These objectives may be quite legitimate ones, and they may be accepted as such by all, or nearly all, members of the community. However, the difference in the distribution of benefits and costs may result in excessive external costs quite independently of the accomplishment of the distributional objectives. For example, suppose that the issue confronted should be that of providing some federal funds to aid the depressed coal-mining area of West Virginia. For such a measure the levy of special taxes on citizens of West Virginia would be largely self-defeating. Nevertheless, it is relatively easy to see that, if such aid is to be financed out of general-tax revenues, a veritable Pandora’s box may be opened. Depressed fishing villages along the Gulf coast, depressed textile towns in New England, depressed automobile production centers in Michigan, depressed zinc-mining areas in Colorado, etc., may all demand and receive federal assistance. As a result, excessive costs will be imposed on the whole population.

One means of eliminating this sort of distortion, which may appear somewhat farfetched because it is novel, would be to require that all such projects be financed out of taxes levied on specific groups in the total population, although not on the same group securing the benefits. For example, if the funds designed for aid to West Virginia were to be collected from special taxes levied on citizens of Oklahoma only, then we could be assured that roughly balancing political forces would determine the final outcome. Excessive external costs world be substantially reduced in this manner, and something roughly similar to the pattern of “general” legislation would emerge. Genuinely depressed areas, considered as such by the whole population, would tend to be provided with assistance without at the same time opening up the whole set of grants to areas not considered to be deserving of assistance. Congressmen from, say, North Dakota or Minnesota, in our example, would be confronted with two opposing partisan interests. Those representing West Virginia would try to secure favorable votes; those from Oklahoma would try to influence the Congressmen in the offsetting manner. Through the logrolling process some solution would be reached, and this solution would more nearly reflect “the public interest” than the alternative one which requires general-tax financing. There could, of course, be no assurance that “optimal” individual decisions would be reached, but it seems relatively certain that a somewhat closer approach to a set of “optimal” collective decisions over time could be produced in this way than under existing rules.

These suggestions are highly tentative and preliminary, as indeed are many which have been advanced elsewhere in this book. The consideration of mutually beneficial constitutional changes aimed at reducing the external costs imposed by the operation of special-interest groups in modern democratic process would seem to represent an extremely important and worthwhile activity for scholars in political science.

The Ethics of Pressure-Group Activity

An analysis requires, first of all, a somewhat more widespread acceptance of special-interest or pressure-group activity as an inherent and predictable part of modern democratic process. In our analysis this activity is a predictable outcome of our fundamental behavioral assumptions. At least in this one respect, the facts of the real world lend support to the confirmation of our assumptions. Scientific progress in the analysis of politics cannot be made until this widespread activity is fully incorporated in the analytical models. Such an incorporation need not commit the analyst to either an acceptance or a rejection of the activity as morally “good” conduct on the part of the practitioners. The economist does not need to say that the individual “should” or “ought to” maximize his own utility; he starts from the assumption that the individual does do so, and that is all there is to it. The student of the political-choice process should do as much; if he does so, the pressure or interest group becomes an essential building block in any political “science.”