The Reason of Rules: Constitutional Political Economy

Our purpose in this chapter is to describe the normative position from which we approach the whole set of issues involving the rules of social order. Through what window do we view the world of social interaction, actual or potential? Until we make ourselves clear in this respect, we may seem to be “speaking in tongues” to those whose perspective differs categorically from our own. Our position is explicitly and avowedly
contractarian. This term alone will identify the conceptual framework to those familiar with classical political philosophy, especially with those works that embody the intellectual foundations of liberal society. To counter the most familiar and pervasive criticisms of this position, we must note that the contractarian construction itself is used retrospectively in a metaphorically legitimizing rather than historical sense. Prospectively, the model is used in both a metaphorically evaluative and an empirically corroborative sense.
*7

The relationship between the contractarian philosophical perspective and the rules-oriented, or constitutionalist, perspective is not so direct as it may at first appear. Section II examines this relationship and briefly discusses possible noncontractarian elements in constitutional thought. The contractarian perspective is grounded in individualistic presuppositions about the ultimate sources of value and of valuation. These precepts are examined in Section III, particularly in contrast to other, more familiar nonindividualistic teleologies. Section IV presents the contractarian paradigm in its most well-known setting, that of ordinary economic exchange. Section V extends the perspective to politics, generally considered. The unanimity requirement is examined in some detail in Section VI. And in Section VII the difference in the choice among rules and choice within rules is related to the unanimity requirement.

II. Noncontractarian Constitutionalism

As we emphasized in Chapter 1, the hallmark of the constitutionalist is the categorical distinction he makes between outcomes generated within defined rules and the rules themselves. Once this distinction is made, the effects of rules in constraining potential outcomes become obvious. Simple recognition of this linkage then prompts attention to the rules of social order as an object of inquiry and tends, almost necessarily, to generate an extended time dimension in all considerations of social policy or social change.

The constitutionalist, so defined, however, need not at the same time be contractarian. Understanding of, and respect for, the distinction between the rules that constrain behavior and the results of actions taken within the rules need not be derived from, or be the basis for, a contractarian position. Perhaps the most obvious example is the extreme conservative, the person who places value on existing rules because they do exist and have existed. Such a person may be constitutionalist in the full sense of the term; indeed, he may claim to be the only true constitutionalist since the ideal is a never-changing set of constraining rules.

A somewhat less extreme but closely related position is based on the recognition of the difference between rules as constraints and action within constraints but at the same time embodies the notion that the rules of social order are not artifactual creations subject to change. In this perspective, rules change slowly during the evolution of society. Although change takes place in the basic structure, it does so only through an organic evolutionary process. Hence, “reform” of the basic rules (of the constitution, broadly defined) is internally contradictory. The ordering function of social rules operates only because they are unchangeable in any directed sense.

A third position may retain apparent constitutionalist elements, but for reasons quite different from those that seem appropriate to the contractarian. A two-level structure of law may be envisaged, with the “higher law” embodying protection for a set of “natural rights” that people possess as human beings. In this interpretation, the “constitution” is that structure of institutions or rules that involves the protection of such natural rights, leaving to ordinary politics all other actions. And whereas the existing constitution may be modified, either in the direction of a closer correspondence with the desired protection of natural rights or in the direction of a further divergence from such idealized protection, there can be no change in the definition of the set of rights as such.

It is not our purpose to elaborate arguments that have been made in support of any one of the three positions sketched. Our intention is merely to indicate that each of the positions may have elements that seem constitutionalist in the common dialogues of the day. The contractarian rejects the first two positions because of the basically negativist implications they generate. If the rules of the socioeconomic-political game are not themselves artifacts subject to constructive change, there is little left to do but to endure the forces of history. The arguments advanced by both the status quo reactionary and the nonconstructive evolutionist are perhaps good counters to the romantic ramblings of those who speak naïvely about human perfectibility, but surely we must be given hope that the institutions of social order are subject to reform and change. If for no other reason, there would seem to be a moral obligation on the part of the social philosopher to provide such hope.

The differences between the contractarian and natural-rights theorists arise from quite other sources than do the differences between status quo reactionaries and evolutionists. The contractarian derives all value from individual participants in the community and rejects externally defined sources of value, including “natural rights.” The discussion in Section II is as germane to the argument against the natural-rights position as it is to all other positions that derive value from external and nonindividualistic sources.

III. Individuals as Sources of Value

The critical normative presupposition on which the whole contractarian construction stands or falls is the location of value exclusively in the individual human being. The individual is the unique unit of consciousness from which all evaluation begins. Note that this conception does not in any way reject the influence of community or society on the individual. The value structure of an isolated human being may be totally divergent from that of such a person described by membership in one or many social relationships. The presupposition requires only that societal or communitarian influences enter through modifications in the values that are potentially expressed by the individual and not externally.

If the individual is presupposed to be the only source of value, a question arises concerning identification. Which individuals are to be considered sources of value? There is no apparent means of discriminating among persons in the relevant community, and there would seem to be no logical reason to seek to establish such discrimination if it were possible. Consistency requires that all persons be treated as moral equivalents, as individuals equally capable of expressing evaluations among relevant options.

From these presuppositions, and these alone, it becomes possible to derive a contractarian “explanation” of collective order. Individuals will be led, by their own evaluation of alternative prospects, to establish by unanimous agreement a collectivity, or polity, charged with the performance of specific functions, including, first, the provision of the services of the protective or minimal state and, second, the possible provision of genuinely collective consumption services.
*8

As noted earlier, the empirical record of the establishment of historical states is essentially irrelevant to the contractarian explanatory argument. The fact that most historical states have emerged from conquest of the weak by the strong does not render unimportant or irrelevant the question, Can the existing state, as observed within the rules that describe its operations, be
legitimized in the broadly defined contractarian vision? To deny the relevance of this question, and at the same time to hold to the contractarian presuppositions, would amount to making the charge that almost all observable states are illegitimate. In this case, the contractarian must join the ranks of the revolutionaries. If, however, within broad limits, the state can be legitimized “as if” it emerged contractually, the way is left open for constructive constitutional reform. Existing rules can be changed contractually even if they did not so emerge.

Note that in the conceptual derivation of the origins of the state just sketched, there is no resort to any source of value external to the expressed preferences of individuals who join together in political community. The state does not emerge to protect “natural rights.” Nor does it reflect or represent the working out of some cosmic force, some will of God or gods. More important, the state does not exist as an organic entity independent of the individuals in the polity. The state does not act as such, and it cannot seek its own ends or objectives. “Social welfare” cannot be defined independently, since, as such, it cannot exist.

To this point, the contractarian defense has been relatively easy. A much subtler confusion arises when we raise questions concerning what it is that individuals seek to gain from political order, even when the basic contractarian presuppositions are granted. Even when the existence of nonindividualistic sources of value is denied, or apparently so, we are still left with individuals’ own attitudes toward their activities as they enter into the idealized contractual dialogue. What do individuals seek when they attempt to reach an agreement with each other on rules governing their behavior and on rules limiting the exercise of state power in enforcing such rules? Do they seek “good”? Do they seek “truth”? Is collective organization viewed as a means or instrument of discovery, whether that to be discovered is described as “good,” “true,” or “beautiful”? Is politics analogous to science, or science to politics?

IV. Contract and Exchange

Rather than plunge directly into the murky discourse suggested by such questions, we shall proceed circuitously by introducing a familiar setting. Ultimately, our interest lies with the participation of individuals in contractual agreement on changes in the most fundamental rules of the socioeconomic-political game in which they live. But a useful analogy is provided by ordinary market exchange, a contractual process in which we all engage and for which economists have well-developed analytic explanatory tools. We need only consider the simplest exchange model: a two-person, two-commodity trade or exchange.

First, let us think of two persons, A and B, each endowed with a bundle containing quantities of two goods or commodities, oranges and apples. For simplicity, let us assume that A initially has all the oranges and that B has all the apples. Both commodities are “goods” in the utility or preference functions of both potential traders, and there is not sufficient abundance in the endowments to satisfy fully the appetites of either person for either commodity. In this setting, trade can be beneficial to both parties. Person A will trade oranges for apples; B will occupy the reciprocal role. The two parties will agree to make exchanges so long as their internal trade-offs between the two commodities differ. When trading stops, at equilibrium, the internal rate of exchange between oranges and apples for A will be equal to that for B and will also be equal to the final terms under which the commodities were exchanged.

This is first-day economics, of course, but we seek here to look at this simple trading process from our contractarian-constitutionalist perspective. The implicitly postulated rules for trade involve the rights of each person to the initial endowments in his possession, along with the prohibitions on force and fraud in dealings. Within these rules, the two traders reached an outcome that can be described by a new endowment, a new imputation or allocation of the two goods between the two parties.

What can we say about the outcomes so attained? We can say that given the initial endowments and the existing rules of the trading game, the outcome is whatever
maximizes value. But note that the source of value lies exclusively within the preferences of the persons who trade. There is no external source of evaluation, and trade as such is not merely a means by which some external, independently existing value is achieved. Individuals make their evaluations of the two commodities only as the trading process takes place, and, without trade, there could be no means of determining what value is at all.

Only through trading agreement can maximum value be attained. But note that this is not the same as saying that value is defined by agreement as such. In the process of reaching agreement, the traders are expressing values they place on the commodities. They are not deriving values from agreement; values are, instead, the elements that lead to agreement.

We deal with this matter in such detail here because there seems to be widespread confusion as to the connection between maximum value, or efficiency, and agreement. Professor Jules Coleman, in a well-informed critique of our position, has distinguished between what he calls epistemic and criterial usage of agreement in the definition of efficiency, or maximum value.
*9 By epistemic usage of agreement, Coleman refers to the instrumental function of the contractual process. Hence, if there did exist some scale of value or valuation external to the traders, the trading process itself might be evaluated in terms of its success or failure in moving toward the maximum on such a value scale. In contrast with the epistemic use, Coleman introduces the criterial usage of agreement, by which he means that value is defined by the agreement per se. In the simple trading process analyzed earlier, we showed that neither of the two uses is involved. No scale of value exists external to the trading process. But in the expression of individual values through which agreement is ultimately reached, the traders are not deriving values from the end state of agreement as such. Their own, individually based values emerge as trade takes place; these values do not reflect feedback from the agreement itself.

The economist, who conceptually observes the trading process and who sees no violations of the basic rules, can assign an “efficient,” or “maximum value,” label to the equilibrium result. In so doing, he is not evaluating the result against any scale external to the participants in the trade, nor is he introducing some value scale of his own. Within the rules, as defined, the trading outcome must always be “efficient,” and there is no way the economist can define an “efficient” allocation independent of trade itself. The economist is forced to bring his own evaluative criteria to bear on the rules of trade rather than on the results of trade.
*10

V. Politics in the Exchange Perspective

We have discussed the simple trading example in some detail because the implications of the contractarian vision of politics can perhaps best be understood through the exchange analogy. In its most general terms, the contractarian paradigm for politics is the exchange paradigm, and analysis proceeds most readily in the familiar illustrations of trading in commodities. So long as the source of value is exclusively located in individuals and there is no differentiation among persons, the whole enterprise of politics can be viewed only as a complex many-person system of exchanges or contracts. Individuals must be conceived to join together to explore and ultimately to agree on the establishment of collective entities or arrangements that prove mutually beneficial.

“Trade” among persons in this setting will not, of course, take precisely the same form as the simple exchange of oranges and apples, both of which are privately partitionable and privately consumable goods. In politics, at the most general level, the result of “trade” among persons will be a set of agreed-on rules rather than a well-defined imputation of goods among separate individuals. At the simplest and most elementary level, that described by the initial leap from Hobbesian anarchy, individuals may reach an agreement to respect the property and person of individuals other than themselves. In such a trade, each participant secures the benefits of order, thereby reducing the need to expend his own resources in defense. In exchange, each participant gives up his own freedom of action in potential exploitation of the property and person of others, as defined in the contractual agreement.

This elementary conceptual example suggests that contractual agreement on rules must precede any ordinary trading of partitionable goods. Unless there exists some mutually acceptable understanding of who has what rights to do what with what, when, and to whom, the more familiar marketlike exchanges cannot be initiated. To return to the earlier illustration, if persons A and B do not mutually acknowledge some property rights in oranges and apples as the initial endowments, they have no basis on which to begin exchange. Political order must, therefore, be antecedent to economic order, even if the exchange paradigm seems more natural when discussed in an economic than a political context.

The elementary conceptual-conjectural example also allows us to introduce several other principles that help clarify the continuing confusion in political-legal discourse. The rules of political order, including the definition of the rights of persons, can be legitimately derived only from the agreement among individuals as members of the polity. There is no basis for the commonly encountered notion that individual rights are somehow defined by governments. Individuals establish governments for the purpose of guaranteeing and protecting the rights agreed on in contract. Independent action by “government,” or by persons who act as agents of government, to modify or to change individually held rights must violate the spirit of the contract. Of course, in the establishment of the political entity, powers of coercion are granted to governments, powers that are designed to prevent criminal trespass and exploitation of rights by internal and external aggressors. In the assignment of these powers, problems of control may arise, problems that are not amenable to easy solution. Once established as sovereign, government may not willingly remain within the limits of its initially delegated authority. To the extent that it exceeds these limits, however, government becomes illegitimate in its actions, even in the gloomiest of the Hobbesian visions of contract. Government may, in this setting, take on itself the role of redefining individual rights, but it does so in explicit violation of its contractually legitimate origins.

The non-Hobbesian vision of contract does not acknowledge that the sovereign must remain ultimately uncontrollable. The rules of political order may also lay down limits within which political units may take action, and, indeed, most explicit discussions of constitutional change involve questions of defining the limits of political rather than personal authority. The importance of making the basic distinction between rules and actions within rules emerges more forcefully in politics than in ordinary personal dealings. In private behavior, individuals are implicitly recognized to remain within their legally defined limits or rights. “The law,” as an institutional structure, prohibits a person from invading the domain of others, or if a person does so invade, the law invokes punishment. Almost all of our ordinary behavior takes place within a well-defined structure of legal rules. In politics, however, the notion that collectivities, governments, also behave and should behave within the constraints of well-defined rules seems less “natural.” The notion that constitutions define the limits of political authority is an abstraction that seems difficult for many to comprehend.

VI. Unanimity as the Contractual Ideal

Individuals are conceived as entering into discussion and ultimately reaching some initial agreement that both assigns separate individual rights and establishes an authority charged with the protection and enforcement of these rights. For this contractarian metaphor to be coherent, the agreement must be conceived to be inclusive. The terms must be accepted by all persons who are to be designated members of the group affected. Contractual agreement among a subset of persons, with terms to be imposed on others, would negate the legitimacy of the whole construction.

In an evaluative usage, an observed set of rules must be such as might conceptually have emerged from general agreement. The basic rules themselves, however, may well prescribe procedures for taking actions, privately or collectively, that do not require the consent or agreement of all members of the polity, either in actuality or evaluatively. Indeed, the distinction among levels of decision, exemplified in our fundamental two-stage emphasis on rules and behavior within rules, helps clarify this feature. There may be general agreement on a rule for taking political action (for example, determining the level of public outlay on education by a majority voting rule in a legislative assembly) that does not require approval by all members of the community. Or there may be unanimous agreement on a structure of legal rules within which persons in their private capacities may be allowed to take actions that other persons oppose (for example, entering into an established industry or profession).
*11

Even if at a highly abstract philosophical level the unanimity basis for establishing the legitimacy of the institutions of social order is acknowledged, on a practical level a requirement for unanimity may seem to be mere utopian romanticizing. Individuals come to the contractual process, even in its most idealized form, with separately generated values and with separately identifiable interests. The analogy with simple trade in partitionable private goods breaks down, or so it may be argued, because in any social or political contract, there must be agreement on the same rule, to be applied to all participants. In the exchange of apples for oranges, by contrast, agreement is facilitated by the simple fact that the two parties need not end up, after trade, with the same potential consumption bundle. If A and B, in the example, were somehow required to agree on a final allocation that assigned equal quantities of each good to each person, it seems likely that this requirement would preclude agreement altogether in many situations.

Quite apart from the “publicness” of any political agreement, however, is the even greater difficulty we confront when we consider the size of the trading group. With simple commodity exchange, or indeed any economic relationship, the agreement reached must ultimately include only two parties or agents in any directly participatory sense: one buyer and one seller for each good to be exchanged. The presence or absence of other buyers or sellers will, of course, affect the terms of trade between any two traders, but 2 remains the magic number for the economic analyst.

With political exchange, however, all parties must agree on terms. If we generalize from the simple economic exchange paradigm, any basis for expecting unanimous agreement on anything seems to disappear. In a situation where a unanimity rule is operative, each person is placed in a position, vis-à-vis all others, fully analogous to that held by a single party in a bilateral monopoly game. The incentives for strategic bargaining behavior seem maximal. At this point, however, the “publicness” of the result, noted before as restricting agreement in the economic exchange setting, partially offsets the incentives for strategic bargaining behavior. If the participants are constrained by the knowledge that any outcome reached must be equally applicable to all of them, they have much less incentive to try to hold out for purely distributional gains analogous to those promised to the successful strategist in the game of bilateral monopoly in goods.

VII. Agreement on Rules and the Veils of Ignorance and Uncertainty

The seemingly convincing argument against the feasibility of applying the unanimity criterion in imputing legitimacy to political rules is substantially mitigated if we recall once again the categorical difference between the choice among rules and the choice among alternatives within rules. For the latter, which we may call ordinary politics, elements of conflict seem to overwhelm elements of potential cooperation because the interests of individuals and groups are well defined and readily identified. The set of starting points, or possible positions of status quo, from which no change could ever be made under a unanimity rule would seem almost all-inclusive. In the economists’ terminology, the Pareto-optimal set would be exceedingly large. This prospect is dramatically modified, however, when the choice alternatives are not those of ordinary politics but are, instead, rules or institutions within which patterns of outcomes are generated by various nonunanimous decision-making procedures. The scope for potential agreement on rules is necessarily wider than that for agreement on outcomes within specified rules.

This result follows directly from the fact that the interest of any person or group is much less easily identified in the choice among rules. It is much more difficult for a person to determine which of the several choice options confronted will, indeed, maximize whatever set of values that person desired to maximize. There are two reasons for this loss of interest identity as we shift to the level of choice among rules or institutions. In the first place, rules are, almost by definition, applicable to a number of instances or cases. That is to say, rules embody characteristics of “publicness” that need not be present in specific political outcomes. As an example, consider the position of a dairy farmer confronting choices at the two levels. He might strongly oppose a specific reduction in milk price supports, since such action will almost surely reduce his net wealth. At the same time, however, he might support a generalized rule that would eliminate political interference with any and all prices for services or goods. The effect of such a rule change or institutional reform on his own net wealth is less determinate in the latter case than in the former.

A related but somewhat different characteristic of rules is that they embody an extended time dimension. The very notion of a rule implies existence through a sequence of time periods. We could hardly describe a game by its rules if they were made up at the beginning of each round of play. Rules tend to be quasi-permanent; they “live” longer than outcomes of decisions made under them. A balanced-budget rule, for example, would be meaningless if it were adopted for only a single budgetary period. Such a restriction would be a “rule” in our sense only if it were known to remain in force for a succession of budgetary periods.

As both the generality and the permanence of rules are increased, the individual who faces choice alternatives becomes more uncertain about the effects of the alternatives on his own position. This relationship is perhaps most clearly illustrated when we postulate that individuals will try, generally, to maximize their expected net wealth. But this standard behavioral postulate of the economist’s is not a necessary part of the analysis here. Regardless of what the participant seeks to maximize, the uncertainty about how particular rules will affect the value of his maximand increases as rules are made more general and more permanent.

The uncertainty introduced in any choice among rules or institutions serves the salutary function of making potential agreement more rather than less likely. Faced with genuine uncertainty about how his position will be affected by the operation of a particular rule, the individual is led by his self-interest calculus to concentrate on choice options that eliminate or minimize prospects for potentially disastrous results. Consider, for example, a political decision rule that would call for the random selection of one member of the community who would be authorized to make all decisions as to who would be put in prison and for how long. Even though each citizen, at the level of constitutional choice, would know that he had the same chance as any other citizen of being elected as the “dictator,” this arrangement would almost surely be rejected. The potential losses from the exercise of arbitrary powers by the one person selected would far outweigh the expected gains. (Chapter 4 develops this point in more detail.)

To the extent that a person faced with constitutional choice remains uncertain as to what his position will be under separate choice options, he will tend to agree on arrangements that might be called “fair” in the sense that patterns of outcomes generated under such arrangements will be broadly acceptable, regardless of where the participant might be located in such outcomes. It was on the basis of this chain of reasoning that John Rawls introduced “justice as fairness.” The affinity of the discussion here with Rawls’s whole construction is apparent.
*12 We have stressed uncertainty rather than ignorance, however, because the design of the choice alternatives themselves may influence the former.

The “veil of uncertainty” may be approached, if never fully realized, if persons are modeled as though they were faced with choices among rules of social order that are generally applicable and guaranteed to be quasi-permanent. By comparison, the Rawlsian “veil of ignorance” is an idealized normative construction, the appropriate starting point for persons when they consider making choices among basic principles of justice. Two objections have been raised to this construction. Can real persons choose as if behind such a veil of ignorance when, at another level of consciousness, they realize who they are? And does the construction have empirical bases in the commonly held feelings of justice? The partial veil of uncertainty, which we use and which was initially introduced in
The Calculus of Consent (1962), is not subject to comparable criticism. It does not require persons who enter into the constitutional dialogues to shift moral gears. Persons are modeled as they are. The design of the choice alternatives must, however, affect their behavior, and in the limiting case, the veil is equivalent to that described much more fully by Rawls.

Another, and ultimately mistaken, objection that has been raised to the Rawlsian construction is not applicable to our own. In Rawls’s formulation, the person in the original position behind the veil of ignorance knows nothing about his own prospective position under the chosen rules of social order—under the selected and potentially operative principles of justice. At the same time, however, the person allegedly knows everything about the general characteristics of the outcomes under such rules. This formalized construction has led several critics to charge that there is no contractarian element in the whole Rawlsian construction, contrary to what Rawls himself suggests, and that all such disembodied persons would automatically agree on the preferred option. This argument is mistaken on its own grounds because it neglects the subjective elements that must be present in predictions concerning the working properties of alternative institutions.
*13 These subjective elements remain, of course, in the uncertainty representation of the calculus; these elements alone make contractual agreement necessary for validation. But, as noted, in the uncertainty representation, the two constructions attain full equivalence only in the limiting case.

VIII. Conclusions

Much of the discussion in this chapter has summarized material developed in earlier works within what may be called the contractarian strand of modern public-choice theory. We felt it necessary to include the material here in summary form, however, in order to maintain the potential interest of those who may not be familiar with the earlier contributions. Without some feel for the contractarian vision or paradigm from which we start and within which we work, the more narrowly focused, less familiar arguments advanced in succeeding chapters may seem to be free-floating irrelevancies.

One of our continuing frustrations has been the apparent unwillingness of our peers to acknowledge the importance of the several principles that, to us, seem elementary. The problem seems to be one of a difference in vision (paradigm, conception, or research program). In this chapter we have sketched our own vision; in Chapter 3 we shall try to describe and criticize the vision of the anticonstitutionalist.