The Reason of Rules: Constitutional Political Economy
By Geoffrey Brennan and James M. Buchanan
- Ch. 1, The Constitutional Imperative
- Ch. 2, The Contractarian Vision
- Ch. 3, The Myth of Benevolence
- Ch. 4, Modeling the Individual for Constitutional Analysis
- Ch. 5, Time, Temptation, and the Constrained Future
- Ch. 6, Politics Without Rules, I
- Ch. 7, Rules and Justice
- Ch. 8, Politics Without Rules, II
- Ch. 9, Is Constitutional Revolution Possible in Democracy
Is Constitutional Revolution Possible in Democracy?
This book is an extended essay in persuasion. We have tried to offer logical bases for imposing constraining rules on the political actions of persons, including ourselves, as present or potential actors in political roles. We hope that our arguments for the desirability of accepting a “constitutional attitude” have been convincing, both in the abstract analysis and in the practical applications of this analysis. Widespread acceptance of such an attitude is a prerequisite of genuine constitutional reform. We would, however, be naïve in the extreme if we failed to acknowledge major barriers to the achievement of any shift in the basic rules of political order. This chapter is devoted exclusively to analysis and discussion of these barriers.
It is, we presume, evident that we must answer the question of the chapter’s title affirmatively. If we did not, the discourse of this book, and of much of our work elsewhere, would have to have been driven by purely aesthetic purpose. In one sense, we share a moral obligation to hold fast to the faith that persons organized in a polity can reform the rules by which and under which they live. Any other belief, it seems to us, would be a counsel of despair. This faith in the potential for reform must, nonetheless, be distinguished from constructive analysis of the difficulties of escaping the dilemmas we confront. If analysis of these difficulties were to yield wholly negative results, faith in any potential for reform might be sorely shaken. We need, if at all possible, to present a scenario in which genuine constitutional change takes place through the activities of men and women we can all recognize. If people were saints or angels, the dilemmas we face would never have emerged in the first place.
We should also emphasize the relevance of the last two words in this chapter’s title. We examine prospects for constitutional revolution
in democracy, which means not only that our attention is limited to democratic polities but, more important, that constitutional change is required to be “democratic” in the sense that it must emerge from the internal workings of politics in which the whole electorate potentially participates, rather than from the dictates of certain “divinely qualified” persons either within or outside the community. We use the phrase “divinely qualified” here because any discrimination among persons with respect to the formation of the basic rules of social order must be legitimated by reference to some external supraindividualistic criterion.
This limit to our analysis excludes from consideration all nondemocratic revolutions, constitutional or otherwise. Such revolutions are always possible, even in countries that have long been accustomed to democratic procedures of governance. A group of persons—a military junta, a single party leadership, a ruling committee, an elitist establishment—seizes power and imposes its will on all persons outside the dominating group. It is relatively easy to model constitutional-institutional reform in this fashion, which amounts to saying that it is easy to play at being God. The academic scholar-scientist, in particular, may be tempted to play such a role by imagining that he holds the revolutionary reins of power. In this dream world, the scholar-scientist can make his own idealized suggestions for constitutional change while still under the guise of participation in an ongoing dialogue with his fellows.
The issue we face is much more difficult than that involved in playing God. Our concern is with the prospect of securing
general agreement on changes in the basic rules of the political game, even on the part of persons and groups who seem to be relatively advantaged under existing institutional arrangements. In other terms, our concern is with prospects for noncoercive and voluntary resolution of the generalized social dilemma that seems to describe modern politics. Honest diagnoses suggest that the dilemma is real, and throughout this book, we have taken these results to be stipulated.
*65 We do not live in the best of all possible constitutional worlds, and here we examine the possibility of escaping into a different one.
II. Pareto-Superior Change and Wicksellian Unanimity
We shall begin by making a summary detour through some fundamental principles of theoretical welfare economics, specifically through the Paretian criterion for the justification of a change or move. Pareto supplied a classification scheme, in which all possible social “situations,” “positions,” or “states” belong to one of two mutually exclusive sets. The first set includes all nonoptimal, or inefficient, positions; the second includes all optimal, or efficient, positions. The latter set consists of all nondominated positions, with dominance being defined for all persons in the relevant community.
*66 No change can be made from a
Pareto-optimal position to any other position without at least one person in the community being harmed. A
nonoptimal position is one from which some change can be made in such a fashion that at least one person is benefited and no one is harmed. This purely definitional classification scheme then allows us to say, formally, that there must exist at least one position within the Pareto-optimal set that can be attained by change from any position in the nonoptimal set, a change that will harm no one. Such a change is defined as
At this purely formal level, we need not be concerned about the precise meaning of a “social state,” “position,” or “situation.” In the orthodoxy of modern welfare economics, a “social state” has been defined almost exclusively in terms of an imputation (allocation or distribution) of endowments and/or goods among persons. Needless to say, such an outcome, or end-state, usage of the Pareto criterion is inappropriate for our purposes in this book and elsewhere, but this concluding chapter is clearly not the place for an extended critique of end-state methodology. In this discussion, the Pareto criterion is applicable to rules, or institutional arrangements within which persons act to generate patterns of outcomes.
Saying, then, that a generalized social dilemma exists amounts to classifying the existing rules of the socioeconomic-political game as nonoptimal, or inefficient. And if we insist on noncoercive and voluntary agreement, we restrict “efficient” or “optimal” to the Pareto meaning. This tells us, in turn, that if our diagnosis of the presence of a dilemma is correct, there must be Pareto-superior changes that are possible. If the rules are nonoptimal, there must be reforms that will benefit everyone.
If we translate the formal Pareto scheme into Wicksellian terms, we can say that if a genuine dilemma exists, it must be conceptually possible to make some change on which
all persons in the community could agree. Unanimous agreement on some proposed change in the rules must be at least conceptually possible. Critics might immediately suggest that unanimous agreement smacks of utopian absurdity, at least in terms of practicable implementation. In our view, however, criticism of this sort amounts to putting things the wrong way around. The relevance and applicability of the Wicksellian-Paretian criterion as a benchmark on which analysis and discussion of constructive constitutional reform must be based are in no way challenged by such criticism. If we are consistent in our determination to limit analysis to prospects for internal change, for “democratic” reform of political arrangements, there is no alternative criterion for the evaluation of proposals.
III. Distributional Limits and Prospective Rules
There are obvious limits on the range of possible Paretian-Wicksellian changes from any defined status quo. This point can be seen most clearly if we shift to the more orthodox setting and define the status quo position as an imputation of valued goods among separate persons in the community. To the extent that a proposed change involves a purely distributional shift, it cannot be classified as Pareto superior, even as a conceptual possibility, if “goods” are defined to include all that persons value. (Of course, if the distribution were defined in terms of physically measurable commodities, Pareto-superior changes might be made.) If we think of some value-numeraire, any change from an initial distribution must reduce value for at least one person, whose agreement could then never be secured. That is to say, if a proposed change involves “taking away” valued goods from some person or persons and “giving to” others, the change could never be accepted voluntarily by those who would be harmed in the process.
A direct implication of this elementary point is that constitutional revolution, in the sense discussed here, would be impossible in a democracy if this revolution were conceived as a fundamental shift in the distribution of commonly valued goods and services or in the distribution of claims to such goods and services.
This distributional restriction on the scope of constitutional change may initially seem quite severe. This impression is misleading, however, and tends to be fostered by the total absence of realism in the definition of any status quo position as
a distribution of goods. It is necessary to keep in mind that our concern is not with potential changes in specific end states, or outcomes, even if such changes were within the realm of the possible. If the relative positions of individuals in sociopolitical order are described by an imputation of valued goods, services, and claims, distributional
reallocation must characterize any change. (No one would expect poker players to divvy up the chips after the game had been played.) Our focus is rather on the potential for changes in the
rules or institutions of social order. These rules provide the framework within which patterns of distributional end states emerge from the interaction of persons (players) who play various complex functional roles. The precise distributional effects of a change in the rules on any identified person or group at any point in time may be difficult if not impossible to predict. A status quo defined only in terms of the rules (the laws and institutions) within which persons act is conceptually very different from a status quo described by a particular distribution of valued goods and claims. (The poker players may all agree on a change in the rules for future play, regardless of the rules under which play has proceeded during earlier rounds of the game.)
Let us digress, briefly, by noting that the proclivity of mathematical economists to force all analysis into mathematically tractable forms has done serious damage to clarity of thought with respect to the end-state-rules distinction. It has proved relatively easy, mathematically, to define distributions of goods in vector language. But how could a status quo be described mathematically if what exists is a set of rules that constrain but do not determine individual choices that, in turn, generate patterns of distributional outcomes? End states lend themselves to mathematical manipulation; processes do not.
Changes in rules are
prospective in their distributional implications, whereas changes in observed distributions themselves are necessarily
retrospective, with reference to the choice behavior of the persons who act to generate the results. Changes in rules that can lay any claim at all to consensual agreement can, at best, modify personal expectations about future distributional patterns. Rule changes cannot modify observed distributions as such, unless, of course, such changes are applied ex post facto.
An example of the distinction is provided by normative tax theory. A collective decision to levy a progressive tax on incomes is a change in fiscal rules. Despite the distributional consequences that might be predicted, such a decision may conceptually be agreed to by all persons in the community, particularly if there is a time lag between the agreement and actual levying of the tax. An individual would know that if he chose to earn income, this income would be taxed at progressive rates. But contrast this with a collective decision to levy a tax on wealth. In this case, persons who have accumulated wealth have made prior choices under other rules. A change in fiscal rules of this sort violates all criteria of “fairness” and could never lay claim to Pareto superiority, even as satisfying a purely conceptual criterion.
Acknowledgment of the prospective nature of any change in rules that qualifies as Pareto superior carries important implications for the design of proposals for reform. To the extent that proposed changes in rules explicitly embody time lags between approval and implementation, the ability of persons who are potentially affected to predict with accuracy the impact on their own identifiable positions must be reduced. Prospective rules introduce uncertainty into the calculus of those who participate in the decision making, and this uncertainty increases with the length of the time lags.
The temporal element becomes even more significant when it involves the
permanence of the proposed rules change. To the extent that persons who must choose among rules do so in either the knowledge or the expectation that any alternative, once chosen, is to remain in effect over an extended sequence of periods, they will be less able to predict how particular rules will affect their relative positions over the same sequence. The logical basis of the differentiation between the decision rule for making changes in the constitution (amendment procedures) and the decision rule for making choices within given constitutional rules lies in the recognition of this temporal distinction.
Proposals for changes in rules that are known to be or expected to be short-lived, even if adopted, are more predictable with respect to their distributional impacts. Hence, agreement on short-lived change may be more difficult to secure. Indeed, the meaning of “rule” implies quasi-permanence; a game whose rules were changed with every round of play would be little different from a game without any rules.
In their book
The Calculus of Consent (1962), James Buchanan and Gordon Tullock used the prospectiveness and permanence properties of constitutional rules as the essential means by which an individual can construct a bridge, of sorts, between short-term, identifiable private interest and long-term, nonidentifiable self-interest, which then becomes “public interest.”
The construction has a close affinity to the somewhat more familiar Rawlsian veil of ignorance.
*68 In the Rawlsian framework, the individual has a moral obligation to choose among principles of justice as if he were behind a veil of ignorance. And at a different level of analysis, the stance so depicted is held to describe empirically widely held notions about justice. That is to say, persons “should” try to adopt the veil-of-ignorance position, and, furthermore, persons do embody some such stance when they seriously consider, and talk about, basic rules for sociopolitical order. The Buchanan-Tullock derivation is, in comparison with the Rawlsian, less demanding on individuals in that, at least in the limit, they are forced by the circumstances of the choice setting to adopt a stance that is equivalent to the veil of ignorance. Because of the inherent uncertainties about their own positions, persons must, on rational self-interest grounds, use criteria akin to the fairness precepts outlined by Rawls. The effect of uncertainty is to mitigate substantially any purely distributional aspects of genuine constitutional choice.
IV. Status Quo Entitlements and Distributional Envy
A second major hurdle to be surmounted in generating consensus on basic constitutional reform is indirectly related to the distributional aspects discussed in Section III and, in one respect, exists precisely because the requisite limits of distributional uncertainty cannot be attained. Realistically, we must recognize that persons will not, and cannot, make constitutional choices behind a veil of complete ignorance, owing either to their inabilities to adopt the moral stance urged on them by John Rawls or to the impossibility of designing institutional circumstances sufficient to make such a stance rational on strict self-interest grounds. To some residual extent, distributional implications of any proposed change in rules can be predicted, including the prediction of differential impacts on identifiable individuals and groups. This fact alone would seem to stymie consensual constitutional reform from the start. Must the question posed in this chapter’s title be answered negatively?
This barrier to potential agreement is not nearly as serious as it may at first seem. The necessary presence of some predictable distributional impact of any proposal for change will thwart general agreement on any simple version of the proposal. But at this point it is useful to recall the earlier detour into Paretian-Wicksellian welfare analytics. If the status quo set of rules is, indeed, nonoptimal, or inefficient, there must exist potential agreement on
some change in structure. The working out of such an agreement, even for conceptual evaluative purposes, may, however, require a complex network that includes various compromises, side payments, compensations, bribes, exchanges, trade-offs—a network aimed precisely at offsetting the predictable adverse distributional properties of the proposed changes. If particular persons, as individuals or as members of groups in the economy, foresee distributional harm as a result of the implementation of a simple change in the rules, the designers-proposers must work out additional elements of a “package deal” that will change the results.
The problem here is not one of difficulty in locating the necessary elements of a complex constitutional “exchange” that will promise net benefits, ex ante, to all participants. There may be many such packages that would command general assent, especially when there are major overall gains to be made by a change in institutional-constitutional structure. The discovery of such mutually beneficial proposals, best considered as complex exchange propositions, is the proper task of the specialist in political economy.
*69 The problem worthy of our attention here concerns the willingness of those who may be direct beneficiaries of structural change to pay the compensations that may be required to secure general agreement.
The problem can be stated differently. All parties in the political community may be unwilling to accept the status quo distribution of entitlements generated under the operation of existing rules. This distribution of entitlements may not be acceptable to many persons as the appropriate starting point from which genuine constitutional reform is to be made. That is to say, there may be no way to get from “here” to “there” until prior agreement is reached on the “here” and the embodied distribution of rights that exist.
Consider land reform in a developing country as an example (and purely as an example, since we make no claim to competence in the analysis of comparative efficiencies of alternative institutional arrangements in such situations). Let us stipulate that there is general agreement among all persons in the community that land reform will increase the overall efficiency of the national economy, efficiency being objectively measured in dollars of gross national output or product. From this fact, it would seem to follow that some reform proposal, in a complex package deal, could be worked out on which everyone could be brought to agree. Such a proposal would have to include compensation to existing owners of extensive landholdings that would be sufficient to secure their acquiescence to a change in ownership patterns. To finance such compensation payments, however, prospective landholders and/or taxpayers generally would have to sacrifice some purchasing power, some share of their net promised benefits. This result might seem less desirable to these groups than the result achieved by a simple “taking” of the landed estates. Despite the acknowledged net gains from the land reform, persons who themselves would be in the net beneficiary groups might remain unwilling to finance the necessary payments to existing landowners, on the grounds that status quo holdings are “unjust.” This attitude, and behavior stemming from it, might prevent any consensus from being reached on constitutional change, even with the widespread knowledge that such a change would generate major overall efficiency gains.
The prevalence of such an attitude toward patterns of holdings, or rights, in the status quo will, of course, depend in part on the individually estimated prospects for successful alternative avenues through which change can be implemented. Respect for the status quo distribution of entitlements and, hence, willingness to pay the compensation required to attain consensus on change depend critically on whether the allegedly “unjust” entitlements and claims can be changed
without the voluntary agreement of existing holders of such entitlements being obtained. The temptation to confiscate privately held capital values overtly through the exercise of power by dominating political coalitions is always present.
Perhaps even more important, the expected value of gains from politically orchestrated “taking” may remain positive, and differentially higher than the expected gains from genuine constitutional change, for many persons who are not at the time, but may expect to be, members of political coalitions capable of carrying out confiscatory action. Indeed, the problem discussed here may be most extreme in those situations in which “political property rights” are most uncertain.
In our land reform example, many persons may consider existing holdings to be “unjust” and hence appropriately subject to overt confiscation (to nondemocratic constitutional change in the terms of our reference in this chapter), but at the same time these persons may be uncertain as to their own potential voice in governmental-political decisions. Despite this uncertainty, however, they may be unwilling to support any scheme that will allow some politically orchestrated “buyout” of existing holdings. The situation of the nominal owners of the estates in the status quo may be similar even if on the other side of the issue. If these nominal owners acknowledge their impotence in the face of identifiable opposition coalitions, they might voluntarily agree to structural rearrangements that include much lower net compensation than straightforward market-value estimates might produce. On the other hand, if these holders maintain confidence in their power to influence political outcomes, they may demand compensation in excess of what other members of the political community will finance, especially given the estimated threat potential that members of the second group think they possess. The impasse that may arise in such settings is analogous to the problem that arises in market transactions when ownership rights to potentially tradable goods are not well defined and legally protected.
The land reform example should not be taken for more than that, an example. Many other examples, such as slavery in the American South in the 1850s, might have been selected. The central point is that ambiguity and uncertainty about the distribution of political rights under a set of constitutional rules may inhibit or even totally prevent changes in these rules, despite a general acknowledgment that such changes would improve welfare. Nondemocratic change in the form of violent revolution or civil war might become the alternative to consensual reform in such settings.
The barrier to genuine constitutional reform analyzed in this section varies in significance with the perceived “legitimacy” and “effectiveness” of the existing political constitution, the status quo set of rules. To the extent that this constitution commands little respect, in part because it is seen to fail in its function of limiting the scope of both governmental and private intrusions into what are widely held to be protected spheres of activity, the direct-confiscation alternative to Pareto-superior or consensual change may seem to offer higher present value to most citizens. The somewhat ironic result is that it may be much easier for a country to achieve genuine constitutional revolution if its citizenry has long adhered to a “constitutional attitude,” fostered by a historical record during which limits on the power of governments have proved effective at least to a degree. If there has been a widely acknowledged separation between “law” and “legislation,” to use Hayek’s terminology,
*71 or between “constitutional” and “postconstitutional”-“in-period” choices, or between “changes in rules” and “play within defined rules,” the willingness of persons to seek somewhat different and more inclusive processes of decision making to change rules, as opposed to processes for political action within existing arrangements, is increased. Conversely, it may be difficult to achieve basic constitutional change in a political setting where governments are seen to have been effectively unlimited in power and authority. Why should an unlimited majority coalition, either existing or potential, ever seek constitutional change through consensus when the same results can be achieved at lower cost?
In these respects, it would presumably be easier for the United States, or any political structure characterized by an effective division of power, to effect changes in its political structure than it would be for the United Kingdom or any parliamentary democracy in which political authority is concentrated in a single legislative assembly.
V. Constitutional Change and Free Riders
The barriers to constitutional reform discussed in Sections III and IV are significant. We should not underestimate either the distributional implications of any change in structure, with their feedback effects on potential consensus, or the effects of uncertainties in the political status quo. Nonetheless, it could be argued that these barriers to reform are less important than the one to be discussed in this section, one that is much more difficult to supersede in any normatively predictive sense.
We refer to the general problem of “publicness,” in the formal welfare-economics meaning of the term, and to the behavioral implications of this characteristic, which are often summarized under the rubric “free rider.” In its most comprehensive forms the question is, How is “public good” produced? How is “public bad” avoided? How is collective action motivated other than via the private interests of individuals?
It is relatively easy to conceptualize or model a small group of players in an ongoing game considering possible changes in the rules of play, the rules that describe the game itself. But what is the analogue in the larger, more comprehensive sociopolitical “game” in which there are literally millions of players? Who are to take upon themselves the personal burden of designing provisional proposals for basic changes in the rules when the promised benefits accrue
publicly, that is, to all members of the political community, and with no differentially identifiable residual claims to the promised “social” profits? What is the constitutional equivalent of the patent law, which guarantees a special, even if limited, monopoly privilege to the inventor and which, in turn, offers incentives for creative effort by all potential inventors? What is the political-constitutional equivalent of entrepreneurial profits, the search for which drives the economy and motivates attempts to locate higher-valued ways of organizing production and combining resources, both within and across markets, broadly defined?
Can general rules be changed in a deliberative process of collective choice, even if there are acknowledged possibilities of Pareto-superior reforms? And if not through deliberative choice, how can general rules be modified? Must we resign ourselves to the acceptance of one of what seem to be only two alternatives? Must we acknowledge that changes in basic institutions can be imposed only nondemocratically, by a group seeking to promote its own interests, the alternative avenue of “reform” that we have refused to discuss in this chapter? Or failing conscious constitutional change and eschewing nondemocratic “reform,” must we fall back on slow, unconscious, and unintended processes of social-cultural evolution, with little more than pious hope that such changes as do occur will lead us toward rather than away from a set of arrangements that might be conceptually evaluated to be Pareto optimal?
Confronted with questions like these, the economist seems likely either to despair or to avoid relevant thought altogether by playing mathematical games. An economic model of behavior that lays any claim at all to predictive power must rely on the uniformity of human nature summarized in
Homo economicus, or economic self-interest objectively defined. Such a model of behavior has served economists well; the model in all its variations explains much of what we know about the political-governmental sphere of human action in addition to the much more that we know and can explain about human action in market relationships. This economic model yields meaningful predictions; testable hypotheses can be derived about the organization and functioning of political parties, coalitions, committees, pressure groups, governmental agencies, and legislatures. Hypotheses about how bureaucrats behave, how budgets are constructed, how regulation operates—these and others emerge from application of the central
Homo economicus postulate. But all such hypotheses rest, finally, on the explanatory power of this self-interest motivational assumption. The professional economist must look at the governmental-political process as driven by the same forces that drive the market process; even when analysis incorporates the recognition that self-interested behavior is not nearly so amenable to simple modeling in political as in market choice.
It should be evident, however, that the basic analytics of “positive public choice” cannot be readily extended to explain changes in the basic rules of political order that are necessarily “public” in scope. The individual actor in the positive public-choice model does not act contrary to his defined self-interest, even if it is not so easily defined as some of the simplistic exercises might suggest. To the extent that “investment” in institutional analysis, design, argument, dialogue, discussion, and persuasion is costly in a personal sense, the individual of the orthodox model will forgo such investment in favor of more immediate gratification of privately directed desires. Why should anyone do “good”? There is no way that economists who stay within the strict limits of the discipline can respond to such a question; they cannot manipulate utility-maximizing actors so as to offer a satisfying response.
VI. The Role of Norms
Economists have great difficulty in moving beyond the rather simplistic, if powerful, models of human behavior grounded in self-interest motivation. We claim no exception to this generalization about our disciplinary peers. Nonetheless, anyone who diagnoses the plight of modern democracy in terms of the existence of a social dilemma described by a set of nonoptimal rules must give up in despair, become a revolutionary, or go beyond the models of utility maximization, nontautologically defined. To hold out hope for reform in the basic rules describing the sociopolitical game, we must introduce elements that violate the self-interest postulate.
If persons do not behave in accordance with their own economic self-interest, objectively defined and measured, on what basis do they act? The hard-nosed positive economist mounts an effective critique of peers in the other social sciences and in social philosophy when they are challenged to produce an alternative model with predictive content. The economist is well equipped to recognize mush for what it is, and when noneconomists hypothesize that persons want to “do good,” he quickly detects the absence of predictive content. To make such hypotheses operational, even at the level of analytic discourse, additional definition is necessary. What, precisely, do persons do when they “do good”? If individuals differ widely in their conceptions of “good,” attempts by each to “do good” amount to little more than random deviations from behavior modeled on self-interest postulates.
If the alternative hypotheses are to carry predictive weight, they must be amended to state something like the following: “People want to do good or to take right actions, and they share a single conception of what is good or right.” The
commonality of the norm, at least over a large number of persons, is a necessary feature of any operationally useful theory of choice or action that moves beyond the strict individualistic models.
Applied to the problem at hand, which is that of deriving some conceptual explanation of why individuals might be expected to seek out, design, argue for, and support changes in the general rules of the sociopolitical order when, by presumption, such behavior would be contrary to identifiable self-interest, it is necessary to resort to some version of “general interest” or “public interest” as the embodiment of a shared moral norm. That is to say, persons must be alleged to place positive private value on “public good” for the whole community of persons, over and beyond the value placed on their own individualized or partitioned shares.
*73 Furthermore, this “public good” that is privately valued must be that state of affairs defined by the interaction of freely choosing individuals, rather than some transcendental notion derived from God or Karl Marx.
If, however, individuals do, indeed, place a positive value on the shared conception of “public good,” the economist-critic can then ask the straightforward question, How could a community made up of such persons ever find itself in the social dilemma that has been diagnosed? Why would a community of such persons ever need a change in the rules of the game, or why is there a need for any rules at all? The appropriate response is equally straightforward. To hypothesize that persons place positive value on a shared “public good,” defined as indicated, does not in itself imply anything at all about this evaluation relative to that placed on the furtherance of individuals’ private economic interests. The economists’ explanatory model reasserts itself at this point and suggests the presence of a necessary trade-off between “public good” and “private good” in the choice calculus of any person who maximizes utility. From this elementary point it follows that the observed degree of “public regardingness” in behavior will depend on the relative costs as these emerge in the institutional setting.
It is precisely here in the argument, or so it seems to us, that a categorical distinction must be made between choices confronted within or under an existing set of rules and choices confronted among alternative sets of rules themselves. In the first of these two settings, that of postconstitutional or in-period choice, the relative costs of choosing courses of action that further the shared “public good” may simply be too high, relative to the increment in “public good” promised to result from such action, to shift behavior significantly away from economic self-interest. In the second choice setting, by contrast, the costs of furthering “public good” may be significantly lower, so much so that the same person who behaves in accordance with narrowly defined self-interest within the given set of rules may well behave in accordance with precepts of shared norms when making genuinely constitutional choices.
The categorical difference here stems from the meaning of
general rules, the generality referring to their application to all members of the body politic. The principle is an elementary one in theoretical welfare economics and in the theory of public finance, and it has been widely recognized, especially since it was emphasized by Baumol in
Welfare Economics and the Theory of the State.*74 Nonetheless, the principle is often overlooked in otherwise sophisticated discussion. A good example is the discussion of public and private charity. Many who have opposed governmental transfer programs refer to the opportunity persons have, in their private capacities, to make charitable transfers to the needy. These critics then go on to infer that arguments in support of governmental transfer programs must arise from a desire to impose coercive taxation on those who would not voluntarily make charitable transfers. Some such motivation is probably present in many of the arguments, but these critics overlook the distinction made here to the effect that persons may, in a collective setting in which their actions are generalized, voluntarily agree to transfer schemes that involve costs to them that would never be borne in a private institutional situation. An individual might voluntarily agree to a tax-transfer scheme that imposes a net individual cost of one hundred dollars, if he knows that all other, similarly situated persons will also bear net costs of one hundred dollars each. The same individual, however, may contribute only fifty dollars, or less, to privately organized schemes having the same purpose in the absence of the political program.
The nature of rules changes ensures that all persons in the political community will be involved similarly, at least in one meaning of the term. (All players are required to play by the same rules, whether these rules are those that now exist or those that might exist after a change.) In considerations of proposals for a change in the rules, therefore, there is much less conflict between what can be called “private” and what can be called “public” advantage. An individual cannot, in nearly so direct a sense, set off his own private gain against public gain as he might do in purely private market choice or in political choice made within ordinary majoritarian institutions for decision making. In either of the latter two decision settings, the behavior of a person in generating “public bad” or in failing to produce “public good” might be motivated strictly by the opportunity to secure private gain. Such an opportunity might not really exist, at least directly, in the genuinely constitutional-choice setting.
The free-rider dilemma that emerges in constitutional choice is of a sort quite different from what enters the choice settings just noted. In constitutional choice, free ridership refers largely to the potential absence of individualized interest in general rules rather than to the presence of any directly contrasting nongeneral or private interest, where nongenerality itself ensures the differentiality. As we have noted, some such identifiable private interest may remain in any constitutional choice; it may not be possible to design proposals for constitutional change that will place persons behind sufficiently thick veils of uncertainty. Nonetheless, a shared moral norm in such a setting has much less “work to do” than it might in a nonconstitutional setting.
VII. Toward a Civic Religion
In this chapter, we have discussed three major barriers to constitutional reform in democratic social order. The thrust of our argument has been that these barriers can be overcome. We do not, however, underestimate the task Western democracies face, and we do not want to seem unduly optimistic. We are asking no less than that the basic rules of the socioeconomic-political game be changed, rules that have been in place for decades, and that these changes be made peacefully while the game continues to be played under the old rules. This order is a tall one indeed, and we should be under no delusion that a constitutional revolution will simply emerge, in revolutionary fashion, without a conscious investment of effort.
Our efforts, in this book and elsewhere, are aimed largely at the academic constituency, at our peers in the social sciences, law, and philosophy. And it is here that we must begin, especially if we accept Keynes’s dictum about the influence of academic scribblers. It is naïve to think that practicing politicians and practical individuals become constitutional statesmen without some coherent set of ideas provided by academe, either before or along with political articulation.
The first step is the achievement of some measure of academic consensus on diagnosis. We must come to agree that democratic societies, as they now operate, will self-destruct, perhaps slowly but nonetheless surely, unless the rules of the political game are changed. In the United States in particular, there seem to be modest grounds for hoping that such a consensus is in the process of construction, notably with respect to the fiscal and monetary rules, which have been the subject of much discussion in the 1970s and 1980s. There is increasing awareness that something other than ordinary politics will be required to generate fiscal and monetary discipline, that a regime is needed that will function in the acknowledged long-term interests of all participants in the body politic. The search for, and discussion of, alternative fiscal and monetary rules has begun.
In more general terms, this book is an expression of the hope that a new “civic religion” is on the way to being born, a civic religion that will return, in part, to the skepticism of the eighteenth century concerning politics and government and that, quite naturally, will concentrate our attention on the
rules that constrain governments rather than on innovations that justify ever expanding political intrusions into the lives of citizens. Our normative role, as social philosophers, is to shape this civic religion, surely a challenge sufficient to us all.
We must redesign our rules, and our thinking about rules, with the ultimate aim of limiting the harm that governments can do, while preserving the range of beneficial governmental-collective activities. We plead with our fellow academicians to cease their proffering of advice to this or that government or politician in office. Good games depend on good rules more than they depend on good players. Fortunately for us all, and provided that we understand the reason of rules in the first place, it is always easier to secure agreement on a set of rules than to secure agreement on who is or is not our favorite player.
The Limits of Liberty (University of Chicago Press, 1975); also see Gordon Tullock,
The Social Dilemma (Blacksburg, Va.: University Publications, 1974).
International Economic Papers, 9 (London: Macmillan, 1959), pp. 39-92.
A Theory of Justice (Cambridge, Mass.: Harvard University Press, 1971).
Journal of Law and Economics 2 (October 1959): 124-38; reprinted in James M. Buchanan,
Fiscal Theory and Political Economy (Chapel Hill: University of North Carolina Press, 1960), pp. 105-24.
The Limits of Liberty (University of Chicago Press, 1975), ch. 2.
Law, Legislation and Liberty, 3 vols. (University of Chicago Press, 1973, 1976, 1979).
Selfishness, Altruism and Rationality (Cambridge University Press, 1982).
Welfare Economics and the Theory of the State (Cambridge, Mass.: Harvard University Press, 1952).