Rational Choice Theory and the Evaluation of Public Policy
Neiman, Max, Stambough, Stephen J., Policy Studies Journal
Theory and Values of Policy Choice
Judging and evaluating the substance, process, and results of governing are venerable preoccupations among scholars. Questions respecting the role of government in the economy are found in the work of the Physiocrats and in Adam Smith's Inquiry into the Wealth of Nations, as well as in the outpourings of contemporary social commentators, politicians, activists, and scholars. In our nation there have been long-standing discourses regarding proper, just, and effective policy. These have occurred in the august realms of scholarship as well as in the boisterous domains of popular discourse ranging across a host of issues, including constitutional principles, slavery, voting rights, protectionism, immigration, welfare, unions, crime, prohibition, capital punishment, and defense policy.
In democracies, the processes associated with how governing institutions address various policy issues and alter public policies always have been primarily an amalgam of various forces. No matter how one approaches the task of explaining why and how public policy is developed and changed, the policy process, in actuality, is disorganized and chaotic. There are appropriate analytical reasons for simplifying things by using some particular framework (Dye, 1987). On the one hand, when it comes to how things really work, there are the various communities of advocates and specialists who are connected to a given policy issue by profession or personal interest (either ideological or material). There are the many actors, in both public and private spheres, who play critical roles in the agenda-building process (Cobb & Elder, 1983) and who formulate and implement policy proposals to address various policy objectives.
While that is happening, of course, public opinion, organized pressures, constituency interests, the personal views of legislators, and the views of those who implement policy all come to bear. We cannot forget, moreover, the role of formal rules and institutional contexts in which these processes take place. Indeed, the consequences of implementing policies are mediated through a host of institutional, behavioral, and process-based contexts. In other words, the politics of initiating policy responses, formulating policy, implementing, and evaluating public policy always has been an open process.
Therefore, in democracies it is through that three-ring circus called democratic politics that we decide which issues are important, which policy responses are chosen, with which methods to implement policy, how we judge the consequences of policy, and whether to terminate policy. In American politics, the process of pluralist policymaking is a highly complex, raucous, crowded carnival of social forces. The results of the process are rarely polices that are well crafted. Rather they tend to be compounds of competing, often conflicting, demands and priorities. For those of us who have special affection for neatness, order, and formalism, democratic policymaking in general, and perhaps more so in the United States, is not a pretty sight (Lowi, 1979).
The concern and worry about government size today is wrapped up with the messy process that usually characterizes policymaking. The increasing distress that people claim afflicts them due to government size is wrapped up with current worries about personal freedom, government costs, and the rise of fringe, paramilitary organizations. Let us eschew the issue of what is meant by government size for the moment and assume we all know what that means, even though it is an extremely complex issue. In any case, in this essay, we are concerned with the implications of rational choice theories as the most important challenge to conventional, pluralist politics.
Although there are many expressions of the current animus toward government and its presumed size, the one we are concerned with here is the emergence of rational models as the basis for deciding when to use government and what governing tools to use. Every governing system must address the following issues: (a) under which circumstances does government deal with previously private matters; (b) which methods does government employ to accomplish these policies; and (c) how does government assess the consequences of what it does for the purpose of refining, altering, or terminating a particular policy? Rational models of policy choice have emerged as the preeminent methods of addressing these issues because conventional democratic politics is viewed as producing too many interventions in previously private matters, poorly designed polices, and ineffective methods of evaluation. In this sense, then, rational models are seen not merely as some bundle of techniques to do policy analysis. The mission of public/rational choice practitioners has become, in our view, far more ambitious; the objective for many devotees of rational models is to supplant the pushing, bargaining, and noise of democratic politics with the putative elegance and parsimony of a rational calculus, the application of which will maximize cost-effectiveness and personal freedom.
At the very time that popular disquiet with "politics" as a means of making governing choices has reached a climax, the infatuation with rational models has grown. Indeed, it is fair to say that the dominant approach to making, implementing, and evaluating public policy is one variation of the rational choice theme or another. Ranging across diverse policy areas, from criminal justice to agricultural subsidies, to housing or medical care, the tools employed by rational choice theorists have come to prevail as the paramount criterion used to judge whether public policies are successful. While there are a variety of reviews and critical assessments of various public/rational choice modes of analysis, the conclusion remains that rational models seem ever more dominant in studies of what government does and how and what it should be doing (Coleman & Fararo, 1992; Green & Shapiro, 1994; Johnson, 1991).
The preeminence of rational choice approaches has popularized such standards of policy performance as returns on investment (cost/benefit analysis) and a host of other concepts critical to the variety of ways in which public interventions in society are justified and evaluated. The concepts of public goods, externalities, market failure, or common property resources have been applied in a wide variety of contexts. The contributions of rational choice theory to the task of evaluating public policy have been substantial and include such matters as: (a) emphasizing the importance of opportunity costs in achieving public benefits; (b) illustrating that there are a broad range of alternative ways of delivering public goods and services, other than hierarchical, government bureaucracies; (c) emphasizing the importance of customer-oriented criteria for assessing satisfaction with government services; and (d) highlighting that marketlike incentives and pricelike mechanisms are superior to rules and coercive sanctions in achieving public objectives.
Despite these achievements, however, there is a sense in which there is a kind of imperialism in rational choice theory - where the elegance and parsimony of analysis overwhelms legitimate concern for the consequences of applying rational choice theory. Steve Kelman (1981) expressed this concern in his examination of the use of incentives to achieve public purposes. Although public/rational choice theories have been assessed in a number of forums, our concern focuses on what we believe is the commitment that many advocates of these theories have to a conservative political agenda in the most fundamental sense. It is our view that, with few exceptions, rational choice theorists begin with a deep repugnance for politics as a means of making choices.
Although it is understood that rational choice theory is imbued with normative commitments, in our view the most important of these derive from the typical rational choice practitioner's loathing of "politics" as process and a kind of celebration of the private exchange of markets. We contend further that this loathing of politics is related directly to the grand tradition of contractarian political theory. Like their social contract ancestors, rational choice theorists reluctantly concede a place for politics. Simultaneously, however, rational choice theorists, again in the fashion of social contract theoreticians, erect barriers to what they believe are unwise, imprudent, and excessive interventions into a mainly felicitous private sector.
Much of the effort of the social contract theorists and their classical liberal descendants is devoted to designing ways to ensure that politics is minimized and does not vulgarize decisionmaking (e.g., tyranny of the masses, especially as this might produce excessive demands for redistribution of wealth or too much regulation of business). The framers of the U.S. Constitution sought to design the institutions for governing so that a variety of objectives could be met. Separation of powers, bicameralism, and property qualifications to vote were seen initially as methods to insulate policymaking from certain political forces that would detract from wise and just policy. The fear of egalitarian, direct democracy was explicit, and a preference for limiting the scale and scope of government, as well as avoiding redistributive public policy, were important features of the Lockean tradition (Hartz, 1955). In this sense, social contract theorists have sought to formulate what we call "extrajudgmental criteria" for assessing when government should do something, how government should do something, and how evaluations of government performance should be conducted.
Public choice theorists, then, are sent off with the following marching orders:
1. Private sector choices are in the main superior to those of the public sector, when such choices are evaluated in terms of efficiency and "consumer" satisfaction.
2. When departures from private sector allocations occur, these should be initiated only when the conditions requiring such intervention are met.
3. Such conditions should be expressed rigorously and should avoid the caprice and expediency associated with conventional politics, even democratic politics.
4. When government interventions are conceded to be necessary, as a result of having met the conditions calling for such actions, then they should be designed as much as possible to replicate the workings of markets, on the supply and demand sides.
Concepts such as market failure, public goods, externalities, and common property resources are developed, then, to provide such rigorous conditions to guide government intervention and the evaluation of such intervention (public policy). Presumably, such concepts will function to limit the occasions on which government intervention is justified, and when such intervention is justified the evaluative tools of rational choice focus on the appropriate standards of assessment. Applied in this sense, then, rational choice theory comprises a flight from politics.
In the end, then, there is a tension between the criteria of democratic policymaking and the technical requirements of "rational" public policy. For example, we suggest that the very tools of rational choice analysis and its values produce far more government involvement than expected. We also offer a number of conclusions regarding the role of democratic policymaking as a means of dealing with some of the concerns of classical liberals and others concerned about excessive government intrusions in previously private matters.
Rational Choice and Public Education
The key point in our discussion, then, is that application of rational choice principles to policy design and delivery might produce results that are in conflict with the objectives of rational choice advocates. To illustrate how this might occur, we explore a policy that has been and continues to be at the forefront of this rational choice reform movement - replacing the currently prevailing mode of public education with one emphasizing school choice. Specifically, we discuss the proposal for the public to give vouchers, through government entities, to parents, which the parents can use for tuition at a wide array of schools. It is not our goal to make a statement as to the overall merit of these policies. Instead, we wish to use this policy as an example of how policies that grow out of the rational choice literature can lead to more government involvement rather than less.
To explore this point, we address two questions. First, do voucher programs create a system in which public goods are ignored or not satisfied? Second, do voucher programs create a system in which government intervention might become greater than it was before? We will make this argument by examining three areas involving the implementation of voucher programs: (a) financing, (b) direct regulation, and (c) distribution of information.
The decline in academic performance of our public schools often has been highlighted as one of the premier examples of failure in public policy (Chubb & Moe, 1990). The work of rational choice scholars and partisans has been important in inspiring many calls for comprehensive reform of our system of public education. The fundamental argument undergirding rational choice school reform programs is that the current public school system is monopolistic. Unless special circumstances exist or unless parents are willing and able to send their children to private schools, in most of the United States a child's school is determined by geography. Rational choice analysts state that this monopolistic control of schools creates a system in which there is little incentive and even fewer opportunities to improve schools. Scholars argue that government is one of the problems and that consumer sovereignty, rather than government bureaucracy and intervention, is the proper way to improve schools.
The most common manner in which the rational choice perspective is used in the field of education is what Richard Hula (1986) refers to as market analogs, where policies are designed to imitate the private sector. In this sense, features of the idealized, private market are seen as desiderata to be replicated wherever possible, even if that has to be contrived through government action. Prices, competition, consumer sovereignty, and individual utility maximization are features of the market that public policy should seek to replicate or reestablish. The purpose of imposing a rational choice solution to a policy area is the perceived or real failure of the current bureaucratically delivered manner in which policy is developed and implemented. Rational choice theorists believe that allowing parents and children to act like consumers and forcing schools to act like competitive producers creates a more efficient and satisfying system of public education. Under most choice proposals those schools providing a quality or satisfying education will be rewarded in the marketplace with higher enrollment, and hence profits. Schools that do not provide a quality education will lose students to better schools and will fail (go out of business). Eventually, the workings of market incentives and market discipline, it is supposed, will engender a system of schools that maximizes the education tastes of citizens as if they were consuming any other market product (Chubb & Moe, 1990; Friedman, 1955, 1962).
Underlying all school voucher proposals is the idea that parents are given a voucher from the government to pay for tuition at any certified educational institution. Many of these proposals allow for the vouchers to be used at approved private schools, as well as at schools operated by the government. Therefore, parents will be able to choose to send their children to the school that best fits their children's needs. It is assumed in these schemes that it is the parents whose educational tastes are being satisfied; the implied accompanying assumption is that satisfying parents' tastes is harmonious with the interests of the children, both near-term and far-term. In this sense, we are not in a position to assess the hazards or benefits resulting from the decisions of parents to send their children to schools that do not teach evolution, severely curtail exposure to modern literature, and/or do not support "frills" like newspapers, orchestras, sports, science fairs, and the like. Similar issues arise when a significant number of parents send their children to schools that exclude controversial topics or issues in their curriculum, such as Native American history or abortion. We mention these things not simply to be alarmists, but to emphasize our view that assuming the values of parents are automatic guarantors of the well-being of their children is often a heroic, if not foolish and tragic, assumption. The argument is not that parents do not know what is in the best interest of their children and that government necessarily knows better. The argument simply is that basing a system on parental goals without knowing these goals (only assuming them) can be problematic.
Commenting on the school-choice program (Center for the Study of Public Policy, 1970; Coons & Sugarman, 1978; Friedman, 1955, 1962), Levin (1991) argues that to understand the implications of voucher proposals one must look at three factors: (a) financing, (b) direct regulation, and (c) distribution of information. We use these suggestions to guide our investigation of school voucher programs in the following sections.
Issues relating to the finance of school voucher systems are much more complex than simply a question of how much the vouchers will be worth, Some of the issues that must be addressed are whether the value of the voucher given to a family should be inversely related to family income, should some consideration be made for travel costs - especially for inner-city families - and whether funding considerations also might involve extra costs for some students who need potentially more costly services such as bilingual education or specialized assistance for various physical and cognitive challenges (Levin, 1991).
Choice advocates argue that an appropriate level of funding can be determined and that the resulting system will be one in which more people have choices and resources are utilized more efficiently. Two assumptions in this logic are important in understanding the true nature of the choice perspective and education funding. First, the rational choice approach assumes that vouchers will save money due to the use of an efficient market mechanism rather than an inefficient government mechanism. West (1991) questions whether the level of savings in voucher plans is as great as proponents argue, due to the fact that the fixed costs of governments collecting and distributing the resources remain since the government is the one issuing the vouchers. Moreover, if payments will vary by the income level of families, then additional and costly mechanisms to verify incomes on a continuing basis will have to be established. On the other hand, if every family, rich and poor, receives the same amount, then there is no redistribution of resources at all, and the prospect will exist that .the school voucher program merely makes the entire school program more expensive. This is because there will be a sudden infusion of resources on the demand side, thereby driving up the prices charged to everyone.
A more important assumption, however, is that these vouchers will allow parents to choose to which schools they would send their children, thus placing schools in competition with other public schools and private schools. Several potential problems must be considered. The first is whether or not parents will be allowed to supplement the vouchers with additional funds of their own. If not, as Chubb and Moe (1990) propose, then the voucher program is, in effect, establishing a price ceiling. As Frey (1992) argues, however, many methods already exist for higher-income families to circumvent this price ceiling. One way is for the schools to establish a system in which private individuals could contribute to an endowment that would support the school. It is not difficult to imagine an arrangement in which the children of individuals who give to this endowment are admitted at a higher rate than those who do not. Insofar as these problems are to be managed, then, they will require government involvement of some sort in school and family finances.
If the government allows the circumvention of this price ceiling as described above, then very little progress in curbing income segregation would be possible. Wealthier families would be able to guarantee enrollment of their children in schools that poorer students could not afford, thus minimizing the potential social benefits of such a plan. An alternative would be to enforce the price ceiling. However, the methods Frey describes to correct this problem pose new questions about the role of government. Governmental attempts to enforce this price ceiling would create a larger role for the government in terms of auditing these private schools and perhaps even personal incomes. This situation is particularly problematic for religious schools, which, to receive these vouchers, will see increased government involvement in their affairs, assuming, of course, that issues regarding separation of church and state are resolved.
In addition to this question about price ceilings, voucher programs also have to address the question of price floors. Any voucher program must state what happens to the additional money if a school charges less than the voucher is worth. Will these schools be allowed to give "rebates" to parents? After all, if there is fierce competition for students, will schools provide incentives for parents to enroll? If schools wish to look better to their customers, to attract other students, might they give rebates to students with excellent records of school performance? If schools are not allowed to grant rebates to parents, then the government is effectively establishing a price floor - assuming this price floor cannot be circumvented as Frey suggested the price ceiling could be.
Additionally, and more importantly, if schools do give rebates to parents - either legally or through circumvention of the law - then one of the key assumptions of the choice approach to education might not be accurate. After all, we are assuming that parents are motivated to provide the best education available to their children. Perhaps this is true most of the time, but clearly, too often, it is not. Families might be tempted to augment family incomes by accepting rebates, even if this sacrifices educational quality. This temptation might be greater among families when rebates can augment a family's income and when that family has financial difficulties. Such rebates might be used to improve diet, housing, or medical care, or perhaps in some cases more frivolous purposes.
We wish to emphasize that our purpose here is not to assault the idea of school vouchers or school choice. It is entirely possible for the government to establish a system that takes into consideration the problems of price ceilings and floors discussed in this and other research. Rather our focus is on the question of whether a system of school choice designed on rational choice principles will reduce the scale of government. Either government can restrict the use of rebates - thus creating a price floor - or government can subsidize further the incomes of lower-income citizens. If not, then this market-oriented approach seriously risks not achieving its stated goals of truly granting consumers the choice to purchase educational services from whatever provider they prefer. Either way government involvement is much larger than rational choice advocates seem to consider.
In addition to the financial aspects of government in voucher programs, the government also would have a major role to play in other types of regulation. There is a public aspect of public education that goes beyond mere funding issues. Other public goods are addressed through a system of public education that may be pushed aside under a strictly market incentive approach. Benjamin Barber (1992, p. 14) discusses this concern by stating: "Voucher systems enhance parental choice and increase competition among different kinds of schools, yet they thwart common schooling and subordinate education's public ends to private market choices." One public end that may be thwarted is that of desegregation.
Scholars have argued recently that choice and voucher programs result in increased segregation (Margonis & Parker, 1995). If there is a collective good to desegregated schools, voucher systems could destroy it. Without governmental involvement in the area of desegregation in voucher systems, schools will need only to respond to the needs of their consumers - individual parents or groups of parents. Therefore, there is no mechanism for the schools to respond to a public good like desegregation, only to the private demands of individual parents. Some research suggests that choice programs can be used - and are used - to reinforce segregation (Moore & Davenport, 1990; Smith & Meier, 1995). This segregation could occur along the usual lines of race, religion, and income, but in the absence of government involvement and regulation, students might be segregated by lifestyle, ideological, and other as yet nonobvious bases. To prevent a return to school segregation, governments might be required to oversee in some important degree the admission requirements of all schools that accept vouchers by establishing standards governing the campus values and curriculum and entrance requirements.
Distribution of Information
A final area in which voucher programs inadvertently may increase the role of the government is in providing information. The market approach to education assumes perfect information on the part of the producer (schools) and consumers (parents). No one expects that perfect information is possible, but a rational, informed consumer is needed for a free market system to work. The imperfect information associated with school choice is of extreme importance to the success of such a system. Catterall (1992) argues that imperfect information among parents about school quality generally follows income and educational levels. There is, of course, a very important racial aspect to this problem, since certain racial minorities also tend to have below-average incomes. Because the relative marginal cost of getting information is so much higher for parents with lower income levels, these parents may not be able to be the informed consumers needed to make the model work. Assuming that education is an important factor in a child's future success, the market failure associated with imperfect information as well as some of the problems mentioned earlier could reinforce a class structure in the United States and decrease social mobility. Might this lead to yet more government involvement to redress these problems?
Because of the problems associated with imperfect information, some agency will need to provide periodic, comprehensive reports about the quality of each school for all parents. In some areas, private enterprise may develop such services. In other areas, the government - most likely state or school board agencies - will be called upon to provide this service. This situation greatly increases governmental activity and oversight of our educational system. For parents to have adequate information about all of the schools, the government would have to evaluate all schools that accept vouchers - including private and religious schools. The collection of this information would go far beyond the current system of accreditation. It would include detailed reports regarding curriculum, success of students in college admissions, and information regarding the kinds of colleges or other schools to which the graduates of voucher-funded schools go. Gathering such information might be costly in obvious ways, but also would engender new relationships between government officials charged with retrieving the information about school quality and with ensuring the reliability of this information. Without this governmental involvement, however, the voucher system would be in danger of making a bad situation worse in terms of educational opportunity and social mobility. A host of other potential difficulties deals with consumer protection issues for both children and parents. There also are additional issues regarding the continued balkanization of the nation, and there are sure to be issues regarding the appropriate role for government in defining what is meant by schools, education, and appropriate curriculum.
Because of imperfect information, price ceilings, price floors, and even the continued problems of unequal distribution of wealth, the rational choice perspective on school vouchers fosters a system in which society must make a choice. It is unlikely that society will ignore many of the issues inherent in implementing a voucher-based system of school choice, and the result might be an entirely new arena in which government becomes involved. If citizens ultimately find that their government must have a role in financing, direct regulation, and information dissemination, then the level of government intervention produced by "school choice" might be in serious conflict with the values of the proponents of these systems.
Regulatory Sanctions as a Key to Smaller Government
In the previous section we focused on the prospect of government's role actually becoming more substantial in unanticipated ways, as we seek to apply a rational choice-driven policy - in this case, school vouchers. Our attention turns now to what may be viewed as a perverse argument: that traditional, coercive forms of government intervention tend towards smaller government. The usual scenario leading to some new government intervention in previously private matters is that a problem ascends to the social and political agenda, advocacy groups emerge, media involvement grows, resistance from the targets of public concern is marshaled, possible responses by elected officials are crafted, law is enacted, implementation is designed, and a variety of impacts results. The tools involved in regulation most often are some kind of coercion, with rules and standards enforced by fines or other sanctions. Public choice theorists really disdain this form of government response, since it tends to displace markets with bureaucratically determined outcomes and presumably reduces personal liberty. The costs of such policies tend to be hidden and diffuse, so that the benefits of regulation are hard to ascertain relative to their costs. The public, moreover, by not bearing substantial regulatory costs directly, is likely to demand more regulation than would be the case if these costs were more apparent and directly felt.
As Wilson (1980) has demonstrated, however, there are times when demands for regulation are unlikely to get transformed into public policy. When regulatory policies are sought, where there are diffuse benefits for a large, public and concentrated costs on some set of power interests, the result usually is a defeat for such demands. Moreover, in the relatively rare situations in which such policies are implemented over a resistant target, there is likely to be on-going interest among the regulatory targets in opposing, altering, terminating, and otherwise weakening the scope, substance, and effectiveness of policies that are seen as costly or harmful. Once the officials who carried the public's interest in the legislative domain are concerned by other matters or displaced by officials with different views and priorities, once the media no longer are interested in dwelling on the reasons that gave rise to public concern, then the political advantage shifts to the opponents of regulation or of other policies that are seen as inimical.
On the other hand, imagine if we were to take to heart the rational choice theorists' laments concerning the use of conventional regulation, and replace rules and regulations with incentives and subsidies. Presumably by increasing the return for complying with public objectives, there will be more efficient compliance by the targets of government policy. After all, we are more likely to do the things for which we are rewarded and we are likely to try to evade and end those policies that inflict costs on us. Government, by not employing coercive methods, presumably would be less inimical to personal freedom. So instead of coercing corporations to comply with clean air standards, we would give tax breaks, direct grants, and honors for achieving such standards. Instead of using affirmative action standards and rules to pressure employers to comply with employment targets and goals, we might provide tax breaks for each percentage increase in the proportion of previously underrepresented groups in the employer's workforce.
No doubt, with "incentives" to comply rather than threatened sanctions, businesses would rush to demonstrate the many things they have done to improve air quality or to reduce water pollution. Employers might rush to discover the many different ethnic, racial, and cultural backgrounds of their employees, to receive the rewards derived from enhancing their workforce diversity.
The key issue here is that the ordinary interest that individuals and organizations have in resisting government interventions is diminished greatly in the face of rewards, incentives, and subsidies. In fact, while the size of government might involve less coercion, government growth actually might accelerate as a consequence of the greatly increased range of things that government will be doing as various actors seek to qualify for subsidies. So, while government might not be more coercive, it might be far more active. In fact, this leads to an important insight: The size of government is a function of how it accomplishes its objectives (coercive v. noncoercive) plus the number of things the government is doing (the range of government activity). We express this in the following equation:
GS = [m.sub.1][A.sub.1] + [m.sub.2][A.sub.2] + [m.sub.3][A.sub.3] + ... + [m.sub.n][A.n]
where GS is total Government Size and [m.sub.1] is the method employed to implement some particular activity or action.
It is possible, then, to have two jurisdictions engaging in the same range of actions yet differing in the size of their governments due to differences in the methods used to implement this range of activity. We are saying that if we reduce the use of methods of coercion we actually might increase the range of government activity. Figure 1 suggests the nature of this inverse relationship, indicating that the more coercive a proposal is, the less likely it is to be enacted or to persist in the face of opposition by those against whom the coercion is directed.
It is difficult to explore conceptually whether a smaller range of coercive activities is "more" government than a wider array of less coercive activity. We do not have adequate conceptual guidance for developing weights and measures for the parameters in the preceding equation. In other words, which activities contribute the most to increasing the scale and/or burden of government is a very muddy issue. How do we compare tax increases to substantive mandates (e.g., comparing an increase in sales taxes to government requirements that political demonstrations must receive prior, official approval)? Did Franco's Spain have a "smaller" government than contemporary Sweden because the former had a smaller social welfare component? Are citizens of the Netherlands more vulnerable to official abuse than citizens of the United States because the latter has a smaller social welfare component? Is every increment of government involvement equivalent to every other government action? Are restrictions or policies governing political participation of citizens and environmental or workplace safety standards equivalent in their relevance for personal liberty?
To complicate matters a bit more, we also argue that there is an inverse relationship between the method ([m.sub.1]) and the action ([A.sub.1]) being implemented. Our claim is that in democratic societies, outside the arena of criminal justice, ceteris paribus the more coercive the method to implement some action the less likely it will be adopted. Obviously, we are more likely to enact tax breaks for installing nonpolluting equipment than we are to enact fines against businesses for not installing nonpolluting equipment. So, as we rely on less coercive methods, we increase the range of things the government does, and while we employ more coercive methods, the number of things the government does is minimized.
Indeed, even government programs that do not involve coercion, but which are administered through annual budgets, tend to come under annual scrutiny by opponents. This is a particularly interesting issue of conventional budget principles and social equity. For example, most programs benefiting the poor require annual budget appropriations and government agencies to administer programs. Over the past 25 years, due to increasing pressures on budgets, many programs for the poor have been cut or otherwise have grown much slower than inflation. Those of us who receive public benefits through the tax code, however, have fared much better, in the main. It is more difficult for the poor to receive their benefits from the tax code, since usually they do not file long-form tax returns. Benefits that are distributed through tax expenditures, however, tend to be less scrutinized, since there is no automatic review of these expenditures in the same way that a program for the poor must experience through the usual budget process. So, while housing programs for the poor have been slashed substantially, just as one example, the tax expenditure for housing programs for the owner-occupied housing population has remained untouched. For example, the estimated foregone revenue (tax expenditure) for just three categories of such items is enormous in the 1997 fiscal year. The estimated total of 1997 tax expenditures associated with interest on mortgage debt, state and local property taxes, deferral of capital gains on the sale of homes, and the one-time permanent exemption from capital gains is far greater than the total cost of all low-income housing programs.
Notwithstanding the current infatuation with the flat tax rate, those who benefit from these tax benefits are not likely to accept quick reductions in these benefits. The relative invisibility and complexity of tax provisions, moreover, make it unlikely that individual tax benefits are likely to be assaulted by someone on mere principle. In short, the apparently less coercive vehicle for providing benefits through the tax code tends to receive relatively little scrutiny through fiscal review, and when it does the review encounters intense opposition from beneficiaries of these benefits. Indeed, it is unclear whether the recipients of tax code largesse even see themselves as beneficiaries of government programs. Instead they see any threats to the tax code as an assault on their disposable income. This belief is similar to that of students at public institutions of higher education who believe that they and/or their parents do not receive government aid yet do not voluntarily pay the nonresident tuition rates.
So it is that the set of programs implemented through the relatively more visible arena of the conventional budget process come to be fodder for the political process. Unlike the beneficiaries of tax code provisions, those who receive direct benefits through annually funded programs are seen as feeding at the public trough or as welfare dependents with no claim on public resources other than those that improve the utility of those whose taxes fund these programs. Naturally, recipients of tax benefits do not see themselves as being funded in part by those who do not qualify for the benefits and must make up the foregone revenue.
During periods of constrained budgets, interests with fewer political resources are less able to defend the benefits they receive through budgeted programs. Put differently, we argue that during periods of budgetary stress, programs that are supported mainly through the conventional political process are more likely to be cut than are programs supported as tax expenditures through the provisions of the tax code. As a corollary: budgeted programs for the poor are more likely to be cut, ceteris paribus, than programs for the nonpoor, since the former tend to rely more on annually budgeted programs. The equity issue is perhaps made more problematic since programs for the poor are not likely to be implemented through the tax code and the tax code is less likely to be mined for budget savings except intermittently as part of some overall tax reform effort, a relatively rare occurrence. One might claim that charitable contributions are encouraged through the tax code - true enough - and that these benefit the poor. However, that depends on how many of these contributions filter through to benefit lower-income people. Remember that charitable contributions include a host of resources that do not always serve the poor directly, if at all (e.g., youth sports teams, scholarships, rescue and emergency organizations, hospitals, and foundations).
Rational choice recommendations regarding public policy often favor noncoercive modes of accomplishing public objectives. As suggested, however, such methods actually might produce a lot more involvement by the public in previously private matters. While these involvements might appear to be less intrusive, and in this sense imply "less government," such incentives-based approaches might increase the range of things through which we apply public influence. It is possible that the increased range of public involvements, as noncoercive as they might seem, still might produce more government.
Perhaps we can press the issue with an additional flight of theoretical fancy as follows: Imagine that the public could use only coercive methods of achieving public purposes. It is our contention that the demand for extending government to some new area would produce greater opposition. Indeed, it would be more difficult to garner support or maintain the zealousness of implementation over the long haul. Hence, ceteris paribus, requiring that any proposed activity of government must employ coercion to accomplish its purpose means that proposal is less likely to be enacted and less likely to overcome opposition as public attention diminishes. Assume, on the other hand, that the only way to accomplish proposed government activity is through incentives and subsidies. Clearly less opposition is likely and more government, albeit with a soft touch and carrot, rather than a mailed fist and stick.
As mentioned earlier, rational choice theorists and empirical evidence are vague regarding what contributes more to government size - the method of doing something (coercion v. incentives) or extending government activity, however it is done, to some new area. Recall that rational choice theorists are concerned about two things: first, that government intervene in matters only when there is some solid extrajudgmental justification (e.g., public goods and market failure); and second, that when such justification exists government should replicate the workings of markets or marketlike devices and rely on prices, rather than rule enforcement and sanctions. We claim that the latter produces many more involvements by government and hence conflicts with the prior, greater concern that government scale be minimized. Surely a government that does an increasing range of things, even through the use of marketlike incentives, can be intrusive, abusive, and burdensome. Hence, the rational choice remedy might not be a remedy at all for the problem of effective governing.
The Irrationality of Taking Rationality Too Seriously, or How I Stopped Worrying About Maximizing and Learned to Love Democratic Politics
Clearly the rational choice approach disdains politics, even democratic politics, as a means of making choices. The usual scene in democratic politics is messy, fraught with emotion, ideology, and self-interest. Democratic procedures, it is implied by rational choice theorists, do nothing to provide appropriate guides regarding when the public ought to address previously private matters and transform them into matters of governance. There are no principles, it is alleged, guiding either the substance of politics or its methods sufficient to limit governments, even democratic ones, from becoming too large, too burdensome, and too great a threat to personal liberty. Contractarian political thought, with its federalism, separation of powers, judicial review, and checks and balances, has been seen as insufficient.
Rational choice seeks to inject policy discipline, through its emphasis on economistic decisionmaking and the variety of recommendations that flow from the theory of public goods and market failure. Insulating decisionmaking from the vicissitudes and wildfires of the usual democratic politics through the use of extraordinary majorities and requirements to balance budgets, limits on taxing authority, and reductions in the revenue base are all methods to erect barriers to the expression of demands by various groups seeking to achieve varying purposes through the tools of governing. These efforts to constrain government growth are contemporary extensions of the fear and loathing of government that has marked American political life from the onset. They are all expressions of the effort to minimize the role of politics, a desire to flee from politics as a means of making choices about what should be done through government, how should it be done, and when given public policies should be modified or terminated. These are expressions of a desire to displace the bubbling cauldron of politics with the cool, orderly extension of rational, economic principles.
But being rational is a difficult thing for individuals and, as has been discussed in many other forums, even more difficult for organizations comprised of a multitude of diverse individuals with different utility profiles. We believe that even in the most elementary sense, the extension of optimization through rational principles collapses of its own weight. For example, there are programs designed to reduce injury to children due to unsafe toys and other products used by children, but are we concerned about the injury only to children? Well, of course, we are not. We also seek to reduce injury to workers and other adults in many different contexts. There are a host of different things we do to improve the physical wellbeing of many different groups. We also seek to improve mental well-being. The fact is that it is impossible to assess and prescribe the optimal array of societal resources and authority so that one can say that, given the information available and time permitted, that such and such an alternative distribution of the current nearly $2 trillion federal budget will optimize social happiness and welfare. There is no rigorous, scientific, or rational way to make such a decision.
The issue can be illustrated by the circumstances of a family involved in raising funds for research into Lou Gehrig's disease, which afflicts approximately 30,000 Americans per year. The family devoted enormous amounts of time to dealing with that illness, because of the death of a parent. The disease is very gruesome and inflicts considerable suffering on those afflicted. Yet, it is a disease that is relatively rare, and a disease with little prospect of a cure in the near future. In a sense, society might have been served better had there been some encouragement for the family to devote its energies to diseases with more victims and with some prospect of progress in treatment and cure, but sentiment and personal gratification had much more to do with the reason for the family's choice of how it expended its charitable impulse. What point might be made by suggesting that if the family wanted to maximize the "good" it did, it would have been more rational from society's viewpoint if it had invested the members' efforts in some other ailment, such as AIDS or heart disease or breast or lung cancer.
But people do not make decisions about where to spend their time and money and energy on the basis of some methodical, rational calculus. They do not review the range of various diseases and select one to fight because of the outcome of some rational choice exercise. Rather, sentiment, personal involvement, friendship, memory, and loyalty are all involved. These family members did not ask themselves, given that they were willing to devote a given quantum of time, money, and energy to make themselves happy, which pattern of work, play, and charity would maximize the bottom line for them in the long run.
In some sense, then, politics is the alternative to rational analysis in making the big choices. We concede the usefulness of using rational models to choose between alternative systems of city refuse collection (since we all agree on the need to clear the garbage); we can use rational models to compare the relative net return to several different ways to reduce drug use (since we all agree on the desire to reduce drug use); and we can compare which system best delivers a defined package of health care (assuming we agree on a desired health care component). But whether we should invest in more research concerning lung cancer, or to improve child nutrition programs, or to reduce workplace injuries, or to minimize traffic accidents, or to incarcerate more criminals, or to subsidize research into tomatoes that can be picked mechanically are not decisions that lend themselves to being combined into some grand rational calculus for some statement about what levels of support each of these activities should receive.
Recall as well that we are concerned also about the maintenance of democratic values. We want to maintain the capacity to participate in politics; we want officials to be both wise and responsive; and we want a fairly broad band of society, comprised of diverse, often disagreeing citizens, to be represented in policymaking. There are few policy areas in which there is a norm favoring less democracy (excepting perhaps monetary policy and decisions involving the prosecution of war). Foisting rational models and obliging us to abide by the analytical results produced from such analyses to the exclusion of democratic norms would be an exceptionally, perhaps unacceptably, high price to pay for policy order or smaller government, such benefits being problematic if they exist at all in the larger scheme of things.
Has politics been such a terrible method for addressing the major issues of the time? We believe that a fair examination of the recent record suggests not. First, there is evidence that even the most oppressive regimes, with virtually no democratic tradition or practice, have given way to democracy. These outbreaks of democracy suggest that the presence of a sizable, tyrannical government might not prevent the onset of liberty and democracy.
The notion that government growth, such as it is in the United States, is some kind of crisis that democratic politics cannot restrain or assess, or that democracy is itself a major source of government growth, is not supported by the wave of antigovernment sentiment and the fairly conservative popular mood that prevails in American life. Raising taxes, launching new regulatory initiatives, or producing new government programs is very difficult. Tax burdens have remained fairly constant, even declining somewhat. The overall proportion of the gross domestic product represented by government spending and taxing has been fairly constant over the past decade, too. Insofar as the public wants "less government," the indications are that democratic politics has done a fairly reasonable job in addressing the issue of government size and burden. Supplanting much of the political process with rational choice algorithms for deciding when to initiate government action and how to design these actions might reduce opportunities for democratic participation and merely create new forms of government involvement in previously private matters. While these increments of government involvement might be seem less coercive in their design and operation, nevertheless they still might not be benign and may have the potential to engender forces for a greater range of government activity.
Barber, B. (1992). An aristocracy of everyone: The politics of education and the future of America. New York, NY: Ballantine.
Catterall, J. S. (1992). Theory and practice of family choice in education: Taking stock - review essay. Economics of Education Review, 11,407-416.
Center for the Study of Public Policy. (1970). Education vouchers: A report on financing elementary education by grants to parents. Cambridge, MA: Center for the Study of Public Choice.
Chubb, J. E., & Moe, T. (1990). Politics, markets, and America's schools. Washington, DC: Brookings Institution.
Cobb, R. W., and Elder, C. D. (1983). Participation in American politics: The dynamics of agenda-building. Baltimore. MD: Johns Hopkins University Press.
Coleman, J. S., & Fararo, T. J. (1992). Rational choice theory: Advocacy and critique. Newbury, Park, CA: Sage.
Coons, J., & Sugarman, S. D. (1978). Education by choice. Berkeley, CA: University of California Press.
Dye, T. R. (1987). Understanding public policy. Englewood Cliffs, NJ: Prentice-Hall.
Frey, D.E. (1992). Can privatizing education really improve achievement? An essay review. Economics of Education Review, 11, 427 - 438.
Friedman, M. (1955). The role of government in education. In R. A. Solo (Ed.), Economics and the public interest (pp. 123-144). New Brunswick, NJ: Rutgers University Press.
Friedman, M. (1962). Capitalism and freedom. Chicago, IL: University of Illinois Press.
Green, D. P., & Shapiro, I. (1994). Pathologies of rational choice theory: A critique of applications in political science. New Haven, CT: Yale University Press,
Hartz, L. (1955). The liberal tradition in America: An interpretation of American political thought since the revolution. New York, NY: Harcourt, Brace.
Hula, R. (1986). Introduction: Market based public policy. Policy Studies Review, 5, 583-587.
Johnson, D. B. (1991). Public choice: An introduction to the new political economy. Mountain View, CA: Mayfield.
Kelman, S. (1951). What price incentives? Economists and the environment. Boston, MA: Auburn House.
Levin, H. M. (1991). The economics of educational choice. Economics of Education Review, 10, 137-149.
Lowi, T. J. (1979). The end of liberalism: The second republic of the United States. New York, NY: Simon.
Margonis, R., & Parker, L. (1995). Choice, privatization, and unspoken strategies of containment. Education Policy, 9, 375-403.
Moore, D., & Davenport, S. (1990). School choice: The new, improved sorting machine. In W. Boyd & H. Walberg (Eds.), Choice in education (pp. 187-223). Berkeley, CA: McCuthan.
Smith, K. B., & Meier, K. J. (1995). Public choice in education: Markets and the demand for quality education. Political Research Quarterly, 48, 461-478.
West, E. G. (1991). Public schools and excess burden. Economics of Education Review, 10, 159-170.
Wilson, J. Q. (1980). The politics of regulation. New York, NY: Basic Books.
Max Neiman is professor of political science in the Center for Social and Behavioral Science at the University of California, Riverside.
Stephen J. Stambough is assistant professor of political science at North Dakota State University.…
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Rational Choice Theory and the Evaluation of Public Policy. Contributors: Neiman, Max - Author, Stambough, Stephen J. - Author. Journal title: Policy Studies Journal. Volume: 26. Issue: 3 Publication date: Autumn 1998. Page number: 449. © 1999 Policy Studies Organization. COPYRIGHT 1998 Gale Group.