The distinction between the IS and the OUGHT may be employed as a demarcation line between two types of social science theories: strictly theoretical-empirical ones on the one hand and normative ones on the other. Although not all theories fall neatly within one or the other of these categories, the distinction may be used as a tool for pointing out implicit or explicit normative elements in theories which claim to be strictly neutral or objective from a scientific point of view. We will employ this demarcation line in the way Max Weber conceived of it in order to analyse the epistemology the emerging public choice approach (Weber 1923, 1949). It is argued that the public choice framework for the analysis of public sector decision-making and implementation is deficient in terms of the distinction between science and values. Is this accusation warranted? Let me state from the beginning that I certainly do not wish to argue that normative theory is inherently bad or that the statement of an ideology is a meaningless' My point is only that we remain clear about the implications of the demarcation between the IS and the OUGHT.

Resumé

The new political economy - public choice - has contributed considerably to a long-awaited inter-marriage between political science and economics. The economic approach to the interpretation of the public sector - politics as well as administration - has resulted in a number of interestingly relevant models. At the same time modelling political behaviour on the basis of the economic man perspective gives rise to puzzling difficulties creating much controversy. The article attempts to establish in principle the ethically neutral and scientifically objective nature of the public choice approach. It elaborates on the crucial concept of selfinterest the public choice behaviour assumptions in order to show that the criticism from the public interest adherents may be countered.

Public Choice and the Virginia School

It is not difficult to find in the Virginia school of public choice highly critical
statements about the nature of politics and the essence of bureaucracy. It
is not only argued that political behaviour sometimes displays deficiencies

of various types, to be pointed out in normal scientific discourse; nor is it only claimed that the lives of bureaux are sometimes despicable, a theme often repeated in the sociology of formal organizations. Much more is implied in the Virginia version of the public choice approach:

The basic structure of property rights is now threatened more seriously than at any period
in the two-century history of the United States .... But there is more of it. We may be
witnessing the disintegration of our effective constitutional rights, regardless of the prattle
about 'the constitution' as seen by our judicial tyrants from their own visions of the entrails
of their sacrificial beats (Buchanan 1977, 93).

Perhaps James Buchanan would defend such a general statement about the misfortunes of modern political life in the welfare state by claiming that it is a piece of neutral observation. However, it is not difficult to find the action emphasis in Buchanan:

'Government failure' against standard efficiency norms may be demonstrated, analytically and empirically, but I seen no basis for the faith that such demonstrations will magically produce institutional reform. I come back to constitutional revolution as the only attractive alternative to the scenario that we have seen bent to act out. In the decade ahead, we shall approach the bicentenary of the constitution itself. Can this occasion spur the dialogue that must precede action? (Buchanan 1978, 368).

One would, of course, be interested in finding out why revolutionary action is of such utmost importance. The answer could not possibly be simply a neutral and objective analysis of the public sector, but would have to include some normative theory of the state. Thus, Buchanan moves from public choice analysis to the quest for public action.

In Gordon Tullock we find the same sweeping generalizations about the
unhappy state of politics. Each and every bureau is a burden for society:

We are saddled with a large and basically inefficient bureaucracy. Improved efficiency in this sector could, looking at the matter economically, raise our national income and improve our rate of growth. Politically, it could both increase the degree of control the citizen, qua voter, has over many fields of our national life and enlarge his personal freedom (Tullock, 1965, 221).

Among other scholars tied to the Virginia school of public choice we find
the same vehement rejection of present-day democracy, not least the usual
majoritarian decision technique:

The expansion in the powers of government - particularly the federal government - and the accompanying erosion of wealth and liberty is not the result of incompetent public officials motivated by evil intentions. Rather the core of the problem emanates from the incentives that ordinary people confront within the prevailing system of majoritarian government (Gwartney & Wagner 1988).

The Virginia school argues that the public sector is inherently suffering
from systemic failure in terms of policy-making and policy implementation.
Political failure is more severe than market failure. The action implication

is that the public sector must be rolled back by means of a new political revolution leading to constitutional contracts limiting the power of the state - the Leviathan of the twentieth century (Tullock 1970; Buchanan 1975, 1977, 1986; Brennan & Buchanan 1980).

What we are here talking about is the Virginia interpretation of the advanced societies with a liberal-democratic constitution. And the content of the interpretation is a fully fledged rejection of the welfare state and its values. What is science and what is ideology in the Virginia interpretation? In the Virginia version of the public choice approach there are not only strong implicit elements of values in the analysis of the welfare state; there is also an explicit normative theory about how these societies ought to be changed. Does the same hold for the public choice approach in general? Since we as social scientists are better able to distinguish between the neutral and objective analysis of the public choice domain and the normative of what public choice ought to be taken, it seems vital to discern whether the same confusion of the IS and the OUGHT is of necessity true of public choice theory in general.

Public Choice and the Public Choice Approach

The domain of public choice is one thing and the variety of approaches to the analysis of this domain are quite another. There is a large variety of frameworks for the understanding of public sector decision-making and implementation, one subset of which constitutes the so-called public choice school. However, within this school there is space for a number of varying approaches, one of which is the Virginia school. I will argue that its distinctive trait is the confusion of the IS and the OUGHT in the form of a peculiar mix of positive and normative state theory.

It is often argued against the public choice approach that it is not unbiased in the sense of scientific neutrality. It scores low on objectivity as it is inherently oriented towards market values. It is critical of the state and welfare spending simply because it favours market allocation and market values out of ideological right-wing motives. This criticism is a serious one and it requires explicit consideration, particularly from those that argue that the public choice approach is a new set of innovative scientific models (McLean 1987). How is this claim about the fundamentally biased nature of the public choice approach to be countered? There are two options: either one accepts the accusation but argues that each and every approach is bound to be biased one way or the other, or one rebuts the criticism. I will strongly argue against the first and for the second; at the same time it is important to take a look at the grounds for the accusation.

It is symptomatic that a Swedish scholar ridiculed the whole enterprise as

a set of 'wallet theories' by invoking the antique and nowadays abandoned concept of the public interest (Lewin 1990). I will try to argue the case that the accusation of scientific subjectivity and a lack of scientific neutrality is not correct with regard to the general public choice approach, although one may find in the peculiar Virginia-school interpretation of public choice strong ingredients of an explicit right-wing ideology based on a normative state theory.

Ethical Neutrality and Scientific Objectivity

The notions of ethical neutrality and scientific objectivity are regulative concepts. They are meaningful even if it is the case that there exist no texts that fulfil the criteria on neutrality or objectivity. Like the notion of truth they are important to the conduct of scientific enquiry because they provide us with criteria with which to evaluate arguments. Just as one cannot show that a theory is true once and for all but detect that a theory is false, so one may use the criteria on neutrality and objectivity to point out the occurrence of biases in arguments, although it may be impossible to vindicate an argument is neutral and objective.

Some argue that ethical neutrality and scientific objectivity do not matter in scientific discourse. The logic of scientific discovery is complete anarchy, meaning that anything goes as long as it works (Feyerabend 1975). Others claim that the criteria of ethical neutrality and scientific objectivity are not feasible, because it is impossible for a theory to be unbiased as it is bound to be value-ingrained (Myrdal 1970). Neither position appears to be particularly convincing. It seems difficult to do away with the distinction between the context of discovery and the context of justification (Scheffier 1967), as scientific arguments do carry a claim to being justified beyond mere personal factors. And though it certainly is true that social science concepts often are value-loaded (Myrdal 1961), this in itself does not prove that neutrality or objectivity is impossible in principle (Nagel 1961). Actually, the continuous detection of bias in social science theories does indicate that ethical neutrality and scientific objectivity are important and valid concepts. Stating that an approach like the public choice perspective is biased is a serious challenge that needs to be considered not simply by responding that each and every approach is subjective.

The case for ethical neutrality or scientific objectivity may be based on either a thin or a thick epistemological theory. A thick defence of these regulative notions would take us into basic issues in meta-ethical argument about the status of values or bring up fundamental epistemic problems about the relationship between perspective and object. A thin defence of the notions of ethical neutrality and scientific objectivity is all that is needed

here, because it is enough to start from an assumption that truth and value are different and that objectivity is distinct from subjectivity - that is the classical Weberian position (Weber 1923). Scientific inquiry is somehow very different from ethics or political ideology - a difference that has to do with ethical neutrality and objectivity in research.

Thus, we have two distinctions, one between the public choice approach in general and other public sector perspectives on the one hand, and another one between scientific objectivity and subjectivity on the other hand. These distinctions may be illustrated by means of a diagram (see Fig. 1). The accusation towards the public choice approach is that it by necessity ends up in combination 111 whereas other approaches to the analysis of the public sector tend to end up in combination 11. In order to evaluate this statement we must take a close look at the basic elements of the public choice epistemology.

What Is the Public Choice Approach?

In order to deal with the objection that the public choice approach is biased - a right-wing perspective on the state, as it were - we need to identify the entire enterprise in a way that makes the accusation of an implicit ideological bias an open question. If we start from the identification of the public choice approach as the science of political failure (Buchanan 1988), then there is indeed a risk that the bias is already there from the start. Claiming that politics as a mechanism for the solution to a practical problem is bound to result in disaster is after all a typical commitment in classical liberal ideology as well as in neo-conservatism. Why would public choice necessarily refer to political failure and not to one single case of political success (Amacher et al. 1976)?

The public choice approach is no doubt oriented towards the understanding
the domain of public choice, i.e. politics and bureaucracy. As
Mueller states in his authoritative interpretation:

Public choice can be defined as the economics of nonmarket decisionmaking, or simply the
application of economics to policitcal science. The subject matter of public choice is the
same as that of political science: the theory of the state, voting rules, voter behavior, party

politics, the bureaucracy, and so on. The methodology of public choice is that of economics,
however (Mueller 1979, 1).

The Mueller identification is not only exceptionally clear but also ethically neutral. It does not imply that public choice is the science of the public sector as misfortune. At the same time it, of course, carries no commitment to some naive model of politics as straightforward achievement of objectives in an efficient manner.

The combination of politics and the economic method is usually singled out as characteristic of public choice. In his The Political Economy of Public Choice, Sugden arrives at a slightly different identification by means of a more complex route. He states:

I attempt to integrate two broad themes in economic theory: traditional Paretian welfare
economics . . . and the theory of social choice .... Both of these bodies of theory. I shall
suggest, are to be understood as analyses of the logic of value judgements that may be
made about public choice . . . they analyse the logic of arguments that can be put forward
to justify particular public decisions, or to justify particular procedures for taking public

decisions (Sugden 1981, ix).

Evidently, welfare economics is more narrow than the economic method itself and social choice broader than politics and bureaucracy. The emphasis on value judgements and justification is misleading as it somehow places moral argument at the heart of the public choice approach. Such an identification would make the problem of ethical neutrality troublesome already from the beginning. The Mueller description of the focus and method of public choice is to be preferred.

If public choice is the economic method applied in the analysis of the public sector, then which are the basic epistemological commitments? Two fundamental notions lie at the foundation of all layers of public choice models: methodological individualism and the homo economicus model (Buchanan 1975a; Mueller 1979; Buchanan 1986; Sugden 1981; Tullock 1988). Thus, we have:

(1) Public sector actors behave as if they maximize their own interests.
(2) All social entities are fundamentally sets of individual actors.

These two methodological themes have been discussed in a large literature (Friedman 1953; Kaplan 1964). Suffice it here to focus on whether they entail a problem in relation to scientific objectivity when they are employed in modelling behaviour in the public sector domain. Is it true that methodological and the homo economicus model implicitly entail a right-wing ideological bias?

The Behaviour Postulate

The doctrine of methological individualism in the social sciences requires

that the fundamental units of analysis be identified as individual actors and that various social aggregates or entities be regarded as somehow reducible to actors and their properties. Buchanan states: 'The basic units are choosingunits, behaving persons rather than organic units such as parties, provinces, or nations' (Buchanan 1984, 13). Buchanan combines the doctrineof individualism with the typical 'economic-man' assumption: 'persons seek to maximize their own utilities, and that their own narrowly defined economic well-being is an important component of these utilities' (Buchanan 1984, 13).

Buchanan explicitly recognizes that the tie between methodological individualism the self-interest axiom is not a necessary relationship, but it lays the foundation for the peculiar public choice perspective of looking at the public sector and all its aggregates in terms of individual choice on a par with market behaviour. There must be a fundamental symmetry between public action and private action: . . . the burden of proof should rest with those who suggest that wholly different models of man apply in the political and economic realms of behavior' (Buchanan & Tollison 1984, 13-14).

By combining the doctrine of methodological individualism with the model of self-interest maximizing behaviour we face the most basic problems the structure of the public sector and the motivation of public action. Are political parties or bureaux or nation-states nothing but sums of persons? Is social choice simply a function of self-interest? Granted that the basic motive is self-interest, why does it have to be only economic interests? What about the role of the public interest or ideology?

As Tullock has recently argued, these serious problems have to be faced within the framework of public choice analysis and not be resolved by means of simple fiat placing the burden of argument with those who question these implications (Tullock 1988, 158-170). Yet, Tullock himself very succinctly stated the selfishness axiom without any qualifications as to how social benefits are promoted in modern society:

It is my opinion that most human beings are (except within their families) to a very large
extent interested in fairly narrow selfish goals .... The point I am making is simply that
the resources they are willing to invest in these goals are customarily very much less than
the amount they are willing to invest to reach straightforward selfish goals (Tullock 1970,

33).

The reductionist bias in the public choice approach locating all behaviour motivation with singular individuals whatever their role or position and underlining the narrow economic self-interests as the decisive incentive is bound to result in difficulties when it comes to the understanding of the public sector and its complexity. However, what may be lost in descriptive realism is gained in analytical power. By employing the standard assumptionsof method powerful implications may be derived from a

strong theoretical core. Deductive strength is gained at the expense of descriptive accuracy. Let us look at each of the two components separately, the self-interest maxim iirst, and then, secondly, methodological individualism.My is that it is not these two assumptions/?er se that create the problem of ideology with regard to the public choice approach. Only if we add other assumptions do we end up in the Virginia right-wing corner.

Interest Maximization

In a fundamental methodological sense all political action is behaviour by individuals. In politics, persons tend to be members of various formal organizations, be these political parties, interest groups or public institutions. with membership in large-scale organizations as the basis for political action individuals become spokesmen for the ideology that formal organizations employ to state their rationale. Now, if individuals are the basic units in the public sector, does it follow that the organizations involved in public action are just sets of individual persons? Buchanan seems to take such a stance, for example, in the debate about social justice in society:

As is the case with efficiency, persons are not likely to express interests in abstract
distributional ideals for the society in general when in political decisions. They are likely,
instead, to seek to further their own well-defined interests (Buchanan 1988, 11).

Politicians in a wide sense, including not simply political-party people but also the representatives of organized groups, do make statements about distributional justice and do argue such notions as the reason for various policy demands. Such abstract distributional ideologies may focus on different of justice as need, desert or rights (Miller 1976). As a matter of simple fact it is not correct to deny that political argument is to a large extent about such contesting distributional ideals. However, one may argue that such abstract notions simply reflect narrow self-centred interests; but whose interests? Why does it have to be the 'well-defined interests' of individual persons?

Let me call this theory about public motivation the 'simplicity theory'. It assumes that public action may be fully accounted for in terms of the welldefinedeconomic of individuals. There are two fundamental problems involved here. On the one hand there is the enormous difficulty tied to the doctrine of selfishness as the basic motivational force of persons. On the other hand, we have the collective-action problem, meaning that public wishes are somehow a function of the aggregation of individual wishes about not only their self-interest but also social states in general. I can see no reason why the public choice approach has to commit itself to the simplicity theory, because it faces tremendous difficulties with regard to the distinctions between selfishness and altruism on the one hand and

between economic and other interests on the other hånd, as well as in
relation to the institutional context of preference aggregation.

If we start from the traditional separation between decision-making and implementation in public choice (van den Doel 1978), we may pinpoint the profound difficulties of the simplicity theory. When persons argue publicly for a policy position whether in elections or in Parliament, are they promoting own well-defined economic interests? Presumably, they argue for policy options that are public in the sense that their implementation would concern each and everyone, not just the person arguing the case. If so, are we not forced to make a distinction between self-centred interests and public interests or some similar kind of separation between private interests and social interests? Or take the case of a bureaucrat or a professional doing service in a public programme. When they act in the implementation of policies, are they ipso facto pursuing the well-defined economic interests of their own? Or if they put a lot of effort into improving existing technologies, would they do that only because their own economic interests are at stake?

Some public choice scholars not only recognize the validity of these
distinctions, but they also claim they are vital in the analysis of the domain
of public choice. Let us quote Michael Laver:

. . . the two general families of evaluation criteria that I shall be using relate to the direct
consumption of costs and benefits by the person making the evaluation and to her vicarious
interest in the consumption of others (Laver 1986, 58).

The distinction that Laver aims at is traditionally expressed by means of the selfishness-altruism separation. Any person would thus have two types of evaluations: selfish and vicarious. However, both types of interests are personal in the sense of accruing to the person in question. Yet there is a need for a second distinction, between consumer-controlled actions and spillover effects (Laver 1986, 65 ff.) Thus, besides the two types of private or personal interests - selfish ones and vicarious or altruistic ones - we must take into account interests related to the existence of social costs and benefits. If so, how can we argue that individuals in public choice settings proceed to act on the basis of their own well-defined economic interests? What does it really mean?

The identification of vicarious interests in combination with social costs and benefits may appear as a restriction on the applicability of the public choice approach to the understanding of the domain of public choice. If it is true, as Mueller states, that 'The basic behavioral postulate of public choice, as for economics, is that man is an egoistic, rational, utility maximizer' (Mueller 1979), then there are only two alternatives. Either one must accept that the public choice approach lacks fundamental concepts for entities that are important in politics or one has to redefine the methodologyof

odologyofeconomics so that it takes into account the variety of interests or preferences. In his Selfishness, Altruism and Rationality (1982) Margolis argues for the first solution, though he notes that the scope of the economic model was never strictly limited to competitive market behaviour:

. . . important areas of economics do not have any simple relation to the empirical study of prices and quantities in markets, and among these are a number of topics (welfare economics, decision theory, public goods) with obvious relevance to broader questions of social choice (Margolis 1982, 10).

Yet, in order to extend the economic method to the analysis of the domain
of public choice Margolis states that we need to add new entities besides
the utility-maximizing individual. To quote:

... we have good reason to expect that a viable formal theory of politics needs to extend the traditional model of rational choice in at least three ways: provision of a central role for public goods; explicit treatment of altruistic motivation; and explicit treatment of the role of persuasion (Margolis 1982, 13).

The maximization axiom results in difficulties, because it is not clear which
interests are being maximized by public sector officials. By combining the
distinctions above we arrive at the possibilities outlined in Fig. 2.

The behaviour axiom of the public choice approach may be interpreted in a narrow fashion meaning that it focuses only on combination I (Fig. 2); however, it may also be interpreted in a broader way to cover all the various combinations as long as they somehow relate to the utility of the actor, the self-interest axiom as tautologically true. The main alternative to the simplicity theory would be to re-examine the structure of preferences. When persons act publicly, why do we have to assume that they either pursue their own interests (selfishness) or the interests of others (altruism)? Why could not one be a means to the other?

Obviously, public behaviour is strongly oriented in terms of social goals whether they take the form of altruistic motivations or the target is social cost or collective benefit. When a premier or a president runs the country, or when a parliamentarian promotes a piece of leglislation, or when a bureaucrat administers a system of rules, or when a citizen takes a stand on the issue of nuclear power, then they all pay attention to interests that are broader than their own well-defined interests. When Ronald Reagan asks for more support from the Congress to the Contras in Nicaragua, then

the motive is not his own economic interests; nor could the refusal of the majority of the Congress to go along with this support be explained by their well-defined self-interests. These policy options may be argued because they are believed to enhance abstract liberty or power of institutions or groups, for which in turn the various actors have selfish or altruistic evaluations. Be that as it may, the doctrine that individuals somehow maximize some utility function consisting of their evaluation of personal or social affairs does not imply that they necessarily seek their own well-defined economic interests in each and every activity.

Methodological Individualism

When the simplicity theory about human motivation is combined with a strong version of the doctrine of methodological individualism, then the difficulties become pronounced. Methodological individualism is the theory that social aggregates of whatever form - mere groups or huge formal organizations - are in reality nothing but sets of human beings (Hayek 1945; Popper 1960). Thus, the properties of social aggregates are reducible to the properties of individuals. The reductionist hypothesis about the construction of social reality has caused much debate about the nature of wholes (Nagel 1980). If we accept the starting-point that there could be no other actors than human beings, must we then also negate the possibility that human organization resulting in organized collective action cannot result in emergent properties that are not reducible to individual beings?

The problem of wholism versus reductionism becomes acute as we enter the domain of public choice. Here, the participants in the arena of policymaking policy implementation are organized collectivities of various kinds: institutions, parties, interest organizations, social movements. How about their motivation? What is it that drives these formal organizations to act and do certain things in the public arena?

The simplicity theory, in combination with a strong version of methodological implies that the individuals that occupy the crucial positions in the domain of public choice watch their own well-defined economic interests (Breton 1974). Thus, public officials seek re-election and private remuneration (Downs 1957, 1967); bureaucrats act in order to maximize their salaries, power and prestige (Niskanen 1971). This is not denying that such interests may play a vital role in politics and bureaucracies. is wrong is the extreme simplification meaning that things have been left out: group interests. When organizations - political parties, interest groups, citizens movements - take action, then more than the welldefined interests of individuals is at stake.

Collective action, which is a very prominent feature in the domain of

public choice, is about group interests. And collective interests may range across the spectrum from the very selfish narrow interests of small lobby groups to widely shared public interests. Such public interests in peace, prosperity, fairness and equal treatment under the law may be promoted in the public sector by means of public choice activities. I take the theory of rent-seeking behaviour to be a recognition of the fact that broad social interests may be enhanced by large collectivities (Buchanan et al. 1980, Tollison 1982).

The typical motivation assumption in the public choice perspective creates when public sector behaviour is to be analysed. Yet it may be defended by the argument that there are limits to the explanatory power of any assumptions. It is well to be aware of how far the self-interest assumption works when explaining public action. Whether it works well depends on whether there are other assumptions that may allow for the same amount or more of integration of knowledge in this area. That it cannot explain everything is not the same as that it fails to explain anything. Moreover, the self-interest axiom is not a normative proposition to the extent that actors in the public sector should only pay attention to their own well-defined interests. In order to move to normative public choice theory we need additional assumptions.

The Wicksell State Ideology

The Virginia version of the public choice approach is not just a positive theory about the public sector based on the two fundamental assumptions of utility maximization and methodological individualism. Crucial in the Virginia interpretation is a normative theory about the state that is straightforward ideology of a neo-liberal kind, or what used to be called Manchester liberalism. It is vehemently opposed to the welfare state and its justification in externalities and equity. The foundation of the normative approach is the elements of a state theory proposed in 1896 by Knut Wicksell in A New Principle of Just Taxation, in which Wicksell launched his special decision rule for policy-making: the unanimity rule. In a very original way, Buchanan has employed the unanimity rule as the basis for a contractarian theory of the state. Considering the fact that the Wicksell idea never had any practical consequences - the article was not even available for a wider Swedish audience until 1987 - this is no small accomplishment (Buchanan 1960, 1986).

Buchanan argues that his neo-liberal theory of the state is at the heart of the public choice approach. This is erroneous. The public choice approach has no special link to any normative theory about the state - its proper size or appropriate programme structure. The insistence on a

distinction between the IS arid the OUGHT is not meant as part of any positivistic philosophy of science. It is interesting, meaningful and valid to do both positive and normative analysis of the public sector. However, the public choice approach as a scientific framework cannot possibly be logically commited to the values that the Virginia school wishes to promote (Buchanan1975b,

The Wicksell state theory derived from considerations of efficiency in taxation argued a strong case for a quid-pro-quo rule on an individual basis. Efficiency demands that each citizen can balance the value and cost of a particular item on the state budget at the margin. Equating marginal value with marginal cost may be accomplished by means of the benefit approach to taxation as long as we do not face public goods. In relation to the freerider the only mechanism that will guarantee optimal taxation for public goods provision is the unanimity rule of the individual-veto principle in a legislative context (Musgrave & Peacock 1967).

Buchanan employs the Wicksellian decision principle to derive two normative rules that he considers are constitutive of the public choice approach: politics as exchange and economic constitutionalism or contractarianism the basis of public policy-making (Buchanan 1977, 1986, 1987). Yet these two normative principles are very different from the two basic assumptions of the public choice approach listed above; they are different, because they orient the entire analysis of the public sector in a right-wing ideological direction with logical necessity. Thus, we have two new principles:

(3) Political interaction is to be basedon voluntary exchange.
(4) Politics as voluntary exchange requires the making of an economic

constitution that is to guide the relationship between the state and the

individual.

It is obvious that the basic assumptions of the public choice approach - (1) and (2) above - do not imply or contain the fundamental principles of the Wicksellian state ideology, (3) and (4). This is not to say that one could not adhere to all four axioms without contradiction; the point is only that the axioms be kept separate as (1) and (2) refer to the IS whereas (3) and (4) deal with the OUGHT.

Politics as Exchange

The axiom of politics as exchange means that each and every public policy must be based on the consent of all the citizens, as unanimity is the criterion through which the policy is in the interest of the citizen. This is certainly, as Buchanan emphasizes, a very optimistic interpretation of the nature of

politics as legitimated by the interests of the citizens. However, this is not positive theory but elements of an ethical theory of the state. In the public choice domain, which the public choice approach is aimed at understanding, politics reveals itself in a number of different ways, from coercion over corporate interests formed at the expense of consumer interests to the few cases where broad citizen interests rule.

The unanimity rule introduced by Wicksell may be given two different interpretations between v/hich Buchanan does not distinguish. On the one hand we have the positivist interpretation meaning that politics is exchange between individuals in a setting of political institutions. Politics is not about the definition and implementation of the common good - some ultimate principle that stands above the interests of individual: 'In the absence of individual interest, there is no interest' (Buchanan 1987, 338). Public choice analysis means the unpacking of the notion of the public interest into the interests of the participating individuals. But if it is admitted that politics is not about the common good, then what prevents a public choice approach from modelling politics as coercion or the exploitation of some interests by other interests? Why is politics as symmetrical as market exchange? Buchanan would answer thus:

The observed presence of coercive elements in the activity of the state seems difficult to reconcile with the model of voluntary exchange among individuals. We may, however, ask: Coercion to what purpose? Why must individuals subject themselves to the coercion inherent in collective action? The answer is evident. Individuals acquiesce in the coercion of the state, of politics, only if the ultimate constitutional 'exchange' furthers their interests. Without some model of exchange, no coercion of the individual by the state is consistent with the individualistic value norm upon which a liberal order is grounded (Buchanan 1987, 338).

This is clearly a confusion of the IS and the OUGHT. Individuals acquiesce in coercive politics because to do otherwise is not in accordance with liberal political values. This will not do as a foundation for a public choice analysis of politics as long as the purpose is scientific neutrality and objectivity.

On the other hand we have the normative interpretation, meaning that only those policies which meet with unanimous consent from among the individual citizens can be called just. But why is this so? That unanimity is the criterion on social justice is obvious beyond all reasonable doubt to Buchanan, but why? Wicksell states:

It would seem to be a blatant injustice if someone should be forced to contribute toward
the costs of some activity which does not further his interests or may even be diametrically
opposed to them (Wicksell 1896, 89).

If one accepts the unanimity criterion on efficiency and justice, then naturally Wicksell's conclusion follows. But why do we have to do that? We are here in the realms of ethical argument where there are contending views about what social justice requires. If Wicksell's position is the only

reasonable one, then we would have to do away with all kinds of welfare programmes that can not be derived from unanimous consent - which one would be thus derivable, one may ask? This is Buchanan's conclusion from applying the Wicksell justice or efficiency rule to the existing welfare state, but it is not scientific analysis. It is normative theory, neither more nor less. One may also be reminded of the fact that Wicksell did make a distinction between political justice or efficiency as unanimity and social justice in an ethical sense. The latter could certainly demand that some vested interests be removed in the name of social justice whether or not there be unanimous consent.

Constitutional Economics

The normative framework in the Virginia public choice interpretation becomes very visible in the theory about economic constitutionalism. The theory of normative public policy is to be a theory about the rules of the political decision-making process which could meet with unanimity. Buchanan states:

Existing constitutions, or structures or rules, are the subject of critical scrutiny. The
conjectural question becomes: Could these rules have emerged from agreement by participants
an authentic constitutional convention? (Buchanan 1987, 341).

Since there exists no system of constitutional revisions on a permanent basis one could only speculate about what citizens would agree to, were they in a constitutional setting. Thus, the idea of a constitutional economics is not an analysis of reality but a mechanism for the expression of political criticism (Brennan & Buchanan 1985).

At the same time as the notion of a constitutional economics is a normative one, it is part of a highly restricted normative theory about politics. It not only admonishes the public choice scholar to evaluate existing politico-economic institutions but it also forbids all evaluative approaches except the unanimity rule. We are not allowed according to Buchanan to employ some ethical criteria and evaluate the state:

There is no criterion through which policy may be directly evaluated. An indirect evaluation may be based on some measure of the degree to which the political process facilitates the translation of expressed individual preferences into observed political outcomes. The focus of evaluative attention becomes the process itself, as contrasted with end-state or outcome patterns (Buchanan 1987, 339).

The distinction that Buchanan refers to - outcome criteria or process criteria - is well-known in ethical argument, but why should a public choice scholar have to accept the solution that Buchanan prefers? There is no scientific basis whatsoever for the claim that we cannot evaluate policies

by means of utilitarian or Rawlsian criteria. I do not mean to say that a debate about the grounds for the evaluation of politico-economic regimes is not meaningful or that the arguments pro and contra various solutions cannot be scrutinized. I am only saying that the public choice approach as such cannot commit itself to any one special ethical doctrine about social justice or the proper economic constitution simply because the fundamental problem involved here is the choice of ultimate values, which question the social sciences cannot resolve by means of their special methodology (Brecht 1959).

Conclusion

The public choice approach has been attacked for not being scientifically neutral and objective. It has been argued that its epistemological core includes the acceptance of political values that correctly belong to a special ideology. Thus, the emergent public choice school would mix the IS and the OUGHT to the same extent as the earlier prevailing political economy, i.e. the set of Marxist approaches. Admitting that it is often difficult to distinguish the IS and the OUGHT and that social science theories often are biased, it is still vital to maintain the concepts of scientific neutrality and objectivity as Weber interpreted them. It must be possible to employ the public choice approach for the understanding of the public choice domain without advocating some political ideology.

Public choice theories should stand on their own, supported by means of the ordinary canons employed in the conduct of scientific inquiry. The two peculiar assumptions in the public choice approach - methodological individualism and the self-interest motivation - afford necessary and sufficient criteria. No ethical assumptions need to be resorted to. This is not to deny that ethical argument is important and that public choice scholars could not possibly engage in both activities, adding, for example, politics as exchange and a very special conception of constitutional economics to the typical public choice assumptions. However, it would clarify epistemological matters if the scientific framework is separated explicitly from the ethical framework. You do not have to adhere to the political values that the Virginia public choice school propagates in order to do serious public choice analysis of the public sector domain.

REFERENCES

Amacher. R. C, Tollison, R. D. & Willett, T. D. 1976. The Economic Approach to Public
Policy. Ithaca: Cornell University Press.

Brecht, A. 1959. Political Theory. NewYork: Free Press.

Brennan, G. & Buchanan, J. M. 1980. The Power to Tax. Cambridge: Cambridge University
Press.

Brennan, G. & Buchanan, J. M. 1985. The Reason of Rules. Cambridge: Cambridge University
Press.

Breton, A. 1974. The Economic Teory of Representative Government. London: Macmillan.

Buchanan, J. M. 1960. Fiscal Theory and Political Economy. Chapel Hill: University of North
Carolina Press.

Buchanan, J. M. 1975a. The Limits of Liberty. Chicago: Chicago University Press.

Buchanan, J. M. 1975b. 'Politics without Romance: a Sketch of Positive Public Choice
Theory and its Normative Implications', Wien: Institute of Advanced Studies. Reprinted in
Buchanan and Tollison 1984.

Buchanan, J. M. 1977. Freedom in Constitutional Contract. College Station: Texas A&M
University Press.

Buchanan, J. M. 1978. 'Markets, States and the Extent of Morals', American Economic
Review 68 (2), 364-368.

Buchanan, J. M. 1986. Liberty, Market and State. Brighton: Wheatsheaf.

Buchanan, J. M. 1987. 'The Constitution of Economic Policy', Stockholm: Nobel Foundation:
Reprinted in Gwartney and Wagner 1988.

Buchanan, J. M. 1988. 'Market Failure and Political Failure', Cato Journal, vol 8.

Buchanan, J. M. & Tollison, R. D. eds., 1972. Theory of Public Choice. Ann Arbor:
University of Michigan Press.

Buchanan, J. M. & Tollison, R. D. eds., 1984. The Theory of Public Choice 11. Ann Arbor:
University of Michigan Press.

Buchanan, J. M., Tollison, R. D. & Tullock, G., eds., 1980. Toward a Theory of the Rent-
Seeking Society. College Station: Texas A&M University Press.

Downs, A. 1957. An Economic Theory of Democracy. New York: Harper & Row.

Friedman, M. 1953. Essays in Positive Economics. Chicago: Chicago University Press.

Gwartney, J. D. & Wagner, R. E., eds., 1988. Public Choice and Constitutional Economics.
Greenwich: Jai Press.

Hayek, F. 1945. The Counter-Revolution of Science. New York: Free Press.

Lewin, L. 1990. Self-interest and Public Interests in Western Politics. Oxford: Oxford UP.

Margolis, H. 1982. Selfishness, Altruism and Rationality. Cambridge: Cambridge University
Press.

Miller, D. 1976. Social Justice. Oxford: Clarendon Press.

Mueller, D. 1979. Public Choice. Cambridge: Cambridge University Press. 2nd edition 1989.

Musgrave, R. A. & Peacock, A. T. eds. 1967. Classics in the Theory of Public Finance. New
York: St Martin's Press.

Myrdal, G. 1961. "'Value-locided' Concepts", in Hegeland, H., ed., Money, Growth, and
Methodology and Other Essays in Honor of Johan Åkerman. Lund: Gleerup.

Myrdal, G. 1970. Objectivity in Social Research. London: Duckworth.

Nagel, E. 1961. The Structure of Science. London: Routledge & Kegan Paul. 2nd edition
1980.

Niskanen, W. A. 1971. Bureaucracy and Representative Government. Chicago: Aldine-
Atherton.

Popper, K. R. 1960. Poverty of Historicism. London: Routledge & Kegan Paul.

Sugden, R. 1981. The Political Economy of Public Choice. An Introduction to Welfare
Economics. Oxford: Martin Robertson.

Tollison, R. D. 1982. 'Rent-Seeking: A Survey', Kyklos 35.

Tullock, G. 1965. The Politics of Bureaucracy. Washington: Public Affairs Press.

van den Doel, H. 1978. Democracy and Welfare Economics. Cambridge: Cambridge University

Weber, M. 1923. Gesemmelte Aufsåtze zur Wissenschaftslehre. Tiibingen: Mohr.

Wicksell, K. 1896. Finanztheoretische Untersuchungen. Jena: Gustav Fischer. Some parts are
translated in Musgrave, R. A. & Peacock, 1958, 1967.