In dealing with the theory of public expenditures, two approaches may be taken. We may take a normative view and explore the role that public expenditures should play in an efficiently functioning economy. Or we may examine the sociology or politics of fiscal behavior, explaining the forces that determine actual expenditure policy in the prevailing historical and institutional context. This essay is directed largely to the normative question, attention being given in turn to three major aspects of expenditure policy—the allocation, distribution, and stabilization functions. In conclusion, brief consideration is given to fiscal sociology or politics.
The central issue in the theory of public expenditures, and, indeed, in the theory of public finance, is how to determine the proper level and pattern of public services. Putting it differently, the question is how available resources should be divided between the satisfaction of “private“and of “social“wants. Looked at as an economic problem, this immediately poses a second issue. If resources are to be used for public services or for the satisfaction of social wants, which private services or satisfactions are to be forgone? As public services must be justified in terms of their opportunity cost, the theory of expenditure-making cannot be separated from the theory of expenditure-financing. Basically, the theory of public expenditures and the theory of taxation are but different sides of the same coin.
Attempts at dealing with this problem have a long and frustrating history. Adam Smith, in Book v of The Wealth of Nations, squarely addressed himself to both its aspects. Certain services, so he argued, must be provided by the state, including the upkeep of the sovereign, defense, some phases of education, and certain public works that require too much capital or are too distant in their payoff to be undertaken privately. The finance in turn should be provided through taxes levied so that all will contribute “as nearly as possible in proportion to their respective abilities; that is, in proportion to the revenue which they respectively enjoy under the protection of the state“(Smith 1776, vol. 2, p. 307 in 1910 edition). Ingeniously embedded in a single formula, this rule contains the roots of both the “ability to pay“and the “benefit“theory, the two approaches that were to provide the major strands of future discussion.
The ability-to-pay doctrine became the dominant view among British writers. It was reformulated by J. S. Mill in 1848 in terms of equality of sacrifice, translated by Edgeworth in 1897 into a requirement of least total sacrifice, and refined by Pigou in 1928 into a choice between equality of absolute, proportional, or marginal sacrifice. As a doctrine, it was attractive both to social reformers, who looked upon it as an instrument of income equalization, and to conservatives, who shied away from the more positive view of public expenditures embedded in the benefit approach.
Notwithstanding its wide and sustained popularity, the ability-to-pay approach has severe shortcomings. By its very nature, it deals with taxation only and fails to establish a link between the tax and the expenditure sides of the budget problem. At best, it can determine how a given tax total is to be distributed among various taxpayers, independently of the budget composition. And even this achievement, unsatisfactory as it is because the crucial problem of opportunity cost is neglected, has become increasingly dubious.
The determination of equal sacrifices involves direct interpersonal utility comparisons. Such comparisons, which were the foundation of welfare economics up to the 1930s, were discarded thereafter as lacking operational content. The new formulation, by which there is a gain in total welfare only if A’s position is improved without worsening B’s, overcomes this difficulty but is also much more limited in application. In particular, it is inherently useless in dealing with distributional problems. If this new formulation is accepted, the ability-to-pay concept merely survives as a reflection of what is considered a socially desirable distribution of income; and once this view is taken, there is no reason why the scope of redistribution should be limited to the financing of public expenditures on goods and services, rather than including transfers as well. [SeeWelfare Economics.]
The benefit approach, from the beginning, had the advantage of linking the revenue and expenditure sides of the budget problem. The question, however, is whether these individual gains can be measured, since otherwise the benefit principle lacks operational meaning. A first answer was given by Smith’s hypothesis that the essential service of government is protection and that the value of protection received may be measured in terms of income. This protection version of the benefit approach was subsequently adopted by many Continental writers. In many instances, it provided a vehicle for imposing severe limits on the scope of budget activity, and through most of the nineteenth century the benefit discussion fell far short of its inherent possibilities.
These possibilities, indeed, did not become apparent until the close of the century when the then new economics of marginal utility calculus came to be applied to the problems of public finance.
The participants in this debate included Sax in 1887, Pantaleoni in 1883, Mazzolo in 1890, Viti de Marco in 1928, and many others; but the most important contribution, from the modern point of view, was made by Wicksell (1896). This was followed by the subsequent elaboration of his work by Erik Lindahl in 1910 and 1919 (see Musgrave & Peacock 1958 for excerpts from these writers).
Wicksell began with the assumption that there exists a “proper“distribution of income, holding that otherwise there can be no just determination of tax shares. Voters are then called upon to vote on various budget proposals, including a wide range of tax-expenditure combinations. For an expenditure plan to be passed, it should be paired with a financing plan so that it receives unanimous acceptance or, at least, acceptance by a substantial majority vote. With remarkable intuition, Wicksell thus advanced a solution under which the budget plan results in a gain in welfare, even as judged by the severe standards of modern welfare economics.
It is important to note that the approach of these writers takes a strictly individualistic view of social wants. Social wants, along with private wants, are included in the preference system of the individual, and social no less than private wants should be satisfied in accordance with the sovereign choice of consumers. But if there is no difference in the psychology of private and of social wants, why need the distinction be drawn? And if private wants are satisfied through the market, why cannot the same be done for social wants, i.e., why is a public budget needed?
The answer is that there is an important difference; but it is technical, not psychological. In the extreme case of a purely social want—or a purely social good provided to satisfy such a want —we are dealing with a good whose benefits are altogether “external,“that is, they are not limited to any particular consumer. This has two important implications that explain why social wants cannot be satisfied through the market mechanism but must be provided for through the budget process.
First of all, note that the nature of social goods is such that all consumers are confronted with more or less the same quantity supplied. The benefits of national defense accrue to all citizens, and smog removal helps everyone in the area. While individual consumers may purchase private goods in whatever particular amounts they find suitable, all must purchase more or less the same amount of public goods. Assuming two taxpaying consumers A and B, A’s demand schedule together with the total cost schedule for public services will determine the price at which the service is available to B, and B’s demand schedule will, in turn, determine the price to A. According to Lindahl, the optimal supply of public services and the determination of proper cost shares is arrived at where both demand schedules intersect. This is the point at which both A and B agree on the amount of public services which they prefer, given the underlying cost distribution. Following Bowen’s version (1948) of a similar view, total demand is determined by adding individual demand schedules vertically (as distinct from horizontally for the case of private goods). Output is determined by the intersection of the total demand and supply schedules; and the offer-price indicated by each demand schedule indicates the proper cost shares of the various demands. Thus the determination of expenditure level and tax-price is again seen as proceeding analogously to that of market supply and market price for private goods.
Restating the argument in terms of general welfare theory, Samuelson (1954) defined a social good as one the consumption of which by A does not reduce its availability for consumption by B. Because of this “joint consumption“feature, the condition of Pareto optimality for the social good is shown to differ from that for the private good. Assuming preferences to be revealed, a planner may determine a set of Pareto-optimal solutions, involving various divisions of total output between social and private goods, and distributions of private goods between A and B. While the Lindahl solution gives one such point on the utility frontier, it is not the only one. The choice among the points on the frontier becomes a distributional issue, to be decided in the context of a general social welfare function. At the same time, the Lindahl solution has certain unique characteristics (analogous treatment of social and private goods from the point of view of the consumer) that render it especially attractive.
Second, note that this formulation, although complex enough, greatly oversimplifies the problem. It assumes that true consumer preferences are, or will be, revealed readily through the voting process. This assumption is unrealistic. In the case of private wants, a consumer must pay at the counter if he wishes to purchase the requisite private good; and in competing with other consumers for the supply of such goods, he is forced to reveal his preferences. The market, as it were, functions as an auction system. In the case of social wants the situation differs. A consumer may enjoy the supply of social goods without having to qualify by paying at a counter. This being the case, he is not induced to bid for such services. But cannot this difficulty be overcome by having the budget office issue a bill and enforce payment? Unfortunately, this is no solution. The real difficulty lies not in enforcing payment of taxes once they are properly deter mined, but rather in determining what and how much should be spent, and how much particular individuals should be called upon to contribute.
In order to induce revelation of consumer preferences, a political voting process is needed, parallel in function to the private voting (bidding) process in the market. The complication arises because—unlike the case of market bidding—voters (consumers) cannot be simply assumed to record their true preferences. Since any particular individual cannot be excluded from the enjoyment of social goods, he will be tempted to understate his true preferences. If the number of voters (consumers ) is large, the effect of any one contribution on the total supply of social goods will be negligible. If the number of voters is small, collusion and imperfections (analogous to interference with competition in the private market) will develop. Strategy will enter, and the result will differ from that which would be obtained under the above solution in which true preferences were revealed.
A dilemma arises because the type of voting mechanism (for example, point voting) best designed to give efficient results in the absence of strategy also lends itself most to abuse by strategy; whereas a system less sensitive to strategy (for example, a crude majority vote) does not permit the voter to record the degrees of intensity of his preferences required for an efficient solution. It is thus left to the “art“of politics to establish a workable system of preference determination and tax assessment, where a constructive compromise is reached between these two sides of the problem. Indeed, this is one of the central tasks of government in a democratic society.
The question arises whether the specific characteristic of the social good lies in the “joint consumption“feature or in “non-excludability.“It appears that the former is more basic because application of exclusion would be undesirable for the social good even if it were possible. Since B’s consumption does not interfere with A’s, it would be inefficient to exclude him.
We have seen that, for reasons inherent in the technical nature of social goods, social wants must be provided for through the public budget. To avoid misunderstanding, the expression “provided for“needs brief interpretation. Budgetary provision for social goods does mean that the supply of such goods and the assignment of their cost must be determined through the tax-expenditure process of government. It does not mean that social goods (i.e., goods supplied to meet social wants) must be produced under public management. Highways and planes may be provided for publicly, yet be purchased from private construction or aviation companies. (The market can function in supplying such goods once government appears as a demander. The problem is in establishing government demand.) Similarly, the government may produce private goods and distribute them through sale on the market, as illustrated by the traditional tobacco monopoly. The issue of public versus private production is the essential issue of capitalism versus socialism, but not of fiscal theory. It must not be confused with the fiscal problem of allocating resources between the satisfaction of social wants and the satisfaction of private wants. Indeed, this fiscal problem exists in the confines of a socialist as well as a capitalist system.
We now turn to some major qualifications or objections that may be raised regarding this statement of expenditure theory.
(1) The argument, so far, has been presented as if there existed a clear-cut distinction between two types of goods—some that are purely private (benefits accruing to the owner only) and others that are purely social (benefits being altogether external). Actually, most goods involve both characteristics to varying degrees. It is hard to find budget items, other than defense, that do not involve private-want components. Education expenditures, in particular, are of the mixed type. Similarly, many privately purchased goods involve social want aspects, be they positive (shade trees next door) or negative (smoke nuisance).
This qualification is quite in order, but it is not in conflict with the above approach. Our argument may be readily adapted to the case of mixed goods, so that budgetary provision is made for that part of the cost reflecting benefits of the social-want type, while private payment is made for that part reflecting benefits of the private-want type. The theory of taxation is thus converted into a more general theory of subsidy, the case of the 100 per cent social good (and the 100 per cent tax finance) being reduced to that of a limiting case only, matched at the other end by the 100 per cent private good (and 0 per cent tax finance).
(2) Our formulation bases the distinction between social and private wants on the technical(externality) characteristics of social goods. As far as the psychology of wants is concerned, both private and social wants are considered homogeneous parts of the individual’s preference system, and both are to be satisfied in line with the principle of consumer sovereignty, i.e., in line with the preferences of the individual consumer. In contrast to this view, some writers (especially in the German literature) have argued that social wants should not be interpreted in terms of individual preferences but should be considered as deriving from collective or community needs. The question is how and by whom these community needs are experienced and determined. If an “organic“view of the state is taken, it remains to be shown how the organism reveals its choice. If the expenditure choice and cost allocation are made by a set authority (tribal chief, king, dictator), the problems here posed are much simplified, but the resulting theory of the public sector is hardly satisfactory in a democratic setting. [SeeConsumer Sovereignty.]
An alternative formulation of the criticism, more acceptable to the democratic setting (Colm 1955), may be that the choice should be made through the democratic process, but that the individual (while participating therein) should be taken to be acting in the social interest rather than in his personal interest. This makes sense where matters of national concern, such as defense policy, are at stake. In such cases the social interest will be reflected in individual preferences. But there are many other instances (e.g., highways for private or commercial use) where the good is neither more nor less trivial than is the ordinary type of private good and where the ordinary motivation of consumer choice is altogether appropriate. The mere fact that externalities are attached to certain goods complicates the process of resource allocation; but it establishes no presumption that such goods serve more noble purposes or that basic preferences for such goods must be set on a different moral basis.
(3) Nevertheless there are instances, even in the most democratic of societies, where individual consumer choice is overruled. On the negative side, minors are not permitted to buy guns, and the availability of drugs to adults is restricted as well. On the positive side, certain services, the supply of which could be made subject to purchase payments (for example, education), are supplied free of charge or subsidized because individual outlays thereon are considered insufficient. Or, redistributional measures are taken through payments in kind (subsidized low-cost housing, for example) rather than through cash transfers.
Such infringements on consumer choice -cannot be banned altogether from a democratic society. Constructively used, they can be a helpful aid in the social learning process. This part of the budget process—which I have referred to in another context (Musgrave 1959, p. 13) as the satisfaction of “merit wants“—cannot readily be fitted into the above theory of expenditure determination. While this is a shortcoming, merit wants are the exception rather than the rule; and their existence does not invalidate the usefulness of our model, which covers the major and more manageable part of the budget problem.
(4) The benefits to be derived from particular expenditure projects are frequently difficult to assess, so that the voter (or his representative) needs expert guidance in making the appropriate choices. It is here that the role of the civil servant and budget expert is of great importance.
Recently, this problem has received special attention with regard to outlays on public investment. Any given public investment should be undertaken only if the rate of return exceeds or equals that of an alternative public or private investment. The problem then is how to measure the rate of return on public investment. This involves a number of important issues. One is the determination of the appropriate rate of discounts, the issue on which much of the recent discussion has focused. Another less explored but perhaps more important matter is the development of techniques by which secondary benefits resulting from a particular project may be measured. Finally, there is the problem of developing rules of thumb (e.g., various types of cost-benefit ratios) that can be used for guidance in actual policy making. [SeeInvestment, article onThe Investment Decision; Water Resources.]
(5) In the modern democratic society, budgetary matters are not decided by referendum but in the legislature. This does not mean that the decision is made without relation to the preferences of the individual voter. The legislator acts by representing his constituents and is elected by exhibiting a platform more or less reflecting their preference patterns. By engaging in coalition with other legislators, he serves to establish bundles of issues on which a consensus can be reached. All this is an important part of preference determination and may be looked upon as a means of expediting, rather than displacing, individual choice. Of course the party mechanism, in some of its manifestations, may also act as an interfering factor.
(6) The assumption that all citizens are confronted with the very same supply of social goods needs to be qualified, especially with regard to regional differentials. The regional aspects of expenditure theory (for example, regional differences in benefit incidence or spillover of benefits between regions with different government) cannot be dealt with here, nor can the qualifications needed as the model of a unitary government (implicit in our discussion) is replaced by one of fiscal federalism. [SeeLocal Finance.]
So far, we have dealt with budgetary provision for goods and services designed to satisfy social wants. This “allocation“aspect of the public household, although perhaps the most basic function of the revenue-expenditure process of government, is by no means the only one. Distributional adjustments must also be considered.
If tenets of modern welfare economics are accepted, economics, as previously noted, has nothing to say on the basic issue of income distribution. While the economic analyst may explore the consequences (with regard to the level of output, growth, and other factors) of various changes in distribution, he cannot compare the merits of alternative distributions of a given output. This could be done only if interpersonal utility comparison is admitted and an operational procedure of comparison could be devised.
However this may be, distribution does present a policy issue. This is obvious in the socialist setting, where the return to capital accrues to the state and wages paid need not equal the return to labor as a planning cost. But even in the most capitalistic of countries, distribution is not left entirely to the ownership of factors (labor as well as capital and natural resources) and the market system of factor pricing. Some degree of intervention is held necessary, if only to provide for the indigent. Beyond this, society may consider the unadjusted state of distribution to be less or more equal than is held desirable and choose to make the necessary adjustments. The necessity for some distributional adjustments is thus generally accepted, even though the desired degree of adjustment is highly controversial. [SeePoverty.]
The most direct and efficient tool for carrying out such adjustments as are desired is provided through the tax-transfer process. Use of the tax-transfer instrument is superior to interference with factor pricing, which gives rise to inefficient resource use in the private sector of the economy. It is superior as well to interference with product prices (subsidies) or redistribution in kind (e.g., distribution of free goods rather than cash), since transfers do not interfere with the option of consumer choice. Such at least is the case unless redistributional objectives coincide explicitly with situations where interference with consumer choice is held desirable in order to satisfy merit wants.
Reasoning along these lines, an orderly budget system would provide for two subbudgets. There would be a tax-purchase budget, designed to provide for goods and services needed to satisfy social wants. Also, there would be a tax-transfer budget designed to make such distributional adjustments as are held desirable. Assuming correction in the direction of greater equality, the latter budget would involve progressive taxes combined with regressive transfers. The distribution of taxes in the former budget would depend on the income and price elasticities of demand for public services. The resulting distribution of taxes in the allocation budget could be proportional, progressive, or regressive, depending on the circumstances of the case.
Given these two subbudgets, one may then think of a combined budget, where the two sets of taxes are combined and netted out against transfers. Assuming the tax distribution under the allocation budget to be proportional to income, with tax transfers toward greater equality in the distribution budget, the net budget would show a progressive structure, with transfers at the lower end of the income scale. In this way, the tax-transfer pattern in the net budget may be interpreted in a rational manner, and the vague and unsatisfactory notion of “ability to pay“is replaced by a more clear-cut system of budget determination.
All this is but a way of formalizing the Wicksellian principle of separating the issue of budget policy for redistribution from that of budget policy for the satisfaction of social wants. This separation has great merit for an orderly view of budget policy. At the same time, it must be granted that complete separation is not possible. Merit-want objectives may coincide with distributional purposes, the assumption being (presumably) that low-income consumers are most in need of consumption guidance. Moreover, we have seen that the very nature of social wants does not assure a single best solution to the budget problem and, in a sense, requires a distributional decision to choose among the points on the utility frontier. How may this be squared with the Wicksellian dictum that a “just“allocation of the cost of public services requires a “just“distribution of income?
As has been pointed out by modern welfare economists, a proper state of distribution must be defined in welfare terms. Assume that there are private goods only. A given state of income distribution may then lead to different welfare distributions (different points on the utility frontier) depending on which pricing rule is used. Since there is more than one efficient rule, a meaningful concept of “proper“income distribution must be linked with a specific pricing rule. In the case of private goods, this is supplied conveniently by uniform price-marginal cost pricing in a competitive market. An analogous argument holds if the allocation and distribution problems are separated in the case of social goods. If a given income distribution is said to be proper, it must be specified as such on the assumption that a given pricing rule is followed with regard to social goods. One such rule is given by the Lindahl solution, where the tax price paid by each consumer is such that his marginal rate of substitution of public for private goods must equal the ratio of public goods prices payable by him to private goods prices.
Having considered the allocation and distribution function, it remains to note the stabilization function of budget policy. In a decentralized market system, there is no assurance that the level of aggregate demand may not at times be excessive (inducing inflation) or deficient (inducing unemployment). To provide the necessary corrective, monetary and/or fiscal measures may be needed. Expansionary fiscal action may be taken in various ways, including expenditure increase, tax reduction, and balanced budget expansion.
In the context of a general theory of expenditure determination, the stabilization function is to be performed in such a way as not to interfere with the previously examined allocation and distribution functions. This suggests that expansionary action takes the form of proportional tax reduction (or transfer increase), while restrictive actions take the form of proportional tax increase (or transfer reduction). Putting it differently, the level of government expenditures on goods and services should be so set as to provide for proper allocation of resources under the condition of full resource use, leaving it to the tax-transfer mechanism to secure full employment; and the use of this mechanism should be such that it will not affect the “proper“state of distribution.
It does not follow, however, that the level of goods and service expenditures should never be varied over the cycle. The very factors that make for changes in the level of private expenditures (for example, changes in investment opportunity) may affect (in the opposite direction) the demand for public services; problems of unemployment may be regionally confined and require regionally focused action; public jobs may be more desirable socially than the public dole, and so forth. Thus, countercyclical expenditure policy cannot be ruled out, even though there is a prima facie argument for tax-transfer adjustments.
Recent emphasis in fiscal policy discussion has been upon the growth, rather than the countercyclical, aspects of stabilization policy. This has important bearing on the allocation function of expenditure policy. Given the level of resource utilization, increased growth requires a higher level of capital formation relative to consumption. This means (a) emphasis on public investment, relative to public consumption, in the allocation function of budget policy; and (b) a mixture of fiscal and monetary policy that is inducive to a high rate of private capital formation. Under favorable conditions (a high level of potential investment in the private sector), this may be secured by combining easy availability of credit with a tight (high-tax) budget policy. Under less favorable conditions, expansion of public investment and special (tax or transfer) incentives to private investment may be needed.
Following the procedure suggested for the distribution function, we may now visualize a third budget, addressed to the stabilization function. In securing high employment and price-level stability, and doing so without interference with the other purposes of budget policy, the stabilization budget calls for proportional taxes when restriction of private spending is needed and for transfers when expansion is required. Whereas the distribution budget is always balanced (taxes on A being used to make transfers to B) and the allocation budget is balanced over time (consumption expenditures being tax financed and capital expenditures being amortized over the life of the asset), the stabilization budget by its very nature is either overbalanced or underbalanced.
Accordingly, we may now think of the over-all budget as reflecting a consolidated version of the three subbudgets, reflecting the net patterns of expenditure and tax payments, as well as the state of net balance. While this is the budget that, in effect, comes to be administered, disaggregation is needed to understand the policy issues involved.
The preceding discussion was addressed to the question of how to secure efficient conduct of the public sector in a democratic society. An alternative approach to the theory of public expenditures is to provide an explanation of the actual conduct of expenditure policy.
This approach, which will be noted but briefly, may be referred to as fiscal sociology or fiscal politics. Karl Marx, in the Communist Manifesto, held that capitalism might be undermined by progressive taxation, just as it would be weakened (and this should be added to place his judgment in proper perspective) by popular education. Adolph Wagner predicted a rising ratio of public expenditures to Gross National Product as a law of social and political development. Wicksell foresaw the possibility that a fiscal mechanism guided by majority voting could lead to the exploitation of the rich by the poor, thereby reversing the earlier concern of social reformers. Rudolf Goldscheid presented a sweeping theory of budget behavior, cast in the context of a Marxist view of history. The wealthy state of feudal society, so he argued, gave way to the impoverished modern state, precisely as the ruling class relinquished control over the state to the people (see Musgrave & Peacock 1958 for excerpts from these writers). Joseph Schumpeter (1918), taking off from Goldscheid, re-examined the change of fiscal forms in the transition from feudalism to capitalism, and so forth.
The present writer has stressed the importance of political interaction between the various functions of budget policy as an explanatory as well as distorting factor in budgetary decision making. Emphasis has been placed by Giinter Schmolders (1955) on the psychological basis of taxpayer behavior, and various attempts have been made— including those by Downs (1957), Buchanan and Tullock (1962), and Black (1958)—to explore the consequences of party behavior and majority voting.
Some have argued that the political process tends to result in an excessive level of public expenditures because voters tend to underestimate the cost of public services, expecting it to be paid by the next fellow. Others have held the opposite, whether it is, as suggested by Galbraith (1958), because the need for private services is overrated in response to advertising, because the political mechanism is biased toward inaction rather than action, or because benefits from public services are less urgently required. Moreover, the intrusion of distribution or stabilization considerations into the allocation budget may lead to a distortion of the expenditure level, rendering it unduly high when expansionary action is needed or when the ruling majority considers budget expansion as a means toward increased redistributional gains, and vice versa under opposite conditions. Finally, it has been suggested by Peacock and Wiseman (1961) that the growth of public expenditures is related to the tendency for increases in tax rates during crisis periods such as war to be nonreversible, thus periodically sucking up the level of public outlays for civilian purposes. None of these explanations is altogether convincing; but they suggest that fiscal sociology, or fiscal politics, is a fascinating, if as yet largely unexplored, area of research.
Richard A. Musgrave
Arrow, Kenneth J. 1951 Social Choice and Individual Values. New York: Wiley.
Black, Duncan 1958 The Theory of Committees and Elections. Cambridge Univ. Press.
Bowen, Howard R. 1948 Toward Social Economy. New York: Rinehart.
Buchanan, James M. 1960 Fiscal Theory and Political Economy: Selected Essays. Chapel Hill: Univ. of North Carolina Press.
Buchanan, James M.; and Tullock, Gordon 1962 The Calculus of Consent: Logical Foundations of Constitutional Democracy. Ann Arbor: Univ. of Michigan Press.
Colm, Gerhard 1955 Essays in Public Finance and Fiscal Policy. Oxford Univ. Press.
Downs, Anthony 1957 An Economic Theory of Democracy. New York: Harper.
Eckstein, Otto 1961 A Survey of the Theory of Public Expenditure Criteria. Pages 439–504 in Universities-National Bureau Committee for Economic Research, Public Finances: Needs, Sources, and Utilization. National Bureau of Economic Research, Special Conference Series, No. 12. Princeton Univ. Press. -” This survey includes extensive references to the literature.
Galbraith, John Kenneth 1958 The Affluent Society. Boston: Houghton Mifflin.
Johansen, L. 1963 Some Notes on the Lindahl Theory of Determination of Public Expenditures. International Economic Review 4:346-358.
Margolis, Julius 1955 A Comment on the Pure Theory of Public Expenditures. Review of Economics and Statistics 32:347-349.
Margolis, Julius 1957 Secondary Benefits, External Economies and the Justification of Public Investment. Review of Economics and Statistics 39:284-291.
Musgrave, Richard A. 1959 The Theory of Public Finance: A Study in Public Economy. New York: McGraw-Hill. -* See especially Parts 1 and 2.
Musgrave, Richard A.; and Peacock, Alan T. (editors) 1958 Classics in the Theory of Public Finance. London and New York: Macmillan. -“Contains excerpts from the more important European writers.
Peacock, Alan T.; and Wiseman, Jack 1961 The Growth of Public Expenditure in the United Kingdom. National Bureau of Economic Research, General Series, No. 72. Princeton Univ. Press.
Samuelson, Paul A. 1954 The Pure Theory of Public Expenditures. Review of Economics and Statistics 36: 387-389.
Samuelson, Paul A. 1955 Diagrammatic Exposition of a Theory of Public Expenditures. Review of Economics and Statistics 32:350-356.
Schmolders, Gunter 1955 Finanzpolitik. Berlin: Springer.
Schumpeter, Joseph A. (1918) 1954 The Crisis of the Tax State. International Economic Papers 4:5-38. -” First published as Die Krise des Steuerstaats.
Smith, Adam (1776) 1952 An Inquiry Into the Nature and Causes of the Wealth of Nations. Great Books of the Western World, Vol. 39. Chicago: Encyclopaedia Britannica. -* A two-volume paperback edition was published in 1963 by Irwin.
Wickseix, Knut 1896 Finanztheoretische Untersuchungen nebst Darstellung und Kritik des Steuerwesens Schxvedens. Jena (Germany): Fischer.