Front Page Titles (by Subject) 7.: The Bridge Between Tax and Expenditure in the Fiscal Decision Process - The Collected Works of James M. Buchanan, Vol. 4. Public Finance in Democratic Process: Fiscal Institutions and Individual Choice
Return to Title Page for The Collected Works of James M. Buchanan, Vol. 4. Public Finance in Democratic Process: Fiscal Institutions and Individual Choice
The Online Library of Liberty
A project of Liberty Fund, Inc.
Search this Title:
7.: The Bridge Between Tax and Expenditure in the Fiscal Decision Process - James M. Buchanan, The Collected Works of James M. Buchanan, Vol. 4. Public Finance in Democratic Process: Fiscal Institutions and Individual Choice 
The Collected Works of James M. Buchanan, Vol. 4. Public Finance in Democratic Process: Fiscal Institutions and Individual Choice Foreword by Geoffrey Brennan (Indianapolis: Liberty Fund, 1999).
About Liberty Fund:
Liberty Fund, Inc. is a private, educational foundation established to encourage the study of the ideal of a society of free and responsible individuals.
Foreword © 1999 Liberty Fund, Inc. © 1967, 1987 The University of North Carolina Press.
Fair use statement:
This material is put online to further the educational goals of Liberty Fund, Inc. Unless otherwise stated in the Copyright Information section above, this material may be used freely for educational and academic purposes. It may not be used in any way for profit.
The Bridge Between Tax and Expenditure in the Fiscal Decision Process
To the individual taxes are the “prices,” the “costs,” of the goods and services that the government supplies for his benefit. This conception of the fiscal structure is central to this study, and our procedure has compared the individual’;s behavior in fiscal choice with that in market or private choice. In the market, the individual selects a preferred quantity at a given price per unit, or, alternatively, he allocates a specific outlay to the purchase of a specific good, which, at the given price, results in a determinate quantity being taken. The selection of a preferred physical quantity automatically determines total outlay, or, conversely, the selection of an amount to be spent automatically determines the physical quantity. In either case, the purchaser is assumed to make only one decision. It is absurd to think of his making two separate decisions, one as to the physical quantity of the good to be purchased and the other as to the total outlay to be made. Given the availability of a good or service at an invariant market price, these two decisions reduce to one and the same.
Institutionally, it is possible for the individual to make market purchases in either of these two ways. I may drive my automobile up to a gasoline pump and order five gallons, or I may, alternatively, order $2.00 worth. In the first instance, my outlay is residually determined by my decision on quantity; in the second, the quantity is determined by outlay. If I am faced with some uncertainty as to price, there need be no unique relationship between quantity and outlay at the moment of my decision. If I order five gallons of gasoline, knowing only that the price falls somewhere between thirty and fifty cents per gallon, total outlay may be anything from $1.50 to $2.50. Or, conversely, if I order $2.00 worth in this situation, I may get anything from four to six and two-thirds gallons.
From our earlier discussion, it is evident that individual fiscal choice resembles the price-uncertainty case here. In that discussion, however, we continue to assume that fiscal choice remained analogous to market choice in that only one decision is taken, with residual determination of either tax rate or total outlay (public-goods quantity). Given a specific tax institution, other than that with invariant tax-price, some residual uncertainty must remain in any individual fiscal choice. The group decision may, for example, be stated in terms of a specified expenditure on public education. This will imply, in its turn, a certain rate of tax on local real property, and this rate is assumed to be residually set by the decision on spending. Or, conversely, the specifically chosen mill rate of tax for education implies a certain revenue total available for spending.
Casual observation of actual fiscal processes, at almost any governmental level, suggests that the “bridge” between the tax decision and the spending decision is not nearly so direct as these earlier models have implied. In many instances, the fiscal process appears to embody a double choice; one a choice or decision as to the size of the public spending program and the other a choice or decision on the rates of taxation. Clearly, the institutional setting that allows this apparent splitting of the fiscal decision into two parts can influence the outcome of decision.
The Necessary Real Bridge
In some real sense, there can be only one independent fiscal decision. To simplify discussion, let us again limit consideration to a single public good or service. Observed ex post, there is a specific quantity of this good or service provided. In the collective decision to supply this quantity, economic resources were committed and these were drawn from other potential employments. What these resources could have produced in alternative uses is the real cost of the public goods or services supplied, and this cost is directly related to the number of units. To the extent that resources are employed in the private sector, any decision to spend publicly directly implies “taxation” that is at least equal in magnitude to the money value of the spending, the measure of the value of alternative product.
The notion that the real cost of public goods or services arises from the decision to commit resources, to spend publicly, is not inconsistent with the earlier point that this decision can be made in two ways. A specific tax of X per cent can be levied with the proceeds dedicated to the provision of the public good or service. Or, Y dollars can be appropriated for spending on the good or service, with a tax sufficient to raise this amount being levied, which may be X per cent. But the logical extension of the real-cost principle suggests that taxation, in and of itself, cannot impose a real cost since there is no implication that the revenues collected are to be spent in providing public services.1 If distributional considerations are entirely left out of account, this extension is a valid one. Within the framework of this study, the real-value approach tends to be misleading, however, because attention here is focused on the choice behavior of the individual and not on aggregative results. To the individual taxpayer, or potential taxpayer, who ultimately makes fiscal choices, the imposition of any tax implies that he will, personally, undergo some cost. He will be largely unconcerned with the macro-economic real variables of the economist. To the extent that a tax is imposed, and the funds are not spent, the only institutional means of eliminating distributional elements entirely would be that of refunding the tax revenues, in the same manner as these are received. In this case, however, no decision to tax could really be said to have been made.
Despite the necessary real bridge that must be present for the whole community between any decision to spend publicly and to impose taxes, the individual as a participant in political choice may not consider proposals to spend funds and proposals to raise taxes as directly interdependent. The degree to which he senses the underlying real interdependence will depend partially upon the institutions through which fiscal choices are made.
The Setting for Individual Choice
We have noted that the individual who participates in collective choice can never be placed in a position that is fully analogous to that which he faces in market choice. He cannot confront a one-to-one correspondence between his own choice behavior and a result or outcome. At best, in collective choice the individual can know only that, if a sufficient number of his fellows agree with him, an outcome will follow from a choice. It will be useful to specify carefully that institutional setting under which the individual’;s position with respect to the two-sidedness of the fiscal decision most closely resembles his position in the marketplace. If the group is faced with a decision as to the quantity of a single public good to be provided from the proceeds of a single specified tax, the individual participant should not find it impossible to construct for himself a personalized real bridge between the benefits side and the cost side of the account. In this limiting case, the residual nature of one or the other side will be recognized. This is not, of course, to suggest that the community outcome will tend to be “efficient,” even in such limiting cases. This will depend in part on the nature of the collective decision rules and upon the relative generality of taxes and of benefits among members of the community.2
Let us now examine the more familiar setting in which proposals are made to spend public funds without specifying the tax sources that are to provide these funds, and, conversely, that in which proposals are made to levy taxes without specifying the public-goods and -services mix that is to be financed with revenues raised. How will the individual behave in this situation? Consider first his reaction to a proposal to spend on a public good that promises to yield him some measurable benefits. If he wholly divorces the spendings decision from the tax decision, he will “vote for” an expansion in outlay to the point at which his own marginal evaluation of the good or service becomes zero. Such a conceptual separation of the two sides of the account will tend to be present insofar as the individual considers both the amount and the distribution of taxes to be settled in a decision process wholly apart from the spending decision. Choice behavior of this sort is not so foreign to real-world experience as it might initially appear. There is much discussion, both in the popular press and in quasi-intellectual circles, concerning “needs” for various public services. Almost universally, these “needs” are measured or estimated independently of costs. This “needs” approach to budgeting is based on precisely the model that is here discussed.3
At the same time that the individual “votes for,” or otherwise supports, public spending programs without substantive regard for costs, the same individual, in yet another capacity, may refuse to support any new tax legislation. This half of the extreme independence model is less familiar to everyday experience, because most individuals are normally aware, at least to some vague extent, that they must accept taxes in order to secure the benefits of public goods. Nonetheless, if the individual treats the two decisions as wholly independent, one from the other, he will refuse to vote for any tax legislation.
A somewhat more realistic model is one in which some cost consciousness informs the expenditure decision, while some benefit consciousness informs the separate tax decision, but in which the two sides of the account are differently weighted. Empirical research might reveal isolated instances, but surely cases are few and far between where legislative assemblies have intentionally voted separately for a level of taxes that is more than sufficient to finance the level of spending separately chosen, debt amortization included. The direction of bias seems evident. The splitting of the fiscal decision into two parts tends to cause a “deficit” between approved spending rates and approved tax rates. Insofar as the expenditure decision fails to take into account the cost side, public services provided will tend to be extended beyond that level which fully informed consideration of alternatives would produce. Conversely, insofar as the tax decision fails to incorporate the benefit side, total revenues will fall short of the amount needed to finance that level of public services that an informed consideration of alternatives would provide. In other words, the gap between approved spending and approved taxes, in such a democratic decision model, will tend to “straddle” the unique tax-spending solution that an “efficient” fiscal decision might produce.
It is to be emphasized that the probable gap between approved spending and approved taxation that is discussed here emerges from the choices of a single individual in an institutional setting that splits fiscal choice into two parts. To this point, we have left out of account the interaction of separate individual decisions in producing a group or community outcome. Our central concern is with the individual calculus, and the gap suggested is between the individual’;s preferred level of spending, as this might conceptually be expressed in a voting choice, and his preferred level of taxation, as this might be similarly expressed. Does not the existence of such a gap, regardless of the institutions of choice, reflect irrational behavior on the part of the individual? If he chooses “rationally” should he not “cut through” the possible institutional maze, regardless of complications, and recognize the underlying real interdependence between the separate decisions? To answer these questions, it is necessary to recall the provisional definition of “rational behavior” suggested earlier. To the individual chooser, rational behavior need not reflect full information for the simple reason that the securing of information is a costly process. In any specific choice situation, there is some “optimal” investment in information gathering which seldom, if ever, will result in perfection. The institutions through which choice must be made can evidently affect this level of optimal investment as well as the degree of perfection in results.
It is useful to look at the apparent splitting of the fiscal decision in this context. Suppose that the individual faces a choice as to his preferred level of spending on a single public good or service or on some designated package of services. He is aware, within limits, of the potential benefits that these goods and services will yield to him. He chooses, of course, under conditions of high uncertainty since he cannot know the outcome of the political process, but he probably can make reasonably accurate estimates for his own personalized share in the benefits from incremental changes in the level of public-goods supply. He will sense that these services must be financed, and he may be able to make some translation into tax-cost terms. But this step will clearly require more investment in information than the comparable estimate on the benefit side. The form of the decision process in effect partially solves his information problem on one side but not on the other. The issue is presented to him in public spending terms. He must make his own translation into tax-costs.
Contrast this situation with that where the same individual confronts a choice concerning the level of taxes to be imposed, without connection to spending levels. In this case, the calculus is reversed. The institution of decision itself partially assists the individual in computing tax-costs but wholly obscures public-goods benefits. It is surely plausible to expect that most individuals will behave differently in the two cases, and, as a result, the gap suggested above will tend to emerge. In the one case the individual is reasonably well informed as to benefits, in the other case as to costs.
Closing the Gap
To shift from an analysis of the individual decision calculus to that of the group requires that some consideration be given to the processing of individually expressed desires through a set of political decision rules. The effects discussed here are surely accentuated when it is recognized that each person in his spending decision will hope that the tax-costs will be shifted onto other members of the group, and vice versa for taxing decisions. It is useful to leave this whole question of group decision-making until a later chapter, however, even though some reference to political outcomes seems necessary. To do this, we may simply assume that the individual that we are discussing is the “median” or “representative” man in the many-person community.
In the aggregate, the potential gap between approved levels of spending and approved levels of taxation must be closed. Throughout this chapter, we are assuming that there exists an over-all restriction of budget balance. And lest the discussion here appear overly abstract, it is worth noting that conflicts of the sort mentioned here are familiar occurrences in real-world fiscal systems. Newspapers carry stories of financial “crises” faced by states and local governments; schoolteachers do not get paid; road contracts do not get let. How are these conflicts, actual or potential, resolved, and how can some prediction as to the manner of resolution be made? There seems to be no general direction of effect that is predictable. In the face of a potential excess of spending over tax revenues, will taxes be increased to meet the deficit, or will spending be cut? The outcome will, in each instance, be determined by the stronger set of rules, dictated in part by constitutional provisions. It is the financial conflict that brings to the surface the necessary final interdependence between spending and tax decisions and makes some resolution of the contradiction essential. This will generate a re-evaluation of both choices, and no general pattern of results can be predicted.
We know that political structures, as they operate, do incorporate institutions that tend to produce this apparent splitting of the fiscal decisions into the two parts. These same structures contain, however, other institutions that have been developed to resolve the potential conflict. Historically, legislative bodies, through which the preferences of individual citizens are most directly represented, have exercised more control over revenue or tax decisions than they have over expenditure decisions. In part this asymmetry has its origin in the development of democratic political institutions out of monarchial institutions. Representative bodies, parliaments, first achieved the power to restrict the tax-gathering privileges of the kings. Before taxes could be levied on the people, representative bodies were given the right to grant their approval. No consideration was given to the spending side of the account because public expenses were assumed to benefit primarily the royal court, at least in the early days of constitutional monarchy. Taxes were viewed as necessary charges on the people, but they were not really conceived as any part of an “exchange” process from which the people secured public benefits. It was out of this conception of the fiscal process that both the modern institutions and the modern theory of public finance developed.
The emerging of modern democratic states dramatically modified the setting for the fiscal process, but only recently has attention been paid to the necessity of revising age-old norms. As royal courts came to be replaced by executives, and monarchies by republics, taxes continued to be viewed as necessary to sustain the expenses of “government,” with the burden of these taxes to be minimized to the maximum extent possible. Surprisingly little recognition has been given, even yet, to the idea that taxes must, in the final analysis, be considered as the “costs” of those public goods and services which provide benefits to the same people who pay taxes.
With the development of modern executive structures, the traditional asymmetry has remained, but, partly because of these structures, the decision conflict discussed above has been mitigated. The executive normally exercises greater control over the budget, over the expenditure plans, than it does over the revenue side of the account. Being somewhat less responsible to the desires of individual citizens than the legislature, in a compartmentalized, differentiated sense, the executive utilizes the expenditure budget as a means through which revenue projections and spending projections are reconciled. And in part it was the prevalence of just such conflicts as those discussed which provided the impetus for the development of modern budgetary institutions. These generalizations are relevant largely to the American political structure, and they are somewhat less applicable to genuine parliamentary systems.
Do Governments Adjust Income to Meet Spending Needs?
It was concluded above that no general direction of adjustment could be predicted when the split-decision conflict arises in democratic political structures. This runs contrary to a time-honored notion in public finance. In some of the earlier works especially, the difference between the government account and the private account was emphasized in a manner that suggested one particular resolution of the conflict. “Whereas the individual or family tends to adjust its expenditures to meet its income, the government adjusts its income to meet its expenditure needs.” If this “principle” has any general validity, there is a basic difference in the way in which the individual behaves in family and in public accounting structures. But does the false-analogy notion here have any claim to validity, especially in representative democracy, where, ultimately, all fiscal choices are made by the individual, directly or indirectly, and not by “government,” as some entity wholly divorced from the citizens? The individual in “voting for” public outlays is “spending” in a manner that is analogous to his spending for private goods. His decision to cover this spending with tax revenues, to approve the levy of taxes on himself and others, is made in order to finance these “purchases” from the public sector. Ultimately, the income constraint applies here just as it does in private spending. What the individual does is to adjust his total spending, private and public, to his income, which itself is adjustable only within relatively narrow limits in the normal case. The traditional generalization concerning the relevance of the income constraint in restricting private spending and its irrelevance in restricting public spending is not applicable. There is no reason why tax revenues should necessarily be adjusted to meet approved patterns of public spending, as implied, rather than spending levels adjusted to meet approved levels of tax revenues. The direction of adjustment will surely depend on the particulars of each situation of conflict.
The primary sources of pressure on democratic legislatures, those for reduced taxation on the one hand and for expanded public spending on the other, arise because of the differentiation among groups in the political community. This distributional aspect of fiscal choice has been deliberately neglected here. The purpose of this chapter is to suggest that, quite apart from the intergroup or distributional conflicts that may arise, the organization of the decision-making institutions themselves may be such that the interdependent fiscal accounts are treated as embodying two apparent choices, the results of which may conflict, even in an individual calculus. It is impossible to predict with accuracy the direction of effect that this institutional influence will impose on fiscal outcomes, or to measure its over-all importance. What can be said is that this apparent splitting of decision, insofar as it is present, tends to create greater uncertainty in fiscal choice than seems necessary. In a balanced-budget context, a decision to spend publicly implies a decision to tax, and a decision to tax implies a decision to spend. Only if the actual institutions of fiscal choice are organized in such a way that this basic truism is reflected in the alternatives confronting the individual participant can these uncertainties be minimized. Much of the modern criticism of the United States Congress is directed at its failure to allow simultaneous consideration of expenditure and tax decisions. Differences of opinion may, of course, arise concerning the most appropriate means of introducing the desired symmetry in the fiscal decision process, in repairing the bridge between taxes and spending. But greater rationality in choosing the mix between public and private goods, on the part of the individual citizen as reflected through the legislative processes, depends critically on some correction of inherited error, both intellectual and institutional.
[1. ]This point has been stressed effectively by Earl Rolph. See The Theory of Fiscal Economics (Berkeley: University of California Press, 1954).
[2. ]If both taxes and benefits are general throughout the community, simple majority decision rules will not distort the outcomes. However, if one side or the other should be differentially general, distortion is introduced. If benefits are specific to particular subgroups in the community, while taxes are general over the whole community, there will be a tendency to expand spending beyond “optimal” levels. The supporting analysis for this conclusion is developed in James M. Buchanan and Gordon Tullock, The Calculus of Consent (Ann Arbor: University of Michigan Press, 1962).
[3. ]For effective criticism, see Charles J. Hitch and Roland N. McKean, The Economics of Defense in the Nuclear Age (Cambridge: Harvard University Press, 1960), pp. 46-49.