firms could be prejudicial to the position of banks as intermediaries holding intangible assets. The double-taxation argument against the taxation of intangibles is found to have little merit, given that property taxation is rationalized on benefit grounds. Because withdrawal of the present immunity possessed by bank-held intangibles could open the door to a wide variety of possible responses by the States— with some electing not to tax intangibles, some electing to tax at preferentially low rates and others choosing to tax at relatively high general property levels-the resulting uncertainty could have profoundly unsettling effects on the banking community and on the process of financial intermediation. It is this last argument that probably is most persuasive in supporting the case for continuation of the present immunity. CHICAGO, March 22, 1971. APPENDIX 9 Comparing State and Local Taxation of Banks and Other Business Enterprises HARVEY E. BRAZER Chairman, Department of Economics, University of Michigan, INTRODUCTION Inter-industry, inter-corporate, and inter-personal comparisons of state and local taxes paid have long been the subject of much interest. They have taken on rather special significance in the case of the banking industry, however, because of the restrictions that the Congress has placed upon the states and their local subdivisions in the taxation. of National banks. Section 5219 of the Revised Statutes (12 U.S.C. 548) and judicial interpretations of it, in fact, frequently have required comparisons, in order to ascertain whether or not the banks have been taxed more heavily relative to "other moneyed capital" and non-financial corporations than the federal statutes permitted. Comparisons have been inspired as well by interest in the question as to how much revenue, if any, is being lost by taxing jurisdictions and saved by banks as a consequence of the constraints imposed by Section 5219 and the extension by the states of similar treatment to state-chartered banks. Moreover, many of the states employ tax rates under income taxes, or taxes measured by net income, that are designed to be sufficiently higher for banks than for other corporations to "equalize tax burdens." At first glance it would appear that tax burden comparisons for banks present no real problems other than those involved in obtaining the desired data, even when it is recognized that they must be seen in the context of the issues involved in comparing tax burdens generally. The slightest reflection, however, soon reveals the existence of some very serious difficulties. These difficulties may be illuminated more readily if, at the outset, we avoid the complexities. introduced by the corporate entity and examine the case of interpersonal tax burden comparisons. Furthermore, recognition of the role of individuals or households as taxpayers should sharpen understanding of and give meaning to the notion of business or corporate tax burden. INTERPERSONAL COMPARISONS We are interested, say, in comparing taxes paid by rich people with taxes paid by poor people, or taxes paid by residents of State A with those paid by residents of State B, or large families versus small 1 The word "burden" is used in this paper for convenience in expression merely as a synonym for "amount paid" or "liability"; it is not meant to be read as implying that taxes are necessarily any more (or le "burdensome" than other charges or prices paid by the individual or corporate taxpayer. families. Clearly the possibilities for drawing comparisons are abundant. What is not so clear is how we ought to go about developing the necessary commensurable numbers. Bases for Comparison Commensurability requires that we relate taxes paid by each of our units of observation to something that will serve, to our satisfaction (even if not to everyone's satisfaction), as a standard of comparison. That is, we are not interested merely in knowing how much in taxes is paid by each of any given set of individuals. Rather, our comparison, to be meaningful, must relate taxes paid to income, wealth, expenditure, perceived value of public goods "consumed," or whatever we choose to designate as the relevant (perhaps "appropriate") base or "norm." Thus if income is selected this implies that we believe that, or at any rate accept the idea that, income defines the individual's role as taxpayer. We should wish, then, to express taxes paid as a percentage of income, and equal, more than, and less than are defined in terms of this ratio. But which of the alternative bases or norms of taxation are we to select? None is self-evidently better than any other until we specify the criterion or criteria by which we are to evaluate the tax system,2 and even then doubts may reasonably be entertained. If we select income as our base we may do so because we believe that taxes should be apportioned among individuals, families, or households according to ability to pay and that this ability is some function of income. But one may also assert, quite plausibly, that ability to pay is more accurately a function of expenditure or of net worth. Irrespective of this choice, however, the notion of ability to pay is entirely subjective in nature and its use implies interpersonal comparisons of utility. One can only, therefore, select income or wealth or expenditure, or some such measure, as an objective proxy for ability to pay and hope that it is a good proxy. What have we then accomplished? As a criterion for tax distribution ability to pay appears to involve nothing more than the use of a highsounding phrase having some persuasive moral force to support one's position with respect to that distribution. Far more straightforward, meaningful, and therefore useful, is the approach that simply states a given objective to be achieved through allocation of tax shares and then selects the tax base best suited to the attainment of that objective. Thus, if it is our objective to use the tax system as an instrument for redistributing income toward greater equality, it follows that whatever the nature of the actual taxes levied, interpersonal tax comparisons must relate taxes paid to income. Then, when we compare taxes as percentage of income among individuals or groups of individuals we may draw conclusions about relative tax levels and about the gap between our objectives and accomplishments. The same exercise might be carried out with respect to other tax bases and policy objectives. 2 For extensive discussions of the issues surrounding the choice of tax base see Richard Goode, The Individual Income Tax (Washington, D.C.: The Brookings Institution, 1964), chs. II and III, and the references cited therein. 3 It may also be argued that demand for public goods and services is a function of income and that the income base for taxation is appropriate to a benefit as well as an ability basis for apportioning tax contributions. The two functions may be expected, of course, to be very different. To this point we have simply opted for the use of an objectively measurable base for interpersonal tax comparisons. Unfortunately, however, "objective" and "measurable" can only be used as relative terms. We cannot, for example, find agreement among economists, attorneys, and accountants as to how properly to define income, let alone measure it once defined. The definitional questions may differ somewhat in substance but they are not avoided by shifting to bases other than income, and the shift will, of course, raise new questions. Measuring Taxes Paid Moreover, the difficulties encountered in defining and measuring the base for use in tax comparisons are joined by truly formidable ones involved in measuring the taxes paid by the individual, family, or household. What is meant by the term "paid by"? Surely this cannot mean simply the tax payments made directly to the various taxing jurisdictions. Quite obviously what we really want to measure is the dollar amount of taxes "borne." By "borne" we mean paid in the sense that the tax incidence is such as to reduce that individual's or unit's income. Thus, of course, taxes an individual actually pays to a governmental jurisdiction may be shifted elsewhere and not "really paid" by him at all, and taxes of which the individual may be quite unaware may, in fact, be borne by him. We are therefore confronted with the thorny subject of tax incidence. Taxes get shifted from those who are initially responsible for paying them to those on whom their incidence falls through the complex process by which the imposition of taxes gives rise to changes in the prices of factors, products, and assets. Our knowledge of the way in which the relevant interactions occur and how individuals and families are affected as suppliers of labor and capital, as consumers, and as asset holders is, at best, sufficient to permit only rough approximations of the answer to the question as to how much in taxes is borne by them.1 Another difficult, but quantitatively less important, problem relates to the definition of "taxes" to be included. Where the charge or tax is in the nature of a price paid in exchange for a specified good, service, or present or future claim, value is received for the amount paid and there is no more reason to count it among "taxes" paid than there is for counting other prices, public or private, paid for goods, services, or claims of whatever kind. Social security taxes, for example, are "contributions" for social insurance, and there is a rough correspondence between the amount paid by and on behalf of a given individual and the value of the benefits he may realize. But the correspondence is only rough and, in greater or lesser degree, all taxes may be said to bear some relation to public service benefits of one kind or another. There is no clean line between taxes "pure and simple" and other governmentally imposed charges. At most we can visualize a continuum along which we may move from one extreme, at which there is a close relation between the amount paid and benefits realized, as in the case of a charge for water service or admission to a public swimming pool, through social security taxes, user taxes or charges like motor fuel For extensive discussion of incidence theory see Richard A. Musgrave, The Theory of Public Finance (New York: McGraw-Hill Book Co., 1959), chs. 10, 15, and 16. taxes, to general sales and income taxes. For purposes of comparing tax burdens we can only select rather arbitrarily a point on the continuum at which we draw the line between taxes and other charges; there is no demonstrably "correct" location for it. In short, we recognize the obstacles in the way of developing interpersonal comparisons of tax burdens, yet acknowledge that such comparisons serve useful and highly relevant public policy purposes. Just as it makes sense as an economic-political choice to tax people according to income, or wealth, or expenditures, so it makes at least as good sense to attempt to measure the relationship between taxes borne by people under a diversified tax system and the base or criterion selected. INTERCORPORATE COMPARISONS Can we, now, extend this conclusion with respect to individuals or families or "people" to banks and, more generally, corporate enterprises and corporations grouped by industrial classification? At one level of response the answer is "no." for all taxes, whether initially paid by individuals or corporate enterprises, have been assigned, through incidence analysis, to individuals. There is, in consequence, nothing left to assign to corporations and hence no intercorporate tax comparisons are possible. Taxes paid by corporations are borne by individuals in their capacities as suppliers of capital, labor, or other inputs, or as consumers of corporate products. Thus it does not make sense to speak of corporate measures or indices of "ability to pay;" the analogy with individuals simply does not hold. Ronald B. Welch put it well many years ago when he said, “. . . ability is an attribute of animate beings, and with all the sovereignty of the State it cannot be injected into the fictitious personality of a corporation." 5 Corporate Tax Incidence and Tax Burdens It must be quickly pointed out, however, that it does not follow that taxes imposed on corporations "do not matter" because they are borne by individuals. They do indeed matter for, as we have noted, in the process of tax shifting relative prices change and the allocation of economic resources is, therefore, different from what it would have been under alternative tax schemes. Thus one may expect that when taxes imposed on industry A are raised relative to those imposed on industry B resources will flow from the more heavily to the less heavily taxed industry, relative prices will follow the direction of tax change, and the new consumption mix will favor the output of the more lightly taxed industry. The nature and extent of the consequences of relative tax changes will depend, of course, on the structure of factor and product markets, on technology, and on the kinds of taxes involved. The main point to be made here is that relative tax loads and changes in them induce important consequences which may be regarded as the intermediate steps or effects giving rise to ultimate incidence changes among individuals. 5 Ronald B. Welch, State and Local Taxation of Banks in the United States (Albany, N. Y.: State of New York, Special Report of the State Tax Commission, 1934), p. 146. 6 This is not the place to develop in detail and with analytical rigor the analysis underlying these assumptions or assertions. See Arnold C. Harberger, “The Incidence of the Corporation Income Tax," Journal of Political Economy, June, 1962. |