Public Principles of Public Debt: A Defense and Restatement

James M. Buchanan.
Buchanan, James M.
(1919- )
Display paragraphs in this book containing:
First Pub. Date
Indianapolis, IN: Liberty Fund, Inc.
Pub. Date
Foreword by Geoffrey Brennan.

Chapter 6Internal and External Public Loans


The third bulwark of the new orthodoxy is perhaps the most vulnerable of the whole structure. This consists in the argument that the external public debt is conceptually distinct in essential and important economic respects from the internal debt. The vulnerability of this proposition is especially evident at this stage following, as it does, the demonstration that the other two sides of the new orthodoxy triangle cannot be supported. From a rather straightforward extension of the arguments developed in the preceding chapters it becomes evident that much of the conceptual difference between the two debt forms disappears. The logic by which we shall establish the converse proposition that internal and external debt are equivalent in many of the important respects is somewhat more direct than that involved in our discussion of the other two propositions. In one sense, therefore, this discussion may be considered as a clarification of what has gone before.

"We Owe It to Ourselves"


Following the procedure of the earlier chapters, we may examine carefully the statements made regarding the internal-external debt distinction. We have already cited Lerner in Chapter 2, but it will be useful to repeat a portion of his argument here.

One of the most effective ways of clearing up this most serious of all semantic confusions is to point out that private debt differs from national debt in being external. It is owed by one person to others. That is what makes it burdensome. Because it is interpersonal the proper analogy is not to national debt but to international debt.... There is no external creditor. "We owe it to ourselves."*52


We should also cite Philip E. Taylor:

To the extent that public debt is held by citizens of the debtor government, "we owe it to ourselves." This is equally true of private debt when viewed in the large. If on the other hand the debt is owed to citizens or governments of other societies, payments on the debt represent deductions from national product, and the standards of national welfare are thereby reduced. This does not mean that funds borrowed from abroad are unproductive to the borrowing economy. It means simply that investment of funds borrowed from abroad produces less net return to the borrowing economy than would similar investment of funds provided at home.*53

Or, if we prefer to cite more venerable authority, Pigou will suffice:

It is true that loans raised from foreigners entail a burden represented in the interest and sinking fund on future generations in the borrowing country. But interest and sinking fund on internal loans are merely transfers from one set of people in the country to another set, so that the two sets together—future generations as a whole—are not burdened at all ... it is the present generation that pays.*54


From these statements the implication is clear that, when a community is faced with a choice between the two loan forms, its choice will determine the location of the primary real burden of the debt. If it chooses the external loan, this primary real burden is shifted to "future generations"; if it selects the internal loan, the present generation will bear the sacrifice.*55


In the analysis of Chapter 4 we showed that the primary real burden is shifted forward even in the case of the internal loan. Thus, in one sense, we have already demonstrated that the distinction between the two loan forms is fallacious. However, in spite of the redundancy and repetitiousness of the analysis, it will be useful to apply the methods directly to the internal-external loan distinction. The basic deficiency in methodology which is common to all aspects of the new orthodoxy comes clearest into light here.

Relevant Alternatives


I shall again assume, only for two paragraphs, the role of an advocate of the currently accepted views.


If a debtor community should be faced with two alternative situations identical in all respects save that in one of these the public debt instruments are internally held while in the other such instruments are externally owned, the first of these alternatives would clearly be preferable. This conclusion would follow whether we conduct our analysis in terms of social aggregates or in terms of individual components. The external debt would require that interest payments be made to foreigners, and such payments would represent deductions from income otherwise disposable. The interest payments on the internal debt represent no such deductions. While taxpayers are no better off, there are now bondholder interest recipients, and the interest payments are mere transfers. The external debt is more burdensome to the community, and to the individuals within it, than an internal debt of comparable magnitude.


This remains true regardless of the relative rates of interest which the two loans might carry. On this logic, the external debt carrying an interest payment of no more than 1 per cent would be more burdensome than an internal debt carrying a rate of 10 per cent. The relative rates of interest do not enter into the argument, and, by implication, into the original choice between the two debt forms. So ends the statement of the ruling conception.


The analysis by which the above conclusions are reached exposes the methodological fallacy already discussed in Chapter 3. Alternatives of the sort mentioned cannot exist either in the real or in the conceptually real world. The relevant comparison for meaningful debt theory is not between two situations which are identical in every other respect than debt ownership. Situations like these could never be present, and could never be constructed except as isolated and unimportant cases. Some other respects than debt ownership must be different, and any analysis which overlooks or ignores the other necessary differences must embody serious error.


In order to define properly the relevant, and realizable, alternatives, it is necessary to examine these at the moment of initial decision or choice, that is, at the moment when the creation of the public loan is only one among several alternative actions which the community could adopt. In such a moment the community faces three broadly defined alternatives. First, it may borrow and in this way finance a public expenditure program. Secondly, it may tax currently to finance the expenditure, and, thirdly, it may neither borrow, nor tax, nor spend. Inflation of the currency may be construed as a tax for our purposes here. Within each of these broad categories there are, of course, many possible subalternatives. We are concerned here only with those within the first. We shall assume that the community has made the decision to undertake the expenditure and to finance it by the borrowing method. Financing by means of ordinary taxation or by inflation has been ruled out. The only remaining decision concerns the form which the loan shall take.


Shall the government debt instruments be marketed domestically or shall they be sold in foreign countries?


If the first alternative is chosen, and the loan funds are secured from internal sources, the public expenditure project will be financed out of current domestic savings which presumably could have been, and in this model would have been, invested productively in the private domestic economy. Let us disregard for the time being the whole question of the productivity or the unproductivity of the public expenditure project. The creation of the internal public debt will act so as to reduce the community's privately employed capital stock by the amount of the loan. Future private income streams for the community are correspondingly lowered, with the precise amount of such reductions depending on the rate of return on private capital investment.


In the contrary case in which the community chooses to float the public loan externally rather than internally, the public expenditure will be financed from foreign savings. The incremental addition to domestic privately employed capital stock is not affected by the process of public debt creation. Therefore, as compared with the internal loan situation, the private income stream in subsequent time periods is higher. To this higher income stream there must, of course, be attached some drainage sufficient to allow the external loan to be serviced and eventually amortized.


It cannot be overemphasized that the internal and the external debt cannot legitimately be compared on the assumption of an equivalent gross income stream in the two cases. The gross income of the community in any chosen future time period cannot be thrown into the other respects which are assumed identical in the two cases and thereby neglected. The external debt alternative must allow the community to receive a higher gross income in the future, quite apart from any consideration at all of the productivity of the public expenditure.

Criteria for Choice


Once this simple fact is recognized, the choice between the two debt forms is somewhat more complicated than the new orthodoxy implies. The community must compare one debt form which allows a higher income over future time periods but also involves an external drainage from such income stream with another form which reduces the disposable income over the future but creates no net claims against such income. The choice must hinge on some comparison between the rates at which the required capital sum originally may be borrowed. The choice between the internal and the external loan should, at this level of comparison, depend upon the relative rates at which funds may be secured from the two sources.


The community should be indifferent between the two loan forms if the external borrowing rate is equivalent to the internal borrowing rate, and if we may neglect the frictional or second-order effects of making the interest transfers. This latter neglect obscures important aspects of the problem here, and we shall discuss it at length later, but it is useful to proceed at this stage on the "equal ease of transfer" assumption. We shall temporarily assume that the making of international transfers is no more difficult than the making of internal transfers.


If the two rates are equivalent under these assumptions, the internal loan would reduce domestic private investment which would, in turn, reduce the future income in any one period by an amount indicated by the magnitude of the loan multiplied by the internal rate or net yield on capital, which is assumed to be the rate at which the government borrows. The external loan would not cause such a reduction in private investment; income in a future period would be higher than in the internal loan case by precisely the amount necessary to service the external loan. Net income after all tax payments and interest receipts are included will be equivalent in the two cases.


The analysis is readily extended to the other possible situations. If the internal or domestic productivity of capital investment exceeds the rate at which funds may be borrowed externally, the community will be better off if it chooses the external loan form. Net income after all debt service charges are met will be higher than it would be if the alternative internal public loan were created.


In the third possible case in which the internal rate of return on capital investment falls short of the external borrowing rate, the community will be worse off with the external than with the internal loan. Net income of the community after debt service will be lower, and the external debt will impose a "burden" in a differential sense. But it must be noted that the "burden" imposed by the external debt in this case is no different from that imposed by the internal debt in the converse situation. The differential burden, or pressure, arises, not at all from the locational source of the loan funds, but from the fact that the community has not chosen the most "economic" or "efficient" source. The borrowing operation is not rational in either case, and the differential burden arises from the irrationality in community choice, not from the "externalness" or "internalness" of the debt.

The Transfer Problem


In reaching the above conclusions, we have employed the "equal ease of transfer" assumption. We assumed that the making of internal transfers and international or external transfers are equivalent in effect. This simplifying assumption may appear to remove the fundamental elements of the internal-external debt comparison. That this is not really the case may be illustrated by reference to the problem of state and local debt. Such debts are normally classified as "external." Yet there is no apparent "transfer problem" in the Keynes-Ohlin sense involved in making the interest payments on these obligations. Such interstate and interregional transfers are presumably effected as smoothly as are the purely "internal" transfers required for servicing the internally held national debt. These results stem, of course, from the existence of the common monetary system and the comparatively free resource mobility among the separate regions of the country. For the individual state, the choice between the external and the internal debt should be dictated purely by market criteria.


This illustration should suffice to indicate that the problem of transfer is a second-order one when the distinctions between the external and internal debt are considered. No attempt need be made, and none is made here, to minimize the possible differences in the two cases when the genuinely international debt is compared with the internal debt. But the point is that such differences stem solely from the institutional framework imposed by the separate national monetary and economic structures. There are, of course, differences between the two loan forms when we introduce these institutional problems. But these differences are not the ones which the new orthodoxy has employed in making the external-internal debt distinction. The fact that the external debt involves a drainage out of a domestic income stream is not the reason that the external debt may be burdensome. And this is the clear implication of the new orthodoxy as the earlier citations prove.


Let us now examine the transfer problem in somewhat more detail and see how it might cause us to modify our conclusions reached above. A transfer problem is created by the necessity of servicing either the internal or the external debt. Insofar as the purchase pattern of those taxed to pay the debt interest differs from that of the domestic bondholders in the first case, and from foreigners in the second, some shifting of resources must take place. The issue concerns the differences in the two cases.


Some assumption must be made about exchange rate flexibility. In a world characterized by free flexibility in exchange rates, the international and the internal transfer problems would be substantially identical. For example, if Canada should decide to undertake a large-scale borrowing program, it would make little difference whether the bonds are sold in the United States or at home. The guiding principle in this case should be the comparative rates of interest. Similarly, if the world economy were characterized by some accepted international monetary system with fixed exchange rates, but each economy in the system enjoyed internal price flexibility, the international transfer need be no more difficult than the internal transfer. To be sure, the debtor community will find it necessary to impose domestic deflation in order to surmount the balance-of-payments problems created by the necessity of transferring interest payments abroad. The point to be emphasized here is that the servicing of an internal debt requires that a similar deflationary effect be automatically imposed on the "taxpayer" sector of the domestic economy.*56


It has been advanced as a conceptual possibility that, either under fixed or flexible exchange rates, extreme values for the elasticities of demand for imports and exports could prevent any transfer of income abroad which is consistent with international equilibrium in balances of payment. Much has been made of this possibility in recent years, but its existence may be questioned.*57 Even if the existence of this possibility is accepted, however, it should be noted that similar extreme values for certain elasticity coefficients in the various sectors of the domestic economy could produce similar results in reference to the internal transfer.


The current world economy is not, of course, characterized by either flexible exchange rates or an international monetary standard. Therefore, certain differential problems are created when "international" loans are considered in this setting. The transfer problem in its classical form may arise, and some premium may be placed on the internal loans. This becomes especially true when the borrowing country obligates itself to service its debt in some international unit of account. In this case, the borrowing or debtor country no longer can retain control over the amount of real income transfer necessitated by the debt service. Action taken in foreign creditor countries can modify the size of the real income transfer.


The point to be made here is not that of minimizing the importance of the transfer problem. Rather it is that of stating that differences in the transfer difficulties provide the only valid reason for making a sharp conceptual distinction between the two debt forms. At the more fundamental level of comparison with which we are concerned here, this transfer difference is an adjustment factor only. It should not be allowed to obscure the essential truth, namely, that in basic respects the internal loan and the external loan are identical.

Sources of Error


What are the sources of the fallacious idea that the external loan and the internal loan differ in fundamental respects? There appear to be two. The first source of the error is found in the general assumption of the classical economists that all public expenditure is unproductive. If this assumption were true, then the external loan would always make future generations worse off than they would be without the loan. But again the relevant alternative must be considered. Similar conclusions would follow when the internal loan is considered. But the failure to see this may have been based on an oversight of the proper capital-income relationship. With the internal loan, the present value of the community's future income stream is directly reduced because of the direct usage of a portion of its current capital stock in an unproductive manner. This present value is not affected, in the aggregate, by the future interest charges which take the form of internal transfers. But with the external loan, the net present value of the community's income stream is reduced, not by any current using-up of resources, but by the necessity of making the future interest payments. The gross value must be adjusted downward by the present value of the interest payments. This adjustment may not be made explicitly, and, therefore, the external loan may appear to carry with it a greater burden.


The second source of the error lies in the failure of the new economics to make a distinction between the real and the monetary aspects of public debt. The "new" approach to the public debt was related directly to the budgetary aspects of the new economics. Discussion was conducted almost wholly in terms of monetary debt obligations, and little attempt was made to separate the real and the monetary sides of the problem. This aspect of debt theory will be thoroughly discussed in Chapter 9 when we introduce the Keynesian assumptions.

Notes for this chapter

Abba P. Lerner, "The Burden of the National Debt," in Lloyd A. Metzler et al., Income, Employment and Public Policy (New York, 1948), p. 256.
Philip E. Taylor, The Economics of Public Finance (Rev. ed.; New York, 1953), p. 187. Cited by permission, The Macmillan Company.
A. C. Pigou, A Study in Public Finance (3d ed.; London, 1949), p. 30.
For example, Murphy, after arguing that the real costs of war cannot be shifted in time, continues: "There are some exceptions to this generalization. The cost of a war might be financed in part from foreign loans and these loans repaid out of the proceeds of tomorrow's labor." (Henry C. Murphy, The National Debt in War and Transition [New York, 1950], p. 80.)
Cf. August Lösch, "A New Theory of International Trade," International Economic Papers, No. 6, pp. 50-65.
See Egon Sohmen, "Demand Elasticities and Foreign Exchange Rates," Journal of Political Economy, LXV (1957), 431-36.

End of Notes

Return to top