Public Principles of Public Debt: A Defense and Restatement
The Analogy: True or False?
In this chapter I shall examine the second bulwark of the new orthodoxy, the false analogy between the public debt and private debt. The analytical model will be the same as that employed in the preceding discussion; this model essentially embodies the "classical" assumptions. I shall limit the discussion here to the internal public debt since it is almost universally acknowledged that the public debt-private debt analogy is more appropriate when external public debt is considered. Chapter 6 will be devoted explicitly to the internal debt-external debt comparison.
The Basis of the False Analogy
Again we may initially seek to determine precisely what the advocates of the new orthodoxy mean when the falsity of the analogy is proclaimed. This in itself is a difficult task; clarity is not one of the characteristic features of the literature in this field. As was suggested earlier, the clearest statements are likely to be found in the works of the more extreme proponents of the new approach. Pedersen states:
The state does not obtain the power of disposal over additional funds, for those funds were already within the realm of its power, and might, in fact, have been obtained through taxation.
Thus an internal loan raised by the state is not really a loan in the ordinary sense, since it possesses none of the essential characteristics of such a transaction.... An internal loan resembles ordinary borrowing only in a purely formal way, and it is obvious that every analogy to private borrowing must be completely false.*45 (Italics supplied.)
As a second example we may cite Nevins' more recent work:
... internal borrowing by a government in no sense represents an addition to the wealth of a society, so interest and capital payments by a government in no sense represent a reduction in the disposable wealth of the society.... A private person may live beyond his income and encroach upon the wealth available to his descendants, but since—apart from the international transfers which are excluded from the present context—a society cannot live beyond its means through the mere creation of monetary debt, such debt cannot constitute a drain on the resources available to future generations.*46 (Italics mine.)
Several points are immediately evident from a first glance at the above statements. The first is that the false analogy argument is not independent of either the future burden argument or the internal-external debt argument, as we have previously noted. Any discussion of one of the three basic propositions must involve some consideration of the other two. Hence, to discuss each of the three propositions separately must introduce considerable redundancy. The second point to be made is that the emphasis is placed on the effects on the national balance sheet. The third point which strikes the reader is the apparent simplicity, and, by inference, the obvious validity of the analysis. This simplicity is important because it serves to mask errors which reasonable scholars would otherwise never have committed.
The Factual Statements
When an internal debt is created, resources for public use are withdrawn from private uses within the economy. Therefore, the creation of debt and the correspondent financing of the public project does nothing toward increasing or adding to the wealth of the society. This is, of course, fundamentally correct as a first approximation and requires no difficult reasoning for its comprehension.*47
The Inferences from the Facts
The new orthodoxy draws incorrect inferences from correct factual statements. Although it is not made explicit in the statement cited above, the inference is clear that the individual (or the public entity which borrows externally) is able, in some way, to increase his wealth by a borrowing operation.*48 The beginning sophomore in accounting can see through the fallacy contained in this inference. The asset secured is precisely offset by the liability incurred (the debt). The balance sheet of the borrower is not affected otherwise. No change in net worth occurs in the objective sense, if we may assume that the borrower does not capitalize favorable expectations immediately. The case is even more clear for the lender. He will write down his one asset item, let us say, cash, which he uses to purchase the private security, to extend the private loan. He will write up another asset item, let us say, loans receivable. No change in net worth takes place. To be sure the borrower secures the command over a greater quantity of resources currently, but this does not affect his objectively determined wealth position. The lender, on the other hand, gives up some current command over resources, but this, in itself, does not reduce his net worth. This command over current resources is replaced with a claim on future resources which carries a present value.
We may now proceed to examine more carefully the accounting aspects of an internal or domestic public borrowing operation. Let us initially assume that the rate of interest to be paid on government securities is equal to the rate of return on private investments made at the margin. This amounts to assuming that the operation is a marginal one for the bond purchaser. Let us further assume that the public project is of such a nature that individuals do not take its value into account when making up their individual balance sheets. We shall also assume that these same individuals do not capitalize future tax payments. These last two assumptions appear realistic enough for most cases of public borrowing.
When the bond purchaser buys the government bond, he draws down some other asset, let us say cash or private bonds, and replaces this with the government securities. No change in his net worth takes place. By assumption, the taxpayer neither includes the capital value of the public asset nor the discounted value of future tax payments in his balance sheet. Therefore, in the aggregate, individual balance sheets are not affected by the public debt creation. We must conclude, therefore, that, for the period when the public debt is initially created, there is no difference between the internal public debt and private debt. There is no fallacy in the analogy to this point.
When the periods subsequent to this initial one are considered, the analogy would seem to be on more shaky ground. How can the payment of interest on a public debt, in itself, represent a reduction in aggregate individual net wealth since domestic bondholders receive the interest paid as taxes by domestic taxpayers? Yet clearly the payment of interest on a private debt represents a drainage from the real income stream of the individual, a reduction in his net worth.
This comparison must be examined more carefully. How are individual balance sheets affected by the payment of interest on a public debt? We shall continue to make the same assumptions about the nature of the project. The payment of taxes, say out of cash, will reduce the asset side of the taxpayer's account. This will be offset by a reduction in net worth on the right side. The receipt of interest will increase the cash position of the bondholder and, on the right side of his account, net worth will be increased. This increase in net worth of the bondholder just offsets the decrease in net worth of the taxpayer. No change in individual net worth, considered in the aggregate, takes place. The analogy with private debt appears to be false.
There is, however, a subtle fallacy hidden in the above reasoning, and it is based on the methodological error against which we warned in an earlier chapter. The analysis reflects a failure to compare relevant alternatives. This may be most easily shown by examining the implications of the conclusions reached above. According to this analysis, no change in aggregate wealth occurs either when the debt is created or when the interest is paid. From this it should follow that, if the public project yields any positive real income at all, the society achieves a net gain in real wealth as a result of the combined debt issue-public expenditure operation. Any rate of return on the public investment greater than zero would be sufficient to justify public investment. A public policy of ever-expanding public borrowing would seem to be indicated. These implications apparently run counter to those reached in Chapter 4, where it was shown that the "productivity" of the public project must be compared with the costs imposed on "future" taxpayers.
There is something obviously wrong here which must be searched out and corrected. Let us recall the assumptions of our model. Resources are fully employed; in order to utilize resources in public employment, some rate of return in private employment must be sacrificed. Therefore, if the project is completely wasteful, the sacrifice in private wealth is not offset by any gain in "social" wealth. But why does this fail to show up when we consider the individual balance sheet adjustments as we have done above?
The answer lies in the assumption that the net worth of the bondholder is uniquely increased by the receipt of interest on the government security. This reflects an oversight of the fact that capital investment has alternative employments in the private economy, and that some increase in net worth would accrue to individuals who are net creditors in the absence of the public borrowing operation. Differentially speaking, it is not proper to offset the increase in net worth of bondholders which is occasioned by the interest receipts against the decrease in net worth of the taxpayers occasioned by the interest payments. The increase in net worth of the bondholders would have occurred without the public debt; only the decrease in net worth of the taxpayers may be attributed to the fiscal operation under consideration. This modification now allows sensible conclusions to be drawn regarding the merits of the debt-public expenditure operation. The differential decrease in net worth imposed upon future taxpayers by the servicing of the debt may be offset against the rate of return on the public project which is debt financed. Quite clearly the project which yields a zero rate of return is not justifiable.
This result is not changed if we modify the assumptions and allow that individuals include both the present capitalized value of the public project and the capitalized value of future tax obligations in their balance sheets. These two are offsetting items, as may readily be demonstrated. Assume now that individuals fully value the public investment project, but that they also recognize the present value of the future tax liability to which they will be subject. The balance sheets of individuals will contain an additional asset item and an additional liability item. Individuals' net worths are not changed by this modification except insofar as these two items differ. But if the present value of future tax payments is fully incorporated into the balance sheet, we cannot then also say that the actual payment of these taxes represents a reduction in individual net worth. The individual will have discounted the necessity for paying taxes; that is, he will have become a Ricardian man. We cannot offset any additional reduction in net worth in periods subsequent to the initial one against the increase in net worth enjoyed by the bondholder who receives interest payments.
The objection may be raised at this point that the bondholder also capitalizes his interest receipts, in fact, that only this process provides a current capital value for the bond. Therefore, it may be objected, if full capitalization of the liability item prevents an interest outpayment from reducing debtor net worth, will not the full capitalization of the asset item prevent the interest receipt from increasing creditor net worth? The answer is no. The debtor-creditor sides of the account here are not symmetrical. The asymmetry stems from the fact that an individual or institution holding a net asset position, that is, who has positive net worth, is in a position to increase his net worth over time. If accounting practice dictates that the individual creditor carry the interest item on an accrual basis, the actual receipt of interest payments will not increase net worth. But net worth between the beginning and the end of the accounting period will have increased by the amount of the interest payment in this case and this will be shown on the asset side of the balance sheet as "accrued interest not yet received." The assumption of full discounting does not modify the conclusion reached. The reduction in net worth of the taxpayer must be offset only against the current value of the public project, not against the current value of the asset held by the lender.
A simple example will serve to demonstrate the whole analysis, although it will perhaps contain some repetition. Let us specifically assume that the public expenditure which is debt financed is completely wasteful, let us say, the funds are used to provide air-conditioning units for the natives of Attu Island. We assume no capitalization of either the future benefits or the future taxes. We have shown that, under these assumptions, no reduction in aggregate individual net worth takes place when the debt is created and the expenditure made. In subsequent periods, the increase in the net worth of the bondholders is just offset by the decrease in the net worth of the taxpayers. Both groups live within the confines of the same economy; therefore the community, taken as a whole, appears to be no richer or poorer by the operation.
The analysis is rescued only when the positions of taxpayers and bondholders are compared in situations with and without the debt, the only comparison which is meaningful and useful for policy purposes. Let us say that the only alternative to debt issue is expenditure contraction; taxation, either directly or through currency inflation, is not possible. If the debt is not contracted, the public expenditure project will not be undertaken. The natives on Attu must do without air conditioners. As a further simplifying assumption, we say that the sacrificed rate of return on private investments is just matched by the return on government securities.
We may now examine carefully the positions of the individual bond purchaser-holder under each of these two possible situations. If the public debt is issued, the individual will transform a private security into a government bond. And while isolated individuals may transform liquid assets such as cash into securities, unless the debt issue generates net dishoarding, the over-all result must be represented by a transformation of private investment funds, actual or potential, into government securities. No change in the net worth of the bond purchaser takes place immediately. When he receives interest on the security, or when interest accrues, his net worth increases. This increase is not greater than the increase in net worth which would have occurred had he devoted his funds initially to private rather than public uses. His position is identical in the two situations. Relaxing our simplifying assumptions, his position will be different only insofar as the interest paid on public debt differs from that which he could have earned on private investment after these comparative employments are adjusted for differences in risk premiums. In the differential sense, the bondholders' net worth is not increased by debt issue, per se.
The story for the taxpayer is, however, different. As we have shown, he undergoes no reduction or increase in his net worth during the period in which the public loan is floated, except insofar as future tax payments are capitalized and reflected in capital values of assets currently held. (We have assumed in this example that such capitalization does not take place.) When interest is paid in subsequent periods, the net worth of the taxpayer is reduced by the amount of the service charge on the debt. If, on the other hand, the public debt is not issued, and the public expenditure abandoned, the taxpayer presumably undergoes no change in his private net worth at either the time of the possible debt creation or in subsequent periods. Thus, we conclude that if the analysis is properly carried out, the public debt does reduce the net worth of the borrower (the taxpayer) during the periods in which interest must be paid, and this reduction is differential in that it takes place only if the debt alternative is followed. Of course, for the taxpayer-borrower, this is only one half of the whole fiscal transaction. His final position will depend, in the debt model, upon the productivity of the debt-financed public project relative to the tax-caused reduction in net worth. If the public funds are spent to provide air conditioners for Attu as we have assumed in this example, the taxpayer will find his net worth reduced when interest payments come due, even when both sides of the transaction are fully taken into account, reduced below what such net worth would have been if the original operation had never been undertaken. If, on the other hand, the public funds are used to ward off disaster, that is, to finance a war, the taxpayer will find his net worth increased on balance in spite of the necessity to service the debt. This indicates that the concept of "productivity" must be used broadly and carefully in this context. If the public funds are used to finance a genuine investment project, the real income attributable to that project may more than offset the reduction in net worth generated by the necessity to pay interest. The basic point to be made in all this is simply that the reduction in net worth involved in making the tax payments must be offset against the productivity of the public project which is financed by the debt. It is improper and misleading to offset this reduction in net worth against the increase in net worth enjoyed by the bondholders. The latter increase must take place regardless of the productivity of the public investment.
From this corrected analysis we may conclude that the public borrower (that is, the taxpayer) is at no time in a position different from the private borrower. The analogy between the two holds good in all of the essential respects. If the private borrower uses his borrowed funds foolishly, he will find that his real income is reduced when the necessity to pay interest arises. If he borrows and invests wisely, he may be better off having borrowed although he still must make the interest payments.
The Source of the Error
As we have suggested earlier, the fundamental error made by the new orthodoxy in making a sharp conceptual distinction between public and private debt is methodological. The analysis reflects a failure to consider relevant alternatives. The neglect of the position of the bondholder in the situation without public debt has caused the interest payments to be viewed exclusively as internal transfers. The receipt of interest by bondholders instead of the receipt of real income from the debt-financed expenditure has been viewed as the appropriate offset to tax payments.
This "transfer payment" approach cannot be extended to private debt because the recipient of interest is, by definition, separate from the payer of interest. When discussing private debt the advocates of the new orthodoxy have consistently stressed the "burden" or "sacrifice" involved in both debt servicing and repayment. For example, Lerner speaks of "nearly everybody who has suffered the oppression of private indebtedness."*49 It is, of course, quite proper to attribute a "burden" or "sacrifice" to private debt if care is taken to limit the analysis explicitly to one side of the financial transaction. Debt represents deferred payment, and any payment, considered apart from its return, entails a "burden" on the debtor. Burden in this context can only mean the differential cost or sacrifice imposed by the necessity of making payment when compared with the hypothetical situation in which the same return is secured without obligation. In this way, any outpayment, current or deferred, involves a "burden," and private debt is in no way unique in this respect.
The neglect of the appropriate offset to tax payment, namely, real income from the public project financed, in considerations of public debt led to a similar neglect in considerations of private debt with the result that the latter has been viewed as necessarily involving some net "burden" or "oppressiveness" even when the full financial transaction is taken into account. This implication is, of course, wholly wrong, and lends support to the usury conception of private finance in which all interest payments are viewed as subsidies to lenders.
The "correct" view of private indebtedness must recognize that the act of borrowing or of lending is a market transaction similar in all its essential respects to other market transactions. Both the borrower and the lender expect, at the moment of contract, to be able to secure advantages from the exchange. The "thing" exchanged in this case is a command over resources at two separate points in time. The borrower secures current command over economic resources, a command which is advanced in time from that which he could otherwise secure without cost. The lender secures a promise of command over resources at some future date. And since resources can be used to produce real income through time, an interest payment or adjustment is necessary to bring the two capital sums into comparable magnitudes. It is commonly stated that interest is a "price" paid for the use of money. This is correct only in a very special sense. The real "price" for the use of money now is the money which must be given up in the future. Interest is the differential between these two capital sums converted into a percentage rate.
The perfectly rational individual may choose to borrow for either consumption expenditure or for investment expenditure. In either case, his decision should turn on a comparison of present values. If, when considering a consumption loan, he estimates the utility value of current consumption to exceed the discounted utility value of the payments stream occasioned by the loan, he should borrow. To be sure, once the initial period is completed he will be "oppressed" or "burdened" by the necessity of having to make the payments. But these "burdens" were presumably fully discounted when the loan was considered, and, on balance, the loan transaction involved an expected utility greater than that which would have been forthcoming without the loan. At least this was the calculation at one point in time. The borrower sacrifices some command over resources when he makes the interest payments on his debt. But this is not different in any way from the usual sacrifice involved in any ordinary expenditure. The family sacrifices hamburgers which it could have consumed when it chooses to buy ice cream instead. The fact that this sort of sacrifice occurs simultaneously with the enjoyment of the alternative product serves to obscure the similarity between this and the sacrifice involved in the interest payment on debt. The family who borrows to buy hamburgers today sacrifices hamburgers in the future in order to enjoy them now. The principle is identical for the two cases.
The analysis of private debts is even more clear when we consider private loans made for investment purposes. Here the individual secures command over resources which he puts to directly productive use. Again his decision, if rational, will turn on a comparison of present values, the capitalized value of a future income stream as compared with the capitalized value of a payments stream. If the former exceeds the latter, there is a net gain to be secured from entering the loan market. These two streams need not be equivalent in their time shapes. It is perfectly rational for the individual to borrow on extremely long term to finance short-term projects or vice versa.
Individuals are not, of course, always rational, and their behavior in creating private indebtedness need not reflect wise decision making. But they are likely to be somewhat more rational in their behavior as private individuals than in their behavior as citizens, or at least no less so. The degree of irrationality in the choice process cannot, therefore, lead to a basic distinction between private and public debt.
The implications of the "transfer payment" theory of public debt for the theory of private debt provides us with one explanation of the widespread acceptance of the false analogy argument.*50 There is no essential similarity between public debt, erroneously conceived, and private debt. Nor is private debt, wrongly viewed, similar to public debt properly understood. If private debt is considered necessarily oppressive, burdensome, and all interest payments usurious and unjustified, then the public debt is not comparable. This re-emphasizes that the analogy between the public debt and the private debt, correctly conceived, holds true in most essential respects, at least in the model which we have been considering.
Real Balance Effects and the Wealth Illusion*51
The demonstration of essential similarity does not, however, imply that the two debt forms are identical in all respects. The differences which are present stem largely from the illusion which public debt creation fosters, an illusion rarely present with private debt. As we have said, neither the public asset which is debt financed nor the liability which the debt obligation represents enters normally into individual balance sheet calculations when the debt is created and the asset purchased. This "compensating" illusion does nothing to affect the net worth of the taxpayer-borrower under static conditions. The two debt forms remain equivalent in their influences on individual behavior.
If the absolute price level changes subsequent to debt creation, this public debt illusion may act to exert some differential influence on behavior through the real-balance or Pigou effect. The failure of the public taxpayer-borrower to consider either the asset or the liability side of the transaction may make him immune to such exogenous changes which serve to modify the real value of the one side relative to the other. Although a fall in the absolute price level will tend to increase the real value of the liability relative to the asset, the taxpayer-borrower may not take this into account in his behavior. On the other hand, the fixed-yield claimant, the bondholder, will be affected. He will tend to increase his spending out of income as the real value of his debt claims increases with a decrease in the absolute price level. For the private debt, the borrower will consider explicitly both his debt-financed asset and his liability. A change in the absolute price level will modify the balance-sheet value for each of these, and, if the asset yields income in real terms, its money value will tend to move with the price level. The real value of the liability will move inversely to the price level. Thus, the private borrower will tend to reduce (increase) spending out of income as the price level falls (rises). To some extent, his behavior will offset the real balance reaction of the lender. This offset may be absent for the public debt due to the presence of the illusion.
In discussing this point it becomes necessary to distinguish carefully between the illusion fostered at the time of debt creation and that which may be present under the long-continued existence of an outstanding public debt. The taxpayer-borrower may, and probably will, discount both future taxes and future yields on the public asset too heavily when he makes his initial decision. But even if he does this, the subsequent necessity of paying service charges may cause him to take tax obligations more fully into account.
Taxpayers who have grown accustomed to a fixed service charge, say $7 billion annually, will make future plans on the expectation that their tax contribution to this total service charge will remain stable. If the absolute price level is then reduced, this fixed money obligation will assume a greater real value. In this sense, the liability side of the public debt does enter into private behavior similarly to private debt, and the introduction of the real-balance effect does not appear so significant as it might at first have appeared.
The difference caused by the real-balance effect does, however, constitute a difference in the two debt forms, which should not be overlooked. Some critics will dispute perhaps my claim that this difference is not an essential or fundamental one. Be this as it may, the point is that this sort of difference is not that which has been emphasized by the new orthodoxy. The difference discussed here is not due to some fallacy of composition. The difference arises solely out of the greater complexity of the collective choice process which tends to render private shares of both assets and liabilities indistinguishable to the individual participant. The importance of this difference for real-world problems is lessened when a careful distinction between real debt and monetized debt is introduced. As later chapters will demonstrate, much of what we call public debt creation is disguised currency creation. In this circumstance, relevant comparison with private debt is modified. For real-debt issue, the possible differences arising from the real-balance effect merit consideration, but these do not require a reversal of the earlier statement that the two debt forms are fundamentally similar.
Notes for this chapter
Jorgen Pedersen, as cited in Alvin H. Hansen, Fiscal Policy and Business Cycles (New York, 1941), p. 142.
Edward Nevins, The Problem of the National Debt (Cardiff, 1954), pp. 22-23. Cited by permission, University of Wales Press.
At a slightly more sophisticated level of analysis, however, even this statement must be qualified. In abstract terms, its validity requires that the marginal productivities of the resources involved be equal in public and private employments, that the operation be "at the margin." This begs more questions than it answers, however, since the whole issue of evaluating resource productivity is immediately raised. By the nature of public goods, market prices are not available to assist in such comparative evaluation. The shift of resources from public to private employments or vice versa can only be adjudged to add to social wealth on the basis of individuals' revealed preferences in supporting the shift. If we assume that a debt creation-public expenditure decision is rationally made, the taxpayer group must be adjudged as having moved to a preferred position. In this sense, therefore, social wealth may be said to have been increased. But the lending as well as the borrowing group must have also benefited, at least in the ex ante subjective sense. Gains from exchange are mutual, and some advantages are expected by both parties to the contract. Lenders assume that they will receive some differential return from the government security (in terms of rate, lowered risk, or other considerations) while borrowers (taxpayers) must be credited with assuming that the public project yields a "social" rate of return in excess of the borrowing rate. The wealth of both groups, which include, of course, many of the same individuals, increases in the subjectively calculated sense. It is evident, however, that this is not the sort of increase denied in the citation quoted in the text.
Over and above the gain from making the exchange which is necessary for any market transaction to be rational. See the preceding footnote.
Abba P. Lerner, "The Burden of the National Debt," in Lloyd A. Metzler et al., Income, Employment and Public Policy (New York, 1948), p. 255.
There have been few dissents in the literature of the past twenty-five years. One dissent from the prevailing view should, however, be noted. Emerson P. Schmidt argued in 1943 that public debt and private debt are essentially equivalent. ("Private versus Public Debt," American Economic Review, XXXIII [March, 1943], 119-21.) The fact that his arguments apparently had little effect indicates the dominance of the "new orthodoxy."
I am indebted to Professor A. Morgner for a brief conversation which has caused me to add this section.
End of Notes
Appendix: An Accounting Summary
In the discussion above, individual balance sheets have been frequently used indirectly. It will perhaps be useful to employ them more specifically here and to diagram some of the analysis in terms of simple balance sheet examples.
The simple T-accounts of Table I require little explanation. It is noted that, in the last set of accounts, the lender is in an equivalent position with and without the debt. But the taxpayer-borrower is in a worse position with the debt than he would be without it. He would be better off had the public investment never been undertaken. This result stems, of course, from the assumption that the project financed is completely unproductive. The situation would be identical for a private borrower who has made a wasteful expenditure which he financed by a private loan.
In Table II we assume that the public project is equally productive with private investment. Here it is noted that all parties to the transaction are in identical situations with and without the public debt. In Table III we assume that the public investment is of greater productivity than private investment. Here it is noted that the taxpayer-borrower is in a better position with the public debt than he would have been without it. The position of the lender is not modified. From these three tables it is evident that it is the position of the taxpayer-borrower that is modified by the relative productivity of the public investment.
In Table IV we retain the assumption of Table II that the public and private investments are equally productive, but we now assume that both the value of the project and the future tax payments are fully discounted into present values which are incorporated into individual balance sheets. The results are the same as for Table II. In any future income period, all parties are in identical positions with and without the debt.
In each of the first four tables, it is implicitly assumed that the alternative to public debt is a failure to carry out the proposed expenditure operation. In order to bring the closeness of the analogy with private debt more clearly into the open, Table V assumes that the same individuals who are the taxpayer-borrowers create a whole set of private loans to finance the same project. As may be seen, the position of the borrower is identical in the two situations.
Many other comparisons could be drawn on simple tables such as these, using other assumptions about relative rates of yield, rates of interest on government securities, and the proclivity of individuals to consider public assets and public liabilities as individual assets and liabilities.
It must be emphasized that these simple T-accounts represent only partial balance sheets. In most cases, a bondholder will also be a taxpayer-borrower. Therefore, to arrive at a composite individual balance sheet, the individual's T-account as a bond purchaser must be combined with his account as a taxpayer-borrower. The separation of these two roles does, however, allow us to clarify the analysis, although such separation here should not be taken to mean that the individual may not in many cases fill both roles simultaneously.
Return to top