IF cs.~; cd c!>:; W«(isH lJ« ~; ~~) (~Jl'\V/ ~ ill) iii\, It\\lf~ C(J) tt ~:b«\\m\ IF'w (i\\m\Cc';li~0(C::C0)

October 14, 1983

Classical Reflections On The Deficit

The primary long-term effect ofdeficits is to Reserve System, which although formally reduce the rate ofcapital formation. Gov­ independent ofthe executive branch, can be ernment borrowing crowds alit private treated for our purposes as part of the gov­ borrowing and causes a lower rate ofinvest­ ernment. When the Federal Reserve buys ment. The lower rate ofcapital formation Treasury debt, itpays with an increase in the hurts productivity, decreases growth, limits monetary base which will be held either as the rise in real incomes and weakens our reserves of depository financial institutions international competitiveness. (Martin or as in the hands of the public. Feldstein, Chairman of the President's This transaction is sometimes called "mone­ Council of Economic Advisers in the Wall tizing the debt," and it leaves the public Streetjournal,)uly 15, 1983.) holding more non-interest bearing govern­ ment liabilities in the form of and Everybody's jumping to the conclusion that fewer interest-bearing liabilities in the form Icrowding out] is going to happen in 30 ofTreasury securities. days ... it's not going to happen in three months. (Secretary of the Treasury, Donald Government's expenditure is constrained to T. Regan, quoted in the Washington Post, equal the sum of taxes plus the sales of debt August 23, 1983.) to the public (including banks) plus the creation of monetary base by the Federal With the prospect of large federal govern­ Reserve. Since deficits are the excess of ment deficits during this year and perhaps government expenditure over taxes, the for several years to come, the question as to budget constraint may also be expressed as: what their effects will be is in the forefront of deficits must equal the sales of debt to the discussions ofeconomic policy. Reasonable public plus the creation ofmonetary base by people evidently differ on the answer. the Federal Reserve. Unfortunately, it is not always clear from their pronouncements what the basisof their Ricardian equivalence judgments is. Close examination shows that The current policy debate centers on what may lie atthe heart oftheir differences whether or not different compositions of are some fundamental conceptual issues in government finance result in crowding out. economic theory-in particular, whether In particular, does issuing more debt and taxation and government borrowing are lowering taxes raise real interest rates and economically equivalent forms of govern­ crowd out private investment? Ricardo gave ment finance, and whether it makes a differ­ an early analysis of the problem in his ence if takes the form of Principles of Political Economy and interest-bearing bonds or non-interest-bear­ Taxation (1821). ing currency and . This Letter builds on a discussion by the nineteenth He argued that if fu lIy understood century English classical economist David that government borrowing only postpones Ricardo to help clarify these issues. the payment o!.taxes, it wou Id not matter at all howdeficits were financed. If the govern­ Government's budget constraint ment chose to finance a given expenditure All federal expenditures must be paid for, through taxes, each could borrow but there is a choice of sources for the enough to pay his taxes (i.e., sell a bond) and needed funds. The Treasury can raise taxes, then would immediately have to pay only or it can go into debt. Its debt may either be the interest and some portion of the princi­ held by the public at large or by the Federal pal on the loan. If the government financed

1 Opinion~ expressed in Ihis newsletter do not necessarilv reflect the vie\vs 01 the manag(~ment of the reelera! Reserve' Bank of San francisco, or of the Board of Covernors of the Federal Reserve Systern. the same deficit by selling its own bonds to taxpayers would save enough to pay only the public, the taxpayers would be taxed to the taxes that cover the interest on the debt pay the interest and the currently maturing and forget about the need to repay the prin­ principal on the government bonds. In either cipal in the future. Debt finance would then case, the taxpayer might pay the taxes with­ appear to be more stimulative than tax out borrowing the money, but if he treated finance when the economy is operating at the amount paid as a loan to himself, he less than full employment. would find that the portion of his income available for consumption would be no less If Ricardo's theoretical argument held in than ifhe had actually borrowed the money. practice, different splits between taxes and Ricardo argued, in effect, that the composi­ debt finance would not affect spendable tion of government finance for the same incomes. Private saving would always level ofexpenditure made no difference adjust to the amount needed to service the since taxpayers can make exactly compen­ debt, and interest rates would not be sating adjustments in their own portfolios. affected. On the other hand, if Ricardo's practical judgment were correct, taxpayers Ricardo's argument for the equ ivalence of would feel richer when government expen­ debt and taxation rests on at least two im­ diture is financed by debt rather than by plicit assumptions. The first is that taxpayers taxes. They would want to spend more, but are economically similar. If this assumption since they would not actually be richer, they were relaxed, then taxpayers' after-tax in­ could only spend more by saving less. To comes would be affected by the method of induce them to save enough to cover the government finance. Suppose, for example, debt, the on government bonds that some people are seen by lenders to be a would have to rise. Higher interest rates greater risk than others and are, therefore, would then discourage private investment. charged a higher rate of interest. Lowering Thus, when there is debt illusion, the more taxes and issu ing an equal amount of debt the method of government finance opts for means that high-risk borrowers have more debtover taxes, the lowerwould be the level cash in hand and pay it back through taxes at of fJrivate investment and the higher the a lower rate than they would pay ifthey had level of private consumption. borrowed it privately. Debt finance would thus affect their incomes. In the same way, The burden of debt if the people who buy most of the bonds Despite showing that debt and taxes could differed from those who pay most of the beequivalent, Ricardo believed that govem­ taxes, substituting debt for taxes would ment should favor taxes over debt even if affect incomes. The second assumption is there were no debt-illusion. His reason was that taxes are not distorting, that they do not that present taxes are immediate and hard to alter the optimal allocation of resources. If escape, but a large debt implying high future this assumption were relaxed, for example, taxes would encourage emigration to avoid if taxes were levied only on consumption paying future taxes. Ricardo's argument can (e.g., sales taxl, then lowering taxes and rais­ be reformu lated to show that the present ing debt would favor consumption at the generation can escape taxes by incurring expense of investment. debt that must be paid by generations that live long after the present one is dead and Ricardo recognized that his argument for the gone. It is said that "the past is another equivalence ofdebt finance and taxation country." The present generation, in effect, was a theoretical one that depended cru­ would emigrate to the past by shifting the cially on accurate anticipations offuture burden ofdebt onto its descendants. That taxes. He believed that, in practice, there burden can be measured by the degree to would be "debt-illusion" in the sense that which the capital stock is smaller because

2 previous generations invested less as a re­ orfixed rate ofreturn, these high rates would sponse to higher interest rates caused by the induce people to hold more government choice of debt financing. bonds and less money. Such a situation is equivalent to a rise in the velocity ofcircula­ This theoretical analysis raises two ques­ tion of money, which would support a tions: First, can the burden of debt actually higher rate of inflation. be shifted onto future generations? That is, are taxes and debt finance equivalent across In practice, we are not facing such a hyper­ generations? Second, is it desirable to shift expansion of government debt. Instead, we the burden ofdebt? The answer to the are facing a large increase in the ratio of second question depends on one's values, interest-bearing government debt to non­ but its importance depends on the answer to interest-bearing monetary base. In the the first question. long-run, since both are nominal liabilities of the government, it may make little differ­ Ricardo believed that the burden of govern­ ence which is used to finance the deficit. In ment debt can be shifted to other genera­ the short-run, however, the government can tions. His implicit assumption is that the sell more debt to the private sector only by debt must eventually be paid off by the tax­ offering a higher rate of return to increase its payers, but that it can be postponed by issu­ attractiveness. That this rise in interest rates ing more debt. If this were done continu­ will crowd-out private investment is the ously, no future generation would be taxed principal fear of those who oppose further to payoff the debt. And each generation increases in the interest-bearinggovernment wouId bel ieve falsely that it was richer, and debt. interest rates would remain high to ensure that savings were sufficient to buy newly Conclusion issued debt. Compared with the recent past, current monetary and in the United Critics of this idea argue that there is some States places relatively more weight on debt size, or some ratio ofthe debt to nation­ deficit finance than ori taxation, with the al income, beyond which no one would deficit being financed more by governrnent want to buy government bonds, that is, at debt than by . What effects which the government has limited "collat­ this policy stance should be expected to eraL" At that point, the debt must eventually have on the economy in the long-run be paid off. How much collateral the gov­ depends on how one decides the main ernment is supposed to have is not clear. issues discussed in this Letter. In other Nevertheless, the critics believe that at the words, which Ricardo does one believe? point at which the governmentcould not sell The "practical" Ricardo expects present any more debt, it would have to cover its policy to crowd out investment, but the deficits by creating monetary base. And they "theoretical" Ricardo expects few ill effects 'ff'l./'~ believe that this monetary expansion could to resu It from the large government deficits cause inflation. currently facing the U.S. at")1 @.,~ Can this implicit distinction between the Kevin D. Hoover and Joseph R. Bisignano ···.l economic effects of an increase ofthe interest-bearing debt and ofthe monetary base be sustained? One could argue, in Oc theory, that an expansion ofthe debtat a rate ~.I much faster than the rate ofgrowth of na­ tional income would require ever-rising interest rates. And since money yields a zero ~J 3 U018U!4Sl?M. 4l?ln • uo8aJO • l?pl?i\aN • 04ePI !!l?MeH • l?!UJOj!!l?) euozpV. e>jsPIV <\:5)~~ Y~'iill\2? Jl~ 'iill\2?~ ~ <\:5) ~ 'iill\2?

1) \\lJ@\llilJ:{lJl\1l!

BANKING DATA-TWELFTH FEDERAL RESERVE DISTRICT (Dollar amounts in Iilillions) Selected Assets and Liabilities Amount Change Change from large Commercial Banks Outstanding from year ago 9/28/83 9/21/83 Dollar Percent loans (gross, adjusted) and investments* 161,743 - 304 - 1,304 - 0.8 loans (gross, adjusted) - total# 141,658 - 294 - 951 - 0.7 Commercial and industrial 43,058 - 208 - 2,675 - 5.8 Real estate 57,113 48 - 509 - 0.9 Loans to individuals 24,724 153 1,206 5.1 Securities loans 2,662 272 23 0.9 U.s. Treasury securities* 7,417 6 837 12.7 Other securities* 12,667 - 16 - 1,191 - 8.6 Demand deposits - total# 39,693 - 582 1,208 3.1 Demand deposits - adjusted 28,669 7 1,915 7.2 Savings deposits - totalt 65,527 45 34,632 112.1 Time deposits - total# 67,111 - 125 - 33,922 - 33.6 Individuals, part. & corp. 61,481 - 4 - 29,474 - 32.4 (Large negotiable CD's) 17,344 - 21 - 20,679 - 54.4 Weekly Averages Weekended Weekended Comparable of Daily Figures 9/28/83 9/21/83 year~ago period Member Bank Reserve Position Excess Reserves (+l/Deficiency (-) 107 98 254 Borrowings 103 118 70 Net free reserves (+)/Net borrowed{-) 4 20 185 * Excludes trading account securities. # Includes items not shown separately. t Includes Money Market Deposit Accounts, Super-NOW accounts, and NOW accounts. Editorial comments may be addressed to the editor (Gregory Tong) or to the author . ... Free copies of this and other Federal Reserve publications can be obtained by calling or writing the Public Informa~ tion Section, Federal Reservt;! Bank ofSan Francisco, P.O. Box 7702, San Francisco 94120. Phone (415) 974-2246.