As a part of the proceedings of the Eleventh Annual Meeting of the Board of Governors of the International Monetary Fund, an Informal Session on “Recent Developments in Monetary Analysis” was held on September 25, 1956. The three papers which were presented at that Session by Dr. M. W. Holtrop, President of De Nederlandsche Bank, Dr. Paolo Baffi, Economic Adviser to Banca d’Italia, and Dr. Ralph A. Young, Director of the Division of Research and Statistics, Board of Governors of the Federal Reserve System, are reproduced below, together with the background paper, “Monetary Analyses,” prepared by the Statistics Division of the Research and Statistics Department of the International Monetary Fund.


As a part of the proceedings of the Eleventh Annual Meeting of the Board of Governors of the International Monetary Fund, an Informal Session on “Recent Developments in Monetary Analysis” was held on September 25, 1956. The three papers which were presented at that Session by Dr. M. W. Holtrop, President of De Nederlandsche Bank, Dr. Paolo Baffi, Economic Adviser to Banca d’Italia, and Dr. Ralph A. Young, Director of the Division of Research and Statistics, Board of Governors of the Federal Reserve System, are reproduced below, together with the background paper, “Monetary Analyses,” prepared by the Statistics Division of the Research and Statistics Department of the International Monetary Fund.

Although there has been a remarkable revival of monetary policy in recent years, some countries still hesitate to take vigorous monetary actions and others are handicapped in doing so by the character of their financial institutions. Pursuit of an active monetary policy may be inhibited by the existence of a large unfunded government debt and the reluctance of the authorities to burden the budget by increasing interest rates to the extent that might be required to check an investment boom. The authorities may also be disinclined to push up interest rates because they do not like to cause sharp fluctuations in the market value of publicly held long-term securities and because they are uncertain whether interest rates can be brought down again quickly, if desired. Even when increases in central bank discount rates are made, there is often delay in transmitting these changes to the lending rates of commercial banks and to mortgage rates and bond yields. In the large part of of the world in which highly developed financial institutions do not exist, some of the ordinary measures for influencing interest rates either cannot be applied at all or have only a limited effectiveness.

In recent years, however, it has become increasingly evident that a flexible means of controlling the volume of investment can play an important part in maintaining internal stability and equilibrium in the balance of international payments. Resort has been had to various direct controls on credit and investment, but these measures lack many of the advantages associated with impersonal and indirect controls, such as variations in interest rates.

This paper considers the possibility of using special tax measures as a partial substitute for interest rate increases as a means of restraining private investment. These tax measures may be broadly classified in two groups: (1) those that apply to borrowing or lending, and (2) those that apply to investment or the sale of capital goods. The measures examined include special taxes and modifications of deductions allowed in determining net income for purposes of general taxation. There have been a few suggestions for taxes on loans, but apparently this approach has not yet been adopted by any country. Among the few examples of the second group of measures are taxes on investment outlays imposed in Sweden and Norway in 1955, and postponement of normal depreciation allowances under the income tax in Canada during 1951–52.

Although there has been little experience with such a policy and only a limited amount of study has been devoted to its implications, it appears that certain special tax measures might operate fairly effectively without involving the drawbacks associated with major changes in interest rates and without encountering some of the difficulties attached to direct controls. In this paper, no consideration will be given to the possibility of offering a stimulus to investment by converting some of the measures into subsidies or negative taxes.1 The budgetary contribution of the special tax measures will not be examined in detail. It may be noted, however, that their efficacy for control of inflation would not necessarily be proportionate to their revenue yield. No revenue would be collected with respect to transactions which were canceled or postponed because of the existence of the special tax measures. On the other hand, if the government should increase its expenditures by the amount of any additional revenue obtained from the special measures, the potential anti-inflationary effect of the measures would be wholly or partly offset—a limitation which applies to any tax increase.

General Comparison with Increase in Interest Rates

In comparing special tax measures with an increase in interest rates, it may be recalled that higher interest rates have traditionally been expected to aid in stabilization not only by discouraging domestic real investment but also by stimulating saving. (In some countries, interest rate changes have been relied on to alter the volume of external investment and the inflow of short-term capital from abroad, but this aspect of stabilization policy will not be considered in the present paper.) Skepticism has been expressed about the sensitivity of investment to interest rates, but there probably would be general agreement that real investment has some degree of interest elasticity, and that changes in interest rates do influence the volume of private capital expenditures. On the other hand, the prevailing opinion now is that the volume of saving and consumption has very little sensitivity to interest rates, at least in the short run.

If it is true that the supply of savings has little or no interest elasticity in the short run, the changes in income of recipients of interest which are associated with adjustments of interest rates perform no important stabilization function. Indeed, if the interest elasticity of savings is zero, any part of changes in interest income which reflects changes in total money income and not a mere transfer between lenders and profit receivers will have a perverse effect from the point of view of stabilization. Lenders will receive additional income when interest rates are raised for the purpose of restricting aggregate demand and will suffer a loss of income when interest rates are lowered in order to stimulate investment demand. The perverse effect may be small in the short run, inasmuch as higher or lower incomes will be earned only with respect to newly acquired assets or contracts which permit adjustments of interest rates. Furthermore, the perverse effect may be partly or wholly offset by the influence of changes in capital value of previously held financial assets, since a change in interest rates causes an opposite movement in the market value of existing financial assets, such as bonds. Holders of these assets probably tend to adjust their consumption, to some extent, to the value of their assets and the amount of their realized or unrealized capital gains, as well as to their current income.

The influence of interest rate changes on the supply of credit—as distinguished from the volume of savings—may also detract from the potential stabilizing influence of monetary policy. Generally, private savers and banks will remain more liquid when interest rates are low than when they are high. When rates are low, the cost of maintaining liquidity is small and the possibility of speculative gains, as a result of a future rise in interest rates, is great. When interest rates rise, the public is likely to reduce its cash balances relative to income, while the commercial banks squeeze their cash reserves by reducing excess reserves and by rediscounting at the central bank.2 If the supply of credit is responsive to interest rates and the propensity to save is not, a part of the potential deflationary or expansionary influence of an increase or decrease in interest rates is counteracted, and the authorities must take stronger monetary actions than would otherwise be necessary to obtain a desired contraction or expansion of demand.

The impact on capital costs and on real investment attributable to higher interest rates can be approximated by special tax measures. The increase in capital costs (or the reduction in investment returns) incurred by investors, however, will take the form mainly of payments to the state rather than to lenders. Some additional interest payments and perhaps some rise in interest rates may be induced by certain of the tax measures. For example, if the profit outlook is good, many investors may wish to carry through their investment plans despite the imposition of an investment tax and may attempt to borrow more in order to cover the tax.3 Increases in interest payments and in interest rates brought about in this way will probably be minor compared with the tax or with the increases that would be necessary to obtain the same deterrent effect through orthodox monetary policy.

The tax approach offers the opportunity of obtaining the positive stabilizing effects of interest rate adjustments with a minimum of unnecessary or perverse effects on lenders’ incomes and a minimum of partly offsetting economization of cash balances and bank reserves and with only minor influence on the market value of outstanding securities. The tax measures have the further advantage of contributing revenue at a time when reliance on orthodox monetary policy requires an increase in budget expenditures on account of higher interest payments on unfunded government debt.

A feature of the special tax devices which may be advantageous in some circumstances is the convenience of approximating by these measures some of the economic effects of an assured future change in interest rates. An increase in interest rates will be an especially strong deterrent to long-term borrowing on the part of those investors who expect the higher interest rates to be temporary; by postponing a long-term project, the investor may anticipate a large saving of interest. If the authorities could give assurance that interest rates would decline at a specified future date, the anti-inflationary influence of a temporary increase in borrowing costs would be powerfully supplemented. It may be difficult, however, to give a firm assurance that interest rates will be lower in a year or so because, if this declaration is believed by the public, arbitrage transactions will tend to establish at once the rate that is expected to prevail a little later. The tax approach is not subject to this complication; it is quite possible to impose a special tax measure for a limited period of time and thus reinforce its influence on long-term investment. Since it is always dangerous to undertake a firm commitment on the future course of policy, the authorities will probably wish to be sparing in their use of assured future changes in tax rates.

Measures Applying to Borrowing or Lending

A tax measure that increases the cost of borrowing or reduces the availability of credit should have much the same effect on the volume of investment as an increase in interest rates. The measure might take the form of a special tax on new loans and security purchases or on interest receipts, a special tax on borrowing or interest payments, or denial of a deduction for interest payments under a general income tax.

The immediate impact of some of the measures would be on lenders, whereas the impact of other measures would be directly on borrowers. It seems probable, however, that the taxes on lenders, if broad in coverage, would be promptly passed on to borrowers, especially under conditions in which the monetary authorities would wish to restrict credit and could raise interest rates by orthodox monetary action. A tax on lending might affect borrowers more promptly than does an increase in the central bank discount rate or open market operations. The tax would offer an easy justification for charging borrowers more, and bankers might not have the same hesitancy in adjusting their loan contracts that they are sometimes said to have when the monetary authorities are attempting to restrict credit by other means. If, for any reason, the taxes on lenders were not fully passed on to borrowers, lending would become less attractive and the availability of credit would be reduced. There would probably be less reliance on informal rationing by bankers and other lenders than in connection with ordinary central bank operations, and this may be considered an advantage of the tax method.

Taxation of loans

One method of taxing lending would be to impose a one-time ad valorem levy on new loans. Lenders could recoup the tax either by increasing interest rates or by adding to the principal a separate charge for the tax while keeping nominal interest rates unchanged. These procedures are equivalent, for they have the same consequences for the effective rate of interest. Thus, for example, if a 5 per cent tax on new loans were fully passed on to borrowers, it would increase the effective interest rate on one-year loans by 5 percentage points regardless of whether the tax was stated as an increase in the interest rate from, say, 4 per cent to 9 per cent, or as a separate charge payable at the maturity of the loan. A onetime ad valorem tax at a uniform rate on the principal would imply an increase in effective rates varying inversely with the maturity of the loan; for example, a 5 per cent tax on the principal would be equivalent to an additional interest rate of 20 per cent per annum on three-month loans, 5 per cent on one-year loans, and 2½ per cent on two-year loans.

If the tax were applied to all new loans, regardless of whether they represented a net increase in credit or renewals of old loans, borrowers would have an inducement to attempt to arrange medium-term or long-term loans when they expected to require a roughly constant amount of credit over a period of time, rather than to rely on a series of short-term loans renewable at maturity. It might be worth while to pay a higher nominal rate of interest on a longer-term loan in order to avoid repeated applications of the tax. The shift in demand might well tend to raise medium-term and long-term nominal interest rates relative to short-term nominal rates.

The impact on short-term and long-term loans could be equalized, at the cost of some additional administrative effort, by applying the tax periodically to outstanding balances or to annual interest receipts or payments.

It is highly unlikely that any country would attempt to impose a tax on all loans or all interest payments or receipts. Some credit operations would be exempted as a matter of policy, and others would probably be excluded because of administrative difficulties.

Exemption would probably be granted for loans to the government and possibly also for loans to other borrowers whom the authorities wished to favor. The exemption of some loans and the taxation of others would establish a two-price system and would tend to divert credit toward the exempt areas. In principle, it would be possible to establish several prices (effective interest rates) by imposing different taxes on loans to different classes of borrowers. The possibility of selectivity would no doubt be one of the attractions of taxation, compared with general monetary operations, but it would have the disadvantage of exposing the system to political pressures for favorable treatment, a danger that has not always been avoided in connection with selective credit controls.

Application to banks and other lenders and to borrowers

Following a suggestion of Professor H.G. Johnson, Professor J.R. Hicks in 1952 recommended a tax on bank advances in the United Kingdom as a means of controlling credit. Hicks suggested that the Chancellor of the Exchequer be given power to impose and modify the tax, and that in order to achieve maximum flexibility the tax not be regarded as a regular budget measure. Lending to the Government would be exempt. Hicks believed that in the United Kingdom the discount houses should be classed as banks and made subject to the tax. Insurance companies and building societies might be allowed to elect to be classified either as banks or nonbanks. If they chose to be considered nonbanks, they would escape the tax but would not be allowed to hold short-term government paper. This approach would be intended to prevent these institutions from undercutting the banks by disposing of short-term government paper to make loans.4 If the tax were restricted to advances, it would not inhibit bank lending through security purchases. One way of closing this loophole would be to apply the tax to all increases in bank assets,5 presumably after deducting increases in holdings of government securities. It might also be appropriate to allow a deduction from the tax base for increases in bank capital.

Restriction of the tax to banks, however, would limit its effectiveness, even if these taxable institutions were broadly defined and were given the exclusive privilege of holding short-term government securities, as suggested by Professor Hicks. Nonbank financial intermediaries, such as insurance companies, might hold substantial amounts of long-term government bonds and perhaps other marketable securities. These institutions would have an incentive to dispose of their securities in order to extend credit to borrowers denied bank loans at the previous interest rate. If the commercial banks and the central banks absorbed the securities at existing prices, interest rates would be held down, but the monetary effect would be the same as if the loans to final borrowers had been extended directly by the banking system. If the banking system did not absorb all of the securities offered by the nonbank financial institutions at unchanged prices, interest rates would be forced up, and the objective of avoiding an increase in the cost of government borrowing would be partly defeated.

A tax on loans could be fairly easily extended to nonbank financial intermediaries, such as insurance companies and building societies, since these institutions are usually subject to a considerable degree of government regulation and are required to maintain adequate accounting records. Real estate mortgages could also be reached when recorded in official records, regardless of whether the mortgagee were a financial institution or an individual. Security issues, on the other hand, could be most easily reached by a tax on the issuer.

It would be more difficult to apply a tax to loans granted by individuals or small firms to other individuals or small businesses. In countries with well-developed financial institutions, these transactions probably represent only a small fraction of total lending; but in underdeveloped countries, the operations of small moneylenders and other individuals are relatively more important. Even in countries with advanced financial systems, consumer credit could probably be reached less effectively than producer credit. The virtual necessity of excluding transactions outside the organized money market might not be a serious weakness of the tax approach, compared with orthodox central bank operations, as interest rates and credit availability for the excluded transactions are not highly responsive to monetary policy in most cases.

Denial of interest deductions under income tax

An arrangement that would apply directly to borrowers would be denial of interest deductions under the income tax. This method would, in effect, impose an annual surcharge on interest payments, thereby increasing the cost of borrowing without directly affecting the return received by lenders. The measure would be powerful only in countries where income tax rates are fairly high and the tax is effectively enforced. Certain administrative problems would be encountered in connection with this apparently simple approach. Perhaps the most serious of these problems would relate to lease contracts which resemble loans. Sale and “lease-back” arrangements have already been developed in various countries, partly for the purpose of avoiding income taxes and direct credit controls. These arrangements usually provide for the sale of a piece of real estate by an owner needing funds to a buyer who agrees to lease the property back to the seller on a long-term basis. The seller obtains funds at once and makes annual rental payments, which are rather similar to interest and amortization on a loan. Ordinary leases less obviously resemble loans, but have considerable economic similarity to them.

Measures Applying to Investment

The second major group of tax measures consists of actions to increase the cost of capital goods or to reduce the return on investment. Sweden and Norway have experimented with special taxes as a means of controlling investment. In 1952, Sweden introduced a special tax on investment in industrial buildings, machinery, and equipment and at the same time imposed an additional duty on production and importation of automobiles and motorcycles. These taxes were allowed to expire in 1954, but in 1955 new taxes were imposed on industrial investment at a rate of 12 per cent and on purchases of new automobiles at a rate of 10 per cent. In early 1955, Norway levied taxes at rates of 10 per cent on new industrial and commercial building, imports of automobiles and tractors, and new contracts for ships of more than 2,500 gross tons.6 Sweden and Norway also raised the central bank discount rate during 1955.7

An obvious difference between a tax on new investment and a tax on loans is that the former, unlike the latter, applies directly to internally financed investment. Although an increase in market interest rates may inhibit internally financed investment, by making the alternative of buying securities more attractive relative to real investment, this effect is generally thought to be slower and weaker than the influence on externally financed projects. Since internally financed investment is an important fraction of the total in most countries, the possibility of obtaining broader coverage by a tax on investment would be an important advantage, compared with either a tax on loans or an increase in interest rates. It would also be possible to apply the tax to the main “investments” of consumers by imposing suitable excises on consumer durable goods and residential construction. On the other hand, it might not be feasible in many countries to tax inventory accumulations, and it would be difficult to apply a direct tax to the investments of small firms.

In the following discussion, it will be assumed that the special measures would not cause a change in the factor cost of investment goods, that is, the prices, exclusive of tax, at which these goods are supplied. In other words, measures applying to purchasers of capital goods are assumed to occasion no change in market prices of the goods, whereas taxes applying to the production, sale, or import of capital goods are assumed to cause market prices to rise by the amount of the tax. (According to the usual theory of tax incidence, the assumption adopted is that appropriate for an excise tax on a good produced under conditions of constant costs per unit.) This assumption is probably not entirely realistic, particularly for an investment tax of broad coverage, but it does not seem seriously misleading. To the extent that the ex-tax prices of investment goods were reduced, the special tax measures would absorb income of producers of these goods and would still be anti-inflationary.

Effect on investment returns

On the incidence assumption stated above, an ad valorem tax on new investment or purchases of new capital goods would raise by an equal percentage the return that must be earned to permit recovery of the amount invested. For example, to permit recovery of invested capital, an asset costing 100 must be expected to yield an average annual return (net of income tax but including depreciation allowances) of more than 50 if its economic life is 2 years, more than 10 if its life is 10 years, or more than 4 if its life is 25 years. A 10 per cent tax would raise to 110 the cost of the capital goods formerly costing 100 and would increase the annual earnings necessary to permit recovery of the investment by 10 per cent—to more than 55, 11, or 4.4 for the illustrations just mentioned.8 Imposition of the tax, therefore, would cause investors to exclude from favorable consideration the range of investment opportunities expected to yield returns between the old and the new (higher) acceptable minimums. This range may cover a large or a small volume of potential investment, depending on the tax rate and businessmen’s expectations concerning profit opportunities.

Comparison with increase in interest rates

In order to draw an exact comparison between a special tax on investment and an increase in interest rates as a means of restraining capital outlays, it would be necessary to know precisely how interest rates influence investment decisions. Knowledge of this subject is unfortunately limited. Three possible attitudes may be considered. Businessmen may think of a rise in interest rates as occasioning additional out-of-pocket costs on borrowed funds; an increase in the imputed cost of using equity capital, as well as an increase in the cost of borrowing; and an increase in the discount rate applicable to expected future investment income. Both the second and third attitudes are based on recognition of the opportunity cost involved in investing capital, but they lead to different evaluations of investment opportunities in some cases and may be usefully distinguished.

For investors who look on interest merely as an out-of-pocket cost, the significance of higher interest rates will vary with the proportion of their capital outlays which is financed by borrowing and the maturity of the credits. Hence no one figure can be computed to represent the change in interest rates that would affect out-of-pocket costs to the same extent as a specified tax on investment. It is evident, however, that in many instances a large increase in the applicable interest rate would be required, since investments are seldom financed entirely by borrowing and the maturity of loans obtained for this purpose is often shorter than the expected economic life of the assets.

For investors who include in their cost calculations imputed interest on equity capital as well as interest on borrowed funds, a tax payable at the time an investment is made would involve not only an increase in the original cost of the capital goods, which would be reflected in higher annual depreciation charges, but also an additional annual carrying cost attributable to the larger size of the investment necessary to obtain a specified income. On the assumption that depreciation is charged on a straight-line basis and is covered by earnings each year, a 10 per cent tax on new investment would increase average annual costs by the same amount as a rise in the interest rate from, say, 5 per cent to 15.5 per cent for an asset with a 2-year economic life, 5 per cent to 7.5 per cent for a 10-year asset, and 5 per cent to 5.9 per cent for a 50-year asset.9 The tax and the interest rate increases would therefore reduce by equal amounts the average net return from the investments.10 For short-lived assets the indicated increase in the interest rate is drastic, but for long-lived assets the increase is comparable in size to the changes that are in practice brought about from time to time by monetary action.

With the discounting technique, an investment tax and an increase in interest rates would be equivalent if they influenced in the same way the relation between the cost of capital goods and investors’ evaluations of the present value of the streams of income expected from various projects. On the assumption that yields accrue at a steady rate from day to day throughout the asset’s life, it can be shown that a 10 per cent tax on new investment is equivalent to raising the discount rate from 5 per cent to 14.8 per cent for an asset with a 2-year life.11 The equivalent increases a for an asset with a 10-year life is from 5 per cent to 7.1 per cent, and for a 50-year asset, from 5 per cent to 5.6 per cent. Illustrative figures for other rates of investment tax are given in Table 1. The taxes and increases in discount rates are equivalent in the sense that, on the assumptions stated above, they would exclude from favorable consideration the same range of otherwise acceptable investment opportunities. The taxes would do so by increasing the cost of capital goods without affecting the expected returns or their discounted value; the increase in discount rates would reduce the discounted value of expected yields without affecting the cost of capital goods.

Table 1.

Increases in Discount Rates Equivalent to Various Ad Valorem Taxes on Investment:1 Initial Discount Rate, 5 Per Cent

(In percentage points)

article image

Approximate values with discount continuously compounded and assuming level returns. See text for explanation of concept of “equivalence.”

Although it seems unlikely that the rather complex discounting procedure is widely used by businessmen in evaluating investment opportunities, there are indications that many large investors make explicit allowance for the imputed interest cost of investing their own funds. The rate at which this allowance is calculated is probably not highly responsive to short-period variations in interest rates, but it may tend to move parallel with market rates of interest over a longer period of time. Even those investors who are not sensitive to changes in interest rates might take into account the existence of a tax on investment, inasmuch as no subtle computations would be required to show that the tax increased the original cost of capital goods.

Despite the difficulty of selecting an appropriate set of assumptions on which to base a numerical comparison of investment taxes and increased interest rates, it may safely be concluded that an investment tax of the order of, say, 10 per cent would have as much effect on the profitability of investment in short-lived and medium-lived assets as would an increase in interest rates that is large in comparison with the changes that ordinarily occur. The influence on long-lived assets would be much less marked but still significant.

The fact that a uniform investment tax would be equivalent to a greater increase in the interest rate for short-lived investments than for long-lived investments does not necessarily imply an objectionable discrimination against the former. In well-developed money markets, short-term interest rates often rise relative to long-term rates during a period of credit restraint that is expected to be temporary. If short-term and long-term rates rise equally, it is generally believed that long-lived projects are more strongly discouraged than are short-lived projects. A special deterrent to long-term investment helps conserve capital and may be desirable for a country that must adjust to a permanent scarcity, but it is less necessary or appropriate in meeting a temporary threat of inflation.

Scope of tax on investment

Like a tax on new loans, a tax on investment could be employed in a selective manner. The tax would not apply to government investments; and private investments to which the authorities wished to give priority might also be exempted. Selectivity would be easier to achieve if the special taxation should take the form of a direct tax on the investor than if it should consist of excises and customs duties on capital goods. Selective exemption of certain end-users from the indirect taxes, however, would not be out of the question.12 It would be comparatively difficult to apply even a direct tax to inventory accumulations, and this kind of investment might have to be exempted as a matter of administrative expediency, even though its control might be highly desirable from the policy standpoint.

Modification of depreciation allowances

Deferment or denial of normal depreciation allowances under a general income or profits tax would be an alternative to a tax on new investment as a means of influencing the volume and timing of private capital outlays. Either deferment or cancellation of normal depreciation allowances with respect to investment carried out during an inflationary period would curtail the amount of tax-free funds available to business firms for purchasing capital goods and repaying loans obtained to finance new investment. Complete denial of normal allowances would reduce the profitability of the investments affected, and when income tax rates were high would turn a prospective profit into a loss on many projects. Deferment of allowances, although a less drastic measure, would adversely influence profit calculations to the extent that investors applied a discount for interest and risk on the deferred allowances.

During the Korean war boom, Canada experimented with deferment of depreciation allowances as a means of deterring unnecessary investment. The plan, introduced in April 1951, established three classes of investments, presumably on the basis of essentiality and relation to the defense program. For one class of investments, the deferment did not apply; for a second class, it applied generally but a government department could issue permits granting exemption in appropriate cases; for the third class, consisting of investments deemed to be of a less essential character, no exception to the deferment provision could be made. The original plan contemplated a 4-year deferment period, but by the end of 1952 the economic situation had eased and the deferment of depreciation allowances was ended.13

Although it is not possible to disentangle the influence of other factors, including direct controls on the use of scarce materials, from the effect of deferment of depreciation allowances, it is interesting to note that the various categories of investment in Canada did show significantly different movements in 1951-52 (when the deferment scheme was in operation) and in 1953 (the year after its termination). Indices (1950 = 100) of the value of investment by firms subject to income tax are as follows:14

article image

An Institutional Problem

A possible institutional disadvantage of taxation as a substitute for changes in interest rates arises from the fact that in most countries taxation is subject to closer and more elaborate legislative controls than are changes in interest rates. Typically, important modifications of taxation require either action by the parliament or at least a formal executive decree issued under conditions prescribed by the legislative body or by the constitution. Action to bring about changes in interest rates, on the other hand, can usually be taken at the discretion of the monetary authorities, subject only to the most general accountability to parliament. For this reason, taxation might be less flexible as a means of regulating investment than is interest rate policy, particularly in countries with developed financial systems and well-established central banks where the monetary authorities can in practice easily use their legal powers to alter interest rates. It is important to recognize, however, that there is often a considerable delay in transmitting the effect of actions by the central bank to final borrowers; the comparable delay in the taking effect of special tax measures is likely to be shorter, which would partly compensate for slowness in obtaining necessary legislative or executive action.

It seems, furthermore, that the distinction between the real effects of changes in interest rates and of special taxes of the kind discussed in this paper is in fact not so profound as to require a great difference in the degree of legislative control. The substance of property rights would not be altered by special taxes to a significantly greater degree than they are in practice altered by changes in interest rates, which affect the market values of securities and other capital assets, and by bank reserve requirements and other credit restrictions imposed by the monetary authorities. Both special taxes and changes in interest rates can also influence the use of economic resources and the distribution of income. These considerations encourage the belief that it might be possible to find acceptable arrangements which would enable special taxes of the kind reviewed in this paper to be used with the degree of flexibility that is essential for successful countercyclical regulation of the volume of private investment.


Mr. Goode, Chief of the Finance Division, was educated at Baylor University, the University of Kentucky, and the University of Wisconsin. Formerly economist at the U.S. Bureau of the Budget and the U.S. Treasury and a member of the faculty of the University of Chicago, he is the author of The Corporation Income Tax (New York, 1951) and of several articles in economic journals.


The possibility of stimulating investment by means of liberalizing depreciation allowances under the income tax is the only phase of the subject which has received much attention. See Richard Goode, “Accelerated Depreciation Allowances as a Stimulus to Investment,” Quarterly Journal of Economics (Cambridge, Mass.), Vol. LXIX (1955), pp. 191–220.


J. J. Polak and William H. White, “The Effect of Income Expansion on the Quantity of Money,” Staff Papers, Vol. IV (1954-55), pp. 398-433.


If the elasticity of demand for investment goods is less than unity, as seems plausible to expect, money outlays for investment (including tax) will increase. With the increase in demand for funds, interest rates will rise unless the supply of credit is perfectly elastic.


J.R. Hicks, Bulletin of the Oxford University Institute of Statistics (Oxford), Vol. 14 (1952), pp. 268–72.


H.G. Johnson, ibid., p. 303.


Certain exemptions to the tax on ship contracts were granted in June 1955, retroactive to February 1955.


Replies to a questionnaire addressed to about 3,000 Swedish industrial companies indicate that the investment plans of nearly 60 per cent of the respondents were not immediately affected by either the monetary measures of 1955 or the investment tax. It should be noted, however, that small firms were exempt from the investment tax. Among firms which did indicate that their investment plans were curtailed, 42 per cent of the reduction was attributed to the investment tax, 30 per cent to the difficulty of obtaining credit, 8 per cent to the interest rate alone, and 20 per cent to a combination of the measures (International Monetary Fund, International Financial News Survey, Washington, June 15, 1956, pp. 390-91, reporting an article in Svenska Handelsb’anken, Index, Stockholm, April 1956). The central bank’s discount rate was increased from 2.75 per cent to 3.75 per cent in April 1955, and the yield of long-term government bonds rose from an average of 3.37 per cent in the first quarter of 1955 to 3.82 per cent in the second quarter (International Financial Statistics, Washington).


With a further allowance for interest and risk, all these figures will be raised, but the generalization about the relation between the minimum acceptable returns with and without tax will still hold.


The increases in average annual costs would be equal when

tC/n + i(tC/2) = Δi(C/2),

where t = the investment tax rate

C = the original cost of the investment exclusive of tax

n = the economic life of the asset, in years

i = the original interest rate.


The percentage rate of return, net of interest, would fall more with the tax, since the amount invested would be larger in this case. Some investors might consider this a reduction in the reward for assuming risk and a further deterrent to investment.


This assumes that discount is continuously compounded. With a 5 percent discount rate, a 2-year asset costing 100 will be just worth acquiring if it is expected that the cumulative return in each year will be 52.5, for the present value of these returns (distributed evenly over the 2 years and subject to continuously compounded discount) is equal to the cost of the asset. A 10 per cent investment tax would raise the cost of similar assets to 110 and, at a 5 per cent discount rate, they would not be worth acquiring unless the expected return were at least 57.8 per year. Thus the tax would exclude from favorable consideration 2-year assets with returns between 52.5 and 57.8. In the absence of the tax, an increase in the discount rate from 5 per cent to approximately 14.8 per cent would have the same effect, since the present value of expected returns of 57.8 per year would be reduced to 100, the cost of the asset.


Exemptions or tax refunds are granted in many countries to certain purchasers of commodities subject to excise taxes or customs duties.


Minister of Finance, Budget Speech, April 10, 1951 (Ottawa, 1951), pp. 12-13; M.W. Sharp, “Deferred Depreciation Reviewed,” Canadian Tax Journal (Toronto), May-June 1953, pp. 277–86.


Derived from M.W. Sharp, op. cit., p. 285. Mr. Sharp—Associate Deputy Minister in the Department of Trade and Commerce, which administered the regulations granting permits for exceptions to the deferment scheme—concluded that the plan “appeared to produce results of the kind the government had intended” (p. 283).


As indicated by a survey of investment intentions.