Date 3 August 1942
Author unknown
Title Alternative Methods for the General Control of Consumption
Description Staff memo, Division of Tax Research, Treasury Department
Location Box 38; Supply and Consumption; Records of the Office of Tax Analysis/Division of Tax Research; General Records of the Department of the Treasury, Record Group 56; National Archives, College Park, MD.
Alternative methods for the general control of consumption


Under the usual proposal for compulsory savings, taxpayers would be required to pay, in addition to and at the same time as their income tax, an additional sum which would be invested for them in government bonds. The amount of the compulsory lending would be determined according to income and family status in the same manner as the income tax. The bonds would be non-transferable, except in certain emergencies. Certain forms of the compulsory savings plan provide for the crediting of the amount of certain types of voluntary savings against the compulsory lendings, so that only the balance need be paid by the taxpayer. Various interest and repayment provisions are proposed, but they are not of importance here. Compulsory savings may be collected at source along with income tax. Compulsory savings does not itself directly limit consumption, but merely imposes institutional and psychological barriers in the way of consuming as much as would be consumed without the tax. Persons who are unable to dispose of capital assets without taking a loss, and persons who cannot borrow except at high cost, even by hypothecating their rights to the repayment of their forced savings, will be induced to curtail expenditure to the amount of income left to them after taxes and compulsory savings. Moreover any attempt by a large number of persons to maintain their expenditures by sale of assets would force down the market price of such assets so that the procedure would become still more costly. Also persons who are accustomed to think of their income on a cash basis are likely to curtail expenditures in accordance with such cash incomes and to disregard to a certain extent their compulsory lendings.


If the offsetting of other voluntary savings against the compulsory lending is extended to all forms of savings, and if conversely disavings are added to the liability, the result is a minimum savings requirement. Whether or not some form of tentative collection at source is used, final determination of an individual's minimum savings requirement cannot be made until after the end of the year. If at that time the savings is shown to have been too low, the deficiency may be collected in the form of a compulsory loan. However, there is apparently no method of insuring that this payment will not be made by liquidating other capital assets, with a further deficiency in savings for the current year, so that the deficiency may cumulate and never be made up. If a penalty is attached to a deficiency in savings, in the form of a fine or supplementary tax payment, the effect is similar to an expenditure tax. If there were some method of assuring that the minimum savings requirement were met, this would be the approximate equivalent of expenditure rationing with sale of ration rights prohibited.


The spendings tax is a tax at graduated rates on personal expenditures in excess of certain personal exemptions. The tax is collected on the basis of individual returns in the same manner as the income tax, after the end of the year. Interim collections can be made on payments of income, at the source, in the same manner as part of the income tax is collected at the source. Rates may be graduated in such a way as to render expenditure at high levels very costly, even prohibitively so, requiring the payment of several dollars in taxes for every dollar of expenditure above given levels. Consumption would be curtailed not only by the direct reduction in disposable income, but by the incentives provided by the fact that money not spent would bear no tax, and even if spent after the emergency would probably bear a lighter tax or none at all because of the probable reduction in the rates or repeal of the tax.


Expenditure rationing involves requiring that all purchases for personal consumption be made with either ration money or money accompanied by coupon to the same denomination. Ration money or coupons would be issued to individuals according to a schedule taking in to account their income, family status, and perhaps other factors such as age, average income over a more or less extensive period, or other special circumstances. As ration money would have to be issued before it could be spend, the income base would have to be some previous period, say the previous quarter or the previous year. For those whose principal income was from a single source, such as wage and salary earners, the wage or salary could be issued in the form of the appropriate amounts of ration and non-ration money directly. Bank accounts, etc., would be segregated into ration and non-ration funds. Exchange of ration money for non-ration money among individuals might be prohibited, in which case enforcement would probably require the filing and auditing of a return on the lines of the minimum savings requirements, or such exchange could be permitted, adding a degree of flexibility to the operation of the scheme.

Comparison of spendings tax, expenditure rationing, and compulsory savings as a means of controlling consumption



Expenditure rationing has the principal advantage that a definite limit can be put on the aggregate of consumer expenditures, and this limit can be changed from time to time as need arises by definite amounts. A "minimum savings requirement" can also set a fairly definite limit on the aggregate of consumer expenditures. A compulsory lending plan, however, places no absolute restriction on consumption, particularly on the consumption of the wealthy, who are free to perform their compulsory savings through liquidation or borrowing on accumulated capital.

On the other hand, while it may be difficult to foretell exactly the effect of a given spending tax rate schedule on total expenditure, the margin of error should be well within the amount of free purchasing power which the price control mechanism can withstand, or which in the absence of price control would lead to a not intolerable price rise.


Both compulsory savings and the spendings tax suffer from the defect that unless some form of collection at source is instituted, their influence is likely to be felt only after a considerable delay particularly with respect to the spendings of the lower income classes. Even with collection at source of a major portion of the spendings tax, the effect on the middle and higher income classes may be largely in the form of a more or less remote liability for which taxpayer may fail to make adequate accrual and so fail to curtail their consumption to the degree desired for a considerable period of time. Even at income levels where the tax is almost entirely collected at the source, the nature of the tax and the extra charge attached to current spendings may not be fully and immediately brought home to the taxpayer.

On the other hand, expenditure rationing involves a completely new and extensive administrative machinery the setting up of which would require extensive preparation, before the plan could be placed in effect, whereas the setting up of a spendings tax, even with collection at source, would require only an addition to the withholding rate before the effect of the tax would begin to be felt, and whatever administrative machinery would be required to audit the final returns could be set up during the taxable year. To the extent that the accrual of the tax liability alone would curtail expenditures, a substantial effect may be felt even before any machinery is set up.


Should it be desired to change the severity of the restrictions on consumption more frequently than at yearly intervals, no one method appears to have any great advantage in this respect over the others. The schedules for the issuing of ration money or other rationing devices can probably be changed on about as short a notice as rates of withholding. The expenditure tax may have a slight edge in that with respect to that portion of the tax not subject to withholding, there need be almost no delay between the announcement of the new schedule and its taking at least partial effect. If the final tax liability is determined on the basis of some arbitrary allocation of the total expenditure for the year to the periods before and after the increase in rates, there may be some need for relief in cases where the taxpayer can show that a substantially larger portion of his expenditure was made in the earlier period than was assumed.



The spendings tax operates with a. high degree of flexibility in determining the amount to be spent by various individuals. Any individual is at liberty to expand his consumption if he is willing to pay a more or less severe tax penalty for the privilege. Conversely) any individual who is willing to curtail his consumption more stringently will obtain a substantial reduction in tax.

Expenditure rationing, on the other hand, will leave very little flexibility if the exchange of ration money for non-ration money, or the sale of rationing coupons or other devices between individuals is prohibited. There will be no possibility for individuals for whom the ration fixed according to the formula adopted happens to be small relative to their needs to expand consumption, nor will there be any incentive for those for whom the ration happens to be greater than the need to reduce their consumption below the ration. Even with the best of enforcement procedures there is likely to be a considerable amount of clandestine sale or exchange of ration money or its equivalent.

If the sale or exchange of ration money or its equivalent among individuals is officially sanctioned, a market should develop in which the rate of exchange between ration money and non-ration money becomes fairly uniform. This would provide a degree of flexibility similar to that under the expenditure tax, but differing in its relative impact on different income groups. Under the exchange of ration money) the money price to be paid for expenditure beyond the ration would be the same in all cases., and the premium paid for consuming less than the full ration would be also constant. Under the spendings tax the cost of additional consumption would bee graduated according to the scale of expenditure; and the scale of graduation can be varied to attain different objectives. From one point of view it would seem that such a graduation in the "rate of exchange" would be preferable in that it would place the pressure for restriction of consumption where it is most needed. There is some danger that a constant rate of exchange would induce a curtailment of consumption at the lower end of the income range that would be more than that which is conducive to efficiency while at the upper end of the income range the rate might be insufficient to prevent the continuance of spending on an unduly large scale. From the point of view of the individual, expenditure rationing may be considered the equivalent of an expenditure tax at a flat rate equal to the rate of exchange established on the market, plus a reduction in the income tax of an amount equal to the expenditure tax on an expenditure equal to the ration, plus a subsidy to lower income groups of the amount by which this tax would exceed the income tax.


If the expenditure ration is determined only according to income, family status, and possible, age, which appear to be the only factors that could readily be applied in a rigid formula, it appears that certain moderate hardships will arise, particularly in the case of persons who have retired and are living on their accumulated savings, and persons whose income is temporarily impaired. Provision for determining the ration on the basis of an average income might take care of some but not all of the cases of temporarily impaired income, and would involve considerable administrative difficulty. Provision for special treatment of those over say 65 might help the retired cases, but a method would be extremely difficult to work out which would grant adequate relief without being overly generous in other cases. This would still not take care of cases where the taxpayer is unemployable or where his most productive years are over at an early age. It is probable that no completely satisfactory solution would be obtained on the basis of the application of rigid rules. Possibly some method of granting relief in individual cases through local boards might be worked out. But this would entail considerable administrative machinery and effort on the part of individuals which might be better devoted to other purposes. There has been sufficient lack of uniformity in the action of local boards with regard to matters where the answer was a simple yes or no; where the answer must be quantitative) there is considerable likelihood that the lack of uniformity would be considerably greater, even with much more definite directives from central authority.

Similar comments apply to a minimum savings requirement.

Permitting the sale of rations would alleviate the hard cases somewhat, but the alleviation would probably be too inadequate to rely entirely on such sales to take care of these cases. This is particularly true if the rationing is stringent enough to drive the price of ration money to a high level. There compulsory savings is so subject to avoidance as to produce few if any hard cases. The only hard cases likely to arise under the spendings tax are the relatively few instances of individuals with large incomes who have committed themselves to a large outlay on personal consumption in the curtailment of which they find exceptional difficulty. This problem arises under expenditure rationing with respect to all income levels and a larger number of cases, but may be more acute in some cases under the spendings tax at the top end of the scale where the rates may be substantially higher than the rate of exchange under rationing. These cases are not likely to be frequent. About the only case that comes readily to mind is the case of the owner or renter on a long term lease of a large estate which is so located that it cannot be adapted to serve other purposes. Even here the hardship is likely to be relative rather than absolute, and usually the owner will already be afforded considerable relief through the exclusion of a part or all of the interest on the value of the property from his expenditure.


Within limits, the relative degree of inequality of the resulting expenditure patterns can be controlled by suitable alterations of the rate schedules in the case of any of the alternative methods. However, expenditure rationing probably cannot out off expenditure at the top of the scale (unless trading in ration money is prohibited as sharply as can the spendings tax, and compulsory savings is still less effective in this respect. On the other hand, at the bottom of the scale the best that the spendings tax can do is leave consumption unaffected through exemptions; expenditure rationing, on the other hand, may give the individual with a small income, who would normally save some of it, an addition to both income and consumption through the sale of surplus ration money, if at such levels the entire income is paid in ration money, or even more so if separate ration coupons or other devices are issued in a fixed minimum per capita amount. However, if the ration money device is used, the effect is extremely limited) and disappears at levels at which individuals normally would spend their entire income; and a flat minimum ration regardless of income has the disadvantage of greatly increasing the total ration issued to the lower income groups) makes a relatively large sale of ration coupons inevitable, and tends to lower unduly the rate of exchange, which in turn further decreases the pressure on consumption at the top of the range.


Comparison of the spendings tax on the one hand and the expenditure rationing or compulsory savings on the other hand must take into consideration the progression of the taxes that may ultimately be levied to pay the interest and principal on the increased national debt that would be left under the latter two alternatives. As this involves a high degree of speculation is to the remote future, this will not be considered here except to remark that it is at least theoretically possible to levy at some future date, when total revenue demands are not so severe, taxes that are on the whole more progressive then additional taxes levied at a time when the revenue possibilities of the higher income brackets have already been heavily exploited.

In general the shift from expenditure rationing to the spendings tax involves taking additional revenue on a progressive scale from the middle and top income classes, and transferring a relatively small amount of income from the very low to the very high income classes. Thus the immediate burden on income is both larger and more progressive under the tax than under rationing.



Compulsory lending involves perhaps the least additional administrative machinery of any of the plans, but it is also the least effective. A minimum savings requirement or the spending tax will require a slight amount of additional administrative machinery, in that new data on savings items will have to be reported and audited. The minimum savings requirement would present somewhat greater difficulties, since some method of preventing continuous postponement of the required savings must be devised. Expenditure rationing would require a large amount of new machinery including either a complex reorganization of banking methods, or the issuance of a large and cumbersome system of vouchers and coupons. If transfer of rations be prohibited, the system would be that much more complicated, and would require a checking of returns similar to those used under the minimum savings requirement.


Compulsory lending again involves the lest serious enforcement problem, in that there is no serious incentive for avoidance. A minimum savings requirement is probably somewhat difficult to handle because of the possibility of cumulative liquidating of other securities and the resulting postponement of obligations. The incentive for permanent avoidance is moderate, being in terms of timing of expenditure rather than evasion of a final obligation, but tends to grow as wealth is accumulated beyond the individual's needs and the urge to attempt to spend it increases. The spendings tax may require fairly careful auditing and investigation, particularly of those subject to the higher rates, in order to prevent evasions. A further problem is that the checking on gifts or exchanges of both money and goods between individuals. The problem of enforcing expenditure rationing is probably the most serious, for although the incentives to evasion are comparable to those of the minimum savings requirement, and less than under the expenditure tax, the intricacies of its operation may prove extremely difficult to police. If there is a prohibition against the sale of rations, the problem is made still more serious; a complete enforcement of such a prohibition's probably beyond any central administration but must be achieved largely through the cooperation of patriotic individuals. The only possible central check is through auditing statements of sources and use of funds made after the end of the year.



The spendings tax will yield a substantial amount of net revenue, even under the most sharply graduated scale that is likely to be adopted. (The only scale that would yield no revenue would be one that was zero up to the point at which consumption is to be curtailed, and prohibitive thereafter. This would be a completely egalitarian distribution of consumption and would be practically out of the question.) Neither expenditure rationing nor a minimum savings requirement will yield any net revenue; they remedy stimulate the non-inflationary purchase of bonds with the money that cannot be spent. From a strictly current point of view there may be little difference; but from the long run point of view there is a substantial difference. Any needless increase in the national debt at this time is undesirable from almost every point of view. From the conservative point of view a large national debt imposes a severe problem of repayment, both the interest and principal. Even though the actual payment be not regard as a burden on the national economy, the process of payment imposes strains on the tax systems and compels resort to less desirable forms of taxation than would otherwise be necessary. The expectation of repayment through the imposition of such taxes will be considered an additional hazard by business men which may inhibit investment and embarkation upon new enterprises. In addition a large national debt has a certain inhibitory psychological effect upon certain groups in and of itself.

Even if the view be adopted that neither principal nor interest on the national debt need over be paid off but that all payments can be refunded as new obligations as they become due, the existence of a large national debt may be at least a psychological barrier to getting an adequate deficit spending program in case such a program is needed to obtain full employment at some future time.

The accumulation of a disproportionately large national debt has still further disadvantages for the point of view of developing a free and stable system of international trade and exchange. Changes in even a strictly internal debt will have some effect on the relative price levels and may interfere with foreign exchange rates. A degree of inflation which might be tolerable from a strictly domestic point of view might have fairly serious repercussions on international trade. And where these effects may be the result of financing a relative small part of war expenditures through taxation, it may prove politically difficult to adjust in a satisfactory manner the disequilibria which may arise.


Of all these programs, the spendings tax is the only one that is acceptable as a permanent part of our peace time fiscal set-up. Thus for this program there is little or no problem of liquidation other than a gradual reduction in the rates to normal levels.

On the other hand, under any of the other programs, the end of the emergency will find large amounts of government bonds in the hands of individuals, probably greatly in excess of the amount that they will be willing to hold as permanent savings. Under these conditions controls cannot be relaxed without in turn producing a severe inflationary tendency. In fact is is difficult to see how any substantial part of these sent up savings can be released into the flow of purchasing power without resulting in a substantial inflationary boom. Two alternatives are present, the first is that the purchasing power be absorbed by taxation for the reduction of the national debt as rapidly as the controls are relaxed. This is not so say that the controls must be kept until the war debt is entirely paid off, but only until it is reduced to moderate levels. The amount of public debt that will be held voluntarily will have been presumably increased because of the increased national income, because of waiver amount of inflation in prices is allowed to remain, and to some extent because of frictions tending to prevent the volume of savings held from falling quite to its previous relative level. The other alternative is that the controls be relaxed gradually enough so that the amount of resulting inflationary pressure is manageable and can be absorbed through administrative controls and general frictions throughout the economic system.

In some case, the non-tax types of controls must, if they are to be effective in preventing inflation, be retained for a much longer period than merely the duration of the war or the period during which an abnormally small part of our resources are being devoted to production for domestic civilian consumption. Consideration must be given to the fact that the administrative costs and the attendant bureaucracy must be endured for a much longer time than is ordinarily thought necessary if any of these courses are chosen, and must be extended into a period in which the patriotic motives inducing their ready acceptance as part of the war effort will be much weaker.


Under the spendings tax, the tax is a permanent transfer, and each new year finds the need for inflationary control much the same, changing only with respect to the relative amount of goods available for civilian consumption.

Under the non-tax types of control, however, each successive year finds the individual with a larger amount of savings of various kinds on hand, which savings will soon amount far beyond the amount that most individuals will consider as being needed. The desire to convert these savings into immediate consumption will thus increase as the individual is trust further and further away from his equilibrium position. This will express itself in increased attempts at evasion and enforcement difficulties. Where under expenditure rationing sale or rations is permitted, this will express itself in a higher price for ration money in terms of non-ration money. This will in turn increase the rigidity of the ration in that individuals who find their ration deficient will find it increasingly difficult or even impossible to increase their ration to the extent that their real needs exceed the ration allowed under the fixed schedule. The need for adjustment of rations by local boards or other agencies will be increased.

A compulsory lending program is likely to break down completely unless auxiliary controls are introduced, since after the accumulation of savings has reached a certain point, the pressure to spend will tend to break through the institutional and psychological barriers to the conversion of the forced loan into spendable funds.


Of the various methods, compulsory lending may be reject as being inadequate, capricious in its effects, and unsatisfactory as to the limitation of the consumption of the wealthy. The prohibition of the sale of rations under expenditure rationing appears to have no practical value aside from a psychological one, and involves serious difficulties. The advantages of collection at source in connection with the spendings tax appear to greatly outweigh any minor increase in administrative difficulties that might result. The minimum savings requirement probably cannot be effectively enforced without recourse to either expenditure rationing or what amounts to a form of spendings tax. This leaves the spendings tax with collection at source and expenditure rationing with sale of rations permitted as the two most acceptable alternatives.

On the basis of effectiveness of control over total expenditure, there is little to choose; rationing is more predictable as to its total effect, while on the other hand the expenditure tax could probably be introduced with less delay. With respect to the distribution of expenditure among individuals, the spendings tax appears to have the edge. Flexibility is concentrated more nearly where it is most needed, so that fewer hard cases are likely to arise; there can be more stringent restriction of consumption at the top of the scale. On the other hand, rationing may provide for a greater apportionment of consumption at the bottom of the scale than is possible with the tax, albeit at the expense of further loosening control at the top of the scale. The effect on the long term distribution of income depends on one's anticipations as to tax program after the emergency.

With respect to administration, rationing would require much new machinery, the tax relatively little; with a corresponding relative difficulty of policing; on the other hand, the incentives to evade would in may cases be greater among the higher income groups under the tax.

From a long run point of view, the tax has considerable superiority. There is little or no problem of liquidation of the controls, the revenue produced diminishes the problems connected with the national debt, the sent up purchasing power to be released is of much smaller magnitude, and there is less of a tendency for flexibility to diminish and incentives of evasion to increase through the passage of time and the accumulation of sent up purchasing power.

On the whole the spendings tax appears sufficiently superior to expenditure rationing to warrant its preference unless some outstanding advantage of expenditure rationing can be adduced which is not covered in the foregoing analysis.

Treasury Department Division of Research

August 3, 1942