Insuring the Future: The Burden of Old-Age Security

I

THE old-age reserve plan in the Social Security Act may best be understood if it is looked upon as a device for solving the fiscal problem of the government resulting from the existence side by side of a government debt approaching 40 billions and an old-age insurance programme entailing heavy future financial burdens. Looking forward to the time when annual old-age pension benefits will have grown to exceed annual payroll tax receipts by 1.5 billions or even more, the country may well be concerned if the budgets of that day shall have to provide not only for a large subsidy to the old-age insurance system but also for interest of more than a billion a year on the government debt. The double load would bear heavily upon the taxpayers.

To solve this fiscal problem the reserve programme would, in effect, use excess payroll taxes to take up existing publicly held government debt, which would then be lodged in the Reserve Account and bear 3 per cent interest to be used for old-age insurance purposes. The old-age insurance appropriations required from general revenue funds would therefore be correspondingly reduced.

Contemplating this programme, we may well ask whether the burden of thus solving the fiscal problem of the government has been placed where it belongs. Does either a direct tax paid by workers on their earnings, or an employer’s payroll tax, which in large measure will be passed on to workers through increased prices of the necessities of life or through retarded wage increases, provide the best method of putting government finances in shape to meet the pension load a generation hence?

Those who support the current-cost plan of operation believe that the fiscal problem should be solved in what they conceive to be the normal way. That is to say, they would balance the budget without reference to the payroll taxes and leave a margin for normal debt reduction.

This might require new sources of taxation. On the other hand, it might not. It is certainly possible that those in authority in Washington will arrive at an understanding of the manner in which our economic system operates, with the result that business activity will reach and remain on a satisfactory level and the budget, including a sum for debt reduction, will come into balance through a decrease in relief expenditures and an increase in revenue from existing taxes.

If the national debt could be greatly reduced in the normal manner over the next thirty or forty years, the interest would be largely out of the way when the time came to make appropriations from general revenue funds to assist in meeting the old-age benefit payments — in which case payroll taxes would be used almost wholly for current benefits. A small margin should be retained to create a moderate contingency reserve — not exceeding, say, two or three years’ current benefit payments — to be drawn upon in years of bad business when tax revenue would fall off but the benefits would not.

II

If, as argued by those believing in the current-cost theory, payroll taxes could be devoted almost exclusively to the payment of current benefits, the old-age benefits in the earlier years of the system could be substantially increased, and workers retiring in those years — that is, workers who are now aged forty-five or over — would receive more adequate benefits. The desirability of this change will be evident from the following table showing the estimated average monthly old-age benefits payable in various years under the present plan.

As we contemplate this series of figures, we should keep in mind the free pensions available under Title I of the Social Security Act providing old-age assistance for those now old and in need, and for those who in the future will become old and be found to have insufficient income. This group in the future will include not only the substantial proportion of the population altogether outside of the insurance plan, but also those in that plan who will receive old-age benefits that are inadequate according to current tests in their respective states, and who will therefore be entitled to supplementary free pensions.

When we realize that even at this early stage the average free pension in nine states is above $25 and in twenty states above $20, it is evident that trouble is likely to develop between the two systems. Why pay taxes and receive a $20 old-age income when Brown down the street, who was not in the insurance plan, receives free a larger amount after the local board in his vicinity has found him to be in need? The situation will be all the more aggravated by the fact that under Title I each member of an aged married couple is eligible for a free pension. In a state where the average free pension per individual is, say, $20, the pension per couple when both husband and wife are aged sixty-five or over is likely to be at least $30. I understand that in California, where the average free pension per individual is now over $32, the average per married couple is over $60.

Unless I miss my guess, this kind of situation will cause a great deal of questioning among those included in the insurance plan. They are going to find it hard to understand the advantages to be derived from paying taxes to receive oldage income after the pattern of the above table when, without paying a cent, those not in the insurance plan are able to qualify for generous free pensions simply by showing that they are in need.

It is interesting that one of the great advantages originally claimed for the insurance plan was that it would eliminate the ‘dreaded’ means test by which it was determined whether or not a person was in need. Well, we do not have to look far to perceive that, even for those in the insurance plan, the means test is going to be with us for a long time to come, since the old-age income under that plan will be relatively so low for those reaching sixty-five in the next twenty years.

III

The basic trouble with the old-age insurance plan as it is now set up is that it relies more upon the principles of individual private life insurance than upon the principles of social insurance. Under individual private life insurance, old-age incomes build up slowly. Time is required to accumulate enough to provide an adequate pension. Hence, if a person waits until after age forty-five or fifty to start to accumulate for old age, it is obvious that he can accomplish his purpose only by paying a large premium each year. The reason why the average monthly old-age benefits are small during the early decades is that they represent payments to be made to those now middle-aged or older, who have relatively few years to go before they reach age sixty-five. Had the principles of social insurance been applied, the benefits in the early years would have come much closer to the ultimate $44 per month instead of beginning at the low rate of $17. In other words, the middleaged and older workers of to-day would not have been penalized so heavily because society had not long ago established a social-insurance system.

We obtain an insight into the way the old-age benefit formula should have been set up by observing what a long-established business corporation would have done if it had put into effect a modern pension plan on January 1, 1937. The corporation would have found that in the case of employees then aged sixty the regular contributions to be made currently on their behalf would provide very small pensions indeed for their retirement five years later at age sixty-five. Assuming that in 1980, a generation hence, the average pension would reach $44 a month, the corporation would have taken from its own funds enough to raise to a figure as near $44 as possible the average pension payable to long-service employees retiring in the early years of the system’s operation. If it could have found the money to do so it would certainly have raised the initial average above $17. That, in a nutshell, is the trouble with the benefit plan in the Social Security Act. It should have adopted average payments in the early years more nearly comparable to the average ultimately to be attained.

But, it will be objected, that would cost too much money. Well, would it? Remember that the number of insured workers reaching age sixty-five each year is relatively small, — in the neighborhood of 300,000 at present, — so that the full quota of persons receiving old-age benefits will not be attained for many years. Suppose, for illustration, that the average benefit in the early years should be set at $30. For about two decades this amount would exceed the average benefit under the present schedule. The cost would rise to about 3 per cent of payrolls by 1950 and would remain below 6 per cent until after 1970. These estimates rest upon the basic data underlying the original calculations in connection with the Social Security Act, and hence cannot be considered final. However, they give an approximate picture of the situation under discussion. As we weigh the social gains that would accrue from paying larger benefits in the early years, does it not become evident that the old-age insurance plan would be improved by revising its benefit schedule to conform more nearly to the principles adopted by business organizations in setting up their own private pension plans?

In considering the broad problem of benefit schedules we must be concerned about the burden that is being passed on to the future if we hold out the prospect of paying benefits amounting to 10 or possibly 12 per cent of payrolls. Would it not be much wiser to increase the present benefits in the first twenty years of operation of the system and to be conservative with reference to the amounts promised after that time? By doing in the near future what we can reasonably afford to do and gaining experience as we proceed, we shall be better able to develop a system likely to prove reliable in the long run.

The British old-age insurance plan is an interesting example of the adoption of thoroughgoing social-insurance principles. During 1926 and 1927, old-age contributions were first collected from the workers. After this two-year qualifying period, benefits were paid to those reaching age sixty-five. The benefits were at the rate of $11 per month per single man, or $22 per month per married couple. These benefits are the same as those paid to insured workers retiring now or at any time in the future. Great Britain did not proceed on the theory that it was wise in a social-insurance plan to promise larger benefits in the future than were paid at the outset. In consequence her programme is likely to prove stable and dependable over the years. Granting that because of our higher wage levels the benefits in this country should be on a more generous scale, and that a uniform benefit would not be satisfactory here, we should have done well to follow more closely the social-insurance principles adopted by Great Britain. Furthermore, the British plan is operated upon a current-cost and not upon an actuarial reserve basis.

IV

Mr. Eliot pays much attention to the year 1960 and relates to that year the well-known estimate that the cost of oldage benefits will eventually reach about 9½ per cent of payroll. As it happens, the estimate indicated that the 9½ per cent would not be reached until 1980 or thereabouts. The estimate for 1960 was only 4.18 per cent, well within the receipts from the then current 6 per cent payroll taxes. Even allowing for revision based upon new data, the 1960 figure would be considerably below 6 per cent. The argument for a reserve under the present plan concerns itself, therefore, with what will happen considerably after the year 1960.

Mr. Eliot takes the position that ‘honest bookkeeping’ requires a reserve account in order to show that the government owes a debt to millions of its people, payable in the future. As I see the problem, honest bookkeeping is not the issue. Rather it is the proposed use of payroll taxes to solve the fiscal problem of the government previously discussed. As we have seen, this places a heavy tax burden upon the wrong groups and makes it necessary to hold down oldage benefits in the earlier years.

I agree with Mr. Eliot that the country should be made aware of the magnitude of the burden it is assuming under the old-age insurance plan. One rather technical way of accomplishing this is to express the liability as a capital sum of tens of billions of dollars representing the net present value of obligations extending indefinitely into the future. Another, and one that is quite easily understood, is to prepare a schedule showing year by year the anticipated receipts from payroll taxes, rising to somewhat more than 2 billion dollars by 1980, and the anticipated benefit payments rising to 3.5 or possibly 4 billions by the same date. The difference between benefit payments and tax receipts represents the amount that must be raised from other tax sources.

In considering this subject of so-called honest bookkeeping we wonder why those who insist that it necessitates a reserve do not become exercised over the joint obligations of the states and the Federal Government to pay free pensions under Title I of the Act. Two estimates of Title I payments were made when the Act was being formulated. One was made by the staff in Washington and the other by the actuarial consultants. The first yielded payments running up to 600 millions a year; the other, payments running up to 1.4 billions a year. The capitalized value of the first set of payments, using 3 per cent interest, is about 18 billions; that of the second, about 39 billions. To date we have not heard it urged that honest bookkeeping requires the establishment of a reserve account for either of these sums.

Perhaps it will be argued that the payments under Title I are not contractual, whereas the old-age benefit payments are. Noting, however, the next to the last section of the Social Security Act which reads ‘The right to alter, amend, or repeal any provision of this Act is hereby reserved to the Congress,’ I doubt if any of the benefits provided by the Act can be considered contractual. In fact, I am convinced that no long-range plan now adopted will be carried out unaltered if a future generation shall find that changes must be made to avoid serious consequences to the economic life of that day.

In setting up a comprehensive old-age security programme based upon socialinsurance principles, it is important not to look upon it as an individual savings or thrift programme analogous to individual private life insurance. It is, rather, a plan under which persons contribute during their working days to the support of the system, and thereby acquire, as it were, a claim upon society to receive support when old age arrives. In essence each generation will devote as much of its current production of food, clothing, and the other necessities of life to the care of those then aged as it deems right and proper. Any attempt on our part now to dictate rigidly what that proportion shall be is likely to go far astray if future conditions turn out to be quite different from what we anticipate.

V

Much ado has been caused by the manner in which the small reserve already created has been handled. It has been charged that workers’ payroll taxes have been used to pay current running expenses of the government rather than to create a real reserve.

In considering this situation it should be kept clearly in mind, as I have said, that the reserve plan is intended to serve in effect as a means of using payroll taxes to transfer publicly held government debt to the Reserve Account so that interest on that debt may be used in the future to pay old-age benefits. When the budget is in balance the theory can be made to work, and our concern then is whether that is the best way to prepare to meet future obligations. But when there is an unbalanced budget the theory falls down.

When the budget is not in balance, and the deficit is at least as large as the available funds in the Reserve Account, it is not possible to reduce the debt in the hands of the public. When the Treasury sells securities to the Reserve Account there would be little point in using the proceeds of the sale to pay off existing debt if it were necessary immediately to turn around and borrow a like amount of new money from the public. Thus there is a basis for the charge that payroll taxes are now being used to pay current expenses. Everything has been done in accordance with the law, but the reserve theory has failed in practice. The net debt of the government has gone up, and there has been an interruption of the programme to reduce the future burden of interest on the publicly held government debt so that old-age benefits may be paid more readily a generation hence. Indeed, the interest burden has been increased so that it is likely to be harder rather than easier to bear the future pension load.

This necessary consequence of an unbalanced budget must give us concern regarding the working out of the reserve plan in general. Large sums of easy money which can be obtained for current expenditure simply by borrowing from the Reserve Account are likely to lead to a chronic state of unbalanced budgets. In that event, what is happening now would become the regular procedure. Treasury securities purchased by the Reserve Account would represent an increase in government debt and not in effect a transfer of existing debt from the hands of the public to the Reserve Account in accordance with the reserve theory.

If the law were amended to provide for the current-cost method of operation, this danger would be largely eliminated. Current tax receipts would be used to pay current benefits with no more margin than necessary to create an appropriate contingency reserve, and the huge excess of income over outgo contemplated under the present plan would not materialize. These considerations constitute a strong argument for the current-cost financing of old-age security supplemented by a normal method of debt reduction.

VI

In conclusion let me sketch in broad outline a method by which we might alter the old-age insurance programme so that it would more nearly meet the needs of the country. First develop a series of trial formulæ providing benefits more in consonance with social-insurance principles than are the benefits now provided. Seek the best possible estimates of the year-by-year cost of these various formulæ and then choose the one which best serves the social needs and at the same time keeps the financial burden within the capacity of the country to bear through taxation. In particular be careful about the ultimate benefit load forty or fifty years hence, when the maximum number of people receiving benefits will have been reached. Set the payroll taxes at rates necessary to pay current benefits, with some such upper limit as 5 or 6 per cent. Until the benefit payments amount to the tax receipts, keep the tax rates low, with no more margin than would be required to accumulate a reasonable contingency reserve. When benefit payments rise above tax receipts, appropriations from general revenue funds will be required. Since the national debt by that time should have been reduced in the normal way and the coverage would have been substantially increased to include groups of workers now excluded, sums that would have been used as interest on the debt could then properly be applied to pay old-age benefits.

Whether the present 2 per cent payroll tax rate should be increased as per schedule to 3 per cent in 1940 would be determined in the light of the new benefit formula to be adopted. Of one thing I am certain. It will be a lot easier to put the increase into effect if old-age benefits are actually being paid in 1940 so that the country may be able to observe some tangible results in the form of old-age income payments. The more than thirty million taxpayers will have to be shown a good and sufficient reason if they are to approve a 50 per cent increase in the tax rate. I doubt if the reserve theory, which attempts to solve the government’s fiscal problem by placing a heavy financial burden in the wrong place, would constitute such a reason in the eyes of workers.