FACTORS INFLUENCING THE CHOICE OF A CONTRIBUTORY TAX UNDER THE INDIVIDUAL-SACRIFICE PRINCIPLE. - Several factors influence the choice of a tax that can be deemed contributory under the individual-sacrifice principle.

INCIDENCE OF THE CONTRIBUTORY TAX. - Any tax levied to make the beneficiary contribute under the individual-sacrifice argument must of course have the correct incidence - if imposed directly on the beneficiary, it must not be capable of being shifted by him, and if imposed directly on someone else, it must be reasonably certain of being shifted to him. In view of the uncertainty as to where a tax will be shifted if it is shifted, and of the tendency of a tax to stay where it is first placed, it follows that a contributory tax is most appropriately placed when it is levied fairly directly on the beneficiary. The requirement that the beneficiary be unable to shift it narrows further the field of appropriate taxes. Here, much depends on the size of the benefits and the consequent else of the tax. As the size increases, a level is reached where no tax can be found that will not be shifted in part by the beneficiary, usually through higher wages. In other words, too severe a saving program will require the involuntary saver to get help from others in the community if he is to live save (perhaps after it has first induced him to trade some saving for some leisure - that is, caused him to work less).

GROUPS WHO ARE TO RECEIVE OLD-AGE BENEFITS. - In practice, the kinds of contributory tax that will be appropriate under the individual-sacrifice principle will depend not only on the incidence of the several taxes but also on the kind of groups that are to be placed under the benefit plan. If, for instance, it were desired to include only owners of real estate, an appropriate tax would be some kind of a real estate tax. If the program is designed to cover everyone, some universal tax, or a universal combination of taxes, is required.

SHARE OF BENEFIT TO BE CONTRIBUTED BY ONE BENEFICIARY COMPARED WITH ANOTHER BENEFICIARY. - Within the group(a) to be covered, the proportion of his (actuarially forecast) benefit to be paid by an individual beneficiary compared to another beneficiary will also influence the election of a contributory tax. For instance, if the proportion is to be equal for all taxpayers, a flat rate tax or a flat tax is useful. If the proportion is to vary, a graduated tax is probably needed. If the proportion is to be erratic emphasis being instead on uniform absolute contributions, a flat poll tax is suitable.

VARIATION OF TOTAL BENEFIT AMONG INDIVIDUAL BENEFICIARIES.- Finally, the choice of a contributory tax (still considering only the individual-sacrifice argument) will depend in part on how the benefits are to vary among individuals. If the benefits are to vary in accordance with the total wages earned, a wage tax is useful, since the share-of-benefit standard just discussed can probably then be net easier than if use were made of a tax on total net income, or property, or expenditures. /27/ On the other hand, if the benefits are to vary in accordance with the total economic history of the individual, some sort of a general income tax may be more useful.

CHOICES ASSUMED IN PRESENT ANALYSIS. - Thus the kind of contributory tax that will be best depends upon a number of decisions that must first be made concerning the philosophy of the old-age benefit plan as a whole.

The present writer has no settled convictions on many of these, but to facilitate exposition the discussion below is based on what seems to be the prevailing attitude, namely, that it is desirable to:

1) Include under the contributory system every adult individual in the country, whether working or not, and regardless of wealth or income (but it must also be assumed that below some level a person becomes unable to contribute, if he is to live to enjoy his benefit payment);

2) That a strict 100 per cent actuarial contribution from each individual, or a uniform percentage share thereof, will not be required, but that departure from this extreme shall not be great:

3) That annual dollar old-age benefits will vary among individuals, according to, but not necessarily in proportion to, variations from one individual to another in average lifetime economic status.

In the light of these assumptions the present payroll taxes and, incidentally, other possible taxes will be analyzed, always from the point of view of the individual-sacrifice principle. /28/

PRESENT PAYROLL TAXES UNDER THE INDIVIDUAL-SACRIFICE PRINCIPLE. - Examined in the light of the tax implications, set forth above, of the sacrifice argument, how do the old-age payroll taxes levied under Title VIII of the Social Security Act appear? Attention will be devoted first to the incidence requirement.

INCIDENCE: "INCOME TAX" PART OF PAYROLL TAXES. - From the point of view of the sacrifice argument, the "income tax" /29/ part of the payroll tax - the tax deducted from the employees wage by the employer - is probably moderately satisfactory with respect to incidence, so long as the rate is low enough. If the rate were as high as 10 or 20 per cent, to use an extreme example, the worker's wage (before tax) would almost certainly have to be increased. He would scarcely be willing or able to stand a forced-saving program of such rigor. But at the present rate of 1 per cent, and possibly even at the prospective rate of 3 per cent, it seems unlikely that the tax is being made up in many instances by a corresponding rise in wages.

However, the fact that it is levied along with a similar tax - the excise tax part of the payroll tax - makes it necessary to consider the effects of a 2 per cent to 6 per cent tax. At 6 per cent the tax is heavy enough to raise serious doubts whether it will leave wages unaffected.

Moreover, the facts that the employer is responsible for collecting the tax and that the tax is on the simple base of payrolls without any allowance for special situations (e.g., family situation) facilitate the shifting of the tax to the employer. It is a simple matter, mechanically, for him to agree with the workers to make the deduction only a nominal one, and pay the tax out of the company's funds, or to raise wages by the amount of the tax. A tax on a more refined base, differing among individuals receiving the same gross wages, and collected directly from the workers, would be more difficult for the employer to assume.

If the "income tax" is heavy enough to cause a shift from direct labor to machinery, it causes complex repercussions that cannot yet be traced satisfactorily, but the possible repercussions include both (a) a greater sacrifice by some of the covered workers than they would undergo otherwise and (b) a sacrifice by some of the uncovered workers. There is no evidence yet whether or not the 1 per cent "income tax" is, by itself or in combination with the 1 per cent excise tax, causing a shift from direct labor. To the extent that it does, it is probably, for reasons just given, an unsatisfactory tax from the point of view of the argument now under consideration.

INCIDENCE: "EXCISE TAX" PART OF PAYROLL TAXES. - The 1 per cent "excise tax," the other half of the old-age payroll tax, is levied on employers. If all employers recouped the tax by reducing /30/ wages (net, to the worker) just as much as they are reduced by the "income tax," the considerations in the paragraphs above would apply. Under a very fluid labor market and almost perfect competition among concerns, this result might be expected. At least the presumption would be against a conclusion that the long-run incidence of the "income tax" could be changed simply by calling it an "excise tax" and by providing that it could not be made the grounds for disturbing wages (net, to the worker) under existing wage contracts, when in all other respects it would be exactly the same as an "income tax".

Actually, wage rates in many industries are probably hard enough to change, profits in relation to the tax are probably large enough, and competition among concerns is probably imperfect enough to allow enough avenues for savings in expense or for enough increases in prices to consumers, so that an excise payroll tax as low as 1 percent may not in many - perhaps most - instances cause directly any change in wages. So far as this condition obtains, the excise tax is quite unsuited for the raising of old-age benefit moneys from the point of view of the sacrifice argument now being discussed. It is merely a general tax on consumers or investors, or on those who were benefiting by the lax business practices of the concern before it was put under the pressure of the tax.

In so far as it causes displacement of labor by machines, it is even worse than most general taxes, from the point of view of this sacrifice argument.

At a high rate - say 10 or 20 per cent, as an extreme example - the incidence of the excise tax, after a few years, would presumably be much like that of the "income tax", whatever that incidence might be. At such a high rate the formal differences noted in the third paragraph above would probably exercise relatively little force. At a rate of 3 per cent - the maximum now scheduled - the question is a very difficult one indeed, but the present writer's guess is that there is a good chance that in a number of industries the excise tax, like the "income tax", will for the most part come out of wages, but that so many other ways of disposing of the tax burden will be utilized in other industries that on the whole it would be unsafe to count on the excise tax as satisfying the requirement of the sacrifice argument.

In summary: if the excise tax is light, it fails (as concerns the argument now under consideration) because it does not affect wages; but if it in made heavy enough to affect wages it is likely to share with the "income tax" of the same heavy rate the disadvantages (from the point of view of the sacrifice argument) of causing a rise in rages or a change in some other items.

SCOPE OF BENEFITS. - The "income tax" and the excise tax can be discussed together here, since they have the same scope.

If it is desired to have the contributory old-age system extend to every adult in the country, the wage /31/ base for the contributory tax is not suitable. Its three major defects, from this point of view, are:

1) Not all persons receive wages, at least in quantity enough to be significant for present purposes. The main classes that are thus left outside the scope of the contributory old-age system by a wages tax are:

     a)   The self-employed, chiefly

          i)   Farmers

          ii)  Active owners of unincorporated concerns

          iii) Professional people

     b)   Those who live largely on investment income.

     c)   Unpaid family workers (chiefly wives and mothers).

2) Of those who do receive wages, some are paid under conditions that make administration of a wages tax very difficult. The three major instances are form laborers, domestic servants, and casual workers.

3) Under constitutional restrictions on the Federal power to tax (at least an hitherto generally accepted) a wages tax cannot reach most of the State and local employees. /32/

RELATIVE SHARE OF BENEFITS TO BE CONTRIBUTED BY INDIVIDUAL BENEFICIARY, AND VARIATION OF TOTAL BENEFIT AMONG INDIVIDUAL BENEFICIARIES. - As indicated above, the particular contributory tax selected is apt to shape the old-age benefit program in two ways: in determining what share of his benefit any one beneficiary will pay, and in determining the extent to which the annual benefit will vary among beneficiaries. Thus, a flat-rate payroll tax fits in well with a plan whereby all individuals are expected to contribute about the same proportion of their benefits, and where the absolute amount of the benefit depends on the total wages an individual receives during his lifetime.

It will be convenient to discuss both of these topics - relative share, and benefit variation - together, in analyzing the payroll tax,

A payroll tax as a contributory tax facilitates, and indeed almost necessarily implies, /33/ a system of old-age benefits where the annual benefit varies among beneficiaries directly with (but not necessarily in strict proportion to) the total amount of wages the beneficiary has received. But is there any reason why the benefits should vary with amount of wages received rather than with, say, the amount of net income received, or expenditures made? Similarly, is there any reason why the amount contributed by beneficiaries should vary with their wages rather than their net income or expenditures?

The present writer has no convictions, and scarcely any tentative opinions, on how the benefits should vary. A few random observations may be relevant, however. Of two persons, A and B, A earning x dollars in wages during his lifetime and B receiving x dollars in professional fees during his lifetime, there seems no evident reason why the latter should be excluded from the old-age system and the former included. So too if B receives x dollars instead from business profits, or, instead, from investment income. These questions, however, are more ones of scope (in terms of individuals covered) than of sharing and variation among covered individuals. That if B, instead, earns 1/2x dollars in wages and 1/2x dollars in fees (or business profits or investment income)? Is there any reason (administrative considerations aside) for treating B differently from A? None of importance occurs to the present writer at the moment. What if B earns x dollars in wages and x dollars in fees, etc.? This case seems reasonably analogous to the immediately preceding one.

The payrolls tax, however, differs from other taxes not only as to the items it includes, but as to the offsets or deductions allowed. Compared to a net income tax, for instance, it fails to allow certain deductions that are used in reaching net income. To be sure, the more important of these deductions have already been assumed in the discussion above by speaking of "profits", a term which here implies business gross income minus all business expenses. There remain, however, under the net income tax as we know it in the United States, certain other deductions. Many, perhaps most of them, are regarded by the present writer as unsuitable for the net income tax as a general tax, and he is therefore predisposed to be unsympathetic toward them in a contributory tax. Such are the deductions for interest paid (as a personal expense), and taxes paid (as a personal expense). The limited deduction granted by the Federal tax for charitable, etc., contributions is in part at least a hidden subsidy, and whatever arguments apply for the general-purpose tax presumably also' apply for the contributory tax. But there are other deductions which have a logical foundation in the relative-sacrifice aspects of the income tax as a general tax: personal (as opposed to business) bad debts, and loss of uninsured personal property through fire or other casualty, or theft. Specifically, for instance: if A and B earn the same total wages, have no other source of income, and have identical family obligations and expenditure patterns except that A loses some money through a bad personal debt and also has his house *** destroyed by fire, should A be asked to contribute on much to the old-age benefit plan as B? If not, should then his benefits be lessened (not proportionately, if the system was only semi-contributory)? (The "should", it will be recalled, concerns only what has been referred to as the sacrifice aspect, since that is the basis of the present section of the old-age tax analysis). Tentatively, the present writer in inclined to answer no to both questions, and therefore to see in the income tax some possibilities for desirable adjustments of the contributory load among individuals that are not possible with less complex tax forms, such as the payroll tax. The benefits to be paid might, as suggested above, be computed on the basis of the beneficiary's gross income, while his contributions were computed on the basis of his net income.

What of family obligations? If A and B earn the same wages but A is married and has children, while B is unmarried and has not even a relative to support, should A pay a smaller amount toward his old-age benefit than B? Should he then receive a smaller benefit? What if B is, of his own free will, supporting some incapacitated non-relatives? Many interesting points emerge here (for instance, the chance that the children will be able to help support A in his old age) but cannot be dealt with adequately in the present analysis. At any rate, if such differences in individual economic status are to be considered very significant, the payroll tax is not as well adapted for the purpose as the nut income tax.

The payroll tax - and here it is like the net income tax - fails to take account of how much the individual has saved on his own account. Should the contribution required of the worker decrease as he increases his savings on his own account? If so, should his total benefits also decrease? If the answer to either question is yes, a tax measured by personal-consumption expenditures is convenient, either to replace or supplement the payroll tax.

ALTERNATIVES TO PAYROLL TAXES UNDER INDIVIDUAL-SACRIFICE PRINCIPLE. - The major taxes available as contributory taxes to take the place of, or to supplement, the payroll taxes are probably limited to the general net income tax (or component parts thereof) and a personal-consumption expenditures tax. Constitutional restrictions *** on the property tax make its use by the Federal Government impracticable, and even if the Constitution were amended, the heavy state and local load on property would probably prevent the Federal Government from using it. Moreover, there would be serious administrative difficulties with a property tax designed to be non-shiftable. A flat rate poll tax would probably be unacceptable unless the benefits were the same for all.

NET INCOME TAX. - An individual net income tax, reaching all sources of income reached by the present income tax law, would be highly satisfactory in principle with respect to scope. In principle, it would reach the self-employed and the RENTIERS who are missed by the payroll tax. Some of them would occasionally show no net income and thus would not be included every year, but only rarely would any one of them pay no tax, for this reason, over a long period of years (e.g., an heir slowly wasting his legacy in an unprofitable business). A more serious problem would be the small rise of the farmer's net income unless the current conception of net income were expanded to cover produce grown and consumed by the farmer and his family (but not his farm laborers). Also, unpaid family workers would fail of inclusion unless the conception of income were expanded similarly. In practice, the administrative task of checking on hundreds of thousands of small taxpayers would be so immense that for some time - perhaps permanently - the cost of collection would be high and an appreciable amount of evasion could be expected.

With respect to farm laborers and domestic servants, the income tax would be no improvement, in coverage, over the payroll tax if the latter were extended as far as it could be. As with the payroll tax, care would be required to include a fair estimate of the value of payments in kind - meals and lodging.).

The income tax would be of only slight improvement over an extended payroll tax as to state and local employees (a few of the exempt employees might have enough *** income, as from investments, to have to make at least some contribution to the old-age fund.

With respect to taxpayers now subject to the payroll tax, the income tax would be inferior, since the complications involved in computing total net income would probably necessitate dropping out of the system certain low-wage workers. However, as noted below, anyone whose gross income is below a certain level cannot in fact be asked to contribute to his old-age pension. Any decrease in current consumption would result in his not reaching old age, (or in his reaching it in such a state that he could not get a reasonable reward for his sacrifice. Hence the dropping-out of low-wage workers dictated by administrative considerations under a net income tax (as opposed to a payroll tax) may largely coincide with the standards set by the individual sacrifice principle.

Administrative difficulties might be lessened if it were clearly understood by the low-income taxable man that his chance to receive a pension depended on his paying his income tax, and if he realized that the system were weighted to give him back much more than he put in (plus interest); in fact, a strong enough weighting might even make feasible some additional income taxation of low-income individuals for general tax purposes. /34/ However, great caution is needed in estimating the strength of the incentive. Many taxpayers would not understand what they would be getting by paying the tax; others would be too careless to take advantage of it; and still others would have such a high rate of preference for the present over the future - so much higher than that represented by the interest rate used in arriving at the "advantage" that they would get - that they would actively resist the tax collector.

Thus the income tax would be an improvement over the payroll tax as to scope of coverage chiefly in bringing into the contributory old-age system some - perhaps most - of the people now left outside because they do not receive wages. Whether there would be a net advantage is open to question, considering the number of low-paid and migrant laborers that might have to be exempted - but considering also the counteracting factor, the impossibility of giving a future reward in return for a present sacrifice when the contributor is at a very low economic level.

The tax would have to be on net income - gross income less expenses directly incurred in obtaining the gross income - to minimize chances of shifting.

If the thesis above concerning the very low-income individual is sound, a personal exemption under a contributory net-income tax is a foregone conclusion. The exact amount of the exemption, whether the exemption should be the same for a married person as for a single one, and whether credits should be granted for dependents are matters that will be passed over, here. They are very important but they raise many collateral questions of old-age benefits with which the present writer is not prepared to deal.

A contributory tax based on net income could be collected slightly if at all at the source. Foreign experience has apparently not been encouraging with respect to the practicability of refunds, etc. (Study of the German experience of the past two decades should be helpful on this point.)

Each the same results as with a net income tax would be obtained (as concerns only the scope of the system) by supplementing the payrolls tax by a series of taxes on fees, investment income, and business profits.

As a standard for contributions by each beneficiary and benefit paid to each beneficiary, the net income base has to the present writer, as indicated above, some advantages over the payroll base. It allows the contribution demanded to go up and down with the taxpayer's fortunes and misfortunes, and with his lack or load of family obligations: and at the same time it furnishes the records for basing the annuities on income before deduction of personal exemptions and other non-business deductions - probably a better base than either mare wages or total net income.

EXPENDITURES TAX. - A contributory tax on personal-consumption expenditures would cover in principle every adult without exception, other than those considered to be below the economic level where savings for old-age is possible. Administratively, it would probably be no more practicable than the income tax for the low-income members of the self-employed group and the RENTIER group.

For reaching domestic servants and farm laborers, likewise, it probably offers no administrative advantage over the wages tax or the general income tax. With respect to those now reached by the payrolls tax, the expenditures tax, like the net income tax, would be inferior administratively.

As a standard for contributions and benefits paid, the expenditures base is useful if the old-age benefit plan is to be linked with the volume of private saving. Those who wanted to save on their own account would - since their expenditures would be correspondingly less - pay less into the benefit fund and likewise, ***, would receive smaller benefits.

SUMMARY. - To summarize the discussion thus far of the question posed at the start of this section: why should the old-age benefits be paid from the proceeds of any particular kind of tax?, it may be said that one reason commonly given is what has been called above the individual-sacrifice argument, and that this argument calls for choice among a payrolls tax, a general net income tax, or a personal-consumption-expenditures tax, the choice among them depending largely on decisions (outside the tax filed) concerning the desired scope of the old-age benefit plan, the extent to which each beneficiary is to contribute to his own old-age benefit, and the extent of and basis for variation in the total benefit among the beneficiaries.

Making Each Business Pay Its May

A second argument that bears on the type of tax to be used in financing an old-age benefit program is an economic one. In a vagus form it says that business ought to count the workers' old-age dependency as a cost just as it counts depreciation as a cost. It follows then that the business ought to put its selling prices high enough to cover an old-age benefit cost: if it cannot, it has no economic justification for existence. /35/

This argument evidently calls for a tax that will, like the other cost elements, be included in the price of the product of the business in question. Thus the tax should not be on net profits, because of the strong possibility that it will then not be shifted. A tax on payrolls - if heavy enough to cause shifting of the tax away from both the worker and the owner - meets this test.

So, however, does a tax on gross income from sales - a sales tax. So does a flat tax. Can this argument provide the basis for a choice among these taxes?

If two consumers' articles sell for the same price (aside from any old-age tax), but the cost of one of them (e.g., fresh vegetables) is almost entirely labor while the cost of the other (e.g., lumber) consists to a greater extent of interest, should the price of vegetables be raised relative to that of lumber, in order to reflect the old-age costs and thus direct activity from vegetable raising into lumbering? If the answer is yes, the payroll is the more appropriate base for the employer's old-age tax than sales. On the payroll base the vegetable business will pay more tax than the lumber business. On the sales base it will pay the same as the lumber business, and no force will be put in motion to remedy what has presumably been an uneconomic allocation of the country's resources - uneconomic because one element of true cost has been omitted in computing sales price.

But what if the vegetable industry replies that it has been providing for the worker's old age, but has done so by giving him enough wages to allow him to save if he wants to, instead of giving him a pension when he reaches old age? Who is to be the judge of what "enough" is in this case? Suppose the vegetable industry agree to pay each worker an annual old-age pension after he retires - but then cuts wages by enough to make up for this added expense. Would it be meeting the "depreciation" charge for old age, if it had not been doing so before?

The farther one pushes into this "depreciation" analysis the more elusive the concept becomes. It does in fact seem impossible to say, except in extreme low-wage cases, whether an industry's wages are covering the "depreciation" charge. /36/ In industry that pays an employee (may the president of the concern) $50,000 a year is surely covering the depreciation charge for his old age even if it does not pay him a cent in pension after he retires. Suppose it is, instead of $50,000, only $10,000 - $5,000 - $1,000 --: at what point does the industry fall to cover the "depreciation" charge?

The lower the wage, the more the likelihood that the charge is not being covered and that, therefore, an old-age tax should be forced into the price of the product. But under a flat-rate payroll tax, the lower the wage the loss the tax. What is needed, in so far as the "depreciation" argument holds at all, seems to be a payroll tax on the employer, the rate of which increases as the annual wage per worker decreases. So far as the present writer knows, no one has ever suggested such a tax.

Attractive as the depreciation argument may appear at first sight, it is scarcely well enough developed yet to be given much weight in deciding what kinds of tax to use in financing an old-age security program.

Checking Undue Growth of Old-Age Benefits

A third argument that bears on the kind of tax to finance an old-age benefit program springs from a fear that if the financing device is not carefully chosen, pressure for increasing both the scope and the size of the old-age benefits may divert to the support of the aged an undue proportion of the nation's income and even involve the Government in serious financial difficulties.

The argument usually runs that these dangers will be lessened if the benefit system is financed, in part at least, by a contributory tax. If the beneficiaries pay the cost - or part of it - they are less likely, it is said, to press for more money, and those outside the system will be less eager to have it extended to them.

Some other considerations, however, are also important. Moreover, it will be seen that great importance attaches to the distinction between an increase in the size of the benefits payable to those already covered by the old-age system (this increase will be called "intensification of the program" below) and an extension of the coverage of the system to include more persons (referred to below as "extension of the program").

First, the decisive factor in promoting or checking either intensification or expansion is, not who is paying the existing cost, but who is to pay the added cost. If old-age beneficiaries are contributing all the cost of an existing system of benefits, that fact in itself may be an effective argument for asking the general public to pay part of the cost, and this in turn may make higher benefits possible. As to those outside the system, however, the situation would offer little encouragement. Thus intensification of the program, but probably not expansion, might be expected. Conversely, if the beneficiaries are contributing no part of the cost under an existing system, they may find they are in a week position to advocate increases in their benefits at the expense of general revenues, but non-covered persons would be eager to join this system that seemingly gave something for nothing. Thus expansion but not much intensification would be the result.

Second, the argument in question (namely, that without the check furnished by a contributory system of financing them, the old-age benefits will grow to absorb an undue part of the national income) makes some assumptions about the political forces involved that are probably over-simplified.

As concerns intensification of the program, the net of circumstances most favorable to this argument calls for a covered group large enough to enforce their will but small enough so that there will be enough outsiders left to pay the bill. If only 1 per cent of the voters are covered, their political chances for intensification are not bright, and if 99 per cent are covered, there are but slight economic possibilities of intensification without expense to the covered group. If, say, 40 per cent are covered and are highly organized, and the other 60 per cent are well-to-do and unorganized, the absence of a contributory check may lead to considerable intensification. Thus the importance of this point depends on the stage of coverage at which the system is resting at the time.

As to expansion of the system, the absence of the contributory check is probably most significant when the coverage is very small. After a substantial majority of the population was covered, the fact that the system was not a contributory one would tend to check its growth rather than promote it, since the dominant part of the voters (excepting cases where the minority is very influential indeed) would be adding to their burden as they brought others in.

The present discussion aims, not to show just what will happen, but to indicate that there is a considerable chance for error in the common assumption that a non-contributory old-age system is bound to run away with itself.

Maintaining Morale of Recipients

Another reason offered for the use of a contributory tax concerns the morals of the recipient. By receiving something to which he feels he has a legal and moral right, something that he has himself worked for and saved for, and by receiving it without having to adduce a certain level of poverty, the old-age pensioner is thought to avoid a disintegration of morale that is assumed to accompany free pensions coupled with the means test. /37/

Since this argument turns on (a) the recipient's own emotional reactions and (b) the use of a means test, it does not apply in cases where the aged have become convinced that they have, in some indirect manner, "earned" a right to non-contributory pensions without a means test, and are powerful enough to enforce their demands. Similarly, even where this state of affairs does not exist, this argument does not necessarily require fully earned benefits in the actuarial sense. The fact that some contribution was made, even though it supplies but a small part of the actual benefit paid, in very likely to give rise to a strong feeling by the contributor that the whole benefit is his by right; the more doubtful question is whether the taxpayer in general will consent to spend so much money on unearned (portions of) old-age pensions given with no means test.

General Summary

Of the four chief reasons commonly given for financing old-age payments by some particular kind of tax is appears from the discussion above that one is virtually worthless, at least so far as it has been developed at present (making each business pay its own way), one calls for very different kinds of tax at different stages of extension and intensification of the program (checking undue growth of old-age benefits), and a third requires only a rather light tax of the particular type (maintaining morale of recipients). Thus only one of the reasons (individual-sacrifice principle) supplies a firm basis for a heavy tax of a particular type over a considerable period of time. In general, it appears that the use of a special, contributory tax rests chiefly on this reason, somewhat on the checking of undue growth, and, for practical purposes, not at all or scarcely at all on the other two reasons.