standard of living on retirement. The standard of living of workers obviously tends to vary with their earnings. Since the purpose of old-age security is to protect men against having to make too drastic cuts in their standard of living when they retire, pensions should vary in some measure with past earnings. Flat benefits, which are not related to earnings but are the same for everyone, would assure that there would be only an arbitrary relationship between men's standards of living before retirement and their standards after retirement. Under a scheme of flat benefits, the benefits that would be about right for some workers would be too small for many others and possibly too large for a third group.

In the fourth place, a Federal system of old-age pensions should be the founda tion of the country's system of old-age security because it can be applied to all members of the labor force and can be made as broad, therefore, as the problem with which it is supposed to deal. This method of meeting the problem does not depend upon the willingness of an employer to grant pensions or upon the bargaining power of unions and their ability to compel employers to grant pensions. Furthermore, a system of old-age insurance can be applied to the self-employed as well as to employees. Since about one out of five workers in the United States is self-employed, it is necessary that the scheme of old-age security be applicable to the self-employed as well as to employees.

III

What are the reasons for believing that the present Old-Age and Survivors Insurance Act is not doing the job expected of it?

The members of the Advisory Council were unanimous in believing that the present Old-Age and Survivors Insurance Act is not doing the job expected of it. There were three principal reasons for this conclusion.

One reason was that, even after 13 years, only about 39 percent of the male workers who reach 65 years of age would receive benefits if retired. Hence, if is plain that many people who need to receive protection from the system are not getting it. The principal reason why they are not receiving protection is that the Old-Age and Survivors Insurance Act applies to only about 3 out 5 jobs. Another reason is that eligibility to receive benefits is determined by the propor tion of time that a man spends in covered employment, not by the proportion of time that he is at work. A man may be quite steadily employed and still not qualify for a pension if he is one of the many workers who move back and forth between manufacturing, which is covered, and agriculture which is not. At the present time, half again as many people are drawing old-age assistance as are drawing old-age pensions- 2.7 million are receiving old-age assistance and about 1.9 million, old-age pensions.

A second reason for believing that the Old-Age and Survivors Insurance Art is not doing the job expected of it was that the average pension is too small and is considerably less than the average payment for old-age assistance. The average pension for a single person averages about $26 a month and the average pension of a retired person with one dependent less than $40 a month. The average monthly payment under old-age assistance is about $44.50. Although the recipients of old-age assistance are half again as numerous as recipients of old-aze pensions, total payments for assistance are nearly 21⁄2 times as large as payt ents for pensions. The principal reason why pensions are small is that the benefit formula is too low even for workers steadily employed in covered industries. It provides a pension of only 40 percent of the first $50 of monthly earnings plus 10 percent of the next $200 of monthly earnings, plus 1 percent of this sum for each year of covered employment, plus additional allowances for dependents In addition, the benefit formula provides that, in computing average month lifetime earnings, time worked in uncovered industries shall be counted, but makes no provision for counting the money earned in uncovered industries

A third reason for believing that the Old-Age and Survivors Insurance Act has not done the job expected of it was that the average monthly pensions under it have not kept pace with the rise in the cost of living or the rise in per capita income. Average monthly pensions have increased 14 percent since 1940 while the cost of living has risen over 68 percent, average weekly wages in manufac turing have risen 117 percent, and the average per capita income of the country has increased 132 percent. During the period that average monthly pensi were increasing 14 percent, average monthly old-age assistance payments more than doubled. The purchasing power of pensions is 32 percent less today than it was 10 years ago.

Pensions ought to bear a more or less constant ratio to the average earnings of persons. Otherwise they do not give people the required help in maintaining their customary standards of living. Part of the explanation of why the increase in the average pension has been far less than the increase in the average wage is that, as monthly earnings rise from $50 a month to $250 a month, the primary pension increases only $10 for each $100 increase in earnings. Thus a man whose average lifetime earnings had risen from $150 a month in 1940 to $250 a month at present would have been entitled to a primary pension of $30 a month then and to a pension of $40 a month now (exclusive of the increment)—an increase of only 33 percent in his pension though his earnings had increased by 67 percent. Another part of the explanation why the increase in pensions has lagged behind the increase in wages is that only wages up to $3,000 a year count in computing the average earnings on which pensions are figured. As wages have gone up, more and more workers earn more than $3,000 a year. In 1948, about one out of four among all workers regularly in covered employment earned more than $3,000. Still another part of the explanation why pensions have increased more slowly than wages is that pensions are computed on the average monthly wages of each worker over his lifetime. Although the average worker is earning more than twice as much today as he was earning 10 years ago, his lower earnings of 10 years ago help bring down the monthly average of earnings on which his pension is computed. All in all, the present methods of computing pensions assure that pensions will be slow to rise whenever wages increase.

IV

Why does the Advisory Council believe that improvement of the act along the lines indicated by its 22 recommendations would enable the Old-Age and Survivors Insurance Act to do the job expected of it and to become the foundation of the country's system of social security?

In the first place, the recommendations of the Council would extend the protection of the act to virtually all of the 25,000,000 jobs not now covered. Obviously, one cannot expect a program to give protection to people whom it does not Cover. Consequently, expansion of coverage of the old-age and survivors insurance scheme is of basic importance. Coverage was originally limited because of the administrative difficulties in applying the scheme to such groups as the self-employed and to some kinds of workers, such as domestic workers and farm laborers. As Mr. Folsom, a member of the original Advisory Council in 1935 has pointed out, it was not the intention to exclude permanently the uncovered workers from the protection of the scheme. Both the Bureau of Internal Revenue and the Social Security Agency have studied carefully the administrative problems of extending coverage. Both agencies have had the benefit of 13 years' experience administering the present act. Both believe that there are today no insurmountable administrative obstacles to extending the insurance scheme to the 25,000,000 jobs not now covered.

H. R. 6000 proposes that about 11,000,000 additional jobs be brought under the old-age and survivors insurance plan. This is a step in the right direction, but it would bring under the act substantially less than half of the jobs that are not now covered. Large and important categories, such as farmers, farm laborers, and a number of professional groups, would still be deprived of protection of the scheme.

The incomplete coverage proposed in H. R. 6000 raises basic questions as to where and how uncovered groups would get security for their years of retirement. Nearly all of the groups left uncovered by H. R. 6000 are groups that cannot be expected to be covered by private pension plans, initiated by employers or negotiated by trade-unions. Apparently the philosophy of H. R. 6000, therefore, is that the uncovered groups should either be able to take care of themselves or should be expected to rely upon charity. Is it realistic to assume that all members of the uncovered groups will be able to take care of themselves? For those who are not, is it fair to expect them to rely upon charity? Have we a right to assume that farmers and the professional people do not need the protection given by the old-age and survivors insurance plan? Is farming so free from economic hazards that farmers should be left to take their chances with old-age assistance? Likewise, are the professions such secure callings that architects, engineers, lawyers, doctors, and dentists do not need the protection which old-age and survivors insurance would give them? Can every lawyer, every architect, every engineer count on reaching the age of 65 or 70 with adequate savings to provide for the years of his retirement? Certainly the people cannot expect old-age and sur

60805-50-pt. 3- -64

vivors insurance to be the foundation of the country's system of old-age security if large parts of the population are excluded from the plan.

In the second place, the recommendations of the Council would liberalize the eligibility requirements for pensions so that older workers who are newly covered by extension of the Old-Age and Survivors Insurance Act to new industries will soon become eligible for pensions. Unless the present provisions for eligibility are modified, all persons covered for the first time in January 1951, who attained age 59 after 1950, would have to have 10 years of coverage before they would be eligible for pensions. Even newly covered persons who reach 65 years of age in their first year of coverage would need seven more years to steady employment (28 quarters) before they could receive pensions. It would obv.ously do little immediate good to extend the act to the 25,000,000 uncovered workers if the eligibility requirements for new workers remain as strict as at present. H. R. 6000 makes a little improvement over the present law, but even H. R. 6000 would require 20 quarters of coverage, or 5 years, for newly covered employees to become eligible for pensions.

The Council recommends that the extension of coverage to new industries le accompanied by a "new start" in eligibility requirements that will require the same qualifying periods for older workers as was required for persons who were the same age when the system began operation in 1937. The Council believes that a minimum of six quarters of coverage should be required. It recommends that requirements for fully insured status should be one quarter of coverage for every two calendar quarters elapsing after the year in which coverage is extended and before the quarter in which a man attains the age of 65 (60 for women) or dies. This would mean that persons who had reached the age of V before coverage was extended could be fully insured by working half the time in the next 3 years. This eligibility requirement, combined with the extension of old-age and survivors insurance to the 25,000,000 jobs now not covered, would raise substantially the proportion of males who are fully insured on reaching the age of 65. By 1955 this proportion would be approximately 66 to 74 percent Unless a very high percentage of persons who reach the retirement age are eligible for pensions, the Old-Age and Survivors Insurance Act obviously cannot do the job expected of it.

In the third place, the recommendations of the Advisory Council would substantially raise the average monthly benefits. It is plain that the Old-Age and Survivors Insurance Act cannot be expected to do an adequate job if the pensions received are less than the average old-age assistance benefit granted after a means test. Nor can the act do an adequate job if the pensions payable under it are so small in relation to average earnings that they fail to prevent a ver drastic drop in the standard of living of retired workers. The primary insurance benefit for a worker with 10 years of coverage under the present act and with average monthly wages of $200 is less than one-fifth of his monthly wage and, if he has one dependent, his total benefits are less than 30 percent of his average monthly wage. Obviously such pensions are far too small.

The Advisory Council has made three types of recommendations designed to raise the average benefit. One recommendation is that the benefit forma të liberalized that benefits be 50 percent of the first $75 of monthly wages instead of 40 percent of the first $50, and 15 percent of additional wages up to the amour, of the tax base instead of 10 percent as at present. Another recommendation is that the benefit base and the tax base be raised from $3,000 a year to $4.30 Still other recommendations pertain to liberalizations in the benefits payable to dependents. The most important of these recommendations is that women may qualify for old-age benefits (either primary or supplementary) at the age of 60 instead of 65. During 1948 only 196 of every 1,000 married men who claimed benefits at age 65 had wives who were 65 or over and entitled, therefore, dependents' benefits under the present law. On the other hand, 565 out of evEST 1,000 married men claiming benefits at age 65 had wives who were at least (a) years of age and who would receive dependents' benefits if the recommendations of the Council were adopted. In other words, the recommendations of the Council would increase by nearly three times the number of cases in whi wives' benefits are paid when the husband retires at the age of 65.

The total effect of the recommendations of the Advisory Council would be ter increase the average pension paid under the old-age and survivors' insurance scheme by about 100 percent. This would be an important improvement. Ne theless, I do not believe that the recommendations of the Council go far enough To begin with, the formula proposed by the Council is not sufficiently liberal particularly for the workers earning above $75 a month. The Council recom

mends, as I have pointed out, that pensions be 50 percent of the first $75 of average monthly earnings plus only 15 percent of the next $275 of monthly earnings. This formula would give a primary pension of $37.50 for workers earning $75, of $48.75 for workers earning $150 a month, and of $63.75 a month for workers earning $250 a month.

H. R. 6000 proposes that the primary pension be $50 for the first $100 of earnings. Several members of the Advisory Council-Mr. Folsom, of the Eastman Kodak Co.; Mr. Rieve, of the Textile Workers' Union, CIO; and Mr. Cruikshank, the social-security expert of the A. F. of L.-have expressed agreement with this feature of H. R. 6000, and I share their views. The benefit formula of 15 percent of wages between $75 and $350 a month does not provide a large enough spread between the pensions received by persons with low earnings and persons with higher earnings, though the council provides a greater spread than the formula in the present law or that proposed by H. R. 6000.1 If pensions are to protect the standard of living of workers who retire, they must be fairly closely related to previous earnnigs. Certainly a rise of only $10 or $15 in pensions as average monthly earnings increase $100 does not provide a very close relationship between earnings and pensions, and is not fair to the skilled workers. The primary pension recommended by the council is about one-third of the earnings of a man making $150 a month, a little more than one-fourth of the earnings of a man making $250 a month, and only 22.5 percent of the earnings of a man making $350 a month. Two members of the council, Mr. Rieve and Mr. Cruikshank, have recommended that the rise in pensions above the basic amount. be 20 percent of additional earnings. I am in favor of at least a 20-percent increase in pensions for each rise in average monthly earnings above the basic amount, but I believe that a 25-percent rise would be preferable. Such a rise would give a man with average monthly earnings of $200 a month a primary pension of $75 (assuming the formula provides for a pension of 50 percent of the first $100, as in H. R. 6000) and a man with average monthly wages of $300 a month a primary pension of $100. Certainly such a spread in the pension received by men with a difference of $100 in monthly earnings is not too large.

Finally, I believe that the limit of $4,200 on the benefit base and the tax base recommended by the council is too low and is unfair to many skilled workers and to foremen and others in the lower ranks of supervision. The upper limit of $4,200 in the benefit base means that no earnings of more than $4,200 a year produce an increase in a worker's pension. There are many skilled workers, straw bosses, foremen, assistant foremen who earn from $5,000, $5,500, to $6,000 a year but under the recommendations of the council these men would receive no greater pensions than men earning only $4,200 a year. Indeed, in the year 1948, 17 percent of the male workers who earned wages in covered industries in all four quarters earned more than $4,200. Why refuse to count any of these earnings over $4,200 in computing the pension that the man receives?

The primary pension for a worker earning $5,000 a year, if the pension were 50 percent of the first $100 and 15 percent of the additional earnings up to $4,200 a year, would be only $87.50 a month. It is not satisfactory to expect all men who receive more than $4,200 a year to depend on individual savings or on company pension plans. About 22 percent of persons with incomes of $5,000 or more in 1949 had liquid assets (bank deposits, Government savings bonds, and stocks and bonds of private corporations of less than $500 and 45 percent had liquid assets of less that $2,000. Business enterprises themselves do not expect these men to depend on individual savings because they establish generous pension plans for executives. But almost all of these company pension plans tie a man to one company and penalize the man who moves from one employer to another. When one considers the serious deficiencies of company pension plans, one reaches the conclusion that skilled workmen and foremen should not be dependent to any considerable extent upon them.

I am not in favor of increasing the tax base and the benefit base for the purpose of increasing the income of the pension fund. Nevertheless, if the benefit base is increased, the tax base should be correspondingly increased because the two should be the same. As a matter of fact, the increase in the tax base above $4,200 would not product much new tax revenue.

1 The present law provides that primary pensions shall be 40 percent of the first $50 of monthly wages and 10 percent of wages from $50 to $250; H. R. 6000, that primary pensions shall be 50 percent of the first $100 of monthly wages and 10 percent of wages between $100 and $300 a month.

V

When the present Federal old-age pension plan was drawn up about 15 years ago, the prevailing opinion was that it should be inadequate that it should provide only a bare minimum of security and that it should be supplemented in various ways. It is understandable that back in 1935, when the idea of a Federal old-age insurance plan was new, many people wished to avoid placing 100 much reliance on it. At that time, the serious disadvantages of private pension plans as devices for providing old-age security were not clearly seen. Nor were all of the developments and problems of old-age assistance clearly

seen.

The time has come, I believe, for a change in our thinking about these matters. It is high time to adopt the view that the Federal pension plan should be so close to adequate that only moderate supplementation by private pension plans or public assistance will be necessary. Of course, substantial supplementation by individual thrift will always be necessary. For example, if one accepts as a rough, but modest, standard that an adequate pension for a worker and one dependent is about half of his average earnings, his standard of living will take a terrific fall on the day when he retires, unless he has accumulated at least a moderate amount of savings to supplement his pension.

If the Old-Age and Survivors Insurance Act were improved along the lines recommended by the Advisory Council with respect to coverage and eligibility requirements, and if the benefit formula were somewhat more liberal than that recommended by the council, the plan would really become the foundation of the country's arrangement for old-age security. Within 3 years after the effective date of these changes the number of recipients of old-age pensions would ex red the number of recipients of old-age assistance, total pension payments would be larger than payments for old-age assistance, and the Federal Government would be able to make substantial reductions in its expenditures for old-age assistance. As a matter of fact, the number of old-age insurance beneficiaries actually drawing benefits has increased during the last several years considerably faster than the number of old-age assistance recipients. Between December 1946 and January 1950, the number of old-age insurance beneficiaries increased by 932,000 in comparison with somewhat over 500,000 for old-age assistance recipients. It is illuminating, however, that the increase in the monthly amount of old-age assistance payments during the same period was more than twice as large as the increase in the total monthly old-age benefit payments under the old-age and survivors' insurance today-vivid evidence of the inadequate benefit formala in the Pension Act.

VI

What would a more adequate system of old-age and survivors insurance cist? The answer to this question depends upon many conditions-how many persons reach the age of retirement, how many become eligible for benefits, how many retire, how long benefits are paid, how much is paid as retirement benefits.

The best way to estimate the costs of an old-age insurance program is as a percentage of pay rolls and of the income of the self-employed, in case the insure ance plan extends to the self-employed. In the next 20 or 30 years there will be a very large rise in pay rolls, partly due to the increase in the labor force and partly due to the rise in wages which must be expected to continue in the future as in the past. In another 30 years, for example, the labor force will increase by over 10,000,000, and if wages rise as rapidly as in the past, hourly earnings will increase between 80 and 90 percent. One must expect that the benefits paid under the old-age insurance plan will bear a more or less constant relationship to average monthly earnings. As wage rates rise benefits will be liberalized from time to time in order to maintain a more or less constant rela tionship between earnings and pensions. That is why the most realistic and conservative way of estimating costs is in terms of percentage of pay rolls and the income of the self-employed.

The actuarial consultant of the Advisory Council on Social Security prepared a low-cost estimate and a high-cost estimate of the recommendations of the council. He estimated that by the year 2000 the expanded program recommended by the Advisory Council would cost from 5.87 percent to 9.70 percent of pay rolls The range in his estimates of level premium rose from 4.90 percent to 7:27 per cent. The more liberal benefit formula which I had suggested would cost somewhat more.