IDA C. MERRIAM

Those of us who played some part in the development of that structure were fortunate. It was exciting and satisfying to work for the Social Security Board. I started to work in the Bureau of Research and Statistics in July, 1936. There were high points and low points in the next 40 years or so, but what I remember now is an overall drive and spirit and integrity of purpose. The 1935 Social Security Act was a major achievement, but it was admittedly incomplete. In recognition of that fact, it included a mandate to the Social Security Board to study and to make recommendations to the Congress, as to what more or what different needed to be done to achieve economic security for everyone.

The Social Security Board took very seriously the responsibility for study and analysis. Research was closely related to policymaking. The early Annual Reports of the Board were far different than the usual accounts of program management. The Board reported on the mammoth administrative task it was tackling successfully, but it reported also the questions that it regarded as most pressing, the research it had under way or planned and, later, the conclusions it drew from that research.

It is difficult today to realize how little detailed information there was in 1935 and through the 1940's and even the 1950's about the characteristics and circumstances of individuals and families. The Social Security program itself, as it grew, provided much information. The Census Bureau and other statistical agencies were encouraged to collect the kind of data we needed, and we undertook some major surveys ourselves in later years.

From the beginning, the scope of the research activities of the Social Security Board was broad. The Committee on Economic Security had turned to the experience of other countries to supplement what little information there was for the United States in 1935 with which to design a social insurance program. The Board continued to study other systems. In 1937 the Bureau of Research and Statistics published the first in what became a continuing series of reports describing Social Security programs throughout the world. The latest edition of that publication was released last year.

Once the Supreme Court had declared the Act to be constitutional, the 1939 amendments had added survivor benefits, the date for payment of benefits had been moved forward, and certain other changes were made, major attention was given to the design of the statistical system for old-age and survivor insurance, and for unemployment compensation and public assistance--then also administered by the Board--giving us now a continuing picture of program operations over almost 50 years. Studies of the feasibility of extending coverage of OASI to all paid employment laid the basis for later amendments. Groundbreaking studies of the fiscal capacity of the States had an influence on many Federal grant programs. As early as 1940, the first year benefits were payable, a small sample of OASI beneficiaries in three cities was interviewed to find out what additional income they had, if any, and how they were living. Similar surveys of beneficiaries were carried out over a period of years, and the results were used. For example, it became clear that widows were much worse off than other beneficiaries; in time, the widow's benefit was increased from 50 percent to 100 percent of the primary benefit.

In 1963, the Social Security Administration undertook the first major governmental survey of all persons aged 65 or over ever carried out in this country. By then, it was no longer sufficient to know the situation of beneficiaries. It had become increasingly important to know how this compared with that of non-beneficiaries and why there were still persons who did not qualify. Congressional Committees and Advisory Councils, among others, kept asking for more and more such information. Special surveys of new beneficiaries (a retirement history survey following a group from before they retired for 10 years to see how their health, income and other circumstances changed over time--these and repeated surveys of all aged persons) were used and brought respect for the research staff and for the Social Security Administration that supported and encouraged such work.

From the earliest days, the Board's vision led to support of another area of research. The important question, they thought, was not "how much is our program spending for a particular purpose?" but "what part of the income loss from retirement in old age, or death, or sickness and disability, is replaced by benefits from all existing programs, public and private?" To even begin to answer that question, one had first to estimate the income loss and then get information on payments under veterans programs, civil service, military retirement, railroad retirement, worker's compensation and private pensions and other private benefits. We worked with other agencies to develop comparable data from our diverse accounting systems. When necessary, we developed new statistical series.

The Social Security program itself introduced a new focus. For example, the Public Health Service has been collecting for some years, detailed data on morbidity--the number of days people were sick in bed or homebound. The concern for economic security led to an additional set of questions that the Public Health Service did not want to tackle. The Social Security Administration had been studying problems relating to disability insurance since the 1940's. In 1966, it mounted a major survey of disabled persons, the first in this country, to get information on the characteristics and circumstances of disabled persons, their remaining work ability and their sources of income. Subsequent surveys added to our knowledge of the many aspects of disability and rehabilitation.

In the health field, a great deal of the early research work related to measures of total expenditures for medical care (the much-quoted national health expenditure series) and the proportion covered by public funds, by private insurance and by direct consumer expenditures. It was the elaboration of these data to show separately expenditures for those over and under age 6s--and the appalling lack of private insurance coverage for those age 6s and over--that led eventually to the proposal for hospitalization insurance for aged OASDI beneficiaries and to the enactment of Medicare.

Economic security implies also some measure of adequacy. What kind of earnings replacement guarantee does society want to give? To show some of the consequences of one or another benefit level, we constructed (and priced with the help of the Bureau of Labor Statistics) standard budgets for older people and later developed the poverty index applying to all age groups (now continued and published by the Census Bureau). Over the years, there was research on a variety of problems related to Social Security financing, the pros and cons of different tax sources, the impact of Social Security taxes on private savings, the importance of Social Security benefits in sustaining consumer purchasing power during economic downturns and thus hastening recovery. Through models and through other types of analysis, we tested the effects of alternative benefit formulas, different ways of adjusting to women's changing roles, and different assumptions as to what could happen to the economy as a whole in future years.

Research doesn't tell the policymaker what to do. It does give him a body of tested knowledge and an understanding of the probable consequences of alternative policy decisions. It takes strong and open-minded leadership to accept, publish and use research findings. For most of the 40 years that I worked in and then directed the research effort, there was a remarkable degree of continuing support by top administrators for research that, in a changing world, could help point the way toward the unchanging goal of economic security for all first laid out in the Social Security Act of 1935.