This case study is one of several that could be used to demonstrate how the policy-planning network operates on issues of concern to the corporate community as a whole. But I have chosen to focus on the origins of the old-age insurance title in the Social Security Act, which is now known as "social security" even though the concept of social security originally encompassed unemployment insurance and various kinds of welfare payments that are included in the act. I focus on old-age insurance for a number of reasons, starting with the fact that its origins may come as a surprise to many readers, but also because the program is now the biggest and most popular government program that serves everyday people. It's also an interesting policy because it has been under heavy attack from corporate leaders and their ideological allies for the past 30 years.

Most people now think liberals and labor leaders created the program because they are the ones who defend it. But as this document will show, the basic principles behind old-age insurance were created and actively supported by the corporate moderates who owned and controlled the biggest and most powerful corporations of the 1920s and 1930s, companies such as Standard Oil of New Jersey, General Electric, and Metropolitan Life Insurance. In fact, government social insurance, including both unemployment insurance and old-age insurance, made enormous business and political sense to corporate moderates from the 1930s to the early 1970s. They only turned against it as part of a more general ideological and political attack on "big government" that began in the late 1970s in the face of a new set of economic and political problems caused by skyrocketing oil prices, stagflation, and the pressures put on government budgets by the social movements that arose in the 1960s.

As will be made abundantly clear, the corporate moderates of the 1920s and 1930s did not act without the advice and help of the experts on social insurance they financed and directed in the fledgling think tanks of that day, especially the first industrial relations counseling firm in American history, Industrial Relations Counselors, Inc. (Kaufman 2003). They also worked with experts who were brought together by the Social Science Research Council, founded in 1923 by the leaders of the Laura Spelman Rockefeller Memorial Fund to generate new policies on a wide range of social, economic, and agricultural issues (Bulmer and Bulmer 1981; Karl 1974). Other charitable foundations that corporate moderates created to shape policy proposals, led by the Rockefeller Foundation and the Carnegie Corporation, further aided these efforts over the next decade and longer.

The corporate moderates and their allies insisted that government old-age insurance had to be based on three principles they developed during several years of experience with private pension plans, especially in conjunction with the major life insurance companies of the time (Klein 2003). First, the level of benefits must be tied to salary level, thus preserving and reinforcing the values established in the labor market. Second, unlike the case in many countries, there could be no government contributions from general tax revenues. Instead, there had to be a separate tax for old-age pensions, which would help to limit the size of benefits. Third, there had to be both employer and employee contributions to the system, which would limit the tax payments by the corporations.

If my claims are as solidly grounded in new archival research as I think they are, it is worth asking why this analysis of the origins of social security is not better known. This question is especially pertinent because newspaper articles and memoirs from the 1930s suggest that my analysis would not come as a complete surprise to knowledgeable political observers who experienced and lived the New Deal firsthand. They knew that a few hundred large companies sat astride the economy and that plantation capitalists ruled the South through the Southern Democrats. They were familiar with the specific corporate leaders and their hired experts that are discussed in this document, and they understood their key role.

However, those who now theorize about the New Deal have overlooked the story that follows for several reasons. First, policy formulation is one thing in the United States and the politics of enactment are another. The planners and experts recede into the background and the politicians take over. This point is all the more important in the case of the Social Security Act because it was formulated by moderate Republicans and embraced by the New Deal. As the years passed, the Republicans did not want to admit any role in it and liberals and labor wanted to claim it as their own.

Second, just as the plan was being discussed in Congress, a Supreme Court ruling in May 1935, almost undermined the rationale for the new legislation and endangered its constitutionality in a case concerning a new government retirement program for railroad employees. The preamble justifying the social security proposal therefore had to be rewritten. The new version emphasized that such legislation would contribute to the "general welfare" of the country, which is permissible under the constitution. In other words, an ideology based in social welfare had to be constructed that stressed "needs," not efficiency and control of labor markets. This change in justification caused the labor-market and industrial relations bases of the plan to be lost from sight, and contributed to the belief that social workers, liberals, and unions had created the social security act (Graebner 1980; Graebner 1982).

Third, many prominent ultra conservative business leaders were highly visible in their opposition, including Congressional testimony against the bill, while its business supporters were less outspoken, leaving experts from think tanks financed by corporate moderates (Industrial Relations Counselors, Inc., and the Social Science Research Council in particular) as its main defenders as the act moved through Congress. Fourth, some of the key archival records upon which my assertions are based only became available through the Rockefeller Archive Center in the 1990s. Fifth, most of the standard histories of the New Deal are focused on Roosevelt, party politics, and the legislative process, with little or no concern for the possible role of business leaders and the foundations and think tanks they directed.

Historical background, 1870s to 1920s

Government old-age pensions for a few of its employees go back to the nineteenth century and veterans of the Civil War, and later their widows and children, received government pensions that lasted into the early twentieth century (Skocpol 1992). Although these government pensions may have given the general idea of old-age insurance some respectability in the eyes of average citizens, this early history of government benefits is largely irrelevant because these pensions did not serve as a precedent. Instead, the starting point was in a few corporations, such as American Express in 1875, that saw pensions as a way to replace superannuated workers with more productive younger workers. Shortly thereafter, other corporations thought that pensions might be an a way to induce loyalty in workers and quell labor unrest, but that idea never really worked because the workers had no legal right to private pensions until the late 1920s. Even then, there were loopholes. Death benefits, accident insurance, and unemployment compensation turned out to be more important in terms of reducing labor strife, but they didn't work very well either (Graebner 1980; Sass 1997). However, the Pennsylvania Railroad, one of the largest railroads in the country, had a full-fledged pension plan for all employees at age 70 by 1900.

The push for old-age pensions actually received its biggest boost from the introduction of state-level accident insurance (workmen's compensation) by corporate moderates and their think tanks in the early 1900s. I won't try to tell that story here, but you can find most of the details in these three sources (Fishback and Kantor 2000; Klein 2003; Weinstein 1968). The plans, which involved the use of private insurance companies to provide legally mandated accident insurance, worked so well in the eyes or corporate leaders and insurance executives that they began to think the same type of approach might work for old-age pensions as well. Two of the three largest insurance companies, Equitable Life and Metropolitan Life, which shared many directors in common with major banks and corporations, began making the analyses necessary to offer group life insurance programs and group old-age pension programs to corporations. Insurance companies soon realized they could do a better job with private pensions than individual corporations could do if contributions were made by both the companies and their employees (Klein 2003; Sass 1997).

The gradual move toward actuarial soundness for old-age pensions received a boost in 1918 from the president of the Carnegie Foundation for the Advancement of Teaching, an early creation of corporate moderates in the policy-planning network. He put the foundation's program for pensions for professors on a better footing by founding the Teachers Insurance and Annuity Association, a life insurance company, which then fashioned the first fully insured pension system (Sass 1997, p. 65). By this point the experience of the insurance companies and the Carnegie Foundation for the Advancement of Teaching also began to have an influence on pension programs for government officials, as seen in the pension program designed for federal civil service employees in 1920 by the Institute of Government Relations, one of the forerunners of the Brookings Institution (Graebner 1980, pp. 77, 87; Saunders 1966, p. 25). In other words, by 1920 large corporations and a key organization in the policy-planning network, the Carnegie Foundation for the Advancement of Teaching, were shaping government insurance programs based on their own principles and experience.

During the 1920s the group insurance plans provided coverage for only a small percent of the elderly who had pensions in the United States. Most people bought old-age insurance from actuarially unsound plans sponsored by fraternal organizations, ethnic lodges, or trade unions, but by the end of the 1920s almost all of those plans had failed. As a consequence of these failures, there was a gradual movement towards support for government pensions by organizations such at the Fraternal Order of Eagles and some local and state union federations, using plans drawn up for them by the American Association for Labor Legislation, a corporate oriented group of academic experts financed by large foundations.

They were joined in these efforts by a new left-oriented group in 1925, the American Association for Old Age Security, which advocated comprehensive social insurance (including old-age pensions) at the state level paid for by general taxes, thereby directly challenging the approach favored by corporations. In the late 1920s and early 1930s as many as 25 states passed legislation allowing for government old-age pensions, usually without any state funding and at the option of individual counties, but few people actually received a pension and the benefits were meager if they did so. These efforts are sometimes credited with preparing public opinion for what came later, but I think that is a very generous assessment, a manifestation of the general desire in the United States--and mainstream academia--to assume that public opinion matters unless proven otherwise.

Despite these grassroots and leftist efforts, the major developments of the 1920s, the ones that impacted the Social Security Act, were being made by the individual corporations that had pension plans of their own and by the insurance companies that made their group programs sounder and less expensive by having both employers and employees contribute. By 1923 Metropolitan Life was confident that it had a group pension plan that was better than anything any one corporation could offer on an equally sound basis. So one of its main spokespersons eagerly presented the new plan to the corporate executives that his company invited to a special conference. However, even though he presented evidence that most corporate plans were unsound, the biggest corporations of the era were not prepared to abandon their own plans, which they still believed to be helpful in controlling their workforces and limiting strikes. Corporations also liked the fact that they had complete control over the plans and did not legally have to pay benefits if they decided not to do so.

But in reality the individual company plans did not help with control of the workforce and they were not actuarially sound. Shortly after the Metropolitan Life conference, for example, a meat packing company went bankrupt, sold its assets, and left its 400 retirees with 14 months of benefits (Sass 1997, p. 57). So it was not long before the Metropolitan Life plan became more attractive to smaller companies, especially when it packaged group old-age pensions with life, health, or disability insurance.

John D. Rockefeller, Jr.

It was at this juncture that a recently incorporated industrial consulting group, Industrial Relations Counselors, Inc., funded in its entirety over the next eight years by the richest man of that era, John D. Rockefeller, Jr., decided to pay more serious attention to company-level old-age pension plans to see if they needed to be reorganized. Rockefeller created the organization in an effort to reduce labor strife, especially in the companies he controlled, and to stave off unions. The result was "employee representation plans" that called for plant-level managers to meet with employee representatives elected by plant workers to discuss working conditions (but not wages or unions). As one small part of this general effort, Industrial Relations Counselors decided to take a more careful look at private social insurance plans, including old-age pensions.

This new emphasis within Industrial Relations Counselors was signaled by the employment of two very well trained independent experts on unemployment and old-age insurance, Murray Latimer and Bryce Stewart, who ended up at the center of the legislative drafting for the Social Security Act in 1934. Latimer, a twenty-five-year-old instructor in finance at the Harvard Business School at the time he was hired in 1926, was born and educated in Mississippi, and received an M.B.A. at Harvard in 1923 before joining the faculty. During his years at Industrial Relations Counselors he helped to establish new pension plans at Standard Oil of New Jersey as well as three other Rockefeller oil companies and an independent steel company, American Rolling Mill. His 1932 book for Industrial Relations Counselors on Industrial Pension Systems in the United States and Canada was well known and respected at the time, and is still frequently cited in historical accounts (Klein 2003; Orloff 1993; Sass 1997). Latimer also did a study of union plans for the American Federation of Labor in 1928-1929, concluding that "the experiments are far from having reached a sound basis and that unless drastic financial reorganization is made they are almost certain to end in failure in the relatively near future" (Klein 2003, pp. 56-57).

Stewart, forty-four years old when he joined the Industrial Relations Counselors staff in 1927, was a Canadian with many years of experience working with employment and labor issues. A graduate of Queens University in Kingston, Ontario, he earned a Ph.D. at Columbia University and first worked as a researcher, chief statistician, and editor for the Canadian Department of Labor, and then as an organizer and director of the Employment Service of Canada. After joining the Industrial Relations Counselors staff, he became its director of research in 1930 and held that position until his retirement in 1952, except for a return to Canada as deputy minister of labor during World War II.

Latimer and Stewart were often joined in their efforts by economist J. Douglas Brown, the director of the Rockefeller-financed Industrial Relations Section of the Department of Economics at Princeton. The son of an industrial executive in Somerville, New Jersey, Brown received his B.A. and Ph.D. at Princeton and taught for a year in the industrial relations program at the Wharton School at the University of Pennsylvania before returning to Princeton as a professor. Brown also worked closely with the industrial relations vice president of Standard Oil of New Jersey, Clarence Hicks. He also met with and sent reports to John D. Rockefeller, III, who oversaw Industrial Relations Counselors, for his father in the late 1920s and early 1930s before turning his full focus to philanthropic endeavors. One of Brown's tasks was to talk with corporate executives around the country and make periodic reports to Hicks and Rockefeller, III.

A pamphlet written for the American Management Association in 1928 by a personal employee of Rockefeller, Jr., best exemplifies the pre-depression thinking about company pensions within corporate moderate organizations in the policy-planning network. According to this detailed analysis, which contains discussions of the moral, economic, and technical issues involved in industrial pensions, a pension is part of a good personnel program. Especially in the case of corporations that have been around for many years, a pension is "a means, at once humane and approved by public opinion, of purging its active payroll of men who, by reason of age or disability, have become liabilities rather than assets" (Cowdrick 1928, p. 10). Pensions also provide the "opportunity to promote their younger subordinates" (p. 11). The pamphlet concluded with the prediction that industrial pensions will be "increasingly valuable to employers" (Cowdrick 1928, p. 21).

Although corporate moderates had a clear interest in old-age pensions by the late 1920s, they had no desire to move toward government social insurance, a point demonstrated in a report by the National Industrial Conference Board in 1931. Based on work by Industrial Relations Counselors employees and a survey of a large number of industrial executives, Elements of Industrial Pension Plans concluded that pension plans were becoming more important in the minds of industrialists and urged that the plans be made actuarially sound, in part through having employees contribute to them. No longer was there any mention of the usefulness of these plans in controlling employees. Now the emphasis was on staving off government plans by demonstrating that industry can "take care of its worn-out workers through pension plans resting on voluntary initiative and cooperation" (Board 1931, p. vi). Even more bluntly:

In proportion as such plans are established and become successful there is thus effected a reduction in the number of dependent aged that must be taken care of by society or the state. The extension throughout the field of industry of pension plans adequate in their provisions, equitably administered, and soundly financed will do much toward removing any real need or excuse for resort to the dubious expedient of state pensions (Board 1931, p. vi).

Most of the corporate moderates' policy-oriented efforts between 1930 and 1932 were directed toward unemployment plans and ways to stabilize employment, but by late 1932 the ongoing depression was starting to take its toll on even the best of the corporate pension plans. More workers were reaching retirement age and retirees were living longer at a time when corporate profits had been flat or declining for three straight years. In addition, low interest rates meant that the investments by corporate pension funds were not generating the cash flow that was needed to pay current monthly obligations. As economic historian Steven Sass (1997, p. 88) concludes, "The Great Depression of the 1930s sent a massive shock wave through the nation's fragile private pension system." This was especially the case for the railroads, which had an older workforce than many industries on top of unsound pension plans. Even a switch to "contributory plans" (meaning that both the workers and the company paid into the pension plan) over the previous three years had not been enough to save the railroad pension plans. But it was not just corporate plans that were in trouble: the handful of small pension plans controlled by the AFL and other unions also began to suffer, as Latimer of Industrial Relations Counselors had predicted they would even before the depression began. In 1932 the AFL dropped its longstanding opposition to government pensions, which was based on the premise that the corporations controlled government.

Frances Perkins

As the depression deepened and Roosevelt took office in March 1933, members of the policy-planning network who had worked on unemployment and pension plans began to be involved in direct service to the New Deal. For example, Stewart of Industrial Relations Counselors became chair of a committee to advise Secretary of Labor Frances Perkins on selecting the members for her Advisory Committee to the Department of Labor. He also served as a member of the Advisory Council of the United States Employment Service and chaired its Committee on Research. At the same time, Latimer provided the Department of Commerce with estimates on the amount of pension income that was being paid out in the country and then became a member of the Advisory Committee of the Department of Labor, where he spent part of his summer months assisting in the revision of the employment and payroll indexes and in making studies which would lead ultimately to the revision of the price indexes.

As this mundane statistical work was grinding along, grassroots efforts by railroad workers, who had been building their movement since 1929, began to pick up momentum when the railroads announced they would be making sudden 10% cuts in both salaries and pensions. In a context where at least 84% of railroad workers had been covered by company pension plans since the early 1920s, and with young workers supporting the older workers so they could move into the senior jobs, the rank-and-file organized on their own because of the lack of interest in government pensions on the part of their union leaders (Klein 2003; Latimer and Hawkins 1938; Sass 1997). In 1931 and 1932, the railroad workers' independent actions--organized as the Railways Employees National Pension Association, which was outside the confines of their cautious union leadership--suddenly gained enough power to force Congress to take legislative action.

In the face of the impending pension crisis in the railroad industry, the Railways Employees National Pension Association was able to convince Senator Henry D. Hatfield, a one-term Republican Senator from West Virginia, to introduce their legislation in 1932. A physician who was a staunch supporter of unions and a former governor of the state, Hatfield had a special sympathy for railroad workers because he had worked for 18 years as a surgeon for the Norfolk and Western Railroad. Significantly, the legislation he introduced called for contributions by workers as well as employers. It also had generous benefits and an option for early retirement. "As pension agitation mounted," concludes sociologist Jill Quadagno (1988, p. 73), "labor leaders began to recognize that their indifference to the pension issue was alienating them from the rank and file, and in the same year they succeeded in inducing Senator Robert Wagner of New York to introduce an alternative proposal." Despite strong opposition from railroad executives, Congress passed the Wagner-Hatfield bill in 1933 (see also Graebner 1980, pp. 171-176).

In short, the railroad workers had won a class struggle with the railroad owners. It may seem small in the larger scheme of things, but it was major for them, and it soon had a much larger impact in an unexpected direction--on the corporate moderates. I underline this point because class conflict on specific issues is often the basic force behind seemingly liberal legislation. There is pressure from workers and then the corporate moderates decide to find a way to accommodate that pressure in a way that makes sense in terms of the ongoing profitable functioning of large corporations.

Although the federal coordinator of transportation advised Roosevelt to sign the railroad retirement legislation because "it is in line with sound social policy," he added that he would have preferred to wait in order to improve it (Latham 1959, p. 160). One of the problems he worried about was the actuarial soundness of the plan, which caused him to bring Stewart, Latimer, and J. Douglas Brown (the Princeton industrial economist whose work was funded by the Rockefeller network) to Washington in late 1933 to serve as members of an Employment Advisory Council that would design the new social insurance system for railroad workers. As Brown tells the story:

The group of us that went down [to Washington] on that centered very much on Industrial Relations Counselors, in New York.... So Latimer and I began working on the old-age protection of railroad workers. We put Hawkins [a student of Brown's] to work on the dismissal compensation. Bryce Stewart worked on the unemployment insurance (Brown 1965, p. 6).

Latimer, Stewart, and Brown lacked the information needed for the actuarial studies on which to base a sound program, and they did not have an army of clerks at their disposal to develop the information. They therefore applied for a $300,000 grant from the recently established Civilian Works Administration and then hired laid-off railroad clerks who had dealt with the relevant employment records for their respective companies--1,500 people collecting records on 400,000 employees and 110,000 pensioners. They also hired a staff of 500 in New York to analyze the data (Brown 1965, pp. 8-9; Latimer and Hawkins 1938, p. 111). The result was a new set of records with the space of a few months, which shows how rapidly state capacity can be created when there is the will to create it.

Latimer, Stewart, and Brown then crafted a plan that was satisfactory to all concerned even though the benefit levels were lower than those originally proposed due to the fact that the study showed the original actuarial assumptions were unsound (Latimer and Hawkins 1938, pp. 123-127). Employers were pleased because they were relieved of the cost of private pensions and their tax rates were lower. Railroad workers accepted the plan because the pensions were satisfactory--in fact, much higher than those later established for the Social Security Act--and there were disability and survivor benefits as well (Latimer and Hawkins 1938, p. 274). In the end the Railroad Retirement Act was a victory for all those who were willing to allow the government to play a role in providing social insurance. Because of this work Latimer was appointed chairman of the three-person Railroad Retirement Board in the summer of 1934.

Strikingly, the railroad worker's success did not lead to similar efforts by other workers, which Quadagno (1988, p. 74) attributes to the division of American workers along craft lines. The best that the American Federation of Labor could do, and its only positive contribution to the process that followed, was to end its opposition to government pension plans at its annual meeting in 1932, as also stated earlier in this document.

However, the lessons from this successful effort were not lost on Latimer, Stewart, and Brown. They began to understand the possibilities for using the group insurance policies developed by the private insurance companies, with whom they were always in close contact, as a model for government insurance plans. They realized they could package old-age pensions--as well as unemployment compensation, which I am not discussing in this document in order to keep it a manageable size--in such a way that they would be compatible with the major concerns of corporate leaders. They also realized that such plans would be far less expensive for corporations than having their own programs, some of which were on increasingly shaky grounds in any case. From this point forward they worked to convince corporate executives, fellow experts, and social workers of the soundness of their ideas. Their efforts are a textbook example of how experts function in the United States, which contradicts the emphasis of mainstream New Deal scholars on independent experts as well as any example could.

There is an even larger point that will be demonstrated in what follows. In effect, the foundations and think tanks created by corporate leaders became part of the government because they paid the salaries of private-sector experts who were de facto state employees. They thereby provided the government with the ability to build new processes and agencies through the utilization of expertise based in private organizations (and especially in Industrial Relations Counselors in the case of old-age insurance). Contrary to most theorists, and the historical institutionalist theorists in sociology and political science in particular, the American federal government does not necessarily build its own capacity, and those who administer the government are not necessarily independent of the corporate community and its closely affiliated policy-planning network.

By November 1933, the experts in the policy planning network who had been working on social insurance plans for nearly four years felt confident enough with what they had accomplished to bring it to the attention of experts just outside their circles. They did so through a small conference in Washington under the auspices of the Social Science Research Council (mentioned in the Introduction as an organization founded in 1923 with funding from foundations to support research and policy development in the social sciences). Twenty-two people attended this conference, representing a wide range of social service organizations as well as government agencies related to social insurance and social provisioning. Fourteen of the twenty-two had served on at least one policy committee of the Social Science Research Council or were connected to the policy-planning network in some other way. Several were affiliated with the local-level policy-planning network created in good part by Rockefeller philanthropies and housed by the Public Administration Clearing House at the University of Chicago (Roberts 1994).

The most prominent representative of the social service organizations was Edith Abbott, a leading reformer during the Progressive Era and since 1921 the dean of the School of Social Service Administration at the University of Chicago. The social welfare representatives also included the director of the Public Administration Clearing House and leaders from the American Association of Social Workers, the American Public Welfare Association, and the Institute of Public Administration.

Harry Hopkins

Perhaps the most important government official present was Harry Hopkins, a former social worker and a close confidant of Secretary of Labor Perkins, who had been appointed by Roosevelt to create the Federal Emergency Relief Administration in May 1933 (Cohen 2009, Chapters 8 and 9). Arthur Altmeyer, Perkins's main assistant on social insurance issues, rivaled Hopkins in importance at the meetings. He had been the executive secretary of the Wisconsin Industrial Commission for many years before joining the New Deal. Also present were Mary Anderson, the director of the Women's Bureau in the Department of Labor, which had jurisdiction over the "mother's pensions" that would become stigmatized as "welfare payments" when they were enfolded into the new Social Security Act; John Dickinson, the Assistant Secretary of Commerce who helped draft the National Industrial Recovery Act just a few months before; and Isador Lubin, a former Brookings Institution employee who had been appointed by Perkins as the Commissioner of Labor Statistics. There were also several experts present who worked closely with government agencies, starting with M. L. Wilson, who played a key role through the Social Science Research Council in formulating the Agricultural Adjustment Act passed earlier by the New Deal, and J. Douglas Brown, the director of the Industrial Relations Section at Princeton who had worked on the railroad retirement plan.

The starting point for the discussions at the Social Science Research Council conference was a document prepared by Stewart listing the nature of the studies needed to understand several problems that had to be resolved to design a comprehensive social insurance program. There was a stress on issues dividing the proponents of various kinds of social welfare, which were most contentious on unemployment insurance for a variety of reasons. In the case of old-age pensions, the draft plan embodied the three basic principles mentioned in the Introduction that the corporate moderates insisted upon: the level of benefits must be tied to salary level, there should be no government contributions from general tax revenues, and there had to be both employer and employee contributions.

Although the attendees were unanimous in encouraging the Social Science Research Council to move forward in refining Stewart's proposal, the liberals and reformers of that era, many of them social workers, did not give their approval without expressing their disagreements with what they called "the insurance crowd," which meant experts such as Latimer and Stewart. This difference showed up most prominently over the issue of funding old-age pensions when Abbott stated her preference for "one welfare statute" that would be paid for out of general tax revenues and "available to all without stigmatizing qualifications" (see Gordon 1994, p. 261 for Abbott's general views; see Witte 1963, pp. 15-16, for the fact of disagreement). This and other differences of opinion show that Stewart and other insurance-oriented experts in the policy-planning network were not liberals in the eyes of the liberals of that era.

The same group of people then met for a second Social Science Research Council conference in early April 1934, to consider a second version of Stewart's proposal. However, they did so in very different circumstances because Senator Wagner had introduced a new state-oriented unemployment insurance bill on February 5 on behalf of moderate reformers. Reformers to the left of the moderate reformers, such as those involved in the American Association or Old Age Security, which had just changed its name to the American Association for Social Security, vowed to defeat the Wagner bill because it was so cautious. They also feared it would undercut their efforts towards more liberal programs in several states, which they thought had a good chance of success. At the same time, most business groups were equally opposed the new proposal for their own reasons.

With this conflict in Congress swirling around them, the Social Science Research Council group once again gave its general approval to the evolving plan that had emerged from the policy-planning efforts over the past several years. Stewart, with the help of a co-author, an economist employed by the Social Science Research Council, then revised his report to take into account concerns expressed at the meeting and to emphasize the group's support for a unified plan--that is, one that included unemployment insurance and benefits for single mothers as well as old-age insurance. They did so in part because Roosevelt, who was in close touch with corporate moderates on social insurance issues, also called for a unified plan. As Stewart and his co-author explained in a formal report to the Social Science Research Council:

In a draft report, revised following the April conference, the unified character of the task of planned protection was developed, and the several phases of relief and social insurance were considered in terms of (a) the problems of planning, administration, and coordination, (b) the present state of knowledge in each field, and (c) further work specifically required for the proper integration of each major segment into a unified program (Stewart and Givens 1934b, p. 1).

The Social Science Research Council group then sent Perkins and Harry Hopkins copies of their conference report in an effort to reinforce the idea that general, not piecemeal, legislation was necessary. From their point of view, their efforts were successful in influencing the creation of the Cabinet-level Committee on Economic Security, as explained in the same Social Science Research Council report of November 16 that I just quoted. I find the following paragraph to be strong evidence that the experts within the policy-planning network were working closely with Perkins and Hopkins to shape the government's agenda:

At the request of officials of the Department of Labor and the Federal Emergency Relief Administration, these materials were made informally available in the formulation of plans for a government inquiry. A draft plan for such an inquiry, developed upon the basis of the exploratory study, was placed in the hands of a Cabinet committee, and these plans have eventuated in the establishment by Executive Order, June 29, 1934, of the Committee on Economic Security. Thus the original project became merged in a major planning venture at the Administration (Stewart and Givens 1934b, p. 1).

Once the Roosevelt initiative was announced, Stewart and his co-author anticipated (on the basis of the liberal social workers' dissents at the two Social Science Research Council conferences) that there might be aspects of the final legislation that would not be acceptable to corporate moderates. They therefore revised their earlier proposal for immediate research funds from the Social Science Research Council to make it a call for a larger study that would begin after the shape of the final legislation became clear. They argued it was very unlikely that any new legislation would be thoroughly satisfactory, which meant that future Social Science Research Council studies would be important in influencing inevitable revisions in the program (Stewart and Givens 1934a, p. 1). Thus, members of the policy-planning network were already preparing for likely amendments--and for shaping the administration of the Social Security Act--well before the plan was finalized and sent to Congress in early 1935 (cf. Fisher 1993).

As this brief history demonstrates beyond any doubt, employees of the policy-planning network were actively involved in developing plans for social insurance (most importantly for my purposes here, old-age insurance) right up until the moment the government process began. They also had plans to draft amendments to the legislation if they thought changes would be necessary. My next step is to determine whether or not they were involved in the drafting processes inside the government and if so, if they had any impact.

The drafting process

Roosevelt announced the plan for a comprehensive study of a program for economic security on June 8, 1934. It would be conducted by a cabinet-level committee, the Committee on Economic Security (CES), chaired by Secretary of Labor Perkins and including Hopkins as the Federal Emergency Relief Administrator, along with the attorney general, the secretary of treasury and the secretary of agriculture (the latter three played minor roles, so I don't introduce them here). The CES was assisted by what was called the "Technical Board," a group of 20 government-employed experts drawn from several different agencies (several of the "government" experts had until recently been employees of foundations, think tanks, and universities).

Walter Teagle (Standard Oil)

Gerard Swope (General Electric)

Marion Folsom (Eastman Kodak)

It also had the input of an Advisory Council on Economic Security made up of 23 private citizens, including several prominent corporate moderates, labor leaders, and social welfare advocates. The business representatives included Walter Teagle, president of Standard Oil of New Jersey; Gerard Swope, president of General Electric; and Marion Folsom, the treasurer of Eastman Kodak -- all of whom had been working with or were aware of the work by Stewart and Brown in the previous few years. The other two business members were involved in a moderate think tank and the Social Science Research Council.

There were also five labor leaders, including the head of the American Federation of Labor, but they attended few meetings and generally had very little impact.

In addition, there was a research staff made up of experts brought in from Industrial Relations Counselors, the Social Science Research Council, other think tanks, and universities. It was the staff's job to draft the proposals to be discussed by the appropriate committees of the Technical Board and the Advisory Council on Economic Security before they were passed up the hierarchy to the CES (and Roosevelt from behind the scenes through his interactions with Perkins and Hopkins). Finally, there was an executive director to lead the staff and serve as secretary to the CES.

Significantly, and frequently commented upon at the time, the most visible left-liberal and leftist reformers on social insurance issues, who had advocated plans at the grassroots for well over a decade and written several influential books, were not included in any part of the formal process, not even as members of the Advisory Council on Economic Security. Like the liberals and the social workers, they advocated protection for everyone without qualification and wanted to finance the program out of general taxes. But they were far less willing to compromise than were the social workers, who despite any disagreements were part of the New Deal coalition through their many connections to Eleanor Roosevelt, Perkins, Anderson of the Women's Bureau, and several other government appointees.

On the basis of a strong recommendation from Altmeyer, Perkins offered the important position of staff director to Edwin Witte, a labor economist who had received his education at the University of Wisconsin and worked as the chief of the state's Legislative Reference Library, a service for state legislators who needed help in writing their bills. Unknown to anyone at the time, Witte kept a diary of the unfolding events to help him keep things straight in a complex situation. The diary became the basis for a memorandum he wrote in 1936 at the request of the Social Science Research Council's Committee on Public Administration, which used it as part of its efforts to help shape the administration of the Social Security Act (Witte 1963, p. xi). Later the Social Science Research Council asked Witte if it could publish his memorandum as a book, which appeared in 1963 and became the basis for just about every analysis of the origins of the Social Security Act since that time. It remains a valuable book, but it is too brief to tell the whole story in detail.

With this background material on Witte and his book in mind, I can begin my discussion of the impact of the policy-planning network on the act itself with the fact that the structure and process for the program and much of its research agenda were developed by Altmeyer and Stewart based on the report written after the second Social Science Research Council conference, as explained by Witte's biographer, Theron Schlabach (1969, p. 99). This conclusion, based on interviews and a statement in the CES files written by the three men for Roosevelt, is consistent with Stewart's(1934b) statement in an Social Science Research Council report that the conference's proposals were made informally available to government officials. In addition, Witte (1963, pp. 15-16) notes that he made "some little use" of the research suggestions in this report "in outlining the fields to be covered by various members of the staff." All in all, this is impressive preliminary evidence that the policy-planning network was an integral part of the governmental process.

To our unexpected good fortune, at this point the Industrial Relations Counselors experts also began to provide corporate industrial relations executives with inside information on the unfolding drafting process through periodic memorandums prepared for the organization's clients. These memorandums give us a new window into the thinking of the corporate moderates that has never been utilized in detail. The first of them, for July 10, provided a thorough overview of how the drafting process would be carried out, concluding that "It is patent that the Administration is determined to develop a program of social welfare to be presented at the next session of Congress, and that broad departures in the field of industrial relations may be anticipated" (Memorandum to Clients, No. 1, July 10, 1934, p. 2). Two paragraphs later it reminded its readers "to prepare for the advent of various forms of social insurance."

Old-age insurance

The corporate moderates and experts associated with Industrial Relations Counselors had extensive day-to-day involvement in the development of the plan for old-age insurance. Describing his search for a staff to study old-age pensions and draft a proposal, Witte (1963, p. 29) reported that "It was agreed by everyone consulted that the best person in the field was Murray Latimer, who was unavailable because he was chairman of the Railroad Retirement Board." However, Latimer was able to serve as chair of the Technical Board's Committee on Old Age Security, an important policy role in itself, and in any case he worked closely with the research staff in drafting the legislation. Latimer also was given the opportunity to recommend a leader for the staff. His suggestion, which is further support for my emphasis on the role of the Rockefeller-financed circle of experts, was J. Douglas Brown, one of his collaborators in the railroad retirement study, whom he also knew through Industrial Relations Counselors and annual conferences at Princeton on social insurance (Witte 1963, p. 3). When Brown decided that he could only give part of each week to the work at hand, Professor Barbara Nachtrieb Armstrong of the Law School at the University of California, Berkeley, was placed in charge of the old-age study (Armstrong 1965, p. 36). Latimer and Brown worked very closely with her, along with Otto Richter, an actuary on loan from AT&T.

Barbara Nachtrieb Armstrong

Armstrong is an intriguing and interesting figure who also figures importantly in my theoretical arguments with the current generation of historical institutionalists in sociology and political science because she is their evidence that independent experts are as important as they claim. Armstrong earned her law degree in 1915 at the University of California, Berkeley, and went to work for California's Commission on Social Insurance. Her report on sickness as a cause of poverty in California also became the basis for her Ph.D. dissertation at Berkeley in 1921. She next taught both law and economics at UC Berkeley for several years before becoming the first woman to be appointed a professor of law in the United States. During the 1920s she immersed herself in the study of European social insurance systems and in 1932 published one of the most respected books in the field at the time, Insuring the Essentials, with the help of a Social Science Research Council grant. She is indeed an example of the kind of independent expert that institutionalist theorists emphasize out of all proportion to their frequency.

Armstrong (1965, p. 38) reports in an oral history that she knew no one in Washington when she was asked to join the research staff and never learned who suggested her inclusion. She says she had received positive letters about her book from Roosevelt intimates and speculates that the president of General Electric, who wrote her a laudatory letter about the book, may have been responsible for her selection (Armstrong 1965, p. 30) However, since she also knew many of the experts in the field and was highly respected for her book, it may be that one of the other experts recommended her.

Originally hired to work on the unemployment compensation program, she was switched to old-age insurance when she arrived because Latimer and Brown would not take on the task. Like them, she favored a nationwide contributory system administered by the federal government. She also had the highest regard and affection for Brown and Stewart, whom she describes as kind and gentle people. After playing a central role in the drafting process, she returned to California, never returning to Washington to testify before Congress, in part because of time pressures, in part because she feared she might be too acerbic as a witness before congressional committees (Graebner 1980, p. 187).

The plan prepared by Armstrong and her colleagues, which contained all the provisions that Industrial Relations Counselors had come to advocate, sailed through the Technical Board's Committee on Old Age Pensions, but its two main features, its national scope and the inclusion of employee contributions, were worrisome to Perkins and the other members of the CES. However, the original plan prevailed on both issues when the CES finally voted. Thus, the process produced a clear policy victory for the approach first developed by the insurance companies and the experts at Industrial Relations Counselors.

However, the plan then faced a different kind of challenge. There may have been some inclination on the part of Roosevelt, Perkins, and Witte to exclude old-age insurance from the final package sent to Congress because they feared that opposition to it might interfere with the passage of the program that mattered the most to them, unemployment insurance. Perkins and Witte always denied there was any such move afoot, but Armstrong, Latimer, and Brown were convinced otherwise. They quickly spoke off the record to reporters to that effect after an ambiguous comment by Roosevelt in a speech to a national conference on economic security in Washington in November 1934, jolted them to attention. "I do not know," Roosevelt intoned, "whether this is the time for any federal legislation on old age security" (Davies 1999, p. 60). The immediate uproar in the newspapers led to assurances by all concerned that old-age insurance would be included in the legislative proposal (Armstrong 1965, pp. 88-89; Brown 1965, p. 13; Schlabach 1969, p. iii, based on his interview with Latimer).

Shortly after the public phase of the controversy ended, the corporate moderates came into the picture in a supporting role through their membership on the Advisory Council on Economic Security. According to Armstrong (1965, pp. 82-83) and Brown (1972, p. 21), they were crucial in convincing Roosevelt and Perkins to retain the old-age provisions in the legislation. As Brown recalled it:

The likelihood of gaining the support of the Cabinet Committee for our proposals was still in doubt. At this critical time, December 1934, help came from an unexpected source, the industrial executives on the committee's Advisory Council. Fortunately included in the Council were Walter C. Teagle of the Standard Oil Company of New Jersey, Gerard Swope of General Electric, and Marion Folsom of Eastman Kodak, and others well acquainted with industrial pension plans. Their practical understanding of the need for contributory old-age annuities on a broad, national basis carried great weight with those in authority. They enthusiastically approved our program. Just as the newspaper writers had carried us through the November crisis, the support of progressive industrial executives in December ensured that a national system of contributory old-age insurance would be recommended to the President and Congress" (Brown 1972, p. 21).

Brown later summarized what he called the "American philosophy of social insurance" in a retrospective book. Echoing Cowdrick in his 1928 pamphlet for the American Management Association, Brown's (1972, pp. 90-91) emphasis was on "the need for a perpetual corporation to assure a flow of effective and well-motivated personnel for the year-by-year operation of the company." More specifically, "retirement programs with adequate pensions became necessary to prevent an excessive aging of staff or the loss of morale which the discard of the old without compensation would involve;" thus, old-age insurance was simply "a charge on current production to be passed on to the consumer"(Brown 1972. pp. 90). This is exactly the conclusion most corporate moderates had reached by the late 1920s. However, it did take the "massive shock wave" of the Great Depression (Sass 1997, p. 88), the grassroots efforts of the Railways Employees National Pension Association, and careful actuarial work by Latimer, Stewart, Brown, and other experts to convince the corporate moderates that they would have to realize their purposes through a narrowly circumscribed government program.

With the arguments over old-age insurance at an end within the CES, much time and energy then was spent over a far more contentious issue, the plan for unemployment compensation. These battles between rival reformers need not detain us here, but I have written about them elsewhere (Domhoff 1996, Chapter 5). Suffice it to say that the more generous and comprehensive plan put forth by the experts from Industrial Relations Counselors was trimmed back through the efforts of more cautious reformers who preferred to have state-level insurance and in any case feared that Southern Democrats would veto national-level unemployment insurance.

It is now time for a brief overview of what happened within the legislative process.

The legislative gauntlet

The CES's overall plan was introduced into Congress in mid-January 1935, with the apparent support of a wide cross-section of the corporate community, including a committee of the National Association of Manufacturers (Brents 1984; Jenkins and Brents 1989). Then, too, a committee of the Chamber of Commerce endorsed the bill in March 1935, while it was still being dissected by Congressional committees (Nelson 1969, p. 214). The plan also had support from reformers and labor leaders. Nevertheless, the proposal had to survive a seven-month legislative gauntlet that included a complete redrafting in the House, near defeats in key Congressional committees, changes in the preamble due to the Supreme Court decision on the Railroad Retirement Act, and an amendment to allow companies with their own large pension plans to opt out of the government plan, which was so controversial that it almost led to a deadlock between the House and Senate. In the end, however, all of the old-age insurance provisions survived even though Southern Democrats insisted upon restrictions on federal regulation of the program.

The Industrial Relations Counselors Memorandum to Clients, No. 5, provided a detailed eight-page overview of the bill to its clients on January 25, eight days after it was introduced. It was very favorable toward the old-age insurance provisions, noting that the contributions to the program were low "as compared with the pension plans of progressive companies..." (Memorandum to Clients, No. 5, p. 10). However, it did worry that the government might not be able to "assure the contractual character of this obligation so long as Congress has the power later to change the terms of the law." It also expressed concern that the bill gave "no recognition to industrial pension plans that have been adopted in several industries, and a number of which have become well established and have accumulated considerable reserves" (Memorandum to Clients, No. 5, p. 10). Within a few days, however, they saw this lack of recognition for private pension plans as a very real opportunity for companies with pension plans.

As Roosevelt and Perkins had feared, most of the discussion in Congress focused on old-age social insurance. This part of the legislation was explained and defended at length and in detail by Witte, Brown, and Latimer. Latimer and others believe that Witte was by far the most creditable witness for the great majority of congressmen (Schlabach 1969, p. 144). However, Brown and Latimer's testimony is of greater theoretical interest here because it stressed labor market concerns, which supports the argument that the program was created with industrial relations in mind, not social welfare(c.f. Graebner 1980, pp. 187-189). For example, Latimer's only concern was that larger benefits might be needed to induce the larger number of retirements that he thought necessary to help improve the unemployment problem.

Responding to the concerns of companies that already had their own pension plans, the Industrial Relations Counselors' Memorandum to Clients, No. 6, for February 1, explained that the government plan would be less costly for these companies. It also proposed that the current company plans could be seen as supplementary to the government program, making it possible to provide more attractive pensions for higher income workers: "The combined cost to companies of the revised company plan and the national plan would presumably be less than the cost of their present plans, since the contribution rates levied by the Security Bill are set below actual cost on the assumption that the additional amounts needed later will be drawn from general tax funds" (Memorandum to Clients, No. 6, p. 1). This memo is very important in terms of current theoretical debates because it shows that Industrial Relations Counselors understood there would be cost savings and opportunities to provide better retirement benefits for executives much sooner than standard historical accounts realize

Memorandum No. 6 then explains why the employees themselves may prefer Social Security over company plans that claimed they would pay higher benefits. These reasons also are an admission of the weaknesses inherent in company pension plans discussed by Sass (1997). It first notes that there is an "absence of real guarantees" in company plans, which is a damning admission if there ever was one. Second, it notes that there had been "widespread cuts in the amounts paid to pensioners and reductions in the rate of pension which have occurred during the past four years," which is an admission that many company plans were not actuarially sound over the long run (Memorandum to Clients, No. 6, p. 2).

Memorandum to Clients No. 6 is also important because it provides the first mention of "contracting out," which would "permit a company to operate a separate plan outside the federal scheme if it is in no way less favorable than that of the government and has its current credits fully financed." The memorandum concedes that such a provision would have "a decided appeal from the industrial relations viewpoint of the individual company," implying that privately controlled pension plans might help the company in retaining and restraining employees, but it then adds that "we understand that the experts who drafted the bill believe such a provision would weaken the effectiveness of the measure for the great number of wage earners who are not under company plans" (Memorandum to Clients, No. 6, p. 2, my italics). In other words, the Industrial Relations Counselors experts, as exemplified by Stewart, had decided that contracting out was not a good idea for the corporate community. The memorandum then added that a separate plan would be burdensome besides: "Certainly the inclusion of the proposed provision would be accompanied by requirements for financial guarantees from the companies of a character that might prove burdensome and difficult to meet and to that extent would lessen its acceptability" (Memorandum to Clients, No. 6, p. 2).

I see such comments as part of the process of disseminating a new perspective in the corporate community. I also see them as the first of several pieces of evidence that the institutional theorist who wrote in the most detail about the origins of the Social Security Act is wrong when she concludes that "almost all of the welfare capitalists" were in favor of contracting out because of their alleged continuing opposition to federal old-age insurance (Orloff 1993, p. 293).

The depth of the Industrial Relations Counselors experts' concern over contracting out is revealed in a letter that Brown sent to Witte shortly thereafter, on February 13, reporting on what he had learned through his discussions of contracting out with corporate executives at the annual meetings of the American Management Association. Once again serving as the eyes and ears of the Rockefeller circle, Brown reported that most of them understood that this provision was not to their advantage. He also had learned that a Philadelphia insurance agent who was lobbying for the idea had very little support among insurance agents or the large insurance companies, with the exception of Prudential and Metropolitan:

The Prudential Company has been rather inept in the matter and I think that you will find that the dozen or more companies other than the Prudential and the Metropolitan are not particularly in sympathy with the tactics of those two companies. I heard in Pittsburgh, however, that the Metropolitan, at least on the surface, is saying the bill will be a boon to the insurance companies in expanding the demand for supplementary group annuity contracts. Both the Prudential and the Metropolitan are somewhat frightened by the threat of investigation of industrial insurance, and may not be as anxious to push the amendment on account of a backfire in this respect" (Brown to Witte, February 13, 1935, Brown Papers at Princeton University Library).

Brown had further encouraging conversations that he reported on in a letter to Witte on February 23. His list of companies lacking interest in contracting out is long and impressive:

I am continuing to receive word from industrial relations executives of their lack of interest in the contracting out amendment. Confidentially, the last word I had was from Art Young, Vice-President of the United States Steel Corporation. I have been in touch also with the American Telephone and Telegraph Company, Socony-Vacuum, Du Pont, United States Rubber, Union Carbide and Carbon, Western Electric, and a number of other companies. The men in question are the chief personnel officers, and since I have known most of them for six or eight years, I have confidence in what they tell me" (Brown to Witte, February 23, 1935, Brown Papers at Princeton University Library).

While committees in the House and Senate were deciding whether to permit the report from the Committee for Economic Security to be voted upon by the full House and Senate memberships, Stewart and the Social Science Research Council hosted a conference in Atlantic City on March 22-23, 1935, based on the earlier mentioned funding request from the spring of 1934, when they had anticipated that amendments to the final legislation would be necessary. This conference unanimously recommended funding for studies of the new social security administrative board. The studies would be carried out by two Social Science Research Council committees, the Committee on Public Administration and the Social Security Advisory Committee, as coordinated by the Committee on Industry and Trade. The Rockefeller Foundation immediately gave approval to this request, which led to donations of $611,000 between 1935 and 1940 (Fisher 1993, p. 139).

Shortly after the Social Science Research Council conference, the Industrial Relations Counselors' next Memorandum to Clients included two attached statements from insurance companies stating their belief that the legislation will "result in renewed appreciated and greater stimulation of life insurance activities both individual and group rather than the reverse" (Memorandum to Clients, No. 9, March 27, 1935, p. 8). This is further evidence that the Industrial Relations Counselors and at least some insurance companies understood the potential of the Social Security Act well before the date claimed by other commentators.

At the same time as Industrial Relations Counselors and the insurance companies were realizing that contracting out was not a good idea, the general CES proposal was being completely rewritten in the House Ways and Means Committee. Minimum benefits and merit hiring of state-level administrators were eliminated at the insistence of Southern Democrats on the committee, which is of course good evidence for an emphasis on their power during the New Deal.

The Industrial Relations Counselors' Memorandum to Clients, No. 10, dated April 10, provided a thorough summary and evaluation of the revised legislation that the Ways and Means Committee introduced into the House on April 4. In addition to changes in the unemployment insurance provisions that the private experts did not like, there were also changes in the plans for old-age insurance that did not meet their expectations. For example, benefits would now be higher for low-wage workers than they were in current company plans and lower than they would be for high-wage workers. Death benefits would now be higher than planned after short periods of employment and lower than after lengthy periods of employment.

When the revised legislation reached the House floor on April 12, it first had to survive two sideshows--one by the Communist Party, the other by a for-profit social movement of the elderly called the Townsend Plan. Both received attention in the media at the time and subsequent attention by historians and social scientists, but in fact neither alternative had any chance of passage. Both plans were far more generous than the Social Security Act, but neither received more than 55 votes. This total has impressed some leftist historians, but in fact most of those votes came from conservatives who opposed any form of government social insurance; there were not more than 20 representatives in the House who stood to the left of Roosevelt's New Deal coalition (Klehr 1984, p. 289). It is therefore unlikely that either the Communists or the Townsend Plan worried any legislators enough to push them into voting for the mainstream Social Security Act (Amenta 2006).

The Senate Finance Committee, which Witte feared as the biggest threat to the legislation because it contained many Southern conservatives, finished its hearings in February, then postponed further action on the proposal until April. It did not approve a report until May 17, after coming within a vote or two of stripping the bill of old-age insurance. However, in spite of Witte's fears, the Senate Finance Committee's bill ended up better than the version passed by the House according to the Industrial Relations Counselors' Memorandum to Clients, No. 12, for May 27, 1935, because it "restored to the bill several features that appeared in it originally but were omitted in the House draft" (p. 1).

The bill also had a new preamble due to a Supreme Court's decision in early May ruling the Railroad Retirement Act unconstitutional. (Not to worry; that act was soon rewritten and passed while the benefits continued.) Since the reduction of unemployment and the efficiency and morale of the workforce were not judged to be "proper objects" of legislation relating to retirement according to the court, the emphasis in the revised preamble was on the general welfare. This change, which obscured the major role of industrial relations experts such as Latimer and Stewart in writing the act, was made because the constitution allows the government to support the general welfare through its taxing power:

Now, to achieve the original purpose, the administration turned to the taxing power and the general welfare clause of the Constitution. In the process, the ideology of social security was given formal sanction. After May 1935, proponents of retirement legislation talked less about efficiency, economy and unemployment relief than about social security and the needs of older workers, which were now a central policy goal rather than ancillary to some larger purpose" (Graebner 1980, pp. 162-163).

Just as the bill was about to pass the Senate, it faced one final obstacle: an amendment to allow the contracting out that was so vigorously opposed by Industrial Relations Counselors and most of the corporate moderates interviewed by Brown a few months earlier. The amendment was formally offered by a conservative Democrat from Missouri, Bennett Champ Clark, so it came to be called the Clark Amendment. From the point of view of the corporate ultraconservatives and Senators who opposed the whole social insurance program, the amendment was a perfect way to undercut the Social Security Act without voting against it. Despite protests from Roosevelt and Perkins, along with actuarial arguments against the amendment by Witte and other experts, it passed the Senate by the wide margin of 51 to 35 on June 19, followed by the Senate's passage of the act in general by a 77 to 6 vote the same day (Witte 1963, p. 106).

Roosevelt then made it clear that he would not sign legislation that included the Clark Amendment because it would create major actuarial and administrative problems, especially when companies--and their pension plans--went bankrupt or when employees left companies with private pension funds before they retired. The standoff led to a two-month delay while the conference committee argued about the issue and searched for a compromise. Congress finally agreed that the bill would be passed without the Clark Amendment, but with the provision that the Clark Amendment would be reconsidered in the next session of Congress after experts had a chance to see if contracting out could be made compatible with the overall system. Roosevelt signed the legislation on August 14.

Industrial Relations Counselors sent out a brief summary of the act's provisions in Memorandum to Clients, No. 14, two days after it passed. The memorandum first repeated its disapproval of the Clark Amendment, concluding "it seems clear that from the practical operating viewpoint such companies would have nothing to gain from the amendment" (p. 1). It then noted that members of the Industrial Relations Counselors staff were meeting with "the representatives of leading insurance companies and other interests concerned primarily with the sections on pensions." Finally, the memorandum announced that the organization already was working on supplemental plans:

Industrial Relations Counselors is now engaged in the formulation of several types of private plans which will supplement the pension benefits provided under the federal scheme and more adequately cover employees in the higher salary brackets. Our recommendations on future procedure may vary as between companies installing a plan for the first time and companies that have operated a formal plan for some years" (Memorandum to Clients, No. 14, August 16, 1935, p. 1, my italics).

This brief memorandum was followed a week later by a longer and more reflective one, No. 15, which nicely reveals the moderate conservative viewpoint and presages the moderates' agenda for dealing with the Clark Amendment. By and large, Industrial Relations Counselors experts were satisfied with the overall result, calling it a program that "will increase mass purchasing power and act as a shock absorber for our economic system" (Memorandum to Clients, No. 15, August 23, 1935, p. 2). The memorandum also said that the old-age pension provisions "were much better drawn than the unemployment compensation phase," a conclusion that comes as no surprise because Industrial Relations Counselors experts--and their ally, Barbara Armstrong--wrote them (Memorandum to Clients, No. 15, August 23, 1935, p. 3). Indeed, I see their satisfaction with federal old-age insurance as further support for my claim that Industrial Relations Counselors experts were involved in shaping it.

In addition, the memorandum also contained some surprisingly moderate and even progressive comments that showed the empirical basis of their policy analyses. For example, the report said that Industrial Relations Counselors' cross-national studies of social insurance systems convinced its authors "that a very considerable proportion of the costs must be borne by the public treasury," which put them in greater accord with the social workers than originally seemed to be the case (Memorandum to Clients, No. 15, August 23, 1935, p. 3).

The summary also contained several criticisms of the Clark Amendment that past memorandums had refrained from mentioning because the oganization wanted "to avoid any comment which might have been misconstrued as being political argument..." (Memorandum to Clients, No. 15, August 23, 1935, p. 6). First, contracting out would be more costly for corporations by as much as 33 to 100 percent. Second, the need to make back payments to the government for "each employee leaving a company before retirement age would subject a company fund to an unpredictable cash withdrawal, which would tend to force investments into a form suitable for commercial banks rather than proper insurance investments" (Memorandum to Clients, No. 15, August 23, 1935, p. 7). Third, private plans would have "burdensome administrative and reporting problems" so that the government could oversee them properly. Finally, the existence of private plans "would tend to weaken the actuarial basis of the government old-age benefit plan" due to the fact that companies with the lowest costs were mostly like to set up their own plans, leaving the government "to deal with the poorest risks" (Memorandum to Clients, No. 15, August 23, 1935, pp. 6-7).

This list of objections to the Clark Amendment was the opening salvo in the effort to make sure that it was not adopted. In the end, no substitute for the Clark Amendment was ever offered, but there had to be a lengthy behind-the-scenes effort to remove it, led of course by Industrial Relations Counselors and committees of the Social Science Research Council.

Once the Social Security Act had passed and the Clark Amendment was buried, corporate moderates, New Deal liberals, and advocates of the Townsend Plan began to argue about how to improve the act, with much of the discussion taking place within Social Science Research Council committees. All parties agreed that the system should be financed on a pay-as-you-go ("paygo") basis rather than by building up a pension reserve, as the act had mandated at the last minute at the insistence of the fiscally conservative Secretary of Treasury. Furthermore, they could agree to a payment schedule that gave a slight boost to low-income retirees and restrained benefits at the top. Liberals, labor, and Townsendites favored these changes because of their concern that low-income people should have enough money to live on. They suited Keynesian economists because they put money into the hands of those most likely to spend it and avoided the drag on the economy that a government pension fund would create.

On the corporate side of the divide, insurance companies pushed for such a plan because it left plenty of room for their profitable private plans for those with higher incomes, and especially for the corporate executive plans that were their biggest customer target. Indeed, this desire was so great that one insurance company even helped finance the liberal reformers who pushed for "adequacy" in old-age pensions (Sass 1997, p. 282, ftn 17). As for the corporate moderates at other corporations, they favored pay-as-you-go in good part because they did not want the government to have any investment funds to manage, which they considered a dangerous precedent that might tempt the Social Security Administration to push for even better pensions.

Several other reforms to the 1935 act, which I do not have the space to discuss here, were carried out in 1939 based on recommendations by a Social Security Advisory Committee headed by J. Douglas Brown and including several members from the Social Science Research Council committees (Fisher 1993; Manza 1995). With the passage of the 1939 amendments, social security had reached just about the form it would have for the next 40 years. (Amendments in 1950 finally added coverage for agricultural and domestic workers, and also increased the size of pensions for everyone by a significant amount.)

Conclusion

The process leading to the drafting and passage of the Social Security Act was long and drawn out, but the final outcome fits well with the class-domination/class conflict theory that underlies this web site. It also provides excellent evidence for state building by the capitalist class through its policy-planning network, which proposed a sophisticated and comprehensive program based on years of experience with private company plans and with group plans developed by life insurance companies. The ultraconservatives in the corporate community went along reluctantly with the proposal on its merits and then turned against it at the last minute on general ideological, anti-New Deal grounds (a story I did not get into in this document). But the ultraconservatives could not stop the program or bring about any changes in it.

Organized labor made its presence felt through the grassroots activity of the Railways Employees National Pension Association, which forced the corporate community to consider government involvement in worker pensions. Then, too, the American Federation of Labor's decision in 1932 to end its opposition to government insurance removed an obstacle to a government program. As for liberals and social workers, they wanted something better and more generous than the corporate moderates' plan when it came to old-age pensions and unemployment compensation, but they ended up as lobbyists for the plan that emerged from the policy-planning network.

As for the plantation capitalists of the South, they wielded a veto power through the Southern Democrats. They were the reason why agricultural and domestic workers were not covered by old-age insurance until a little later even though an early study by the Committee on Economic Security research staff said that such coverage was feasible (Alston and Ferrie 1999). I also think the substance of the legislative battles in Congress show that the Southern Democrats were the main reason why unemployment insurance was placed under the control of the states, but I did not have space to discuss that issue. Southern Democrats also eliminated the civil service requirements for the staff that administered the programs and any minimal federal standards for payment levels, which both the Industrial Relations Counselors experts and the more cautious reformers favored. This allowed the representatives of the plantation capitalists to put their local cronies in charge of agencies and keep benefit payments low enough to maintain full control of their workforce (see Quadagno 1988, for detailed information on the powerful impact of the Southern Democrats on this legislation).

For all its shortcomings, old-age insurance is perhaps the most popular program ever developed by the federal government. And ever since social security benefits were first indexed for inflation in the early 1970s--with the approval of President Richard M. Nixon, by the way--they have reduced the previous high incidence of poverty among the elderly. But despite its conservative origins and great success, the system has been under constant attack since the 1990s by ultraconservative and moderate conservative experts from the Cato Institute, the Heritage Foundation, the American Enterprise Institute, and the Brooking Institution who want to privatize at least part of it and/or limit benefits. Using projections based on very low and unlikely estimates concerning the rate of economic growth, they claim that the system may not be solvent in 30 or 40 years. Their scare campaign through the media has convinced many young adults that the system may not be around when it comes time for them to retire. The ultraconservatives claim social security is a bad "investment" for people because it does not earn a high enough "rate of return." They therefore suggest that people be allowed to withdraw from social security and invest their money for retirement purposes through Wall Street (see Rogne, Estes, Grossman, Hollister, and Solway 2009, for information on how the corporate community has misled the public on social security).

The defense of the system has fallen to the liberal-labor coalition, which is joined on this issue by two potent lobbying forces, the American Association of Retired Persons and the National Committee to Preserve Social Security and Medicare. The defenders note that social insurance is a communal concept that insures a decent life for those who happen to live to an old age, not an individual investment strategy. They argue that any future deficits can be averted with a combination of very small changes, such as extending the payroll tax to include income over $107,000 a year (this 2009 figure will go up a little bit each year). They point to the long downturn in the stock market in the 1970s and the recent crash of the stock market to suggest that the stock market is not always as rosy as it was in the 1990s, which means that many people might not have what they thought they would a few years from now if the conservatives and ultraconservatives in the policy-planning network have their way.

If the corporate community ever succeeds in "reforming" social security by increasing the retirement age or reducing pensions, they will have pulled off one of the most massive peaceful transfers of wealth from the middle class to the upper class in Western history. That's because any cuts in payments or increases in the age of retirement would mean that the people who started paying higher payroll taxes in 1983 that were supposed to guarantee their security until 2042 won't get their fair return. Instead, the money supposedly "set aside" or "invested" for social security has been used to finance the very large tax cuts to the corporate rich over the past 25 years. Now that it is time for the Department of the Treasury to put the money it borrowed back into the Social Security Administration, the White House and Congress may decide not to honor this obligation.

In other words, the money the government has received from increased payroll taxes since 1983 has been used to help fund the federal government's yearly expenses, making lower taxes for high-income people possible without running even larger deficits. Put still another way, if benefits are cut, it is the equivalent of defaulting on some of the $1.2 trillion in Treasury bonds that President Reagan and other political leaders, Democrats as well as Republicans, told taxpayers they were purchasing with payroll taxes (Baker 2001; Baker and Weisbrot 1999).

And yet a step in the direction of such "reforms" seemed to be emerging early in the Obama Administration when the president talked with Senate Republicans about setting up a "bipartisan task force" to propose changes that might lead to later retirement ages, higher payroll taxes for average workers, and even caps on the amount that could be spent for Social Security (Calmes 2009; Greider 2009). Such a task force, which would have been staffed by economists from the policy-planning network, some of them already working in the Obama Administration, was strongly opposed by spokespersons and lobbyists for the liberal-labor coalition, which in turn generated opposition by liberals in the House and Senate. But according to a conservative New York Times columnist, the hope for a bipartisan bargain is not completely dead. Based on his conversations with "four senior members of the administration" he concluded that President Obama is "extremely committed to entitlement reform and is plotting politically feasible ways to reduce Social Security as well as health spending" (Brooks 2009, p. A23). He added that he "had the impression they'd be willing to raise taxes on the bottom 95 percent of earners as part of an overall package."

So, old-age pensions have surprising origins that fit my theory better than rival theories, but now they are under attack by the current incarnation of the corporate community, which grew far richer between 1980 and 2008 than anyone ever could have imagined, thanks in part to the tax cuts for the rich that were partly paid for by looting the Social Security Administration's investment portfolio (Baker, 2001; Greider, 2009).

For further information on how to preserve and improve social security in reasonable ways, see the National Committee to Preserve Social Security and Medicare at ncpssm.org.

For information on how to become involved in student activism to preserve social security, see Students for Social Security at studentsforsocialsecurity.org.

References

Alston, Lee J. and Joseph P. Ferrie. 1999. Southern Paternalism and the American Welfare State. New York: Cambridge University Press.

Amenta, E. 2006. When movements matter: The Townsend Plan and the rise of social security. Princeton: Princeton University Press.

Armstrong, Barbara N. 1965. "Oral History." in New Deal Oral History Project. New York: Columbia University.

Baker, Dean. 2001. "Defaulting on the Social Security Trust Fund bonds: Winner and losers." Center for Economic and Political Research, Washington.

Baker, Dean and Mark Weisbrot. 1999. Social Security: The Phony Crisis. Chicago: University of Chicago Press.

Board, National Industrial Conference. 1931. Elements of industrial pension plans. New York: National Industrial Often Board.

Brents, Barbara. 1984. "Capitalism, Corporate Liberalism and Social Policy: The Origins of the Social Security Act of 1935." Mid-American Review of Sociology 9:23-40.

Brooks, David. 2009. "When Obamatons respond." Pp. A23 in New York Times. New York.

Brown, J. Douglas. 1965. "Oral History." in New Deal Oral History Project. New York: Columbia University.

Brown, James Douglas. 1972. An American philosophy of social security: Evolution and issues. Princeton: Princeton University Press.

Bulmer, Martin and Joan Bulmer. 1981. "Philanthropy and Social Science in the 1920s: Beardsley Ruml and the Laura Spelman Rockefeller Memorial, 1922-29." Minerva 19:347-407.

Calmes, Jackie. 2009. "Obama finds resistance in his party on addressing Social Security " Pp. A15 in The New York Times.

Cohen, Adam. 2009. Nothing to fear: FDR's inner circle and the hundred days that created modern America. New York: Penguin.

Cowdrick, Edward. 1928. Pensions: A Problem of Management. New York: American Management Association, Annual Convention Series, No. 75.

Davies, Gareth. 1999. "The unsuspected radicalism of the Social Security Act." Pp. 56-71 in The Roosevelt years: New perspectives on American history, 1933-1945, edited by R. Garson and S. Kidd. Edinburgh, Scotland: Edinburgh University Press.

Domhoff, G. W. 1996. State autonomy or class dominance? Case studies on policy making in America. Hawthorne, NY: Aldine de Gruyter.

Fishback, Price V. and Shawn E. Kantor. 2000. A prelude to the welfare state: The origins of workers' compensationChicago: University of Chicago Press.

Fisher, Donald. 1993. Fundamental Development of the Social Sciences. Ann Arbor: University of Michigan Press.

Gordon, Linda. 1994. Pitied But Not Entitled. New York: Free Press.

Graebner, William. 1980. A History of Retirement. New Haven: Yale University Press.

—. 1982. "From Pensions to Social Security: Social Insurance and the Rise of Dependency." in The Quest for Security, edited by J. N. Schact. Iowa City: Center for the Study of the Recent History of the United States.

Greider, William. 2009. "The looting of Social Security." The Nation, March 2, pp. 12-15.

Jenkins, J. Craig and Barbara Brents. 1989. "Social Protest, Hegemonic Competition, and Social Reform." American Sociological Review 54:891-909.

Karl, Barry. 1974. Charles E. Merriam and the study of politics. Chicago: University of Chicago Press.

Kaufman, B. 2003. "Industrial Relations Counselors, Inc.: Its history and significance." Pp. 31-112 in Industrial relations to human resources and beyond: The evolving process of employee relations management, edited by B. Kaufman, R. Beaumont, and R. Helfgott. Armonk, NY: M.E. Sharpe.

Klehr, Harvey. 1984. The Heyday of American Communism. New York: Basic Books.

Klein, Jennifer. 2003. For all these rights: Business, labor, and the shaping of America's public-private welfare state. Princeton: Princeton University Press.

Latham, Earl. 1959. The Politics of Railroad Coordination, 1933-1936. Cambridge: Harvard University Press.

Latimer, Murray and S Hawkins. 1938. "Railroad retirement system in the United States." in Murray Latimer Papers. Washington George Washington University Library, Special Collections.

Manza, Jeff. 1995. "Policy experts and political change during the New Deal." Ph.D. Thesis, Sociology, University of California, Berkeley.

Nelson, Daniel. 1969. Unemployment Insurance: The American Experience, 1915-1935. Madison: University of Wisconsin Press.

Orloff, Ann Shola. 1993. The politics of pensions: A comparative analysis of Britain, Canada, and the United States, 1880-1940. Madison: University of Wisconsin Press.

Quadagno, Jill S. 1988. The transformation of old age security: Class and politics in the American welfare state. Chicago: University of Chicago Press.

Roberts, Alasdair. 1994. "Demonstrating neutrality: The Rockefeller philanthropies and the evolution of public administration, 1927-1936." Public Administration Review, pp. 221-228.

Rogne, Leah , Carroll L. Estes, Brian R. Grossman, Brooke A. Hollister, and Erica S. Solway. 2009. Social insurance and social justice: Social Security, Medicare, and the campaign against entitlements. New York: Springer Publishers.

Sass, Steven A. 1997. The promise of private pensions: The first hundred years. Cambridge: Harvard University Press.

Saunders, Charles B. 1966. The Brookings Institution: A fifty-year history. Washington: Brookings Institution.

Schlabach, Theron F. 1969. Edwin E. Witte: Cautious reformer. Madison,: State Historical Society of Wisconsin.

Skocpol, T. 1992. Protecting soldiers and mothers: The political origins of social policy in the United States. Cambridge, MA.: Harvard University Pressf.

Stewart, Bryce and Meredith Givens. 1934a. "Planned protection against unemployment and dependency: Report on a tentative plan for a proposed investigation." Social Science Research Council, New York.

—. 1934b. "Project report: Exploratory study of unemployment reserves and relief (economic security)." Social Science Research Council, New York.

Weinstein, James. 1968. The corporate ideal in the liberal state. Boston: Beacon Press.

Witte, Edwin E. 1963. The Development of the Social Security Act. Madison: University of Wisconsin Press.

printable version of this page

All content ©2012 G. William Domhoff, unless otherwise noted. Unauthorized reproduction prohibited. Please direct technical questions regarding this Web site to Adam Schneider.