6 6 Providing Employment Assurance for American Workers
Our greatest primary task is to put people to work. This is no unsolvable problem if we face it wisely and courageously. It can be accomplished in part by direct recruiting by the Government itself, treating the task as we would treat the emergency of a war, but at the same time, through this employment, accomplishing greatly needed projects to stimulate and reorganize the use of our natural resources.
—Franklin D. Roosevelt, First Inaugural Address, March 4, 1933
Of all the programs and legislative reforms comprising the Roosevelt Administration’s response to the Great Depression, none is more emblematic of the New Deal than the era’s direct job-creation programs—most famously the Works Progress Administration (WPA) and Civilian Conservation Corps (CCC). It might be supposed, therefore, that when Progressive economists consider whether the New Deal has anything to teach us about responding to the problem of unemployment, these are the initiatives that would attract their attention. But the only attribute of these programs that appears to interest Progressive economists today is their macroeconomic effect—as if the only thing they have to teach us about combating unemployment is the negative lesson derived from the Roosevelt Administration’s failure to engage in enough deficit spending to end the Great Depression as quickly as it could have. The jobs provided to WPA and CCC workers simply do not register in their assessment, either because they assume those jobs are already accounted for in the multiplier effect of the deficit spending that created them, or because their Keynesian bias has blinded them to the possibility of addressing aggregate shortfalls in employment by means other than accelerating economic growth.
This chapter argues that this view of the New Deal’s direct job-creation programs misses the most important lessons they have to teach us. It misperceives the multiple objectives they served in combating the labor-market effects of the Great Depression. It ignores the social welfare benefits that employment in the programs provided to millions of unemployed workers and their families; it discounts the value of the goods and services they produced for the nation’s communities; it overlooks the ability of programs like the WPA to enhance the effectiveness of Keynesian anti-cyclical measures; and it disregards the ability of such programs to achieve the ultimate goal of Keynesian economic policy—sustained full employment.
To fill these gaps in our appreciation of the Roosevelt Administration’s response to the problem of unemployment the chapter (1) reviews the history of the New Deal’s direct job-creation initiatives; (2) describes the policy vision that motivated their establishment; (3) discusses the advantages of this strategy for achieving genuine full employment;1 and (4) explains why the New Deal strategy would have would have constituted a better response to the “Great Recession” than the Keynesian strategy advocated by most progressive economists.
When the Roosevelt Administration assumed office in early 1933, a consensus existed across the relevant political spectrum that some form of government intervention in the economy was necessary to meet the relief needs of the population, reduce unemployment, and facilitate a return to prosperity. The continuing debate concerned the form this intervention should take. The Roosevelt administration was eclectic and pragmatic in the strategies it pursued, guided by varied and often conflicting visions of how the economy should be structured. It also had to contend with other centers of power and interest, both inside and outside government.
The policies that emerged in this context reflected compromises, not the pure application of a particular ideological agenda. The strategies pursued had various goals—to “prime the pump” of business activity, increase consumption, stabilize the financial system, increase the money supply, ease the availability of credit, promote “business confidence,” reduce industrial strife, or introduce a measure of economic planning into the management of the economy. Ironically, in light of his reputation as a spendthrift, the one firm belief Roosevelt held regarding federal economic policy was that balancing the federal budget—if it were possible—would hasten the economy’s recovery. Roosevelt was not alone in holding this view, and he made sure it was well represented in his administration—principally in his choice of Secretaries of the Treasury. When his advisors finally agreed the time was right to rein in spending, the result was the disastrous recession of 1937–1938 that delayed the economy’s full recovery by several years.
Despite their differences, however, the New Dealers shared a common view of the general nature of the nation’s joblessness problem. This view directly contradicted the presumption embedded in the nation’s existing Poor Law system (and supported by the teaching of classical and neo-classical economists) that joblessness was a voluntary condition.2 The New Dealers believed that joblessness was caused by a lack of jobs, not by a failure on the part of jobless individuals to seek or accept work. They believed that cutting wages would increase joblessness, rather than reduce it, because of its depressing effect on consumer purchases. The goal of government initiatives to combat joblessness should be to close the economy’s job gap, not to correct the supposed moral failings of jobless individuals or to put pressure on them to seek and accept presumably available work. Concerns about the negative effects public assistance might have on jobless individuals persisted, but they were overwhelmed by concerns about the negative effects of joblessness itself. The New Dealers believed that society had an obligation to offer aid to persons denied the opportunity to be self-supporting, and that the stigma associated with the receipt of such assistance under the nation’s existing Poor Law system was inappropriate.
Consistent with the eclecticism of the Roosevelt administration, a variety of reforms were initiated to address the joblessness problem. Some of these reforms focused directly on the problem of unemployment, such as the administration’s direct job-creation initiatives. Others focused on problems that had an indirect effect on unemployment, such as the widespread use of child labor or the lack of old age pensions. All of the reforms they implemented pursued a common strategy consistent with the New Dealers’ shared view of the nature of the unemployment problem. Their goal was to narrow the economy’s job gap either by increasing the number of jobs available or by reducing the number of job seekers.
The most obvious strategy devised to achieve this goal was to use public funds to create jobs. They did this in two ways. The first was to increase federal funding for contracted public works. The second was to establish public employment programs for needy workers in which the government itself acted as the employer. In addition to the direct job-creation effect of these initiatives, it was believed they would stimulate job creation in the private sector by increasing both consumer purchasing power and capital goods orders.3
The New Deal’s contracted public works initiative was implemented mainly through the Public Works Administration (PWA), established in the summer of 1933 with an initial $3.3 billion authorization.4 This was a sizable fiscal commitment for a government whose expenditures in 1933 totaled only $3.4 billion5 in an economy whose GDP equaled only $56.4 billion.6 Congress intended for these funds to be fully committed, if not fully spent within two years, so the net fiscal stimulus contemplated was on the order of 2–3 percent of GDP a year. All PWA project hiring was done by private contractors.
The program took much longer to spend its authorization than was contemplated, and other job-creation programs were adopted in the interim, but the PWA remained the primary source of funding for large-scale public works in the United States between 1933 and the end of the decade. Reauthorized and granted additional funding several times during this period, the PWA made grants totaling about $2.3 billion to state and local governments and another $1.8 billion to other federal government agencies for the support of public works construction projects. State and local governments contributed another $1.9 billion to this effort, though a significant portion of their contribution was initially financed with loans furnished by the PWA.7 Thus, a total of approximately six billion dollars was spent on PWA projects between mid-1933 and the end of the decade, an average of about 1.3 percent of GDP annually.8
Between their establishment in mid-1933 and March 1939, PWA projects furnished approximately 1.7 billion hours of direct employment and paid wages averaging seventy cents per hour ($10.59 in 2012).9 The maximum work week for individuals employed on these projects was originally set at thirty hours, though this was later raised to the industry standard of forty hours.10 Based on this work week, the program provided an average of 183,204 jobs at any point in time, a figure that amounted to only three-tenths of 1 percent of the nation’s labor force.
For each dollar spent on direct labor costs, however, PWA projects spent an average of $1.79 on materials, and the U.S. Bureau of Labor Statistics estimated that approximately 2.5 jobs were created in producing those materials for every job created within the PWA itself.11 Adding these jobs to the average of 183,204 jobs provided directly on PWA projects results in an estimated employment effect of 485,000, or about 0.9 percent of the nation’s labor force at the time.
This estimate of the program’s employment effects is still incomplete, however, because it does not include the full multiplier effect of program expenditures. If the tail-end of the multiplier effect of PWA expenditures equaled fifty percent of the 485,000 jobs attributed to the program’s direct purchases of labor and materials,12 the overall employment effect of the program equaled about 750,000 jobs, or about 1.5 percent of the nation’s labor force—in exchange for an annual investment that we have noted averaged about 1.3 percent of GDP between 1933 and 1939. The program was phased out during World War II, with its functions transferred to the Federal Works Agency in 1943.13
Although the difference is often overlooked, it is important to distinguish the PWA’s conventional public works model (which relied on private contractors funded by the government to hire workers and carry out projects) from the New Deal’s direct job-creation initiatives (in which the federal government itself hired workers and carried out projects). The New Deal’s direct job-creation effort was embodied in four major programs: the Civilian Conservation Corps (CCC) established in the spring of 1933; the Civil Works Administration (CWA), established as an emergency initiative using PWA funds in the winter of 1933–1934; the WPA, established in 1935; and the National Youth Administration (NYA) also established in 1935 (originally as the youth division of the WPA but later reorganized as a freestanding program).14 The CWA lasted only four months, but the CCC, WPA, and NYA all continued to operate until the early 1940s, when they were phased out as spending on World War II and a dramatic expansion of the armed forces pushed the unemployment rate down to the genuine full employment level (1.9 percent in 1943).
Why, if Roosevelt believed in balancing the budget, was he willing to engage in deficit spending to fund programs like these? The principal reason was that there was one thing Roosevelt considered even more important than balancing the budget. That was the duty of government to fulfill its obligation to secure what he viewed as the right of every member of society to economic security. Simply stated, although he may have thought that balancing the federal budget was the road to recovery, Roosevelt believed the fulfillment of the government’s social welfare obligations to the American people took priority over that goal.
Roosevelt’s views on this point were always clear. In his principal campaign address on the federal budget in 1932, candidate Roosevelt excoriated President Hoover for his administration’s failure to balance the federal budget, but he also made it clear that he was willing to operate in the red to meet human needs. After summarizing his strategy for cutting government expenses, he offered the following caveat.
At the same time, let me repeat from now to election day so that every man, woman and child in the United States will know what I mean: If starvation and dire need on the part of any of our citizens make necessary the appropriation of additional funds which would keep the budget out of balance, I shall not hesitate to tell the American people the full truth and ask them to authorize the expenditure of that additional amount.15
This attitude created an opportunity for the president’s social welfare advisors to play a particularly prominent role in shaping his administration’s response to the problem of unemployment. Joblessness was not just an economic problem to be solved by promoting the economy’s recovery. It was also a social problem that required the immediate intervention of government to provide for the needs of unemployed workers and their families. The most important and creative aspects of the New Deal’s response to the problem of unemployment were viewed by the New Dealers themselves in this light—as social welfare rather than economic policy measures—and the social welfare strategy embodied in those measures was conceived and developed by the President’s social welfare advisors rather than his economic advisors.
Harry Hopkins was Roosevelt’s public relief administrator. A social worker with strong administrative experience, he was appointed by the president to head the Federal Emergency Relief Administration (FERA) established in the spring of 1933 to distribute $500 million in federal funds appropriated by Congress to shore up the nation’s state-based and locally administered public relief system.16 Hopkins’s principal aide was another social worker, a white Southerner named Aubrey Williams, who grew up in poverty and was a lifelong champion of civil rights.17
Frustrated by the narrow range of reforms the FERA was empowered to make in the nation’s existing social welfare system, Hopkins and Williams developed an alternative model for the delivery of public aid to the unemployed. In a conceptual memo outlining their plan, Williams wrote that “relief as such should be abolished.”18 Instead, the unemployed should be offered real jobs paying good daily wages, doing useful work suited to their individual skills. In other words, instead of offering the unemployed public relief, they should be offered quality employment of the sort normally associated with contracted public works. However, to minimize both the cost of the undertaking and the amount of time needed to launch it, the government should serve as its own contractor, and the projects undertaken should be both less elaborate and more labor-intensive than conventional public works.
In late October 1933, Hopkins pitched a job-creation proposal to President Roosevelt based on the model he and Williams had devised. Disappointed by the slow pace in getting the PWA up and running, and concerned about growing political unrest among the unemployed, Roosevelt was quick to embrace Hopkins’s proposal. A week later, the Civil Works Administration (CWA) was formally established by executive order, with Hopkins at its head and an initial budget allocation of $400 million diverted from the PWA.19
The program was funded only through the winter of 1933–1934, but it still stands as the largest public employment program ever established in the United States. With a peak employment of 4.3 million in a labor force of fifty-one million, the CWA provided employment to about 8.4 percent of the nation’s work force during its short existence.20 The administrative task of establishing the CWA—which moved from nothing more than an idea to a fully operational program with four million employees in about ten weeks’ time—was gargantuan. It employed seven and one-half times as many people as the rest of the federal government (civilian and military) combined.21 A program of similar relative dimensions in the United States today would have to create almost thirteen million jobs.
Also, though it fell short of realizing the policy goals Hopkins and Williams had formulated for the reform of public aid for the unemployed—the creation of a program that offered unemployed workers jobs that were indistinguishable from regular employment and devoid of associations with public relief—it embodied enough features of that model to mark a sharp and definitive break with the nation’s existing public relief system.
The most important constraint Hopkins and his associates faced in implementing their reforms was that they had only enough funding to create 4 million jobs at a time when 1) there were over 12 million unemployed workers22 and 2) they had a continuing responsibility under the FERA to provide support for existing recipients of public assistance, about 40 percent of whom were families headed by unemployed workers enrolled in locally administered, poor-quality work relief programs of the sort Hopkins was intent on replacing.23 If the CWA had been funded and authorized to provide jobs for all unemployed workers, the existing population of unemployed relief recipients could have been directed to apply for work in the program on the same basis as other unemployed workers. However, with only enough funding to create about 4 million jobs, the decision was made to reserve approximately half of the positions in the CWA for public assistance recipients (so that all FERA work relief participants could be transferred to the CWA) while using normal skill- and experience-based hiring criteria (subject to a statutorily mandated preference for war veterans) to fill the other half.24
Another constraint limiting Hopkins’ options was that the CWA’s initial $400 billion in funding was subject to the same statutory restrictions as the PWA. These included a requirement that the funds be used only for the planning and execution of construction projects. To allow for the employment of persons for whom such work would not be suitable, Hopkins used his limited pot of FERA funds to establish a parallel Civil Works Service (CWS) program for white-collar and professional workers (both male and female),25 and a Women’s Division (transferred to the CWA from the FERA) to provide non-construction projects for working-class women and to encourage the hiring of women in the CWS and in non-construction positions in CWA construction projects.26
Although it accounted for less than 5 percent of total program employment,27 the CWS was quite successful and included some of the CWA’s most innovative projects.28 The Women’s Division was far less successful. The U.S. work force was approximately 25 percent female at the time, and while confirmatory data are hard to find, there is no reason to believe they suffered less unemployment than men during the 1930s.29 Nevertheless, only 12.4 percent of all CWA applicants were women, and only 7.5 percent of all program jobs went to women.30 Hopkins’ goal was to provide jobs for about 400,000 women—approximately 10 percent of total expected program employment—but the Women’s Division struggled in its efforts to recruit women and to develop appropriate projects to employ them. The Division’s staff consisted mainly of FERA personnel with backgrounds in voluntary charity work, and they tended to hold more traditional views of the functions of work relief than the engineers Hopkins hired to run the program’s construction projects. This contrast was especially marked at the state and local level, and the Women’s Division’s very capable director, Ellen Woodward, had trouble getting her field staff to develop projects other than the sewing rooms associated with traditional public assistance. It is hardly surprising, therefore, that working-class women had trouble figuring out whether the CWA was offering them work or just a continuation of conventional public assistance.31
The Women’s Division had more success developing appropriate work projects in collaboration with the CWS. These collaborative projects employed women in a wide range of endeavors, including nursing, education, social work, public health, child care, library work, and the collection of statistical data.32 But given that the CWS created a total of only 211,000 jobs, it could hardly make up for the more limited employment opportunities provided women in the rest of the CWA.
African Americans and other people of color were also shortchanged. Discrimination on the basis of race or color was prohibited in the application of eligibility and wage standards, but the eligibility rule was a hard one to enforce at the local level where the actual hiring occurred. Moreover, segregated work assignments and, in some instances, wholly segregated work projects were tolerated. It was also common for skilled minority workers to be discriminatorily categorized as unskilled. The CWA staff in Washington did not direct or approve of this discrimination, but they did not object to it, either.33
On the other hand, non-white workers were paid the same as white workers with the same job classification, and this was enough to precipitate significant political opposition to the CWA in the South, where whites resented the idea of blacks’ being paid the same wages they received, and employers relied extensively on cheap black labor. The experience of African Americans with the CWA was accordingly mixed, particularly in the South, where discriminatory practices were most common—and since almost 80 percent of the African American population of the United States still lived in the South at the time, it was the way the program functioned there that defined the experience most African Americans had with it.
Despite these shortcomings, 9 million people applied for the approximately 2 million CWA jobs that were not reserved for public relief recipients. This is a stunning figure, since national unemployment stood at less than 11 million at the time outside of FERA work relief programs.34 To emphasize the non-relief character of this hiring, it was performed by the newly organized United States Employment Service (USES) rather than local relief offices. At the same time, however, it is worth noting that local relief offices were swamped with new applicants for public aid, since job seekers quickly realized that qualifying for relief was a surer means of getting a CWA job than applying for one through the USES.35
Special hiring procedures were also adopted for skilled craftsmen. Instead of requiring applicants for these positions to apply through the USES, unions were allowed to refer their members in accordance with customary procedures for the trades in question. More importantly, the CWA agreed not to fill these positions from among USES applicants unless a local union failed to refer enough qualified workers. In other words, the CWA formally adopted a union-shop policy for the skilled trades; however, local CWA administrators often ignored this policy unless local unions insisted on its observance.36 It should be noted, however, that while this policy was supportive of unions, it reduced the access of African Americans to skilled work in the CWA because the craft unions that benefited from the policy were among the most racially discriminatory in the American labor movement at the time.37
The CWA’s wage policies broke even more decisively with traditional relief practices than its eligibility standards. Customary practice in work relief programs had been to limit an individual’s earnings to the family’s “budget deficiency”—the difference between the family’s available resources and their “need” as determined by local relief officials. Consequently, the number of hours an individual was required to work in a traditional work relief program depended on the size of the individual’s budget deficiency, and this was true of FERA-funded work relief projects as well.38 Thus, despite a minimum wage (30 cents per hour) that would have generated a twelve-dollar weekly income for a forty-hour work week, actual earnings on FERA-funded work relief projects averaged less than five dollars per week in the period immediately preceding the establishment of the CWA.39
No individual working-hour limitation existed under the CWA. Hourly wage minimums were generally higher than those paid in FERA work relief programs, but the more important difference was that everyone worked the same number of hours. The result was that average weekly earnings among CWA workers (about $15, until program funds began to run short in mid-January) were three times as great as the benefits received by individuals enrolled in FERA-funded work-relief projects.40 In fact, average CWA earnings during this period were the highest of any New Deal direct job-creation program.41
The CWA’s hourly wage scale was the same as the PWA’s.42 It was also national policy (though frequently ignored by program administrators at the state and local level) to recognize locally negotiated union contracts in the construction trades as determinative of prevailing wage rates.43
The hourly rates paid by the CWA were controversial because they were often higher than what employers in particular regions (especially the South) or industries (especially agriculture) were accustomed to paying. What this debate tended to ignore, however, was that the program’s 30-hour work week limited the earnings of program employees below what their hourly wage rates suggested. Moreover, as program funds began to run short in mid-January, the program’s work week was shortened to 24 hours in order to spread the remaining work as widely as possible. As a result, average program earnings declined from about fifteen dollars per week to about $11.30 per week, compared to average weekly earnings by privately employed workers of about $20 in 1933.44
Although Hopkins’s goal was to give unemployed worker jobs that utilized their existing skills, statutory and practical limitations made this impossible in most cases. First, as previously noted, statutory restrictions limited the CWA to construction projects. Second, a pre-existing FERA policy required that projects be performed only on public property. Third, no project was supposed to be undertaken that would duplicate work normally performed by state and local government employees. Fourth, no projects were supposed to be approved that could qualify for funding by the PWA.45
Project selection was further constrained by timing issues and the desire to maximize the program’s employment effect. This meant projects had to be labor-intensive and capable of completion in a short period of time. It also meant they could not require significant advance planning or be hard to shut down on short notice. Finally, project selection was subject to weather and political constraints.
Although the CWA hired its own work force and administered its own projects, the program relied on other government agencies (at the federal, state, or local level) to sponsor the projects it undertook. The involvement of state and local government sponsors was limited to providing plans for the projects and contributing the cost of the materials and supplies used to complete them. At the federal level the CWA assumed all program costs but still relied on the sponsoring agency to propose and plan the projects it undertook.
The quality of the sponsored projects varied widely. First, by taking over all FERA-funded work projects from the local relief officials who had been administering them, the CWA burdened itself with an initial portfolio of poor-quality activities. While the CWA gained direct administrative control over these projects, it took time to implement significant improvements in their quality.
A second large group of projects originated with suggestions for new undertakings by local government officials. Approval authority for these projects was exercised by state CWA administrators whose review of the projects was often cursory. The quality of these projects varied greatly. Where sponsors had already developed plans for suitable construction projects, the activities tended to be quite successful and provided good value. Where advance planning had not been completed, the results were less satisfactory, though the CWA’s newly recruited and very competent Engineering Division was, over the life of the program, able to achieve steady improvement in the quality of the program’s construction work.
A third large group of projects originated at the federal level. These projects were sponsored by a variety of federal government agencies, including the Treasury Department, the Departments of the Interior and Agriculture, the Commerce Department, and the War Department. Most were developed in collaboration with CWA staff and also required the approval of a special office established within the Engineering Division that vetted them for quality control purposes. By all accounts, the CWA’s highest quality projects were found in this group.
As for the type of work performed, the single largest category of CWA projects consisted of road work, which accounted for 35 percent of all project expenditures and employed close to half the program’s entire work force. The road work consisted mainly of minor repairs and improvements rather than new construction, and in many rural areas this was the only type of CWA work available.46
The CWA administration was not happy with the predominance of road work in the program’s activities since it was associated both historically and in the public’s mind with the kind of work relief the CWA was supposed to replace. Indeed, a significant proportion of these projects were taken over from FERA-funded programs. The difference between these earlier programs and their CWA counterparts was not immediately apparent to the public walking or driving by a CWA work crew. The fact that these projects were more visible than other, higher quality projects also made it more difficult for the CWA leadership to explain the innovative character of the CWA to the public.47
Nevertheless, it would be wrong to underestimate the social utility of this work at a time when state and local governments lacked the funds needed to maintain, let alone improve, the roads and highways upon which local commerce depended. The CWA built or improved over 250,000 miles of roads. In rural areas, improving farm-to-market roads was a priority. In urban areas, significant repairs and improvements were undertaken—such as a Chicago project that employed 11,500 CWA workers to lay brick pavement in a major street-improvement project.48
The next-largest category of CWA projects consisted of construction and repair work on public buildings. Accounting for about 15 percent of project expenditures, approximately 60,000 public buildings were repaired or constructed, two-thirds of which were schools. Public health and sanitation activities constituted another major activity. Almost 2,300 miles of sewer lines were laid or repaired, swamp-drainage projects to fight malaria employed 30,000 CWA workers, and 17,000 unemployed coal miners were employed sealing abandoned coal mines to protect ground-water supplies. CWA workers also were employed in emergency disaster relief—either fighting floods or assisting in post-flood cleanup and repair work.49
Other CWA project categories included improvements to public recreational facilities and to public transportation and utility systems. The program constructed 4,000 athletic fields, 2,000 playgrounds, 350 swimming pools, and 150,000 privies. Surprisingly, the CWA built 469 airports and improved another 529, but this was the dawning of the aviation age, and the facilities in question mainly consisted of unpaved landing fields.50
Because the CWS was not limited to construction projects and employed professionals, the projects it undertook were more varied. And since most of these projects were sponsored by federal government agencies, they also benefited from the attention of the CWA’s Washington staff. Professional associations also assisted in the design and management of many of these projects.
Education projects provided jobs in local schools for 50,000 laid-off teachers. Another 13,000 kept small rural schools open through the winter, while 33,000 were employed in adult education and nursery school programs. Adult education classes staffed by CWS teachers were attended by 800,000 people during the winter of 1933–1934, and 60,700 preschool children attended CWS nursery schools. The latter were generously staffed and provided warm clothes, hot meals, medical care, and parent-education services in addition to childcare.51
Twenty-three thousand CWS nurses staffed a nationwide child health study, and 10,000 more were employed in a variety of other programs. The U.S. Coast and Geodetic Survey sponsored a triangulation and mapping project that employed 15,000 CWS workers. An aerial mapping project charted hundreds of U.S. cities and employed another 10,000 CWS workers. The National Park Service and the Library of Congress undertook a survey of the nation’s historic buildings that provided work for 1,200 draftsmen. Over 70,000 people were employed in CWS pest-eradication campaigns, and a group of ninety-four Alaskan Indians was employed restocking the Kodiak Islands with snowshoe rabbits.52 Descriptions of CWS cultural projects are described in Chapter 8.
The single largest category of CWS employment, though, consisted of work performed on statistical surveys. The Department of Commerce employed 11,000 CWS workers to conduct a census of real property in sixty cities. An Urban Tax-Delinquency Survey documented the fiscal condition of 309 cities. The CWA’s own Statistical Division employed 35,000 CWS workers to collect and record data and documentation concerning program operations, labor market conditions, and the nation’s public relief problem.53
The establishment of a program as large, as complicated, and as innovative as the CWA within a span of weeks was a major administrative achievement. A War Department engineer assigned to study the program compared it favorably to the country’s mobilization effort in World War I.54 A New Deal historian has called the CWA “one of the greatest peacetime administrative feats ever completed” in the United States.55 It was a remarkable experiment, more ambitious in its goals than any other New Deal employment program. It contemplated nothing less than the replacement of means-tested work relief with a promise of public employment paying good wages and doing work of genuine social utility.
The quick demise of the program shows how controversial that idea was. Criticism of the program by conservatives was fierce, and support for it within the Roosevelt Administration was by no means unanimous. The president’s own support for the program was ambiguous. He was not yet ready to commit the federal government to the operation of a massive, year-round public employment program for unemployed workers. Still, Hopkins and his associates continued to work toward their broad reform goals, hoping that by stepwise movement they could win the political support needed to establish a more sustainable, if less ambitious, version of the CWA. We shall see that this is exactly what they achieved with the establishment of the WPA in 1935.
Following the termination of the CWA, the FERA once again assumed responsibility for funding the nation’s relief effort. Hopkins and his associates continued to direct that effort, and even though the CWA was politically unsustainable, they made a concerted effort to preserve as much as they could of the CWA model in FERA work relief programs. But, once again, the FERA had to rely on local relief officials to administer the work relief projects it funded, with employment on those projects limited to persons on relief. The common work week established under the CWA was also lost, as hours of work were once again determined by applying the “budget deficiency” principle. Work relief participants worked as many hours as were necessary to “earn” their family’s public assistance grant. At first, the thirty-cents-an-hour minimum wage rate that had previously applied to FERA-funded work relief programs was reinstated. However, under political pressure from low-wage employers, particularly in the South, the thirty-cents- per-hour minimum was abandoned in November 1934. Henceforth, wage rates depended exclusively on locally prevailing wages, with no minimum wage floor.56
Under Hopkins’s leadership, the FERA established a number of special programs that experimented with new forms of work relief. One was a part-time employment program for needy college students who might otherwise have been forced to drop out of college—the prototype for today’s College Work Study program. Another was a Transient Program that provided both residential camps and work for a population that today would be referred to as “homeless.”57 Perhaps the most interesting of the FERA’s innovations during this period, however, were “production for use” projects undertaken in cooperation with the Federal Surplus Relief Corporation (FSRC). The FSRC purchased surplus agricultural commodities in an effort to maintain farm prices and hence the livelihoods of farmers. These commodities were processed by FERA workers (including the canning of fresh fruits and vegetables and the production of clothing and bedding from surplus cotton) and then distributed to relief recipients outside market channels. This distribution was accomplished on an “over-and-above” basis, to prevent local relief agencies from counting the commodities received from the FSRC in determining the budget deficiencies of relief recipients.58
Business interests strongly opposed the FERA’s “production for use” initiatives. One frequently voiced complaint was that the projects competed with private producers and hence reduced unsubsidized employment—despite the fact that the commodities at issue were distributed at no cost to people who otherwise would not have been able to afford them. The real source of business opposition was more accurately reflected in complaints that the projects challenged the private enterprise system and constituted a dangerous step down the road to government control of all industry. This opposition was strong enough that Hopkins was forced to abandon the production-for-use model by the time the WPA replaced the FERA in the spring and summer of 1935.59
In June of 1934, shortly after the termination of the CWA, President Roosevelt appointed a Cabinet-level Committee on Economic Security (CES) to develop a comprehensive social welfare strategy that would be responsive to the American people’s desire for “some safeguard against misfortunes which cannot be wholly eliminated in this manmade world of ours.”60 Chaired by Secretary of Labor Frances Perkins, the other members of the committee were Secretary of the Treasury Henry Morgenthau, Attorney General Homer Cummings, Secretary of Agriculture (later Vice-President) Henry Wallace, and Harry Hopkins.
The committee’s report, officially delivered to the president in January 1935, proposed a two-legged social welfare strategy—one leg to address the economic security needs of those members of society who could support themselves if adequately paid work were available for them, and the other to address the economic security needs of people who were either unable or not expected to be self-supporting.61
The first leg of this strategy adopted the plan Hopkins and Williams had developed and tested with the CWA. Specifically, the report proposed that the federal government provide unemployed workers with “employment assurance” in both good times and bad.
Since most people must live by work, the first objective in a program of economic security must be maximum employment. As the major contribution of the Federal Government in providing a safeguard against unemployment we suggest employment assurance—the stimulation of private employment and the provision of public employment for those able-bodied workers whom industry cannot employ at a given time. Public-work programs are most necessary in periods of severe depression, but may be needed in normal times, as well, to help meet the problems of stranded communities and overmanned or declining industries.62
The second leg of the CES strategy relied on the establishment of a series of income-transfer programs designed to insure that everyone who was either unable or not expected to be self-supporting would be assured a reasonable subsistence. The best-known of the programs proposed by the CES to perform this function was the contributory old-age pension system that subsequently came to be known as Social Security; but the CES report included proposals addressing a range of other income security needs as well.63
Given Hopkins’s role in developing the CES’s employment assurance proposal, it is hardly surprising that it reflects the reform vision he and Aubrey Williams had previously developed. At the same time, that vision is enhanced in the CES report by its linkage to reforms proposed to meet the economic security needs of persons who cannot rely on work for their support. If fully implemented, the two legs of the social welfare system proposed by the CES would have effectively secured what subsequently came to be called the “right to work” and the “right to income security” in international human rights law.64
Given the expansiveness of the CES’s employment assurance proposal, it is striking how far short of that goal the Roosevelt Administration aimed its implementation efforts—especially when compared to the virtually complete incorporation of the CES’s income security proposals in the Social Security Act.65 Established by executive order and a budgetary enactment a few months before the Social Security Act passed,66 the WPA did not even attempt to provide work for all unemployed workers. Instead, it aimed only to provide work for unemployed workers who qualified for public relief. Moreover, in contrast to the reasonably good wages paid by the CWA (an average of $15.00 per week at first for 30 hours of work), the WPA paid a below-market “security wage” designed to provide an incentive for program workers to accept regular employment when it became available.67
It is true that Hopkins did his best, with partial success, both to minimize the onerousness of the program’s means test and to raise program wages over time. Still, the program’s eligibility requirements and low wages constituted a step backward (compared to the CWA) in Hopkins’s efforts to sever the job creation he promoted from its continuing associations with public relief. The program also failed to eliminate the gender and race discrimination that tarnished the CWA.68
On the other hand, the program’s administrative structure and the very substantial contributions it made to the country’s physical and social wealth demonstrated Hopkins’s sharp break with the past. As was true of the CWA, the vast majority of WPA projects were sponsored by state and local governments. These sponsors were responsible for providing all drawings for construction projects, and paid an average of about 20 percent of total program costs—usually in the form of purchases of supplies and materials. It was the WPA, though, that hired and supervised the program work force.69 The WPA also adopted the CWA model of having federal agencies sponsor some of the program’s more innovative projects—like the program’s famous Art, Theatre, Music, and Writers’ Projects.70
The WPA staff worked proactively with local governments and non-governmental organizations to develop new ideas for projects, especially in the non-construction fields. Since the WPA possessed a wealth of information concerning widely diverse projects operated under a variety of circumstances, WPA staff served as a viaduct for the dissemination of new ideas and best practices throughout the country. Program evaluators from the National Resources Planning Board who studied the WPA’s operational model described it as achieving the advantages of both decentralized and centralized government control. The projects selected responded to local needs, while national standards of performance in carrying out the projects were maintained.71
This administrative structure permitted the WPA to operate its work projects on an “enterprise” rather than a “relief” model—notwithstanding the program’s low wage scale and need-based eligibility standards. Once they qualified for the program, applicants were hired to fill specific job classifications on the basis of their experience and qualifications relative to other qualified applicants. When hired, they became federal employees who received a monthly paycheck in exchange for work performed—not a local welfare grant based on their “budget deficiency.” If the income they received from their WPA earnings left them poor enough to qualify for relief (a fairly common occurrence for WPA workers with large families), they could still apply for public assistance from their local relief agency—on the same basis as privately employed workers and without involving the WPA in any way. Though low, WPA wage rates did vary with skill level and by region; a common work week was observed; and normal workplace discipline was maintained. WPA workers were covered by workers’ compensation and were permitted to organize unions.
What accounted for the WPA’s hybrid character—with one leg still stuck in the public relief tradition, while the program’s organizational structure reflected the CES’s reform goals? Moreover, why did Roosevelt allow the CES to advocate an employment assurance plan he was not prepared to fully implement (and which he might not have been able to get through Congress if he had tried)? It’s a puzzling question. The CES also developed a national health insurance proposal, but when it became clear that the American Medical Association had both the intention and power to scuttle it, the president told the CES to drop the proposal from its published report.72 Why didn’t he tell the CES to scale back its job-creation proposal to match what he was prepared to implement?
One possibility is that Roosevelt was hesitant to commit to the employment assurance goal because he did not know how much it would cost to achieve it. No one knew in 1935 how many unemployed workers there were in the country because the government had not yet begun to collect unemployment data. The size of the fiscal commitment required to provide “employment assurance” to the nation’s work force was therefore unknown, and given how many people had applied for work on the CWA—9 million in a matter of weeks—the president may have wondered whether the federal government really could take on the task. Any concerns he had on that score would have been encouraged by fiscal conservatives within his administration who applied constant pressure on him to restrict federal spending.
It is also possible that he remained unpersuaded that the federal government should assume responsibility for the task even if it could afford to do so. He knew the business community would be unalterably opposed to the idea—based on its concerted opposition to the CWA. He also knew, though, that the CWA model lacked strong institutional support among liberals. The social work profession, which one would have thought Hopkins and Williams represented, had considerable misgivings about the CWA experience because it dispensed with the individual casework that social workers viewed as the sine qua non of their professional role in assisting the poor.73 The engineers and managers who had administered the CWA’s work projects were strongly invested in completing the projects, but they did not view their mission as providing work for the unemployed.74 Union members were happy for the jobs the program provided, but the role of unions in the CWA model was both unclear and unsettling for union leaders accustomed to representing private sector employees.75 Civil rights organizations supported job-creation efforts but were disappointed by the Roosevelt Administration’s failure to rid them of racist practices.76 The list could go on. There simply was no organized liberal constituency that placed job creation for unemployed workers at the top of its agenda, and the president surely knew that.
In any case, Roosevelt made it clear in his 1935 State of the Union message, delivered as he was preparing to forward the CES report to Congress with his recommendations, that the “program for putting people to work” he would advocate was limited to creating security-wage jobs for those unemployed workers who qualified for public relief.77 He made no mention whatsoever of the CES’s employment assurance goal—which was thereby abandoned before it was even published.
But why, then, let the report be published with that recommendation intact—a recommendation that served as the foundation for the CES’s entire economic security program? It may simply be that the report had already gone to press by the time Roosevelt made his decision to ignore the committee’s employment assurance proposal. Another possibility may be the importance Roosevelt attached to the principle of societal obligation on which the employment assurance proposal was based—even if he was not prepared to press for its adoption. Roosevelt believed very strongly that society had an obligation to its members to insure their ability to support themselves. In a widely reported campaign address delivered to the Commonwealth Club of California in the fall of 1932, then–presidential candidate Roosevelt explained his views on this matter in the following terms.
Every man has a right to life; and this means that he has also a right to make a comfortable living. He may by sloth or crime decline to exercise that right; but it may not be denied him. We have no actual famine or death; our industrial and agricultural mechanism can produce enough and to spare. Our government formal and informal, political and economic, owes to every one an avenue to possess himself of a portion of that plenty sufficient for his needs, through his own work.78
Roosevelt made it clear in the same speech that he assigned primary responsibility for securing this right to those he referred to as the “princes of property,” who “claim and hold control of the great industrial and financial combinations which dominate so large a part of our industrial life.” Only if they failed to fulfill their responsibility, he continued, would it fall upon government to “assume the function of economic regulation … as a last resort.” He also made it clear that he had not yet given up hope, in the fall of 1932, that the “princes of property” would fulfill their duty in this regard. “As yet there has been no final failure, because there has been no attempt, and I decline to assume that this nation is unable to meet the situation.”79
Four years later, in 1936, Roosevelt made it clear in his acceptance speech at the Democratic National Convention that the “princes of property”—whom he now referred to as “the royalists of the economic order”—had failed the test he set for them four years earlier; and in doing so he once again referred to the right to work as the touchstone by which the legitimacy of their power should be judged.
The royalists of the economic order have conceded that political freedom was the business of the government, but they have maintained that economic slavery was nobody’s business. They granted that the government could protect the citizen in his right to vote, but they denied that the government could do anything to protect the citizen in his right to work and his right to live.80
In short, the assertion of the CES that government should assume the duty of providing jobs for workers whom the private sector could not employ was an article of faith for Roosevelt, even if, as “Politician in Chief,” he knew better than to ask Congress to authorize him to do it, and as “Fiscal Officer in Chief” he thought it was not something he could justify funding with additional deficit spending. Still, we might speculate that the broader commitment the CES advocated was close enough to his heart that he was willing to let it remain in the CES report.
It is also important to note that, even with the limited mandate given the WPA, it made a far greater dent in the nation’s unemployment problem than is generally recognized. This accomplishment is obscured by the unemployment statistics commonly reported for the New Deal period, which count workers employed in these programs as unemployed rather than employed. If workers employed in direct job-creation programs are counted as employed (as they are in unemployment statistics today), we see that the nation’s unemployment rate dropped from 22.9 percent to 10.0 percent during President Roosevelt’s first term in office, rather than the commonly reported drop from 23.0 percent to 17.0 percent. The difference between the two sets of figures is attributable entirely to the effect of hiring by the FERA, CWA, and WPA. The WPA alone reduced unemployment by 4.4 percentage points during its first year of operations—8.9 percentage points, if we count as unemployed the workers who received public relief on a “budget deficiency” basis in locally administered work relief programs replaced by the WPA.81
As the CWA’s accomplishments during its brief, four-month existence illustrate, the New Deal’s direct job -creation programs also enriched the country with a profusion of new public goods and services at a time when normal budgetary considerations would have ruled out virtually all such undertakings. The forced idleness of the nation’s work force was turned into a wealth-producing asset in service of the public good. As Leighninger points out, the incalculable legacy of this public investment in infrastructure, health, education, recreation, environmental conservation, and culture has been little recognized, even by historians.82 Indeed, had the government not undertaken such a vast public-investment program, the United States almost surely would not have been as prepared as it was to enter World War II. Nor would we have emerged from the war with as much economic momentum as we did. Conventional public works spending by the PWA accounted for only about 30 percent of the New Deal’s total investment in the country’s physical and social infrastructure during the 1930s. It was the era’s direct job-creation programs that accounted for most of it—approximately 70 percent of the total.
Moreover, by tapping this idle productive capacity, the federal government gave back to millions of unemployed workers and their families precisely what the Depression had taken from them—the opportunity to support themselves in dignity. The jobs provided by the New Deal made it possible for them to put their broken lives back together again while they waited for the private economy to recover. The value ordinary people attached to this is indicated by the 9 million applicants for the CWA’s 2 million non–means-tested jobs.
Nevertheless, among progressive economists, the prevailing view of the New Deal’s direct job-creation strategy is that it was nothing more than a delivery device for a Keynesian fiscal stimulus. Noting that it was World War II spending that finally brought the Depression to an end, these economists see no reason to emulate the Roosevelt Administration’s direct job-creation strategy because there are far easier ways to boost aggregate demand. Of course they should advocate for the allocation of stimulus spending to socially useful purposes, but except for differences in the multiplier effect of different categories of stimulus spending, they can cite no economic reason to prefer one type of deficit spending over another. As Keynes himself famously noted, if there are political difficulties that prevent a sensible allocation of stimulus spending, engaging in otherwise wasteful spending would be preferable to doing nothing at all. A government could even bury bank notes in bottles at the bottom of abandoned mines, fill the mines with trash, and then invite capitalists to dig them up.83 From this perspective, there is no reason for progressives to pursue the administratively challenging and politically fraught direct job-creation strategy when responding to a recession. It’s not that progressive economists argue against using the strategy. They simply ignore it when fashioning their policy proposals.
Ironically, this tacit dismissal of the direct job-creation strategy overlooks the role played by military enlistments in achieving full employment during World War II. Between 1943 and 1945—a three-year period during which the nation’s civilian unemployment rate fell below 2 percent—an average of 10.9 million able-bodied workers were employed in the U.S. military.84 That was three-and-a-half times the average number employed in the New Deal’s direct job-creation programs between 1936 and 1940. In short, even though it was not the reason for the military buildup, the achievement of full employment during World War II actually did rely on direct job creation—with the military replacing the New Deal’s direct job-creation programs as the vehicle for providing the jobs.
Still, the Keynesian critique of the New Deal response to the Great Depression is clearly correct in its condemnation of the Roosevelt Administration’s fiscal conservatism. Figure 6.1 shows the U.S. unemployment rate from 1933 through 1947. The top line shows the rate as it is normally reported—that is, with persons employed in the New Deal’s direct job-creation programs counted as unemployed. The lower line shows the unemployment rate as it would be reported today, with persons employed in these programs counted as employed.
The first thing Figure 6. 1 illustrates is the disastrous effect of President Roosevelt’s ill-conceived attempt to balance the federal budget in 1937. It took the economy three years to recover from the economic reversal precipitated by that action, confirming Keynes’s argument concerning the beneficial effect of deficit spending.
The second thing the figure shows, however, is the dramatic impact the New Deal’s direct job-creation programs had on the level of unemployment over and above the beneficial multiplier effect of program spending on the private sector. The employment effect of a direct job-creation program has two components. The first is the program’s direct employment effect—the jobs that are created in the program itself. The second is its indirect (i.e., multiplier-induced) employment effect on private-sector hiring. It is this dual employment effect that makes the direct job-creation strategy such a potent means of responding to economic downturns.85
Figure 6.1 U.S. Unemployment Rate from 1933–1947.
If, instead of funding programs like the CWA and WPA, the Roosevelt Administration had used the same amount of stimulus money to pay for other types of government benefits (such as unemployment insurance), the private sector probably would have recovered at the same pace shown by the top line in Figure 6. 1; but it would have sacrificed the additional, direct job-creation effect shown by the difference between the top and the bottom lines in the figure.
The third thing Figure 6. 1 shows is that Keynesian critics of the New Deal are correct in noting that the Roosevelt Administration could have achieved a full recovery from the Great Depression far more quickly if it had engaged in more deficit spending during the 1930s. What the figure also shows, however, is that the best way to have spent that money would have been to implement the employment assurance proposal contained in the CES report.
If, in 1936, the WPA had been expanded to provide 4.7 million jobs instead of the 3.3 million actually provided, the economy’s unemployment rate could have been reduced to the full employment level of 2 percent in 1937, rather than waiting for wartime spending (and the draft) to do it in 1943. If they had also spent enough to increase the WPA work week to a standard 40 hours for full-time workers, and increased program wages to market levels, those jobs would have been fully comparable to their private-sector counterparts. The additional spending required to achieve that goal probably would have doubled the New Deal’s direct job-creation budget—from $2.6 billion (2.2 percent of GDP) to $5.2 billion (4.4 percent of GDP, or the equivalent of about $690 billion in 2012). The fiscal stimulus provided by that additional spending would have caused private sector unemployment to decline more rapidly as well; and if the mistaken 1937 attempt to balance the federal budget had been avoided, the private sector probably would have fully recovered by 1939 instead of 1943—and possibly with lower overall levels of spending on direct job-creation than were actually incurred during that period.
Why would this strategy have been superior to the standard Keynesian strategy—i.e., spending stimulus dollars on budget items other than the WPA? Both strategies would have achieved full recovery from the Great Depression at about the same rate. The difference is that the standard Keynesian strategy would have forced unemployed workers and their families to wait for that recovery in order to get their lives back on track, while the direct job-creation strategy would have given them the benefits of full employment immediately rather than years later.
As noted earlier, President Roosevelt was naturally inclined to view the “right to live” and the “right to work” as entitlements that society has a duty to secure for its members. In light of that predisposition, it required no great conceptual leap on his part to describe the New Deal’s social welfare goals (which were most fully expressed in the CES’s 1935 report) in similar terms. The rhetorical turning point came in Roosevelt’s 1941 State of the Union Address to Congress—his famous “four freedoms” speech. In it, he identified “equality of opportunity” and “jobs for those who can work” as the first two things the American people expect of their government; and he famously included “freedom from want” as one of the “four freedoms” World War II was being fought to secure, not just in the United States and Europe, but “everywhere in the world.” This was “no vision of a distant millennium,” he declared, but a “definite basis for a kind of world attainable in our own time and generation.”86
At the time the president delivered his Four Freedoms speech, the National Resources Planning Board (NRPB)—a research and advisory body housed in the Executive Office of the President—was engaged in a detailed assessment of the social welfare initiatives undertaken by the Roosevelt Administration since 1933 as a guide for planning the social welfare institutions it believed the country would need at the end of World War II. In other words, it was tasked with carrying on the work of the CES.
Recognizing the link between “freedom from want” and the goals of its own planning initiative, the NRPB undertook the task of developing a nine-point “Declaration of Rights” that would “translate” the “freedom from want” into a list of specific economic and social entitlements. Drafted in close consultation with President Roosevelt, this declaration was first published in a pamphlet bearing the title “Our Freedoms and Our Rights.” The first two rights listed in the declaration were the “right to work” and the “right to fair pay.”87
President Roosevelt’s framing of the war effort as a human-rights struggle also inspired the American Law Institute (ALI) to undertake the drafting of a “Statement of Essential Human Rights.” A highly influential and quintessentially mainstream organization of judges, lawyers, and academics dedicated to the improvement of the law, the ALI convened an international drafting committee to enumerate the human rights on which a lasting peace could be based following the end of World War II. The effort took three years of study and discussion, and the final “Statement,” like the NRPB Declaration, recognized a range of economic and social entitlements as human rights—including the right to work and the right to reasonable wages, hours, and other conditions of work.88
In 1944, with the end of the war in sight and a presidential election scheduled for the following fall, Roosevelt chose to use his State of the Union Message to restate his belief that the employment and social welfare entitlements his administration had sought to secure during the preceding eleven years were in fact human rights that the federal government had a duty to secure. Invoking both the country’s Declaration of Independence and its Constitution, Roosevelt opined that the United States “had its beginning, and grew to its present strength, under the protection of certain inalienable political rights” secured by the country’s Constitution. However, as the nation grew and its economy industrialized, these rights (which he described as “our rights to life and liberty”) proved inadequate to “assure us equality in the pursuit of happiness.” Claiming that “these economic truths have become accepted as self-evident,” he asserted that “we have accepted, so to speak, a second Bill of Rights under which a new basis of security and prosperity can be established for all—regardless of station, race, or creed.” He then listed the economic and social entitlements included in this second bill of rights—a refinement of the NRPB’s earlier nine-item list, and, like it, beginning with the “right to a useful and remunerative job” and the “right to earn enough to provide adequate food and clothing and recreation.”89
Framing these entitlements as “rights” rather than “economic policy objectives” may not guarantee that the government will actually fulfill its obligations, but there was nothing wide-eyed about the New Dealers’ willingness to make the commitment. They believed that policies were available that actually could secure the rights in question—beginning with the right to decent work. As President Roosevelt stated in his Four Freedoms speech, it was not a “vision of a distant millennium.” It was an agenda for reaching an immediate and achievable goal.
Beginning with the key insight that jobs could be provided for unemployed workers in the same way that other social-welfare benefits could, the New Dealers had fashioned a strategy for providing employment assurance that was eminently doable. When wartime spending appeared to confirm Keynes’s premise that a market economy could spend its way to full employment without going to the trouble of operating programs like the WPA, it merely reinforced the confidence of progressives in their ability to secure the right to work while giving them a less contentious way of describing their objective—the achievement of full employment.
The tragedy is that they forgot they had two strategies for achieving full employment and also forgot the association of that goal with securing the right to work. Fully embracing the Keynesian full-employment strategy, they abandoned both the New Deal direct job-creation strategy and the rights-based claims the New Dealers asserted to describe their employment and social welfare policy goals. Generations of progressive economists came to believe that their job in the battle against unemployment was to induce the economy to grow fast enough to achieve full employment—without ever asking how successful they were in securing the right to work.
The CES had a more balanced view. The Committee’s report recognized that the “stimulation of private employment” was an important part of the strategy for providing workers with employment assurance, but they saw no reason to rely exclusively on the success of that effort. The Committee’s employment assurance strategy was shaped by social workers who knew from long experience that the scourge of unemployment had never been limited to economic downturns, nor to slackers angling for a handout. Harry Hopkins expressed their view in the following terms:
[People] suggest that we make relief as degrading and shameful as possible so that people will want to get “off.” Well—I’ve been dealing with unemployed people for years in one way and another and they do want to get off—but they can’t, apparently, get “off” into private industry. Well—if they can’t get off into private industry, where can they turn if they can’t turn to their government? What’s a government for? And these people can be useful to America; they can do jobs no one else can afford to do—these slums, for instance. No private concern can afford to make houses for poor people to live in, because any private concern has got to show a profit. Why, we’ve got enough work to do right here in America, work that needs to be done and that no private concern can afford to touch, to lay out a program for twenty years and to employ every unemployed person in this country to carry it out.90
Therefore, while acknowledging the government’s obligation to do what it could to stimulate private-sector employment, the CES report recognized the need for the federal government to back up that effort with a commitment to provide “public employment for those able-bodied workers whom industry cannot employ at a given time,” and to do so, not only in “periods of severe depression,” but in “normal times” as well.91
Given the nature and scope of this commitment—and particularly its focus on filling whatever employment gap the private sector failed to close—it may be fair to say that the principal goal of the New Deal’s direct job-creation strategy was not to end the Great Depression but to help people survive it. It was a social-welfare strategy whose synergy with Keynesian economic policy was not perceived at the time—for the simple reason that the strategy was conceived and implemented before Keynes’s General Theory had even been published.
The progressive economists who embraced Keynes’s work a few years later lack this excuse for failing to recognize the advantages of using a direct job-creation strategy to deliver a Keynesian fiscal stimulus to a depressed economy. Moreover, their blindness to the social-welfare contribution of the New Deal strategy also prevented them from truly comprehending the value of the New Deal strategy apart from its fiscal impact.
The true irony, though, is that the full implementation of the CES’s employment assurance strategy would have automatically incorporated the benefits of Keynesian countercyclical policy without anyone knowing it was doing so. The reason for this is simple. Spending on a direct job-creation program designed to secure the right to work would be naturally countercyclical in its effect—just like spending on unemployment insurance, another social welfare program conceived without the benefit of Keynesian economic theory. In other words, even if the countercyclical virtues of Keynesian fiscal policy had never been discovered, the New Deal’s direct job-creation strategy could have functioned perfectly well to both secure the right to work and dampen the business cycle.92 Unfortunately, the same is not true of Keynesian measures implemented in ignorance of the virtues of direct job creation.
In February of 2009, President Obama proposed, and Congress enacted, the American Recovery and Reinvestment Act (ARRA), a two-year $787 billion Keynesian fiscal stimulus. It was estimated at the time that the ARRA would “save or create” 3 to 4 million jobs, and subsequent analyses have confirmed that the programs did indeed perform as predicted, even though its job-creation effect was disguised by the fact that the recession’s negative effect on employment levels was greater than the ARRA’s positive effect.93
I have argued elsewhere that if the New Deal’s direct job-creation strategy had been pursued with the same resources, it could have reduced the national unemployment rate either to its pre-recession level of 4.5 percent or to the genuine full-employment level of 2 percent in less than a year, while providing a substantially larger fiscal stimulus than the ARRA furnished to the private sector.94
The job-creation program I modeled would have provided not only officially unemployed workers, but also involuntary part-time workers and discouraged workers with their preference of either part-time or full-time employment producing relatively labor-intensive public goods and services.95 I assumed the jobs would have paid the same wages as similar jobs in the private or regular public sectors of the economy, and program participants would have been provided the same health insurance benefits as federal government employees. To the extent possible, persons employed in the program would have been offered positions comparable in skill level and responsibility to those they previously occupied. Employment in the program would also have been treated as regular employment for tax purposes, for establishing eligibility for government benefits, and for asserting legal rights—including the right to unionize.
Based on these assumptions, I estimated the program’s average job cost at $46,800 annually.96 However, because such a program would generate additional government revenues (e.g., income and payroll taxes) and savings (e.g., reduced unemployment insurance and Medicaid expenditures), the additional funding required to cover its true “net cost” would have averaged only $24,189 per job. Based on these figures, the budgeted cost of creating the 12.2 million jobs that I estimated would have been needed to reduce the nation’s unemployment rate to its pre-recession low of 4.5 percent would have been approximately $571 billion the first year, but the program’s net cost would have been only $295 billion. Reducing the nation’s unemployment rate another 2.5 points to the genuine full employment level of approximately 2.0 percent would have increased the program’s first-year net cost by another $130 billion, bringing the program’s’ total net cost to $425 billion.
How much the program would have cost the second year would have depended on the size of the program’s multiplier effect and how quickly private sector investment recovered from its recessionary levels. My model projects that the multiplier effect of program spending would have induced private sector employers to create 4.9 million jobs over and above those created in the program, with most of that job creation occurring during the program’s second year. That means the program would have needed approximately that many fewer jobs during its second year of operations, for a net program savings of about $170 billion, compared to the program’s first-year cost. The program’s total two-year net cost accordingly would have been about $679 billion—about a $100 billion less than Congress voted to spend on the far-less-effective ARRA.
Why is the New Deal strategy so much more effective at creating jobs than the standard Keynesian strategy exemplified by the ARRA? There are several reasons, but the two most important ones are what I call the direct job-creation strategy’s “twofer” effect (as in “two for the price of one”) and the government’s ability to create jobs at a lower average cost in such a program than the cost of stimulating the private sector to create the jobs.
The “twofer” effect of a direct job-creation program refers to its ability to create jobs both directly and indirectly.97 When you use stimulus dollars to pay for SNAP benefits (food stamps), the government in effect pays for two benefits. The first is the benefit that food stamps provide to their recipients. The second is the private-sector job-creation induced by the government’s expenditures on the program (assuming additional deficit spending is used to fund the benefits). When you use stimulus dollars to provide jobs in a direct job-creation program, both of the effects you purchase consist of jobs. First, you get the jobs provided by the program itself, but you also get the private-sector job-creation induced by the government’s spending on the program.98 This is the program’s “twofer” effect. Add to it the value to the community of the additional public goods and services produced by the direct job-creation program, and you have a third benefit for taxpayers.
The same effect occurs, of course, when the government uses stimulus dollars to purchase goods or services from private businesses and the businesses hire unemployed workers to produce them; but that brings us to the second major reason why the direct job-creation strategy is more effective in creating jobs than the Keynesian strategy. It turns out, for a variety of reasons, that a job can be created in a direct job-creation program for about half of what it costs the government, on average, to induce the private sector to create one. The reasons for this advantage include the following:
1. The tendency for private businesses to increase the hours of work of already employed workers rather than add new workers to their payrolls—especially during a recession;
2. The tendency for direct job-creation programs to employ more labor-intensive methods of production than the private sector uses, and to deliberately choose labor-intensive over capital-intensive projects;
3. The fact that the money used to create jobs in a direct job-creation program is not diminished by the earnings paid to owners of private businesses or the depreciation allowance the owners of those businesses retain as compensation for the use of pre-existing capital goods that are not replaced (or whose replacement is delayed until after the recession is over);
4. The fact that unemployment tends to be concentrated among lower-wage workers rather than higher-wage workers, so that the average wages paid by a program that provides jobs only to unemployed workers tends to be lower than the average wages paid to a cross-section of the national labor force—the typical profile of employment generated by a stimulus-induced increase in the economy’s overall rate of economic growth.
Taken together, the “twofer” effect and the lower cost per job of the direct job-creation strategy mean that a billion dollars spent on direct job-creation is likely to produce two to four times as many jobs as the same expenditure on unemployment insurance and SNAP benefits (which are representative of the types of stimulus spending that generate the biggest multiplier effect on private-sector employment). Compared to types of stimulus spending with a small multiplier effect (e.g., the retention of the so-called Bush-era tax cuts) the direct job-creation strategy is ten to twenty times as cost effective.99
As if this were not enough, the New Deal direct job-creation strategy has four other economic advantages over the standard Keynesian strategy for combating unemployment. First, its job-creation effect is achieved much faster. It is “front-loaded” in the sense that most of the jobs attributable to the strategy are provided immediately—or as soon as the program can be up and running—whereas the peak employment effect of a standard Keynesian stimulus takes about eighteen months to achieve.
Second, the New Deal strategy has a natural tendency to target its job-creation effect on the individuals, population groups, and communities that most need jobs. The burdens of joblessness are unequally distributed, with disadvantaged population groups bearing substantially more than their fair share of the pain. Unfortunately, the job-creation effect of conventional Keynesian stimulus strategies does little or nothing to correct this imbalance. The same economic forces that cause private-sector job losses to be concentrated among disadvantaged workers tend to direct private-sector job gains away from them. A direct job-creation program that provided work for all job seekers would disproportionately benefit the members of disadvantaged population groups, and it would do so automatically. A direct job-creation program that provided work for only some of the unemployed could achieve the same goal with eligibility requirements that take the length of time a person has been unemployed and his or her need for work into consideration in allocating employment opportunities.
The third advantage of the New Deal strategy is that it delivers its private-sector fiscal stimulus in a way that is likely to maximize its anti-cyclical effect. The revenue losses that businesses suffer during a recession flow primarily from rising unemployment rather than whatever economic problems triggered the recession in the first place. Businesses that are otherwise healthy and well-managed lose their customers because their customers lose their jobs. The fiscal stimulus provided by a direct job-creation program would reverse this process. The resumption of ordinary consumer spending by re-employed workers would fill the very same gap in the balance sheets of local businesses that rising unemployment created in the first place. Why does this matter? Under conventional stimulus approaches, the government increases spending in places and in ways (e.g., new infrastructure projects) that bear little immediate connection to the losses most businesses have suffered. Eventually, the multiplier effect of the stimulus spending spreads through the economy, but it takes time, and this delay can mean the difference between life and death for stressed businesses. Because it would deliver its fiscal benefits to the very same segments of the economy that suffered income losses as unemployment grew, the direct job-creation strategy would help insulate otherwise healthy firms from the harmful effects of the recession, and in doing so, it would provide a more stable foundation for the resumption of self-sustaining economic growth.
Finally, the New Deal strategy is much better designed than the Keynesian strategy to stop recessions from feeding on themselves. The rapid deployment of a large, direct job-creation program at the beginning of a recession could reduce the depth of the recession more than the deployment of a large Keynesian stimulus package like the ARRA. This is because, as just noted, most of the job losses and attendant economic harm that occur during a recession are the consequence of earlier job losses rather than being linked to the tendencies or events that triggered the economic contraction in the first place. It is this downward spiral of job losses leading to further job losses that turns a business correction into a recession. If the initial job losses associated with a recession could be stopped from triggering further job losses, periodic slowdowns in economic activity would still occur. There might even be recessions, but they would not be as deep. By offering immediate re-employment to laid-off workers, a direct job-creation program would prevent their job losses from triggering further job losses. That alone might be enough to stop a recession in its tracks, but even if it did not, it would lessen its severity.
Sadly, the employment-assurance leg of the CES’s overall strategy for securing everyone’s right to an adequate income has languished, with progressives failing even to understand what their predecessors abandoned as a result of their pursuit of the seemingly easier path to full employment promised (but only rarely delivered) by the Keynesian strategy. The advantages of the direct job-creation strategy over the Keynesian strategy are so clear and so great that the attachment progressives continue to show to the latter seems almost perverse. It is long past time for progressives to shed their Keynesian bias and start learning from the New Deal.
1. The definition of full employment (or “genuine” full employment) used in this chapter is grounded on William Beveridge’s widely cited 1944 description—provided we ignore his gendered language. “It means having always more vacant jobs than unemployed men, not slightly fewer jobs. It means that the jobs are at fair wages, of such a kind, and so located that the unemployed men can reasonably be expected to take them; it means, by consequence, that the normal lag between losing one job and finding another will be very short.” William Beveridge, Full Employment in a Free Society, London: G. Allen & Unwin Ltd. (1944), 18. While my usage of the term is consistent with Beveridge’s, I define it more precisely as equivalent to securing the “right to work” recognized in the Universal Declaration of Human Rights. See Philip Harvey, “Why Is the Right to Work So Hard to Secure,” in The State of Economic and Social Human Rights: A Global Overview, Alanson Minkler, ed. (New York: Cambridge University Press, 2013), 135–172. As a practical matter, I believe this requires the achievement of an unemployment rate in the neighborhood of 2 percent. See Philip Harvey, Securing the Right to Employment: Social Welfare Policy and the Unemployed in the United States (Princeton, NJ: Princeton University Press, 1989), 13.
2. For a more detailed account of these views and a discussion of their historical origins, see Philip Harvey, “Joblessness and the Law before the New Deal,” Georgetown Journal on Poverty Law and Policy, 6 (1999): 1–41.
3. National Resources Planning Board, Security, Work and Relief Policies (Washington, D.C.: U.S. Government Printing Office, 1942), 234, note 42.
4. National Industrial Recovery Act, sec. 201 et seq., Statutes at Large 48 (1933), 195 et seq.
5. U.S. Bureau of Economic Analysis, National Income and Product Account Tables, Table 3.2 (Federal Government Current Receipts and Expenditures), accessed May 6, 2013, available at http://www.bea.gov/iTable/iTable.cfm?ReqID=9&step=1#reqid=9&step=3&isuri=1&910=X&911=0&903=87&904=1933&905=1933&906=A.
6. U.S. Bureau of Economic Analysis, National Income and Product Account Tables, Table 1.1.5 (Gross Domestic Product), accessed May 6, 2013, available at http://www.bea.gov/iTable/iTable.cfm?ReqID=9&step=1#reqid=9&step=3&isuri=1&910=X&911=0&903=5&904=1933&905=1933&906=A.
7. Because of the limited fiscal capacity of state and local governments at the time, many were not in a position to finance their required 55–70 percent share of the cost of PWA-funded projects. To address this problem, the PWA not only established a bonding unit that purchased state and local government securities to finance their share of the cost of PWA projects, it helped state governments rewrite their laws to allow for the necessary borrowing and actively assisted both state and local governments in creating agencies to undertake PWA-funded financing and construction activities.
8. Arthur E. Burns and Edward A. Williams, Federal Work, Security, and Relief Programs, Works Progress Administration, Division of Social Research, Research Monograph XXIV (Washington, D.C.: U.S. Government Printing Office, 1941), 70–71.
9. Public Works Administration, America Builds: The Record of PWA (Washington, D.C.: U.S. Government Printing Office, 1939), 19.
10. Ibid., 86.
11. Ibid., 18–30.
12. This is roughly equivalent to assuming a multiplier of 1.5 for program expenditures.
13. Franklin D. Roosevelt, “Executive Order 9357—Transferring the Functions of the Public Works Administration to the Federal Works Agency,” July 2, 1943, available at http://www.presidency.ucsb.edu/ws/?pid=16423.
14. For more detailed descriptions of these four programs than this chapter provides (and of the PWA), see Burns and Williams; and National Resources Planning Board, Security, Work and Relief Policies (Washington, D.C.: U.S. Government Printing Office, 1942).
15. Franklin D. Roosevelt, “Campaign Address on the Federal Budget” in Samuel I. Rosenman, comp., The Public Papers and Addresses of Franklin D. Roosevelt, Vols. 1–5 (1928–1936) (New York: Random House, 1938), Vol. 1, 810.
16. Federal Emergency Relief Act, Statutes at Large 48 (1933), 55.
17. Williams was particularly hated by Southern Democrats in Congress, who succeeded in blocking his elevation to head the WPA when Hopkins resigned from the post to become Secretary of Commerce in 1938, and again when Roosevelt nominated him to head the Rural Electrification Administration in 1945. He lived long enough, though, to attend the March on Washington in 1963 and died in 1965 railing against the Vietnam war to a former protégé from his New Deal days—Lyndon Baines Johnson. See John Salmond, A Southern Rebel: The Life and Times of Aubrey Willis Williams, 1890–1965 (Chapel Hill, NC: University of North Carolina Press, 1983).
18. Bonnie Fox Schwartz, The Civil Works Administration, 1933–1934: The Business of Emergency Employment in the New Deal (Princeton NJ: Princeton University Press, 1984), 36.
19. Franklin D. Roosevelt, “Executive Order 6420-B—Establishing the Federal Civil Works Administration,” Nov. 9, 1933, in Franklin Delano Roosevelt, The Public Papers and Addresses, Vol. 2, 456.
20. Burns and Williams, 31, Table 2.
21. Susan Carter et al., eds., Historical Statistics of the United States: Millennial Edition (Cambridge, UK: Cambridge University Press, 2006): series Ea932 and Ea933, accessed April 24, 2013, available at http://hsus.cambridge.org/.
22. Ibid., series Ba474 and Ba477.
23. Burns and Williams, 133, Supplementary Table 2.
24. Schwartz, 43–44.
25. Burns and Williams, 32; Schwartz, 133.
26. Schwartz, 156–180.
27. National Resources Planning Board, Appendix 5.
28. For descriptions of some of these projects, see Schwartz, 133–139.
29. The only authoritative comparison of male and female unemployment rates that I am aware of from the era is from the 1940 census, which found an identical rate of 9.6 percent for both male and female labor force participants. Carter et al., Historical Statistics, series Ba353 and Ba354.
30. Schwartz, 179.
31. Ibid., 158–164.
32. For descriptions of CWS projects developed for women workers, see ibid., 172–178.
33. John Charnow, Work Relief Experience in the United States (Washington DC: Committee on Social Security, Social Science Research Council, 1943), 40–41.
34. Carter et al., series Ba475.
35. Charnow, 11; Schwartz, 43.
36. Schwartz, 105–109.
37. Richard L. Worsnop, “Racial Discrimination in Craft Unions,” Editorial Research Reports 1969, Vol. II (Washington, D.C.: CQ Press, 1969), accessed April 29, 2013, available at http://library.cqpress.com/cqresearcher/cqresrre1969112600.
38. Burns and Williams, 26.
39. Charnow, 51.
40. Burns and Williams, 34.
41. When the WPA was established in 1935 it provided average monthly earnings of about $50 (Charnow, 52). Over the seven-year life of the program, average WPA earnings increased. In 1940, the average was about $55 per month (Burns and Williams, 62). In 1942 the average was over $60 per month generally and over $65 per month on certified defense projects (Charnow, 52). The CCC paid all enrollees $30 per month plus subsistence valued at another $30 per month (ibid., 122–123). Earnings in FERA-funded work relief projects increased following the demise of the CWA, averaging about $28 per month (ibid., 51).
42. Schwartz, 117–118. The minimum hourly rates for unskilled workers under this scale were $.40, $.45 and $.50, depending on the area of the country in which the program operated, with the exception of highway construction work, for which the usual rates paid by state highway departments were used, subject to a national minimum of $.30 per hour. The minimum rates for skilled workers were $1.00, $1.10, and $1.20 per hour, depending on the region. Charnow, 58, note 23.
43. Schwartz, 118–125.
44. Charnow, 51; Burns and Williams, 33–34.
45. Charnow, 103–106, 108–109; Burns and Williams, 32–33.
46. Schwartz, 182.
47. Ibid., 53–54.
48. Burns and Williams, 34; Harry Lloyd Hopkins, Spending to Save: The Complete Story of Relief (New York: WW Norton, 1936), 120–121; Schwartz, 182.
49. Schwartz, 183–184; Hopkins, 121–122.
50. Schwartz, 59, 183, 185; Hopkins, 122; Charnow, 82; Searle F. Charles, Minister of Relief: Harry Hopkins and the Depression (New York: Praeger, 1974), 65.
51. Schwartz, 136, 176–178, 186; Hopkins, 113.
52. Schwartz, 135, 174–176, 183, 186.
53. Ibid., 69, 138, 178; Burns and Williams, 35.
54. Arthur M. Schlesinger, Jr., The Coming of the New Deal (Boston: Houghton Mifflin Co., 1958.), 271.
55. Charles, 65.
56. In some regions, wage rates were reduced to as little as $.10 or $.15 an hour by local relief officials. Earning on FERA work relief projects averaged between only $5.50 and $7.25 a week from the spring of 1934 (when the FERA program was reinstated following the demise of the CWA) through the end of 1935 (by which time it had been replaced by the WPA). Burns and Williams, 11; National Resources Planning Board, 44.
57. Burns and Williams, 43, 45; National Resources Planning Board, 44.
58. Nancy Rose, Put to Work: Relief Programs in the Great Depression (New York: Monthly Review Press, 1994), 65–70.
59. Ibid., 76–80, 85–87.
60. Committee on Economic Security, Report of the Committee on Economic Security, reprinted in The Report of the Committee on Economic Security of 1935 and Other Basic Documents Relating to the Development of the Social Security Act (Washington, D.C.: National Conference on Social Welfare, 1985), 70.
61. See Philip Harvey, Securing the Right to Work: Social Welfare Policy and the Unemployed in the United States (Princeton, NJ: Princeton University Press, 1989), 16–20.
62. Committee on Economic Security, 74–75.
63. These included a non-contributory old-age pension system, an unemployment insurance system, an income support program for needy children who lacked a father’s support, a state-administered system of residual relief for persons of working age who were unable to be self-supporting, and a range of public health initiatives, including a national health insurance system (though this latter proposal was withheld from the final report when it became clear that it lacked the political support necessary to be enacted). See ibid.
64. United Nations, Universal Declaration of Human Rights, G.A. Res. 217A (III), 1948, arts. 22 and 25. It is important to distinguish the right to work proclaimed in the UN’s Universal Declaration of Human Rights and generally recognized in international human rights law from the willful misappropriation of the term by anti-union groups in the United States to describe laws outlawing union security agreements.
65. The only noteworthy deviation from the CES’s income security proposals was the exclusion of agricultural and domestic workers from the Unemployment Insurance and contributory old age pension systems established by the act. The CES report opposed any such exclusions, though it supported some others (e.g., public employees). It is often asserted that Congress excluded agricultural and domestic workers from these programs to appease Southern Democrats who wanted to deny the benefits in question to African Americans, but this view has been challenged on the grounds that it was Secretary of the Treasury Henry Morgentheau who promoted the exclusion out of concern for the difficulties involved in collecting payroll taxes from the employers of these categories of workers. The exclusion had in fact been originally proposed by the CES staff on the grounds later cited by Morgenthau, but the CES itself overruled that recommendation because Hopkins and Frances Perkins felt strongly that the programs in question should be universal. Morgenthau, himself a member of the CES, undercut the official CES recommendation in his Congressional testimony, citing concerns by the Internal Revenue Service. See Larry DeWitt, “The Decision to Exclude Agricultural and Domestic Workers from the 1935 Social Security Act,” Social Security Bulletin, 70, no. 4 (2010): 49–68.
66. Emergency Relief Appropriations Act of 1935, Pub. Res. 74-11, 49 Stat. 115 (1935); Franklin D. Roosevelt, “Executive Order 7034—Creating Machinery for the Works Progress Administration,” May 6, 1935, accessed July 23, 2013, available at http://www.presidency.ucsb.edu/ws/?pid=15053.
67. Burns and Williams, 61–64.
68. See Rose, 100–104.
69. Burns and Williams, 39, Table 7.
70. For a slide-show introduction to the WPA in general and to its various art projects in particular, see Franklin D. Roosevelt Presidential Library and Museum, “FDR, the WPA and the New Deal Arts Programs,” accessed July 23, 2013, available at http://www.fdrlibrary.marist.edu/pdfs/ppDIRwpa.pdf.
71. National Resources Planning Board, 393–394.
72. See Social Security, the Official Website of the U.S. Social Security Administration, Social Security History, “CES—Unpublished Study on Health Insurance,” accessed May 1, 2013, available at http://www.ssa.gov/history/reports/cesmedical.html.
73. Schwartz, 221–225.
74. Ibid., 234–237.
75. Ibid., 102–128.
76. Dona Cooper Hamilton and Charles V. Hamilton, The Dual Agenda: The African American Struggle for Civil and Economic Equality (New York: Columbia University Press, 1997), 9–27.
77. Franklin D. Roosevelt, “Annual Message to the Congress, January 4, 1935” in The Public Papers and Addresses, Vol. 4, 18–22; National Resources Planning Board, 237.
78. Franklin D. Roosevelt, “Campaign Address to the Commonwealth Club of California, September 23, 1932,” in The Public Papers and Addresses, Vol. 1, 754.
79. Ibid., 754–755.
80. Roosevelt, “Acceptance of the Renomination for the Presidency, Philadelphia, PA, June 27, 1936,” in The Public Papers and Addresses, Vol. 5,233–234.
81. Author’s calculations using data from National Resources Planning Board, Appendix 1; and Carter et al., eds., series Ba470, Ba474, Ba477.
82. Robert D. Leighninger, Long Range Public Investment: The Forgotten Legacy of the New Deal (Columbia: University of South Carolina Press, 2007), xv. That legacy included schools and university buildings, courthouses and prisons, bridges, viaducts, tunnels and dams, hospitals and clinics, waterworks and incinerators, ports and zoos, golf courses and tennis courts, stadiums and auditoriums, botanical gardens and museums, fairgrounds and farmers’ markets, city halls and fire stations, parks and trails, shelters and lodges. For a quantitative accounting of some of these public works, see National Resources Planning Board, 342, notes 4, 5.
83. John Maynard Keynes, The General Theory of Employment, Interest, and Money (New York: Harcourt Brace, 1936), 129.
84. Robert Goralski, World War II Almanac: 1939–1945 (New York: Putnam’s Sons, 1981).
85. For a more extended discussion of this point, see Philip L. Harvey, “Back to Work: A Public Jobs Proposal for Economic Recovery” (New York: Demos Policy Brief, 2011), 20–23, accessed May 3, 2013, available at http://www.demos.org/sites/default/files/publications/Back_To_Work_Demos.pdf.
86. Franklin D. Roosevelt, “The Annual Message to Congress, January 6, 1941,” in Samuel I. Rosenman, comp., The Public Papers and Addresses of Franklin D. Roosevelt, Vols. 6–9 (1937–1940) (New York: Macmillan, 1941), Vol. 9, 670–72.
87. The balance of the list consisted of: (1) the right to adequate food, clothing, shelter, and medical care; (2) the right to security, with freedom from fear of old age, want, dependency, sickness, unemployment, and accident; (3) the right to live in a system of free enterprise, free from compulsory labor, irresponsible private power, arbitrary public authority, and unregulated monopolies; (4) the right to come and go, to speak or to be silent, free from the spying of secret political police; (5) the right to equality before the law, with equal access to justice in fact; (6) the right to education, for work, for citizenship, and for personal growth and happiness; and (7) the right to rest, recreation, and adventure; the opportunity to enjoy life and take part in an advancing civilization. National Resources Planning Board, “Our Freedoms and Rights,” quoted in Marion Clauson, New Deal Planning: The National Resources Planning Board (Baltimore, MD: Johns Hopkins University Press for Resources for the Future, 1981), 183–184.
88. The other economic and social rights recognized in the ALI Statement were the right to education, the right to adequate food and housing, and the right to social security (including access to medical care and compensation for loss of livelihood). American Law Institute, Statement of Essential Human Rights (New York: Americans United for World Organization, 1945), 281–286.
89. Franklin D. Roosevelt, “Message to the Congress on the State of the Union, January 11, 1944,” The Public Papers and Addresses, Vol. 13, 40–41.
90. Quoted in William E. Leuchtenburg, The FDR Years: On Roosevelt and His Legacy (New York: Columbia University Press, 1995), 74–75.
91. Committee on Economic Security, 70.
92. For a description of how the direct job-creation strategy could be funded by a polity that operated under a balanced-budget constraint, see Philip L. Harvey, “Securing the Right to Work at the State or Local Level” (Berkeley, CA: Institute for Research on Labor and Employment Big Ideas for Jobs Report, 2011), accessed May 3, 2013, available at http://www.philipharvey.info/securing.pdf.
93. Alan Blinder and Mark Zandi, “How the Great Recession Was Brought to an End” July 27, 2010, accessed August 11, 2010, available at http://www.economy.com/mark-zandi/documents/End-of-Great-Recession.pdf; Congressional Budget Office, “Estimated Impact of the American Recovery and Reinvestment Act from January 2010 through March 2010,” May 2010, accessed May 3, 2013, available at http://www.cbo.gov/sites/default/files/cbofiles/ftpdocs/115xx/doc11525/05-25-arra.pdf.
94. See Harvey, “Back To Work”; Harvey, “Securing the Right to Work at the State or Local Level.”
95. My assumed program budget includes one dollar of funding for non-labor costs for every three dollars spent on employee compensation.
96. This average assumed that approximately 85 percent of all program jobs would be full-time (averaging forty hours per week) and the balance, part-time (averaging twenty hours per week). The average budgeted cost per full-time-equivalent (FTE) job would have been $53,300.
97. It should be noted that the same is true of spending devoted to hiring regular public-sector workers, so using stimulus dollars for that purpose would be as effective as the operation of a direct job-creation program.
98. This indirect job-creation effect is attributable to government purchases of materials and supplies for the program and to purchases of privately produced goods and services by program workers with their program wages.
99. Harvey, “Back to Work,” 6, Table 2.