influence within the New Deal. Critical to the success or failure of either side was their explanation of how the Great Depression had happened in the first place, what was causing the continuing crisis, and what the most effective federal action would be to bring it to an end. The outcome—the victory of both the WPA and direct job creation—was a genuine turning point in the New Deal. It opened the way for more substantial fiscal intervention and a larger public presence in the private economy, a more profound connection between the New Deal and the “one-third of a nation” struggling with economic deprivation, and greater development of American job policy.
The Other ERA
In the winter of 1935, the New Deal emerged from a period of uncertainty and drift caused by ongoing conflicts over the NRA’s industrial codes, the crop-reduction policies of the AAA, and the nature and extent of economic intervention needed to produce recovery. The 1934 midterm elections provided an opening for a new burst of activity: fourteen more seats in the House and ten more seats in the Senate gave the Democrats a filibuster-proof two-thirds majority, with a much stronger progressive wing than had been the case during the Hundred Days. At the same time, the need for a new round of intervention was stark: ten million workers were on the unemployment lines, eighteen million Americans were on relief, and industrial production was stubbornly sluggish.4 Beginning with the dramatic paired introduction of the Social Security Act and the Emergency Relief Appropriation Act, accompanied by hard-hitting presidential messages that spoke of an intensified commitment to the poor and downtrodden, the Roosevelt administration began pushing forward with a new agenda.
Historians refer to this period as the Second Hundred Days and the start of a Second New Deal.5 The Emergency Relief Appropriation Act was central to the moment. Introduced within a package with the Social Security Act, the ERA if anything drew more press attention with its record-setting request for funds, hailed by President Roosevelt in his presidential message as “a permanent program … based on the principle … that the right to work is the birthright of every citizen” that would protect “the unemployed, the rural groups, and others not benefited by the social security act.”6
Debate over the bill was fierce, albeit one-sided—but, in the end, the size of the Democratic majority enabled an easy triumph: 317 to 70 in the House and 66 to 11 in the Senate.7 Indeed, as Edwin Amenta and Drew Halfmann have argued, the passage of this bill was part of a formative process of wielding the Democratic majority into a working coalition, as rural Southerners and Westerners traded their votes on the relief bill in return for favorable votes on agriculture bills from urban Democrats from the Northeast and Midwest.8
The ERA gave President Roosevelt an unusual amount of flexibility in deciding how to spend the $4.88 billion. The act appropriated the funds “in order to provide for relief, work relief, and to increase employment by providing for useful projects … to be used in the discretion and under the direction of the President” and authorized spending for highways, streets, rural rehabilitation and relief, conservation and irrigation, rural electrification, housing, the Civilian Conservation Corps (CCC), and a host of other purposes.9 The act granted Roosevelt wide latitude in prescribing rules for “the employment of labor … to persons receiving relief,” in regards to wages and hours, as well as in creating any new agency to oversee such work, authority that would later be used to establish the work program’s key features, from the NEC’s ACA to the WPA itself.
Characteristically, Roosevelt established an institutional framework that divided authority among his subordinates, while giving himself maximum flexibility. The new Division of Applications of the NEC, headed up by Frank Walker, would receive the project applications, pouring in from all forty-eight states and thousands of municipalities, counties, and relief administrations, and pass them on according to their merits. The new WPA, led by Harry Hopkins, would analyze the applications for their capacity to increase employment and then provide a list of relief clients in the locality of each project. The PWA under Harold Ickes was supposed to be the main planning and building institution, undertaking the heavy lifting of the work program by carrying out the public works, using the manpower assigned to it by the WPA.
Theoretically, this system should have worked smoothly. Local applications for projects would flow into the Division of Applications, which would sift out the good proposals from the bad. The WPA would certify that enough labor was available in the area to undertake the project, and the PWA would start building. However, two factors would destabilize this institutional setup and give direct job creation advocates a chance to dominate the agenda. First, there was a fundamental difference in philosophy between the hiring and building sides of the system. Ickes and his administrators in the PWA did not believe that relief workers (often unskilled or semiskilled workers who had been unemployed for several years) were qualified to efficiently carry out the complicated and demanding work of the kinds of “heavy” construction that would make for valuable, long-lasting, and, above all, “self-liquidating” public works that could repay the cost of construction through user fees (toll bridges and tunnels, for example).
Ickes and his civil engineers preferred to work with experienced building contractors, including the future corporate leviathans of Bechtel and Halliburton, which had the skilled workforce, technical expertise, and machinery needed to construct complex structures such as Boulder Dam. For their part, Hopkins and his advisors in the WPA viewed traditional public works as too slow and cautious, ultimately unsuited to the task of reducing unemployment. After all, when major projects were steered to construction firms, the accompanying dollars went to men already employed, and construction firms were reluctant to do more than add on workers at the margins, let alone hire the millions of Americans in desperate need of work.
Second, FDR made a number of structural decisions that created tension within the work program. The WPA, as the certifying authority for public works projects, was in a position to choke off the PWA by refusing to certify their projects on the grounds that they were not employing enough unemployed workers. Moreover, at Hopkins’s behest, Roosevelt had, through an executive order, granted the WPA the authority to “recommend and carry on small useful projects designed to assure a maximum of employment in all localities.”10
This provision of the executive order may have been intended to provide President Roosevelt the leeway to speed the work program up. After all, Secretary Ickes had been slow to launch the PWA in 1933–1934 while Hopkins had created the Civil Works Administration (CWA) in record time. In any case, the loophole gave Hopkins an opening to proceed with direct job creation, independent of the PWA’s authority. Finally, the NEC itself retained the ultimate authority to allocate funds to projects, leaving the WPA, PWA, and the Division of Applications without any clear hierarchy of authority among them. Accordingly, each party could advocate for itself on equal terms before the NEC.
It is hard to tell from the historical record to what degree FDR was aware of the impact that these decisions would have. Certainly, the split into three agencies was consistent with Roosevelt’s tendency to make lieutenants compete with each other and rely on his decision making as a means of keeping a handle on the sprawling dimensions of the New Deal.11 Especially in regards to the “small useful projects” rule, it may have been the case that FDR knew what Hopkins would do with it, and that he accepted direct job creation as a “plan B” if Ickes took too long to roll out his public works. From the president’s draft press release on the work program, for example, we can see that “the Works Progress Administration … [is] responsible for the honest, efficient, speedy, and coordinated execution of the Work Relief Program as a whole … in providing employment [for] the maximum number of persons in the shortest time possible.”12
The ultimate structure of the work program, however untidy it might have been, provided a mechanism for the resolution of political conflict over ERA funds. Applicants before the ACA would face an adversarial setting in which to make a case for their share of the finite lump of ERA funding. The ACA itself would provide a majoritarian solution by voting requests for allotments up or down, and the work program’s separate parts would proceed independently. This institutional structure would also be a major spur to the creation of rival camps, as it encouraged both sides to repeatedly justify their own proposals, attempt to win over the people who would be casting votes on the future