Reagan- and Clinton-Era Reforms: Early Characterizations of Race and the Welfare State

It is no coincidence that accounts of the origins of the welfare state emerged in the late 1980s and 1990s, when welfare programs and social policy became the source of deep and significant divides in American politics. Scholars have now connected this moment to the fracture and breakdown of the New Deal political order, following a period of Democratic Party rule and triumphant political liberalism. By the late 1980s, amid narratives about a significant rise in crime (particularly in cities and urban centers) and after the ascendance of figures like Richard Nixon and Ronald Reagan, many components of the New Deal state had come under fire. It was convenient and plausible for many to blame urban unrest and race riots on the party that had championed civil rights reform in the 1960s, criticizing affirmative action and welfare programs.

Seeking to deflect this line of attack, scholars in the late 1980s began to construct a more historically accurate account of the US welfare state’s development. In December 1991, in a special issue of the Nation, Frances Fox Piven and Richard A. Cloward stated the issue plainly: “The current political wisdom, even among some on the left, is that blacks and liberals are to blame for the Democratic Party.”¹¹ Piven and Cloward, as well as their fellow contributors to this special issue (including Adolph Reed Jr, Julian Bond, Rogers Smith, and James Forman Jr), attempted to recast this characterization by highlighting the structural flaws of the New Deal Democratic coalition, particularly when it came to race. Their goal was to defend affirmative action and critique the Democratic Party’s approach to welfare reform.

The same tenor and critique clearly motivate two of the earliest scholarly arguments in this vein: Jill Quadagno’s The Transformation of Old Age Security and The Color of Welfare. In the earlier volume, Quadagno takes a broad view of welfare state development, asking “why the American welfare state developed later and less comprehensively than its European counterparts.”¹² She identifies three primary explanations: the strength and power of the private sector, significant divisions between craft and mass-production workers, and the “dualism” of American economic development in the North and the South.¹³ Quadagno’s focus on regional dualism in American political development in particular provides a comprehensive account of the features of the Southern plantation economy, which connects to her arguments about race, Southern lawmakers, and the welfare state. While the lines of class conflict in the Northern economy were drawn between capital and labor, the South’s “labor-intensive agrarian sector” formed a distinct mode of production, not feudal or capitalist but a “plantation mode of production” that was most “characteristic of the cotton-growing counties.”¹⁴ And, importantly, it was a mode of production that relied heavily on African American workers. The Southern agricultural sector sought to preserve control over the plantation mode of production, which it was able to do given the critical committee leadership of Southern Democrats in Congress, particularly in the House of Representatives, and in the Democratic Party.¹⁵

Especially in the early phase of New Deal policymaking, Quadagno traces Southern lawmakers’ attempts to restrict the size and scope of federal social insurance programs, both by insisting that states be granted greater control over program administration and by arguing for a wide variety of exclusions. Yet Southern lawmakers were not alone in weakening insurance programs; Quadagno also argues that business was critical in pushing toward a contributory (not redistributive) social security framework, and that organized labor was too divided internally to provide an alternative.¹⁶ Ultimately, Quadagno contends that the resulting welfare state reflected fundamental class-based hierarchies. However, she also points to other elements of social stratification. Referring to relative access and benefits provided in old-age assistance and disability insurance, she emphasizes that “the quality of benefits varied, not only along lines of class but also by race and gender.”¹⁷

But Quadagno’s conclusion is not that racial stratification in program benefits was an effect of class-based inequities; she argues that Southern lawmakers were motivated by concerns that broad, national legislation would “undermine planters’ paternalistic control over tenant labor, particularly black labor,” and that “southern congressmen had no intention of letting federal funds go directly to black workers.” In support of this claim, Quadagno points to what has become one of the most frequent examples of Southern lawmakers’ racial preferences being written into the law: the fact that agricultural and domestic workers were excluded from the SSA and the NLRA. As of 1935, she points out, “three-fifths of all black workers, most of whom resided in the South, were employed in agricultural or domestic service.”¹⁸

However, disparate impact does not necessarily prove discriminatory intent. While it was true that the 3.5 million black workers who were excluded from the SSA constituted the majority of black workers, many of whom lived in the South, they did not constitute the overall majority of workers cut out of federal insurance programs. As Social Security historian Larry DeWitt has documented, these agricultural and domestic worker exclusions also prevented nearly twelve million white workers from benefiting from the old-age insurance program.¹⁹ Although Southern lawmakers employed the same kinds of local control arguments in marshaling opposition to federal attempts to roll back Jim Crow laws and later to resist implementation of desegregation following Brown v. Board of Education, their objectives in this case are not best understood through the lens of racial exclusion.

Instead, my own research has found much stronger support for a version of Quadagno’s dualism thesis. The division between the agricultural and industrial sector, however, transcended regional lines in important ways. In national policy debates over federal agricultural and labor regulation, a variety of representatives of the agricultural sector argued forcefully (and successfully) for agricultural workers to be excluded from industrial policy for fear of “dual administration” under differing policy domains. Policymakers, including some New Dealers, accepted this distinction and allowed agricultural workers to be segmented from the industrial workforce. Importantly, this delineation was not challenged by organized labor, which instead remained largely silent in this policy fight.²⁰

Broadening the lens beyond the Southern Democrats brings into sharper relief the reasons their policy proposals were ultimately accepted by other national politicians. As David Kennedy suggests, Roosevelt’s willingness to devolve authority to the states was rooted in “the constitutional doubts that overhung any federal initiative,” which “created a tightly confining matrix within which the Committee on Economic Security planners were compelled to work.”²¹ Roosevelt needed to narrowly tailor policies to avoid the constitutional challenges that had brought down the first wave of New Deal policies. Despite Quadagno’s attention to these dynamics elsewhere in the book, she does not provide the same context to help make sense of the agricultural and domestic workers’ exclusions.

Racial stratification and the racism of Southern lawmakers were not the primary focus of Quadagno’s 1988 volume. However, in The Color of Welfare, published in 1994, she moved away from her focus on class and the development of the welfare state, arguing instead that “means-tested programs of the American welfare state had less to do with maintaining class divisions than with maintaining racial segregation.”²² Quadagno’s shift in analytical framework was likely an effort to align with other Clinton-era welfare critiques. In the introduction, she acknowledges the significance of urban and social unrest in the 1990s, including the Rodney King riots, and criticizes President Clinton’s failure to reject Republican arguments linking the root of the unrest to “liberal social programs of the 1960s” that unfairly benefited racial minorities.²³ Instead, she argues, the triumphs of the Great Society and War on Poverty resulted from their use of social programs designed to promote equal opportunity for African Americans through community-action programs, job-training programs that forced skilled unions to integrate, and, most important, affirmative action policy. These social programs were not special favors but rather necessary corrections to a social welfare state established in the New Deal period that had created “impediments to racial equality.”²⁴ The primary culprits, she argues, were not only Southern Democrats, who refused to “support any welfare programs that would place federal funds in the hands of black sharecroppers,” but also President Roosevelt and the New Dealers, who agreed to exclude African Americans from core Social Security programs in order to secure Southern support. While Southern Democrats were the architects, it was New Dealers’ acceptance of those exclusions that cemented them into law.

These assessments lead Quadagno to a broader critique of the civil rights legacies of the Northern Democrats and Roosevelt, beyond his capitulation in social welfare provision. Quadagno argues that, in an effort to “stabilize his unwieldy coalition of northern workers and white southerners,” Roosevelt refused central demands of civil rights activists at the time — namely, to support anti-lynching and anti-poll-tax legislation in Congress. Roosevelt’s refusal to support civil rights priorities in the 1930s and ’40s is frequently cited as evidence of the segmentation of racial and economic liberalism in this period of partisan history. Eric Schickler has effectively documented how slow, state-level shifts led the Democratic Party to embrace national commitments that reflected principles of both racial and economic liberalism in the 1960s.²⁵ However, Schickler’s own analysis also indicates that, regardless of Roosevelt’s civil rights policies, African Americans were more supportive of New Deal federal jobs and social welfare programs than white Americans, which supports the point that, despite its limitations, the New Deal was a transformative program for African Americans.²⁶ Further, recent scholarship has highlighted other important aspects of Roosevelt’s civil rights legacy. Kevin McMahon documents Roosevelt’s creation of the Civil Rights Section in the Department of Justice and his focus on and support for civil rights efforts in the federal judiciary.²⁷ Sidney Milkis points to the importance of Roosevelt’s 1938 purge campaign — an electoral effort to push Southern Democrats out of office. While ultimately unsuccessful, the campaign shows that Roosevelt and Northern New Dealers were engaged in concerted efforts to rein in the power and influence of Southern Democrats.²⁸ These accounts point to a more complicated and fraught relationship between Roosevelt and the Southern Democrats.

Despite these efforts to purge Southern Democrats, Quadagno describes Roosevelt as primarily capitulating to them. She argues that Roosevelt actively wove “racial inequality into his new welfare state,” again pointing to the exclusion of agricultural and domestic workers from old-age insurance and the lack of national standards for unemployment insurance. “These omissions were not random,” she notes. “Rather they reflected a compromise reached with southern Democrats over the structure of the welfare state.”²⁹ Quadagno does not provide her own empirical evidence for these claims of exclusionary intent behind Southern Democratic demands and Roosevelt’s response. Instead, she references two other accounts of the development of the US welfare state: James Patterson’s America’s Struggle Against Poverty and Edward Berkowitz’s America’s Welfare State.³⁰

Yet neither of these accounts provide evidence that racial exclusion was the primary factor that led to the agricultural and domestic workers’ exclusions. Patterson argues that the failure to offer a comprehensive and national unemployment insurance program resulted from liberals’ inability to build support for the idea that, since “unemployment was a national structural problem, it should be funded by federal subsidies from general tax revenues.” Their second failure was not to secure the inclusion of the classes of workers most in need — “employees of small firms or to domestic and agricultural workers,” all of whom were excluded. Instead, unemployment mirrored private pension funds, which had been developed in the 1920s, and reflected an overall desire for economic efficiency over broad inclusion.³¹ Reflecting on the consequences, Patterson highlights that “women, blacks, and migrants suffered especially from this policy.”³²

Similarly, Berkowitz’s discussion of the exclusion of agricultural and domestic workers points to New Dealers’ and social insurance architects’ acceptance and adoption of the “popular prejudice” that farms and farmers functioned very differently than large industrial employers. In particular, planners accepted the idea that farmers “kept poor records of their payroll, and in many cases hired live-in help, which meant that part of the laborer’s income took the form of room and board.” Exclusion of these workers, then, was an effort to “start modestly” and prioritize “effective enforcement” of the plan.³³

Both Patterson and Berkowitz provide compelling evidence that the exclusion of agricultural and domestic workers was intentional, not random, and that it resulted in an uneven national social insurance system. The delineation between the industrial and farm economy certainly served the interests of the largely farm-based economy in the South. But these exclusions did not have exclusively Southern roots. For example, Mary Poole documents the influence of Wisconsin economists on members of Roosevelt’s Committee on Economic Security. Wisconsin’s system for unemployment insurance, which was established in 1932, served as a model for the federal architects of the SSA, and importantly the Wisconsin program excluded agricultural and domestic workers.³⁴ Richard Rodems and H. Luke Shaefer document that agricultural workers were similarly excluded from welfare programs in other Western democracies.³⁵

Despite this being a broader tendency, Quadagno focuses on race as the primary and unchanging barrier to social welfare in the United States. Even after the political influence of the South faded, Quadagno claims, race continued to be “the defining feature of the American welfare state in the 1960s.”³⁶ Yet it was during the 1950s that Congress amended the SSA to include agricultural and domestic workers in the programs. Quadagno’s failure to acknowledge this essential change significantly undercuts her argument, and it also elides the critical point that Southern Democrats’ resistance shifted significantly during the New Deal period. Economic historians Lee Alston and Joseph Ferrie argue that, at a certain point, Southerners no longer marshaled the same resistance to old-age benefits and other welfare programs.³⁷ Instead, as later sections will discuss, Southern resistance was trained on labor policy.

Racial Intent Versus Racial Effect: Competing Causal Narratives

To understand the shift in Quadagno’s causal arguments, it is useful to consider two subsequent accounts that sought to further document racial segmentation in the welfare state and, in so doing, opened two potential directions for subsequent scholarship. Robert Lieberman’s Shifting the Color Line and Michael K. Brown’s Race, Money, and the American Welfare State each explore New Deal welfare state development and its impact on African Americans. Although they draw on similar bodies of evidence, their distinct arguments lead to fundamentally different assessments of the New Deal. While both authors examine policy development and implementation, Lieberman’s emphasis on racial intent as a primary causal force in the design and implementation of New Deal programs runs against Brown’s narrative, which underscores that the racial effects of policies resulted from an array of influences. These distinct historical and empirical trajectories point toward different political conclusions.

In Shifting the Color Line, Lieberman, aligning closely with Quadagno, argues that “race inhibited the development of a strong, unitary, centralized welfare state in the United States.” Here Lieberman adopts a dialectical understanding of race, arguing not only that race stymied welfare state development but also that welfare state development in turn “helped reshape the politics of race and the place of racial minorities in American life,” fixing the position of African Americans in contemporary society. Lieberman’s use of this dialectical frame to make sense of the politics of race provides a clear illustration of what sociologist Karen Fields and historian Barbara Fields have described as a tendency to describe a variety of identities, experiences, and social practices under the umbrella term “race.” In particular, using “race” as a shorthand produces a “weird causality,” conflating ideas about group characteristics and identities (race) with the social practices that act on and cement perceptions of those identities (racism).³⁸ Ultimately, Lieberman’s account suffers from this causal confusion. While he points to specific actions of policymakers and administrators, he also claims that those same policies contain a set of “racially relevant, or race-laden, exclusions,” which does little to describe the cause or intention behind them.³⁹

Lieberman’s account begins with the construction of social insurance in the SSA of 1935. He argues, like Quadagno, that social insurance “sorted Americans by class, classifying their target populations by occupation and work status.” This class segmentation, he posits, also led to further sorting into racial categories, “in keeping with the prevailing racial norms of the 1930s and the institutional structure of American politics.” The result was a class-based and racially discriminatory welfare state by design.⁴⁰ He examines the design and administration of three separate policies under the SSA: old-age insurance (OAI), Aid to Dependent Children (ADC), and unemployment insurance. Each of these policies followed a different trajectory of statutory and administrative excision and inclusion.

Overall, Lieberman identifies two primary strategies opponents used to make Social Security less inclusive: limit federal oversight and regulation and shift implementation to the state and local levels.⁴¹ His primary evidence for this claim is, once again, the exclusion of agricultural and domestic workers from both the SSA and the NLRA. Lieberman asserts that exclusion based on job category was an effective strategy for limiting the OAI program and points to its exclusion of agricultural and domestic workers, which disproportionately impacted black workers, particularly in the South.⁴² However, for programs like ADC, where occupational categories were not the determining metric for public assistance, the result was quite different. Lieberman notes that “blacks were not systematically excluded as they were from social insurance benefits”; rather, it was the case that “blacks throughout the country were more likely than whites to be among [ADC’s] beneficiaries.” Importantly, Lieberman’s evidence indicates that African Americans disproportionately benefited from programs that were not means-tested around job categories. However, he goes on to argue that for programs like ADC, the “race-laden” administration still worked to limit the program’s impact on African Americans.⁴³

Lieberman’s evidence here seems to support the idea that programs that did not discriminate along class lines (either by means testing, directed exclusions, or program limitation) were ultimately more inclusive of African Americans. Where programs were “sorted by class,” they became narrower and more limited in ways that also, consequently, limited the welfare state along racial lines. Yet from this base of evidence, Lieberman’s conclusion that “race and class were mutually constitutive in the making and growth of the American welfare state” does not clearly align with his overall thesis that “race inhibited the development of a strong, unitary, centralized welfare state.”⁴⁴ Instead, Lieberman’s primary evidence introduces some causal confusion into his account. At many points, Lieberman inverts his original claim that beneficiaries were sorted by occupation and class and posits instead that the primary point of sorting and excluding was to divide the population by race.

Lieberman’s discussion of the exclusion of agricultural and domestic workers from OAI is a clear example of this inversion. In trying to construct large, federally managed, and inclusive welfare programs like OAI, New Dealers faced opposition from Southern Democrats. The most pronounced debate over inclusivity, Lieberman reports, was over the “question of inclusion or exclusion of agricultural and domestic workers,” who were ultimately excluded from OAI.⁴⁵ While these were occupational exclusions, he points to occupational data from the 1930s in order to demonstrate that those exclusions had “a racially imbalanced effect, and this effect was national, touching the North as well as the South.”⁴⁶ Lieberman’s claim is more guarded than that of Quadagno and other scholars. However, by foregrounding the racial implications of an exclusion based on occupation, he begins to invert the race-class ordering and the logic he claims to follow. He points toward the racially stratified consequences of this exclusion as evidence of racial motivations of the legislation’s architects and, perhaps more importantly, its opponents. While Lieberman acknowledges that those opponents, Southern members of Congress, were not “monolithically conservative,” he nevertheless contends that their racial preferences were the primary driver behind a racially circumscribed welfare state.⁴⁷

Michael Brown poses a direct challenge to Lieberman’s characterization of Southern lawmakers. While many of the most powerful Southern Democrats, like Senators Howard W. Smith and Harry F. Byrd, showed considerable support for the relief programs that would benefit the struggling Southern economy (the same programs that initially disproportionately benefited African Americans), these programs are noticeably absent from Lieberman’s account. Brown shows that Southern support for the New Deal’s early relief efforts began to wane when national efforts became focused on social insurance, which Southern lawmakers feared would increase taxes and remove local programmatic control by instituting nationalized benefits standards.⁴⁸ Alston and Ferrie provide further support for Brown’s critique of Lieberman, documenting the shift among Southern members who supported early relief measures and later opposed the expansive, federally managed programs in the SSA they believed would threatened their “paternalistic system of southern labor relations,” which was defined by low labor and supervision costs.⁴⁹ Brown also argues that Lieberman overstates the stranglehold that the South had over Roosevelt, instead suggesting that Roosevelt held leverage but chose not to use it because of his preference for fiscally circumscribed policies and regressive taxation.⁵⁰ Ultimately, Brown doesn’t contend that race and racism were irrelevant factors in the development of the American welfare state, but he cautions that the assumption that “racial exclusion” was the “sole problem confronting African-American families risks oversimplifying the problem.”⁵¹

Like Quadagno, Brown also situates himself in the debates over 1990s welfare reform. “Had we created a more capacious and comprehensive welfare state,” Brown argues, “there would have been more racial equality within the welfare state, and racial hostilities would be less likely to serve as a channel for conflict over social policy.”⁵² In order to understand the racial dynamics undergirding the ’90s welfare debate, Brown claims, it is necessary to understand the initial construction of a less “capacious and comprehensive” US welfare policy during the New Deal era. He characterizes this as a poorly funded set of fragmented welfare programs that relied heavily on means-tested programs, private health insurance, and a “historically weak commitment to full employment.”⁵³ In short, racial stratification was an important effect of New Deal welfare provision but not its only limitation. While there were some exceptions, and some programs provided more universal and comprehensive benefits across the population, Brown argues that the “pattern of social provision” that emerged from the New Deal was a version of “truncated universalism.”⁵⁴ He identifies two primary fetters on welfare state development: race and money. Despite building a political coalition around the promise of New Deal universalism, where class identity “overwhelmed ethnic and racial status,” the welfare state “concealed a racial bifurcation” in its new policies and programs.⁵⁵

Brown’s analysis begins in 1933, in the earliest phase of New Deal policymaking, a period that Quadagno and Lieberman both pass over, which was characterized primarily by massive relief payments aimed at lessening the extreme consequences wrought by the Great Depression. These relief payments targeted those most impacted by the economic downturn, which meant they disproportionately benefited African Americans, who were overrepresented in the lowest rungs of the economy. During the second major wave of New Deal policy development, which included the SSA of 1935 (the origin of many welfare state programs), Brown’s evidence shows that while many African Americans were excluded from old-age and unemployment insurance through “statutory occupational exclusions,” they were, once again, disproportionate beneficiaries of some means-tested programs.⁵⁶ However, he also demonstrates that despite being major beneficiaries of these early programs, African Americans’ overrepresentation in relief and means-tested programs produced a negative backlash and stigmatization of black Americans as overly reliant on the welfare system. For example, when he compares the transfer processes for individuals receiving different forms of government assistance, he finds that while African Americans were disproportionate beneficiaries of relief, they were transferred from relief programs to other benefits programs, like those within the Works Progress Administration, at much lower rates than their white counterparts.⁵⁷ Here Brown provides a clear basis to connect 1930s welfare development to the political realities of the 1990s. Yet as Brown’s careful and thorough documentation of the effects of social policy was taken up by Ira Katznelson, as the next section will show, key arguments and evidence were mischaracterized, and important distinctions between the accounts of Brown, Lieberman, and Quadagno were lost.

“Labor Became Race”: Popular Conceptions of the New Deal

Ira Katznelson’s When Affirmative Action Was White, published in 2005, has become one of the most popular interventions and commonly cited books in these debates. Its primary goals are to examine the ties between social and racial policy and to underscore the influence of Southern Democrats — “guardians of racial segregation” — on Roosevelt and Harry S. Truman.⁵⁸ Katznelson admits that his use of “untold” is polemical, particularly since he sought to draw together various strands of existing scholarship that, in his opinion, had “not been brought together sufficiently.”⁵⁹ His conclusion, as suggested by his provocative title, is that the New Deal was profoundly influential in building the American middle class, but that those public policies were “crafted and administered in a deeply discriminatory manner,” leaving African Americans out of the nation’s first affirmative action programs.⁶⁰ In contrast to the works of Brown and Quadagno, neither the preface nor the introductory chapter of When Affirmative Action Was White provide any sort of link to Clinton-era welfare debates. The political stakes Katznelson underscores is the failure to live up to the goals Lyndon B. Johnson elaborated in his June 1965 address, “To Fulfill These Rights,” a challenge that, according to Katznelson, grew out of the New Deal period and persists to this day.⁶¹

This shift in political target is significant, as Katznelson does not directly intervene in the same scholarly debates over welfare state development that Brown and Quadagno engage. And while he follows much of the logic and impetus of Lieberman’s focus on racial intent, Katznelson is also less focused on program administration and the broader structural impact of the welfare state. He is instead more attentive to the legislative origins of social policies and programs, placing particular emphasis on identifying the actors who actually pulled the strings of exclusion. He claims that, while they never admitted it, Roosevelt and Truman’s efforts to create universal programs were cut off at the knees by the Southern Democrats’ stranglehold on Congress and national politics. However, where Quadagno, Lieberman, and Brown were more cautious about assigning blatantly racist motivations as the key or sole factor driving policy design and implementation, Katznelson claims that the SSA, NLRA, and GI Bill were explicitly “crafted in a discriminatory manner,” constituting the core of America’s first affirmative action project: building the white middle class. Following Quadagno and Lieberman’s cues, Katznelson also fixes the blame on Southern Democrats in Congress, who were allowed to “dictate the contours of Social Security” by insisting that these laws be “tailored to meet their preferences, most notably their desire to protect Jim Crow.”⁶² However, despite these bold and provocative claims, Katznelson lacks clear and concrete evidence of racist intent driving Southern opposition to labor regulations, instead providing interpretations of congressional debates that point toward a different set of goals.

Katznelson foregrounds three core mechanisms in his account: the inclusion of race-laden provisions in New Deal policies, the failure to include antidiscrimination proposals, and the uneven administration of the resulting policies. He moves through the early, middle, and late New Deal periods, drawing together the work of other scholars and adding some novel analysis. For example, his discussions of the early New Deal reconstruct Brown’s assessment of relief distribution (under the Federal Emergency Relief Act, or FERA) and his consideration of the middle period draw heavily on Lieberman’s commentary on the SSA provision. Katznelson’s own analysis, in this book and in other work published in the same period, focuses on labor policy, primarily the 1935 NLRA and the 1938 Fair Labor Standards Act (FLSA).

Strikingly, in several places, Katznelson uses Brown’s evidence to construct completely contradictory conclusions. For example, Katznelson does not acknowledge that FERA beneficiaries were disproportionately African American, a crucial empirical and analytical point for Brown. Brown argues that FERA was “potentially the foundation for a centralized welfare state” and provoked consistent tension and conflict between Harry Hopkins, the federal administrator, and state administrators, whose relations were “frayed and hostile.” Brown acknowledges that there were limits to efforts to maintain federal standards: Hopkins and his chief deputy “had to tailor their relief policies to accommodate the demands of southern plantation owners for cheap farm labor by curtailing relief payments to agricultural laborers and sharecroppers.” However, Brown claims that this was the exception to an overall “radically centralized” program that was best characterized by sharp conflict between federal and Southern administrators in providing relief payments that disproportionately went to black Americans.⁶³

Katznelson quotes this exact passage from Brown in his own discussion of FERA, arguing that “neither the size of benefits nor the pattern of [FERA] distribution was standardized” and painting Hopkins and his deputy’s tailoring to Southern plantation owners as the rule, not the exception.⁶⁴ Katznelson omits Brown’s point that the program was “radically centralized,” instead framing FERA as being, on the whole, decentralized and shaped to Southern demands. While he draws on Brown’s evidence, Katznelson does not acknowledge the difference in their conclusions.

Beyond this, the new evidence Katznelson supplies emphasizes disparate outcomes but without careful consideration of the political economic context. For example, Katznelson argues that local discretion allowed for lower relief rates for black people as opposed to white people in ten Southern states, including some counties in the state of Georgia that excluded black Americans altogether from their relief rolls. In addition, he draws heavily on Gunnar Myrdal’s An American Dilemma as well as Richard Sterner’s The Negro’s Share. (Sterner was a lesser-known Swedish researcher who was part of Myrdal’s research team.) Katznelson relies on Myrdal and Sterner’s evidence to paint a picture of racial differences and “unequal patterns.” Katznelson’s use of the Swedish researchers points to what Adolph Reed Jr and others argue has become a standard trope in race relations research starting in the mid-twentieth century. Myrdal and his Swedish research team are frequently cited in this body of research, which also emerged during the Clinton era, that emphasizes disparate impacts and outcomes while eclipsing consideration of broader structural and economic forces. Reed has emphasized that these studies are characterized by a “naïve, often evasive pursuit of a notion of scientific objectivity that eschews political discussion of the sources of inequality.”⁶⁵

Katznelson is more faithful to Lieberman’s evidence and argument in his characterization of the second wave of New Deal policymaking, particularly on the creation and implementation of the SSA. He posits that the social welfare policies included in the SSA were overall less generous to African Americans, since social insurance programs distributed benefits based on income level and African Americans were in the bottom rungs of employment. Second, he also nods to the exclusion of agricultural and domestic workers in place until 1954, when Republicans gained control of Congress and eliminated the exclusion. Katznelson’s argument here is somewhat perplexing: he asserts that it was increasingly conservative Republican control of Congress that finally loosened the Southern Democrats’ stranglehold control over the lawmaking process.⁶⁶ Are we to understand, then, that it was Republican lawmakers who came to the aid of the disproportionately black agricultural and domestic workforce? This conclusion seems out of step with the political inclinations of Katznelson’s work and particularly discordant with the well-documented partisan realignment that took place in these decades, which ultimately led conservative Southern Democrats to merge with Republicans.

Along those lines, Katznelson’s most significant empirical contributions in the book can be found in his discussion of New Deal labor policy, which has been less thoroughly examined than the welfare state provisions but is deeply connected to them. Here Katznelson again argues that Southern Democrats “traded votes” of support for New Deal labor legislation in exchange for the exclusion of agricultural and domestic workers.⁶⁷ These groups of workers were included in Senator Robert Wagner’s original draft legislation, but in another version of the bill put forward by six senators, three of whom were Southern Democrats, agricultural and domestic workers were explicitly omitted from the bill’s definition of “employee.” That these exclusions were intentional and deliberate is clear. However, Katznelson fails to present compelling evidence that the primary intention was racial exclusion. On the contrary, he acknowledges that “there was no debate on the Senate floor that explains the motives or purposes behind the exemptions.”⁶⁸

While it is true that the racial motives of policymakers are not made plain in these debates, my research identifies two other motivating factors. The agricultural industry, which had significant interest and impact on the debates over the 1933 National Industrial Recovery Act (NIRA) and the 1933 Agricultural Adjustment Act (AAA), argued that these two frameworks would result in double oversight for the agricultural industry, making it needlessly subject to two sets of regulations. To avoid this issue of “dual administration,” they argued that they should be subject only to regulation under the AAA. Southern and Northern lawmakers’ acceptance of this argument meant that the agricultural industry was able to evade federal labor regulations, which were part of the NIRA framework but not the AAA one (a fact that members of the agricultural industry explicitly acknowledged in their congressional testimony).⁶⁹ In sum, there is evidence as to the motivations and interests of those who sought to exclude their sector from being subject to labor regulations; however, that motivation was not primarily racial exclusion in these early debates.

Katznelson makes a more concerted effort to document the racial motivations in debates over the second wave of New Deal labor policy but fails to present clear evidence for his claims. He argues that votes on labor legislation had become a vehicle to protect Southern racism and that opposition to labor bills was rooted in a narrowly construed fear that stronger unions would increase civil rights protest. Rather than acknowledging the fact that stronger unions might transform the economic system and power structures more broadly, he concludes that, for Southern Democrats, “labor had become race.”⁷⁰ He expands these arguments in a 2005 article published with Sean Farhang in Studies in American Political Development that examines the shifting positions of Southern Democrats on labor proposals over three distinct periods of New Deal policymaking. In the piece, Farhang and Katznelson provide a clearer description of the process through which “labor had become race,” yet their evidence fails to provide a compelling foundation for that claim. First, they argue that the only reason Southern Democrats signed on to prewar labor legislation like the NLRA was because it included carve-outs for agricultural and domestic workers.⁷¹ But beyond the fact that those occupational groups were excluded, the authors include very little evidence that this was in fact the bargain.

Furthermore, their account of the exclusion of those occupational groups indicates that Southern lawmakers, serving as champions for the agricultural industry, primarily sought to keep any and all labor regulations out of agricultural production.⁷² They do not link this objective in a direct or indirect way to racism or racial motivations, beyond the fact that the agricultural sector included many black workers. Considering the earlier resistance to federal labor regulations in the NIRA and AAA, the more plausible conclusion from these debates is that the agricultural industry had a vested interest in curtailing labor organizing in the South (as well as in other parts of the country) and that Southern lawmakers readily acted upon this concern.

Farhang and Katznelson point to more explicit discussion of the racial motivations of members of Congress when they move into a discussion of the FLSA, which they argue was a pivotal moment when the Democratic coalition in Congress fractured. They claim that similar exclusions in the FLSA were intended to avoid setting a “floor on wages,” which would have required employers to increase wages in the lowest-paid industries and address the “wide wage disparities between African American and white wage workers.”⁷³ This suggests, and their evidence supports, that there was rhetoric around the wage differentials between white and black workers deployed in conservative and Southern Democrats’ assault on the FLSA. However, Farhang and Katznelson do not draw a clear connection between those debates and the overall “fear of civil rights organizing” they claim drove opposition to the FLSA. On the contrary, the passages they reference reveal an enduring interest in reducing, avoiding, and eliminating federal labor standards. This evidence indicates a more compelling and coherent explanation: Southern opposition grew as federal standards and regulation expanded, a conclusion that is clearly supported by Brown, Alston, and Ferrie. Employers’ desire for greater regional and local control over their labor markets was no doubt in part fueled by an interest in maintaining racial wage differentials in addition to a whole host of factors allowing employers and industry to depress wages and control workers. The fact that this hostility toward labor standards continued to grow over the next decade, which the authors clearly document, even as lawmakers removed Social Security exclusions for agricultural and domestic workers in the 1950s, seems to suggest that Southern lawmakers’ preferences and resistance to federal intervention had once again shifted target.

Conclusion

In sum, departing from a genre of welfare state development literature aimed at uncovering the limits of the New Deal’s brand of universalism, Katznelson’s When Affirmative Action Was White popularized a version of New Deal political development that skirted critical consideration of the political economic context. It offered a more reductive account of the motivations of federal and regional policymakers and administrators in order to illustrate that racial segregation and discrimination not only remained intact after various waves of New Deal policymaking but also were the primary motivator for the architects of those laws. In 2013, Katznelson published the follow-up Fear Itself: The New Deal and the Origins of Our Time. Drawing on the same arguments and evidence from his previous work, he introduces the provocative and pithy idea that Southern Democrats formed a “Southern cage” that allowed them to restrict policymaking in the interest of maintaining the South’s racial system. Despite the popularity of When Affirmative Action Was White, Katznelson also claims that this fear of and deference to the Southern racial structure continues to be “the most overlooked theme in almost all previous histories of the New Deal.”⁷⁴ However, as Dylan Riley notes in his review of Fear Itself, Katznelson again fails to provide compelling evidence that it was racial preference that drove increasing opposition to the New Deal in the South. Riley counters that labor control, which served to maintain control over black workers, is a more historically grounded explanation for Southern lawmakers’ behavior.⁷⁵

Despite these empirical and analytical issues, Katznelson’s account, which builds upon and reinterprets more careful conclusions put forward by Lieberman, Brown, and Quadagno, has become the most popular rendering of the racial legacy of New Deal programs. Returning to the example of Representative Ocasio-Cortez and Briahna Joy Gray, the links between their arguments and Katznelson’s analysis are abundantly clear, particularly in the focus on the ascendance of the middle class and the emphasis on racial intent and racism undergirding New Deal programs. The growth of these ideas has ultimately served to obscure the New Deal’s legacy and limit our understanding of the development of one of the most important episodes in American political development. By pinning the curtailment of social welfare provision and labor law development on the racial motivations of Southern Democrats, Katznelson’s analysis offers rhetorical support to the characterization that the New Deal was racist. This line of argument rests on relatively weak empirical grounds, making its theoretical claims suspect.

This essay is not an invitation for progressives to uncritically celebrate the New Deal state and overlook its significant limitations. Rather, it is an invitation to be concrete and specific in the conclusions we draw regarding its limitations and clear about how they inform progressive politics. One way to do this is to recognize that, while many of the programs comprising the New Deal welfare state adhered to and re-created racist social demarcations, where the New Deal succeeded in constructing broader, more universal programs, those programs disproportionately benefited African Americans. The exchange between Gray and Ocasio-Cortez makes clear that the emphasis on racial motivation has seeped into contemporary discourse in ways that ultimately pose a challenge to efforts to build broad support for a more robust welfare state, conflating uneven effects with a racist or otherwise discriminatory intention. Such claims lead to confusing attempts to appeal to the broad popularity of the New Deal while at the same time condemning the program’s inherent racism. And, more critically, these claims do nothing to further the goals of those attempting to make a more generous social welfare state, instead suggesting a certain futility in efforts to build a more robust set of social democratic institutions geared toward advancing the common good — an argument the Right has eagerly (and easily) capitalized on.