The Early Origins of the Civil Rights Movement in the United States: An Analysis of the Growth of the NAACP
The Early Origins of the Civil Rights Movement in the United States: An Analysis of the Growth of the NAACP
- Daniel Aaronson, Daniel AaronsonFederal Reserve Bank of Chicago
- Jala Abner, Jala AbnerFederal Reserve Bank of Chicago
- Mark BorgschulteMark BorgschulteEconomics, University of Illinois at Urbana-Champaign
- and Bhashkar MazumderBhashkar MazumderFederal Reserve Bank of Chicago
A newly digitized panel of county-level branch activity of the National Association for the Advancement of Colored People (NAACP) is used to describe the potential factors underlying the expansion of political participation in the American South, with a particular emphasis on the short period from the late 1930s through the 1940s. This period has long been recognized for its significant progress in reducing sizable racial gaps in labor market outcomes. But little work in economics has considered the role of political participation in shaping that progress. As the preeminent civil rights organization prior to the 1950s, the NAACP provides a natural lens in which to explore the expansion in political activism during this crucial period. Associative evidence suggests that a few potential channels could be especially worthy of future study, including the role of demographics, increased human capital, expansion in labor demand driven by wartime efforts, reduction in racial violence, latent political activism, and expansions in political and social networks, all of which have been highlighted in a variety of history and social science literatures. However, careful causal empirical work does not currently exist on these factors. Filling in this hole is important for providing compelling evidence on the origins of the 20th century’s most important U.S. political movement, as well as adding to a growing literature in political economy and development economics which examines the role that grassroots activism has played on economic growth and income inequality around the world.
- Economic History
- Health, Education, and Welfare Economics
- Labor and Demographic Economics
- Public Economics and Policy
The 1940s has long been recognized as a period of significant progress in reducing racial gaps in labor market outcomes (e.g., Goldin & Margo, 1992; Smith & Welch, 1989; Wright, 1986). Many studies have considered the causes of this progress with a particular focus on the role of human capital investment, public sector interventions, and World War II (WWII) (e.g., Aaronson & Mazumder, 2011; Aizer et al., 2022; Ferrara, 2022). However, the economics literature has been largely silent about the role of political forces, especially prior to Brown v. Board of Education in 1954, even though the civil rights movement, the most important political movement in the United States over the last century, was already well underway.1 That lack of attention stands in contrast to a number of excellent recent studies that have examined the causal role of grassroots activism on economic growth and upward mobility following recent global political movements like the Arab Spring, the Tea Party Movement, Brexit, and the Hong Kong protests (e.g., Bloom et al., 2019; Campante & Chor, 2012a, 2012b; Cantoni et al., 2019; Madestam et al., 2013).
With that in mind, this article highlights the descriptive factors underlying participation in the National Association for the Advancement of Colored People (NAACP), the preeminent early civil rights organization. Founded in 1909, the NAACP’s mission was to challenge racial disenfranchisement and segregation. Much of the success during their first 3 decades involved legal cases litigated from their New York City and Washington, DC offices. But a good part of the funding for these efforts, as well as its grassroots activism, sprang out of a network of branches spread throughout the country that slowly redefined the nature of the organization. A broad literature across social science considers the role of the NAACP as the canonical example of an organization that serves to identify test cases, focus public attention, and mobilize resources in the face of dispersed local demand for political action (e.g., McAdam, 1999; Morris, 1986). The NAACP’s local branch activity is therefore a potentially relevant indicator of the degree of local political participation in the early days of the civil rights movement.
The county-level facts about branch activity between 1919 and 1956 are laid out in the section titled “Growth of the NAACP.” After a large membership drive during World War I (WWI), branch activity stagnated for much of the 1920s and 1930s, especially in the South where the number of branches remained below 1919 levels for nearly 2 decades. However, starting just before 1940, there was a sharp increase in new branches and a significant upswing in the survival of incumbent branches, leading to substantial growth in membership throughout WWII and the years immediately following. Indeed, over roughly a decade, the number of NAACP branches grew by nearly 700% in the South and just over 400% in the non-South. According to the 1940 Census, roughly 70% of the Black population in the United States lived in the South in 1940. Therefore, the tremendous growth of the NAACP’s presence in the South connected a large share of the Black population to the emerging civil rights movement. By the mid-1940s, nearly 60% of Black households had an NAACP branch in their home county.
Some of the main correlates of the rapid growth in the South from the late 1930s to the mid-1940s are explored in the section titled “Determinants of Growth in the South.” The focus on the South is partly because of the dramatic expansion of NAACP branches there but also because that region was so critical for Black economic progress (Heckman, 1990). A range of county-level economic, political, and demographic variables that roughly correspond to the hypothesized factors behind the sharp increase in political activity during this era are examined. However, to be clear, causal claims about these relationships are not made. Instead, statistical associations are described to lay the groundwork for future work. Social scientists and historians have discussed many, but not all, of these factors in previous studies, and these literatures are referenced throughout.
The first set of factors considered include race-specific demographics and economic characteristics, such as population, urbanicity, marriage rates, occupational income, and education. Unsurprisingly, important positive associations are identified between NAACP branch growth and the size of the Black population, as well as urban share. There is also a strong positive correlation between NAACP activity and the education levels of Black Americans and occupational income levels of White Americans. These findings are consistent with Campante and Chor (2012a), in that areas where the skills of Black workers were raised but racial income gaps persisted are the same areas where economic frustration likely grew and political activity followed.
Another hypothesized reason for the explosive growth in NAACP branches around WWII relates to sudden increase in the demand for Black workers. The theoretical effect of a labor demand shock on political activism is ex-ante ambiguous. An increase in wages could lead to a reduction in the amount of time spent on political activity (a substitution effect). However, an increase in resources from higher wages also has an offsetting income effect that could lead to an increase in political activity. Our empirical analysis shows that changes in local wartime production spending were positively associated with branch formation, even after conditioning on other economic, political, and demographic factors.
Many historical accounts rightfully emphasize the constant threat of political violence faced by Black communities in the early 20th century, particularly aimed at civil rights organizations (e.g., Tolnay & Beck, 1995). By WWII, this threat may have moderated. Public lynchings significantly declined over the preceding decades (Seguin & Rigby, 2019), and the NAACP identifies the 1930s as the key period in which its efforts to combat lynching finally shifted public opinion against the practice. During the war, many young White men who traditionally served as the enforcers of the racial hierarchy in the South left to serve in the military. Racial tensions may have also been reduced by wartime messaging that encouraged national unity and other shifts in attitudes related to the War. Nevertheless, the threat of terror and violence against civil rights organizers clearly continued for many decades after the war. We provide empirical evidence that a reduction in the threat of violent reprisals, as captured by contemporaneous lynchings and the enlistment of White soldiers, is associated with some of the expansion of local NAACP branches during the War.
The growth in NAACP activity around WWII may also be related to preexisting local patterns of political activity. Measures such as physical proximity to the nearest NAACP branch, previous voting activity in the area, previous support for the Communist party, and previous support for the Republican party are examined. Some modest evidence suggests that counties which had a higher historical Republican party vote share experienced more rapid growth in NAACP branches, consistent with prior political activity playing some role in laying the groundwork for the later surge in NAACP growth.
Finally, some prominent researchers emphasize the importance of social networks to successful political movements. Three are considered—church attendance, the existence of Historically Black Colleges and Universities, and residential racial segregation—but only church attendance and racial segregation measures are correlated with increases in NAACP branches and only when considered individually. Together with other variables, they are no longer economically or statistically significant. However, given the importance of social factors in many models of the civil rights movement (McAdam, 1999; Morris, 1986) and other global activism (Bursztyn et al., 2021; Cantoni et al., 2019; Manacorda & Tesei, 2020), and the difficulty of capturing their power with admittedly coarse measures, these associations clearly deserve to be at the top of the list of empirical investigators.
In summary, a number of factors are associated with NAACP branch growth during the short period when the organization rapidly expanded around WWII, including demographics, human capital, labor demand, decreasing racial violence, and latent demand for political activism. Our own recent work (Aaronson et al., 2022) provides a research design for identifying the causal impact of human capital investment on NAACP growth. In our view these other important channels, often studied but not within a compelling statistical framework, should be considered as well, partly to better understand this vital era but also to gain a better understanding of political movements and their economic impact more broadly.
A Brief History of the NAACP
The NAACP was chartered in 1909 by a biracial group of prominent activists intent on challenging racial disenfranchisement and segregation.2 An earlier all-Black organization led by W. E. B. DuBois, called the Niagara Movement, was struggling financially and otherwise. After a brutal anti-Black riot in Springfield, Illinois, in 1908, DuBois and several other Niagara members formed an alliance with a handful of White Northern activists that led to the NAACP’s founding. The White members were critical for accessing financial and political networks, in sharp contrast with the experience of the Niagara Movement.
The NAACP charter, published as the organization was incorporated in 1911, underscored a commitment to using various forms of resistance and judicial activism to secure equality of rights, eliminate discrimination, and increase opportunities in schools, courts, and the labor market.3 In the early years, the most successful of these efforts involved organizing protests and boycotts, most prominently against the 1915 film Birth of a Nation, and collecting data on, monitoring, and campaigning against lynching in the South. An in-house publication, The Crisis, edited by DuBois, played a significant role in raising awareness about the NAACP’s activities, as well as reporting relevant news and culture.
Ultimately though, the organization’s principal successes were achieved through the courts. The NAACP’s legal arm, led for many years by Thurgood Marshall, backed numerous civil rights lawsuits starting as early as the mid-1910s, and eventually culminating 40 years later in the overturning of the “separate but equal” doctrine in Brown v. Board of Education. Prior to that groundbreaking ruling, the NAACP was heavily involved in breaking down racial disparities in public sector labor markets. For example, NAACP lawyers regularly represented Black employees, particularly at the postal service, in employment discrimination lawsuits. Donohue et al. (2002) highlight the impact that NAACP litigation had in forcing local school districts to equalize Black and White teacher pay, thereby helping raise Black teachers’ salaries and boosting an array of measures of access to school quality.
These activities were primarily centralized at New York City headquarters and the law offices in Washington, DC. But there was also a recognition that local grassroots activism was critical. Therefore, plans were made early on to establish branches throughout the country. These local branches also served as the principal means to raise funds for the national office, and especially its legal and legislative teams. Half of the annual $1 per person membership dues collected by local branches were sent to the national office, with the remaining half used to fund local operations and organize events and activities. It is this local activity that is described next in order to gauge temporal and spatial variation in political participation during the early years of the civil rights movement.
Growth of the NAACP, 1919–1956
The first major push to construct a nationwide network of local branches occurred during World War I (WWI), when the NAACP expanded its reach beyond Northeastern and Great Lakes cities to the South and West. Taking advantage of the energy created as Black soldiers returned from the war, the number of NAACP branches grew from 80 in 1917 to over 300 in 1919, with the expansion encompassing over 90% of the Southern branches active at the end of 1919. Membership increased tenfold to over 90,000 during this drive.4
Data on branch activity from 1919 to 1956 comes from administrative files of the NAACP’s national office, which are held at the Library of Congress. Information from 1919 to 1938 was taken from internal reports that record location and annual membership at the branch level. Because the membership and dues data are missing for portions of the sample period, and inevitably include measurement error, the focus is on the presence of a branch in a county. After 1938, public annual reports on local branches were used for years that were available in the archives: 1942, 1946, and most years from 1950 to 1956. The archived records of the NAACP Directors of Branches provided insight into the context of branch activities. These individuals were responsible for corresponding with the local branches, facilitating interaction between local branches and national offices, and traveling to local areas to aid membership drives and revive lapsed branches.
Figure 1A plots the number of branches, separately for the South and non-South regions, between 1919 and 1956. Although the NAACP was the main civil rights organization throughout this period, its local reach nevertheless declined after WWI and was stagnant for much of the 1920s and 1930s. That was especially the case in the South, where branches typically declined year-to-year for over a decade and remained below 1919 levels through 1938. There was little growth outside the South either. As a result, the fraction of Black people living in a Southern county with an NAACP branch fell from around 40% in 1919 to under 30% by 1930 (Figure 1B); again, the decline in exposure was especially notable in the South.5 Figure 2A–C further emphasizes the interwar decline in local participation through a series of county-level maps of the number of NAACP branches per 100,000 Black residents for selected years.
Although new branches continued to open every year, entry (green shading in Figure 3, panel A for the South, panel B for the non-South) was offset by a significant amount of exit (red shading). The blue area, representing net branch growth, highlights the reality that the primary activity of an NAACP branch prior to World War II (WWII), and local civil rights activism more generally, was the fight to exist. For example, only 15% of counties that opened a branch in 1919 kept a branch open for a decade (Figure 4).6 Those that opened by the end of the 1920s were more likely to last but still faced, on average, a 10-year survival rate of under 40%. The threat of political violence, especially in the South, limited the scope of local action and the onset of the Great Depression restricted resources. Instead, local activity, as described by the Director of Branches, tended to concentrate on activities from the early days—that is, fundraising and monitoring lynching threats and occurrences, as well as other violence against Black people. Reports of violence did serve to identify cases of interest for the legal defense arm. High-profile legal cases were the primary successes of this period and indeed their pace increased with the rise of the Roosevelt court in the late 1930s.
The turning point was the late 1930s. At the national level, the Roosevelt Administration called upon the leadership of the NAACP to support its mobilization policies, granting significant concessions to prevent a national strike. Legal victories, powerful in principle but constrained by the realities of enforcement, began to pile up. The growing power of the national office was matched by a surge in local support reflected in Figures 1–4 by the sharp increase in new branches throughout the country and a significant upswing in survival of incumbent branches. In the South, the number of branches expanded by more than sixfold in less than a decade, and the share of Black residents in a county with a branch more than doubled. Outside the South, branches also became more ubiquitous. Newspaper references to the organization skyrocketed (Figure 5). This period of extraordinary growth, covering the late 1930s to the late 1940s, transformed participation in the U.S. civil rights movement and therefore is the focus of the section “Determinants of Growth in the South.”
The activities of local branches changed a great deal during this time as well. Ella Baker, a legendary figure in the civil rights movement, took over as traveling secretary in 1938 and Director of Branches in 1943. Baker’s letters describe her unfailing belief in the importance of grassroots support for the organization, which she did not find reciprocated by national leadership. She describes the local branches’ demand for action on school quality, desegregation of public facilities, and jobs, harkening back to the original charter but with less focus on political resistance as in the early years. Her letters convey her conviction that growth in the early 1940s reflected a surge in demand for legal and economic justice that the NAACP could help promote, rather than a shift in the strategy of the national office.
Growth continued through the organization’s crowning achievement, Brown v. Board of Education, in 1954. At that point, nearly 70% of Black Americans lived in a county with an NAACP branch, including roughly 90% of Black residents of urban counties and 60% in rural counties. Around this time, several other historically important organizations formed, including the Congress of Racial Equality and Martin Luther King Jr.’s Southern Leadership Conference, which tended away from judicial activism and more toward organizing and mass protests. Naturally, this led to some dwindling in the popularity and influence of the NAACP. However, even as the civil rights leadership began to disperse, nearly all civil rights activism remained tethered to leaders involved with the NAACP in some way. Indeed, according to interviews with activists, direct action on civil rights activism still required consent of the NAACP into the late 1950s (Morris, 1986).
Determinants of Growth in the South, 1938–1946
The exponential growth in the South that began in the late 1930s and lasted for roughly a decade represented a profound transformation of local representation in national civil rights activism. But little is known analytically about the determinants of this growth. To dig further, a series of reduced form regressions of the form Yi = α + βXi + εi are presented. The variable to be explained, Yi, is defined as whether a Southern county that did not have a NAACP branch prior to 1938 picked one up by 1946. If so, Yi is coded as 1, and if not, it is coded as 0. The sample is restricted to the 1,070 counties in the South that did not have a branch before 1938. The matrix Xi reflects a set of covariates that may help statistically explain whether a county opened a branch between 1938 and 1946. These covariates are included in the regressions simultaneously in order to gauge the extent to which a single variable (say, the average years of education among a county’s adult Black population) is associated with NAACP branch growth, after accounting for the effects of other variables (say, the size of the county’s Black population). In that sense, these regressions can be thought as of as a “horse race.” To make the race transparent, each covariate is transformed to have mean 0 and standard deviation 1 so that the relative economic importance of each covariate is reflected by the size of β.
It is critical to emphasize that these results should be interpreted as partial correlations between NAACP growth and the indicated variables, not causal relationships. Although many of the variables are closely related to hypothesized causal factors from the literature, assessing and purging the estimates of endogeneity would require research designs that move well beyond the regressions’ descriptive scope. Moreover, disentangling the relative roles of the various causal factors presents an even greater challenge. That said, these results may prove useful for assessing promising directions for future research.
Table 1 reports results in groups of related covariates organized by potential explanations for the NAACP’s take-off during the late 1930s and 1940s: (a) demographic and economic factors (including population characteristics, education and income), (b) wartime production, (c) threats of violence, (d) prior political activism, and (e) social interactions and networks. The final column (5) of the table runs a modified version of a “kitchen-sink” regression using the least absolute shrinkage and selection operator (LASSO) procedure. The LASSO model is a statistical technique that drops variables with little predictive power as a way of preventing overfitting. Therefore, what remains are variables selected by the model because they are important predictors, though still not necessarily causal, of local NAACP branch growth.7
Table 1. Correlates of NAACP Branch Formation in the South, 1938–1946
Log White Population, 1940
Log Black Population, 1940
Log Other Population, 1940
Marriage, Black (%)
Marriage, White (%)
Log Income Black, 1940
Log Income White, 1940
Education Black, 1940
Education White, 1940
WW2 Mil. Spending
Threats of Violence
WWII Enlistment, Wh. (%)
WWII Enlistment, Bl. (%)
Prior Political Activism
Log(Distance to Closest)
Vote/Vot. Age Pop., 1940
D(Comm. Above Average)
Republican 1868-1916 (%)
Social Interact. & Networks
Religious Membership (%)
Hist. Black Church (%)
Note: Authors’ calculations using data on branches collected from NAACP’s records stored at the Library of Congress. Further information about the covariates are provided in the text. Columns (1) to (4) are from Ordinary Least Squares (OLS) regressions; column (5) is from a least absolute shrinkage and selection operator (LASSO) regression that initially included all the covariates in Table 1. *, **, and *** = statistically significant at the 10, 5, and 1 percent level. NAACP = National Association for the Advancement of Colored People.
Demographic and Economic Forces: 1940 Population, Education, and Income
The first set of determinants are the 1940 characteristics of the local population. These characteristics likely reflect a combination of very long-run features of counties and more recent developments over the preceding decades, such as the rise in the level of education of Black individuals. A key point is that these features can be thought of as underlying latent forces for political change. While they were in place for a considerable period, it was only the advent of World War II (WWII) that catalyzed them, thereby enabling them to play a role in the expansion of the NAACP. This section discusses three subcategories in turn: demographics, education, and income.
One obvious factor that may play an important role in the growth of the NAACP is simply the sheer size of the population of a county and its racial composition. A reasonable hypothesis is that areas with larger Black populations would likely demand greater civil rights. Column 1 of Table 1 shows that a 1 standard deviation increase in the log of the 1940 Black population was associated with a 0.16 increase in the likelihood of having an NAACP branch. Even with a full set of controls (column 5), a 1 standard deviation increase in the log of the 1940 Black population was associated with a 0.17 increase in the probability of having an NAACP branch, the largest coefficient among all covariates considered. There is also an indication that a more diverse population, as reflected in the log of population that is neither White nor Black (“Other”), is also positively correlated with greater NAACP activity and is statistically significant.8 There is no statistically significant association with a county having a higher White population.
Perhaps not surprisingly, since NAACP branches were largely formed in cities and towns, there is a strong positive and statistically significant association between the share of the population in a county that is urban and NAACP growth during this period even when the full set of controls is included. Sundstrom (2007) highlights that racial discrimination in Southern wages was higher in more urban areas, which could be an explanation for higher demand for NAACP activity in these areas. Additionally, NAACP membership dues of $1 per year may have constrained membership in more rural areas. Finally, Black individuals may have felt safer to be politically active in urban areas. In contrast, family structure as captured by the share of the adult population that is married has no association with NAACP growth.
A vast literature in political science has hypothesized that education should play a prominent role in political activities. Bömmel and Heineck (2020) point to several theories. First, education may increase civic engagement by helping individuals develop civic skills and knowledge, which lowers the cost of participation and allows them to pursue effective strategies for engagement. Second, more schooling could also improve the ability to gather and process politically relevant information. Third, the participation of the well-educated in social networks can potentially reinforce individual-level effects.
There is a very large empirical literature that attempts to estimate whether there is in fact a causal effect of education on measures of political activity. One important challenge in these studies is the endogeneity bias that can arise when other factors such as personality traits lead individuals to both obtain more education and be more civic minded. More recent studies that employ either quasi-experimental methods or other advanced statistical techniques to address this concern appear to find mixed results.9 It is also possible that the effects may vary in different contexts as education may affect several outcomes simultaneously with ambiguous results on political activity.10 Therefore, it is somewhat of an open question as to whether any positive correlation between county education levels and NAACP growth are truly causal.
Campante and Chor (2012a) also posit a theory that may be relevant to the growth of the NAACP and the possible role of schooling of Black Americans. They argue that it is important to consider the opportunity cost of engaging in political, rather than personally productive, activities. In contexts where there is a high payoff to education, the link between education and political activity may be more muted than in situations where there is a lower payoff to receiving more schooling. Given that there was enormous discrimination against Black workers, we may anticipate that those areas with high levels of education among the Black population, and high levels of labor market discrimination, may have been areas that experienced the most civil rights activity.
Two studies echo the idea that discrimination prevented greater income growth despite increased human capital among Black Americans in the South in the early to mid-20th century. Mohammed and Mohnen (2021) find that improvements in education among Black men did not improve their occupational standing and that this was related to specific occupations where they tended to be excluded. Bhalotra and Venkataramani (2015) similarly find that early life exposure to sulfa drugs that combatted bacterial infections had large long-run effects on the labor market outcomes of Whites but not Blacks and explain this difference as due to labor market discrimination.
Turning back to Table 1, there is a very strong positive association between the average level of education of Black adults (between the ages of 20 and 50 years) in a county and NAACP growth that is robust to the inclusion of the full set of covariates. A 1 standard deviation increase in years of education is associated with a 0.14 increase in the probability of gaining an NAACP branch, the second largest coefficient obtained in column 5. The coefficient on White adult education levels is also positive but smaller and statistically insignificant. Related work in progress studies whether the Rosenwald schools built largely during the 1920s may have contributed to the later increase in NAACP activity during WWII, as many cohorts of students who attended these schools entered the prime of their work life during the late 1930s and 1940s (Aaronson et al., 2022). Preliminary findings suggest that the expansion in schooling for Black Americans that took place in the South for cohorts born after World War I (WWI) may have contributed to the increase in NAACP activity that is also picked up in Table 1. Further, there is an important interaction effect between the level of exposure to Rosenwald schools in a county and the degree of labor market discrimination, as proxied by occupational income gaps between White and Black workers, on NAACP activity. These results are consistent with Campante and Chor (2012a), and suggest that it is precisely in the areas where the skills of Black workers were raised but their labor market aspirations went unfulfilled, that experienced greater NAACP activity.
Another possibly important and related factor in explaining increased political activity is income. A simple theory might suggest that Black people with higher incomes would be more likely to engage in political activity as they would have greater means and lower costs of participation. In that case, one might expect a positive association between the mean income levels of Black workers and NAACP growth. On the other hand, Campante and Chor’s (2012a) model would suggest that this association may not hold as higher-income Black Americans would face a higher opportunity cost of engaging in political activity. In fact, here, there is no correlation between the average income levels of Black individuals in a county and growth in NAACP activity.11 However, there is a strong positive association between the income levels of White individuals and gaining an NAACP branch between 1938 and 1946. This also fits the Campante and Chor model, as the higher income levels of White workers may proxy for greater discrimination against Black workers (Mohammed & Mohnen, 2021).
Wartime Production and Labor Demand
WWII generated massive changes in the U.S. economy, shifting labor supply and technology to such a degree that it fueled years of postwar growth (Acemoglu et al., 2004; Gross & Sampat, 2020). These changes triggered a massive decrease in wage inequality, referred to as the Great Compression (Goldin & Margo, 1992), including dramatic declines in top income and wealth inequality (Kopczuk & Saez, 2004; Kopczuk et al., 2010).
The Great Compression was also associated with a convergence in Black-White income, as described in Smith and Welch’s (1989) seminal review of midcentury Black progress. Subsequent work has disentangled race-specific factors from aggregate forces. For example, Maloney (1994) and Margo (1995) argue that Black gains in the 1940s arose due to a combination of broader trends embodied in the Great Compression and race-specific factors such as Black-specific occupational and migration flows and increases in Black education. More recent work has attempted to distinguish between progress during and after the War. For example, using retrospective work histories drawn from workers in six cities outside of the South, Collins (2000) finds that Black Americans’ movement into higher-skilled occupations and war-related industries drove a significant share of progress during this era. The persistence of wartime changes in the labor market is supported by evidence on women’s labor supply as well (Acemoglu et al., 2004; Goldin & Olivetti, 2013).
While overall wage convergence drove Black economic progress during and after WWII, evidence is mixed on how much these effects extended to the South. Collins (2001) looks more closely at the movement of Black workers into manufacturing, where wartime regulations may have reduced discrimination. Consistent with this hypothesis, Black employment gains during the War outside the South correlate with anti-discrimination actions taken by the Federal Employment Practice Committee (FEPC), which handled complaints against federal contractors. However, these effects are not found in the South, consistent with the lack of cooperation of local officials with the FEPC. For example, Jaworski (2017) finds little evidence that wartime investment drove industrial development in the South, and further, that Black populations declined in areas with more wartime investment, consistent with Black workers being denied access to these high-paying jobs. Moreover, unions are thought to have played an important role in driving the Great Compression (Collins & Niemesh, 2019), and the South’s hostility to unions kept membership notably low in this region (Farber et al., 2021).
However, in contrast to these studies, Aizer et al. (2022) find persistent gains in economic outcomes of Black Americans around the war, including in the South. They theorize that manpower shortages drove large changes in Black employment, and eventually, reduced discrimination in hiring. Consistent with the manpower-shortage theory, Ferrara (2022) attributes over a quarter of Black progress in the South in the 1940s to the deaths of White soldiers during the war. Thus, the war likely caused convergence in Black-White incomes in the South despite little enforcement of anti-discrimination regulations and minimal progress in civil rights.
Regardless of the war’s impact on Black economic gains, the theoretical effect of a labor demand shock on political activism is ex-ante ambiguous. An increase in the wage will increase the opportunity cost of activism, which some models highlight as an important determinant of mass action (Campante & Chor, 2012a, 2012b). Alternatively, labor demand shocks may affect the resources available to organize and increase the incentive to fight for jobs, particularly when a population faces discriminatory barriers to desirable occupations. An important distinction, particularly given our data, is whether changes in labor demand occur locally or nationally, and hence work through migration. A long tradition in economics emphasizes the importance of the threat to migration as a key element of local Black communities’ ability to extract public goods (see, e.g., Hornbeck and Naidu (2014) for a fascinating example of the importance of Black outmigration).
In the Wartime Production section of Table 1, we report the association between local wartime spending and NAACP branch formation from 1938 to 1946. With no other conditioning variables, a 1 standard deviation increase in war-induced spending is associated with an increase in the likelihood of branch formation of nearly 0.11. This estimate is perhaps implausibly large and may reflect the fact that other key covariates such as the urban share of the population are not included. Wartime spending, however, is selected by the LASSO model used in column 5, indicating that it remains an important covariate even after accounting for other covariates such as local demographic and economic characteristics. The coefficient of 0.03 is considerably smaller than the raw correlation and marginally statistically different from zero. Nevertheless, changes in labor demand during the war were positively associated with branch formation, with a similar economic magnitude to some of the other important nondemographic determinants.
Threats of Violence
A third set of determinants relates to threats of violence faced by the local Black community (Tolnay & Beck, 1995). WWII may have opened the door to civil rights political activism by removing the threat of violent reprisals from local White communities, at a time when Black soldiers’ resolve to fight discriminatory behavior upon returning from the war had increased.12 The decline in violence could have happened for at least two reasons. First, the war may have removed the young White men who traditionally served as the violent enforcers of the racial hierarchy in the South. A second possibility is that war-inspired patriotism encouraged national unity that reduced racial tensions independent of the presence of White youth.
A number of papers have emphasized the extent to which White violence against Blacks has persistently impacted Black political participation. Most of this work uses lynchings, the most egregious and public-facing violence used to perpetuate segregation, as a marker for local racial conflict. Cook et al. (2018), who provide a thorough review of the social science literature, document a strong correlation between lynchings and segregation, conditional on local demographics. Jones et al. (2017) show that lynchings in the post-Reconstruction South reduced local Black turnout in subsequent elections. This effect was long-lasting, with detectable effects on the registration and voting of Black Americans decades later (Williams, 2022).
Table 1’s subsection labelled Threats to Violence reports the partial correlation between a Southern county’s likelihood of gaining an NAACP branch from 1938 to 1946 and local lynchings. While lynchings predict NAACP branch formation—as would be the case if the war removed the perpetrators of this violence—the results are insignificant with no demographic or economic covariates included. Moreover, the full set of partial correlations in column 5 shows that lynching enters with a negative sign. A negative effect of lynching is consistent with a continued threat of White violence during this period; it is possible that the war reduced the threat of violence, but the negative coefficient suggests that an easing of the threat of violence was likely not the primary cause of Southern NAACP branch growth during the war.
The second two rows of the Threats to Violence subsection report the correlation between White and Black enlistment with NAACP branch formation. White enlistment is robustly correlated with branch formation, consistent with the Threats of Violence hypothesis. However, this evidence should be interpreted with caution, as the relationship may also be indicative of a change in local labor supply. Ferrara (2022) shows that high rates of White fatalities during the war improved economic outcomes for Black Americans through the creation of new employment opportunities and resulting skill-upgrading. Black enlistment shows the opposite pattern, with higher rates of enlistment correlated with lower levels of branch growth. These results suggest that the labor supply story may not explain the coefficient on White enlistment, as one may expect the enlistment of Black workers to have more direct labor supply effects than the enlistment of White workers. When all regressors are included, Black enlistment is dropped from the LASSO specification, while lynching and White enlistment are retained. This provides further evidence that a reduction in the threat of violent reprisals during the war is associated with the contemporaneous expansion of local branches.
Prior Political Activism
A simple explanation of the WWII-era NAACP growth is that it consolidated preexisting demand for activism. Although the story of the civil rights movement often begins with the Birmingham bus boycott or Brown v. Board of Education, the struggle for civil rights began before Emancipation, and the history of slavery is replete with examples of resistance and rebellion. If the growth around WWII consolidated long-running albeit constrained demand for activism, one might expect NAACP branch formation to be correlated with previous political episodes of Black activism in local areas.
The economics literature on pre-WWII civil rights activism is necessarily limited, as most efforts to promote civil rights in the South were quashed. For example, one way to organize was through unions, but unions were massively underrepresented in the South (Farber et al., 2021), and White hostility to unionization of low-wage Black workers likely played a significant role (Friedman, 2000). Related to this, in order to keep access to cheap workers, landowners did whatever they could to resist Black outmigration, likely further complicating organizing.
The “Prior Political Activism” section of Table 1 reports four correlations, each reflecting a different era of political progress or retrenchment. The first coefficient, the logarithm of the distance to the nearest NAACP branch, is informative about the geographic interrelationships between branches. The negative coefficient of −0.039 can be interpreted to mean that an NAACP branch is more likely to appear in counties that are closer to another county with a branch, consistent with a model of diffusion and spread through exposure, rather than one in which branches satisfy local demand and suppress further nearby growth. One should not overinterpret this association, however, as the distance to branch coefficient is dropped in the LASSO model, suggesting that distance to the closest branch is a poor predictor of growth once other area characteristics are considered.
The second row of the “Prior Political Activism” section reports the correlation between voter turnout in 1940 and NAACP branch formation. If voter turnout reflects political engagement, then one might expect areas of NAACP growth to have high voter turnout. Alternatively, turnout could be low due to suppression of Black voters. In the simple models, voting is negatively associated with the formation of NAACP branches, consistent with the suppression story. However, voter turnout loses much of its predictive power in the LASSO model, consistent with population and other area characteristics explaining most of that association.
The third row examines the relationship between NAACP branch formation between 1938 and 1946 and votes for the Communist party in the late 1920s and 1930s. Communist party support in a county could potentially capture higher latent demand for civil rights activity. In particular, this exercise uses an indicator that a Communist party presidential candidate in a Southern county ever received a larger vote share than the national average for that Communist party candidate. Historians trace a substantial portion of classical civil rights activism to intellectual forebearers in the 1930s, especially Communist activists in the labor movement (Hall, 2005; Sullivan, 2009). For example, the Communist party represented the nine Black youth falsely accused of rape in the Scottsboro Trial in 1931 and 1932. In 1932, The Crisis sponsored a symposium with newspaper editors, finding largely positive assessments of the Communist party (Lift Every Voice, p. 154). However, by the early 1940s, the party had lost considerable credibility and many Black workers left. Therefore, one potential story is that civil rights activists switched their allegiance to the NAACP once the Communist party fell into disrepute. The regression analysis finds small, positive, and statistically insignificant associations between Communist party vote shares and NAACP branch formation. The LASSO model retains this variable, but evidence for its cross-sectional economic importance remains lacking.
Finally, the fourth row of the “Prior Political Activism” section examines the correlation between NAACP branch formation and Republican vote shares in the pre-NAACP era. The Republican party championed Reconstruction-era policies that targeted assistance to freed slaves, such as the Freedman’s Savings Bank (Stein & Yannelis, 2020), and was the political party more likely to support civil rights between the Civil War and WWI. There is consistent but modest evidence that a history of support for the Republican party predicts Southern NAACP branch growth between 1938 and 1946. The correlation is of similar magnitude as that for WWII military spending and measures of violence. Therefore, there appears to be some evidence of an economically relevant association between growth in grassroots political participation and latent demand for civil rights activism, which the NAACP managed to consolidate in the 1938–1946 era.
Social Interactions and Networks
A growing literature in economics and sociology explores the role of social interactions and the formation of social networks in political movements, starting with the seminal work of Morris (1986) and McAdam (1999) on Black resistance to discrimination (see Kitts, 2000 for a good review) and broadening to other global activism (e.g., Bursztyn et al., 2021; Cantoni et al., 2019; Manacorda & Tesei, 2020). Of relevance for the civil rights era, economists have also recently explored the importance of social ties among Southern Black Americans post-emancipation (Chay & Munshi, 2015), including a series of papers on the importance of connectiveness in determining movement during the 20th century’s Great Migration (Black et al., 2015; Derenoncourt, 2022; Stuart & Taylor, 2021).
Historical accounts of social interactions and networks during the early civil rights era focus on the role of Black churches (Morris, 1986, 2014; McAdam, 1999). Indeed, McAdam highlights the role of the Black church in organizing civil rights activities as among the most important evidence of what sociologists call the political process model, or the idea that opportunities for change must be present in order for a political movement to move forward. In other settings, researchers have found that church attendance has a positive effect on political participation (Gerber et al., 2016). Despite its prominent role in U.S. history, quantitative work on Black churches in a causal framework is scarce. A rare exception is Chay and Munshi (2015), which argues that crop and farming patterns caused higher levels of social connectedness in areas that farmed cotton, tobacco, rice, and sugarcane, as compared to areas in which Black farmers farmed wheat and corn. In turn, more connected communities led to stronger Black churches and more political participation.
The regression analysis under the subsection “Social Interactions and Networks” of Table 1 examines how Black institutions may have impacted NAACP branch formation. It begins by looking at religious organizations. In the first row, religiosity is represented by the share of the local population that attended church in 1936. There is a positive association between church attendance and NAACP branch formation; however, it is dropped from the LASSO model. In the second row, attendance is refined to historically Black churches but it also has no predictive power. While the lack of an association between church attendance and NAACP growth was surprising, it is consistent with the observation in McAdam that Black churches were rarely engaged in civil rights activism before 1930, and that urban Black churches in the South only became a force in the civil rights movement in the 1940s, simultaneous to the rise of the NAACP. Therefore, perhaps these two institutions were distinct factors.
The third row reports the correlation between the location of Historically Black Colleges and Universities (HBCUs) and NAACP branch formation. While HBCUs enrolled at most a small share of the Black population, a disproportionate share of the leadership of the NAACP and other civil rights organizations was college educated, and students of HBCUs played important roles in the direct action of the mature civil rights movement (Morris, 1986, 2014). It is, therefore, somewhat surprising that HBCUs enter with no association in column 4 and a negative correlation in the LASSO model in column 5. One possible explanation is the relatively limited number of HBCUs and the struggling finances of these colleges at the time. A more prosaic possibility is that the 1940 education controls dominate the HBCU variable as a measure of Black education, leaving behind residual variation that captures nonlinearity or other second-order relationships between education and activism.
The final row of the panel reports the association between local residential segregation in 1940 and NAACP branch formation. Residential segregation is both a symptom and cause of racial conflict and could reflect latent demand for civil rights activity. The effects of residential segregation on long-run outcomes are the subject of active research (e.g., Boustan, 2010; Derenoncourt, 2022; Massey & Denton, 2019). Less is known about the direct connection between residential segregation and political activism. Column 4 shows that residential segregation strongly predicts NAACP branch formation in this era, with more segregated counties gaining a branch. However, the apparent association with segregation disappears in the LASSO model, suggesting that segregation may proxy for the share of the local Black population or other confounders. This finding is consistent with Cook et al. (2023), which finds that the areas in which Green Book guides identified nondiscriminatory lodging and service providers for travelers were only weakly associated with NAACP activities.
A newly digitized panel of NAACP branches is used to describe the potential factors underlying the expansion of political participation in the American South from the late 1930s through the mid-1940s. This period has long been recognized for its significant progress in reducing racial gaps in labor market outcomes, but little work in economics has considered the role of political participation in shaping that progress. Associative evidence suggests that a few potential channels could be worthy of future study, including the role of demographics, human capital, labor demand, violence, and latent political activism, all of which have been highlighted in a variety of history and social science literatures. However, careful causal work does not currently exist on most of these factors. Filling in this hole would be important, not only to provide compelling evidence on the origins of the 20th century’s most important U.S. political movement but also to add to a growing but still nascent literature examining the role that grassroots activism has played on economic growth and income inequality around the world.
The views expressed here do not necessarily represent the views of the Federal Reserve Bank of Chicago or the Federal Reserve System.
- Cascio, E., & Lewis, B. (2022). Legal activism, state policy, and racial inequality in teacher salaries and educational attainment in the mid-century American South. NBER Working Paper 30631.
- Daugherity, B. J. (2016). Keep on keeping on: The NAACP and the implementation of Brown V. Board of education in Virginia. University of Virginia Press.
- Francis, M. M. (2014). Civil rights and the making of the modern American state. Cambridge University Press.
- Aaronson, D., Borgschulte, M., & Mazumder, B. (2022). Schooling and political activism in the early civil rights era (Working paper). Federal Reserve Bank of Chicago.
- Aaronson, D., & Mazumder, B. (2011). The effect of Rosenwald schools on Black achievement. Journal of Political Economy, 119(5), 821–888.
- Aaronson, D., Mazumder, B., Sanders, S., & Taylor, E. (2021). Estimating the effect of school quality on mortality in the presence of migration: Evidence from the South. Journal of Labor Economics, 39(2), 527–558.
- Acemoglu, D., Autor, D., & Lyle, D. (2004). Women, war, and wages: The effect of female labor supply on the wage structure at midcentury. Journal of Political Economy, 112(3), 497–551.
- Aizer, A., Boone, R., Lleras-Muney, A., & Vogel, J. (2022). Discrimination and racial disparities in labor market outcomes: Evidence from WWII (Working paper #27689). National Bureau of Economic Research.
- Almond, D., Chay, K., & Greenstone, M. (2006). Civil rights, the war on poverty, and Black-White convergence in infant mortality in the rural South and Mississippi (Working paper). MIT.
- Berinsky, A., & Lenz, G. (2010). Education and political participation: Exploring the causal link. Political Behavior, 33(3), 357–373.
- Bhalotra, S., & Venkataramani, A. (2015), Shadows of the captain of the men of death: Early life health interventions, human capital investments, and institutions (Working paper). SSRN.
- Black, D., Sanders, S., Taylor, E., & Taylor, L. (2015). The impact of the great migration on mortality of African Americans: Evidence from the Deep South. American Economic Review, 105(2), 477–503.
- Bloom, N., Bunn P., Chen, S., Mizen, P., Smietanka, P., & Thwaites, G. (2019). The impact of Brexit on UK firms (Working paper #26218). National Bureau of Economic Research.
- Bömmel, N., & Heineck, G. (2020). Revisiting the causal effect of education on political participation and interest (Working paper #13954). IZA.
- Boustan, L. P. (2010). Was postwar suburbanization “White flight”? Evidence from the Black migration. The Quarterly Journal of Economics, 125(1), 417–443.
- Bursztyn, L., Cantoni, D., Yang, D., Yuchtman, N., & Zhang, Y. J. (2021). Persistent political engagement: Social interactions and the dynamics of protest movements. American Economic Review: Insights, 3(2), 233–250.
- Campante, F., & Chor, D. (2012a). Schooling, political participation, and the economy. Review of Economics and Statistics, 94(4), 841–859.
- Campante, F., & Chor, D. (2012b). Why was the Arab world poised for revolution? Schooling, economic opportunities, and the Arab spring. Journal of Economic Perspectives, 26(2), 167–188.
- Cantoni, D., Yang, D. Y., Yuchtman, N., & Zhang, Y. J. (2019). Protests as strategic games: Experimental evidence from Hong Kong's antiauthoritarian movement. Quarterly Journal of Economics, 134(2), 1021–1077.
- Chay, K., & Munshi, K. (2015). Black networks after emancipation: Evidence from reconstruction and the Great Migration (Working paper). Yale University.
- Collins, W. (2000). African-American economic mobility in the 1940s: A portrait from the Palmer survey. The Journal of Economic History, 60(3), 756–781.
- Collins, W. (2001). Race, Roosevelt, and wartime production: Fair employment in World War II labor markets. American Economic Review, 91(1), 272–286.
- Collins, W., & Niemesh, G. (2019). Unions and the great compression of wage inequality in the United States at mid-century: Evidence from local labor markets. Economic History Review, 72(2), 691–715.
- Cook, L., Jones, M., Rosé, D., & Logan, T. (2023). The Green Books and the geography of segregation in public accommodations. Quarterly Journal of Economics,138(1), 37–102.
- Cook, L., Logan, T., & Parman, J. (2018). Racial segregation and Southern lynching. Social Science History, 42(4), 635–675.
- Dee, T. (2004). Are there civic returns to education? Journal of Public Economics, 88(9–10), 1697–1720.
- Derenoncourt, E. (2022). Can you move to opportunity? Evidence from the Great Migration. American Economic Review, 112(2), 369–408.
- Donohue, J., & Heckman, J. (1991). Continuous versus episodic change: The impact of civil rights policy on the economic status of Blacks. Journal of Economic Literature, 29(4), 1603–1644.
- Donohue, J., Heckman, J., & Todd, P. (2002). The schooling of Southern Blacks: The roles of legal activism and private philanthropy, 1910–1960. Quarterly Journal of Economics, 117(1), 225–268.
- Farber, H., Herbst, D., Kuziemko, I., & Naidu, S. (2021). Unions and inequality over the twentieth century: New evidence from survey data. The Quarterly Journal of Economics, 136(3), 1325–1385.
- Ferrara, A. (2022). World War II and Black economic progress. Journal of Labor Economics,404(4), 1053–1091.
- Friedman, G. (2000). The political economy of early Southern unionism: Race, politics, and labor in the South, 1880–1953. The Journal of Economic History, 60(2), 384–413.
- Gerber, A., Gruber, J., & Hungerman, D. (2016). Does church attendance cause people to vote? Using blue laws’ repeal to estimate the effect of religiosity on voter turnout. British Journal of Political Science, 46(3), 481–500.
- Goldin, C., & Margo, R. (1992). The great compression: The wage structure in the United States at mid-century. The Quarterly Journal of Economics, 107(1), 1–34.
- Goldin, C., & Olivetti, C. (2013). Shocking labor supply: A reassessment of the role of World War II on women’s labor supply. American Economic Review, 103(3), 257–262.
- Green, D., Aronow, P., Bergan, D., Greene, P., Paris, C., & Weinberger, B. (2011). Does knowledge of constitutional principles increase support for civil liberties? Results from a randomized field experiment. The Journal of Politics, 73(2), 463–476.
- Gross, D., & Sampat, B. (2020). Inventing the endless frontier: The effects of the World War II research effort on post-war innovation (Working paper #27325). National Bureau of Economic Research.
- Hall, J. D. (2005). The long civil rights movement and the political uses of the past. Journal of American History, 91, 1233–1263.
- Hanushek, E., Kain, J., & Rivkin, S. (2009). New evidence about Brown v. Board of Education: The complex effects of school racial composition on achievement. Journal of Labor Economics, 27(3), 349–383.
- Heckman, J. (1990). The central role of the South in accounting for the economic progress of Black Americans. The American Economic Review, 80(2), 242–246.
- Heckman, J., & Payner, B. (1989). Determining the impact of federal anti-discrimination policy on the economic status of Blacks: A study of South Carolina. American Economic Review, 79(1), 138–177.
- Henderson, J., & Chatfield, S. (2011). Who matches? Propensity scores and bias in the causal effects of education on participation. The Journal of Politics, 73(3), 646–658.
- Hornbeck, R., & Naidu, S. (2014). When the levee breaks: Black migration and economic development in the American South. American Economic Review, 104(3), 963–990.
- Jaworski, T. (2017). World War II and the industrialization of the American South. The Journal of Economic History, 77(4), 1048–1082.
- Jones, D. B., Troesken, W., & Walsh, R. (2017). Political participation in a violent society: The impact of lynching on voter turnout in the post-reconstruction South. Journal of Development Economics, 129, 29–46.
- Kam, C., & Palmer, C. (2011). Rejoinder: Reinvestigating the causal relationship between higher education and political participation. The Journal of Politics, 73(3), 659–663.
- Kellogg, C. F. (1973). NAACP: A history of the National Association for the Advancement of Colored People, volume I, 1909–1920. Johns Hopkins Press.
- Kitts, J. (2000). Mobilizing in Black boxes: Social networks and participation in social movement organizations. Mobilization: An International Quarterly, 5(2), 241–257.
- Kopczuk, W., & Saez, E. (2004). Top wealth shares in the United States, 1916–2000: Evidence from estate tax returns. National Tax Journal, 57(2), 445–487.
- Kopczuk, W., Saez, E., & Song, J. (2010). Earnings inequality and mobility in the United States: Evidence from Social Security data since 1937. Quarterly Journal of Economics, 125(1), 91–128.
- Madestam, A., Shoag, D., Veuger, S., & Yanagizawa-Drott, D. (2013). Do protests matter? Evidence from the Tea Party movement. Quarterly Journal of Economics, 128(4), 1633–1685.
- Maloney, T. (1994). Wage compression and wage inequality between Black and White males in the United States, 1940–1960. The Journal of Economic History, 54(2), 358–381.
- Manacorda, M., & Tesei, A. (2020). Liberation technology: Mobile phones and political mobilization in Africa. Econometrica, 88(2), 533–567.
- Margo, R. (1995). Explaining Black-White wage convergence, 1940–1950. ILR Review, 48(3), 470–481.
- Massey, D., & Denton, N. (2019). American apartheid: Segregation and the making of the underclass. Routledge.
- McAdam, D. (1999). Political process and the development of Black insurgency, 1930–1970. University of Chicago Press.
- Meier, A., & Bracey, J. H. (1993). The NAACP as a reform movement, 1909–1965: “To reach the conscience of America.” The Journal of Southern History, 59(1), 3–30.
- Milligan, K., Moretti, E., & Oreopoulos, P. (2004). Does education improve citizenship? Evidence from the United States and the United Kingdom. Journal of Public Economics, 88(9–10), 1667–1695.
- Mohammed, S., & Mohnen, P. (2021). Black economic progress in the Jim Crow South: Evidence from Rosenwald schools (Working paper). Northeastern University.
- Mormino, G. (1994). GI Joe meets Jim Crow: Racial violence and reform in World War II Florida. The Florida Historical Quarterly, 73(1), 23–42.
- Morris, A. (1986). The origins of the civil rights movement. Simon & Schuster.
- Morris, A. (2014). The Black church in the civil rights movement, . In Disruptive religion: The force of faith in social-movement activism (pp. 29–46). Routledge.
- Pelkonen, P. (2012). Length of compulsory education and voter turnout: Evidence from a staged reform. Public Choice, 150(1–2), 51–75.
- Persson, M., Lindgren, K.-O., & Oskarsson, S. (2016). How does education affect adolescents’ political development? Economics of Education Review, 53, 182–193.
- Seguin, C., & Rigby, D. (2019). National crimes: A new national data set of lynchings in the United States, 1883 to 1941. Socius: Sociological Research for a Dynamic World, 5, 1–9.
- Sitkoff, H. (1971). Racial militancy and interracial violence in the Second World War. The Journal of American History, 58(3), 661–681.
- Smith, J., & Welch, F. (1989). Black economic progress after Myrdal. Journal of Economic Literature, 27(2), 519–564.
- Sondheimer, R. M., & Green, D. (2010). Using experiments to estimate the effects of education on voter turnout. American Journal of Political Science, 54(1), 174–189.
- Stein, L., & Yannelis, C. (2020). Financial inclusion, human capital, and wealth accumulation: Evidence from the freedman’s savings bank. The Review of Financial Studies, 33(11), 5333–5377.
- Stuart, B., & Taylor, E. (2021). The effect of social connectedness on crime: Evidence from the Great Migration. Review of Economics and Statistics, 103(1), 18–33.
- Sullivan, P. (2009). Lift every voice: The NAACP and the making of the civil rights movement. New Press.
- Sundstrom, W. (2007). The geography of wage discrimination in the pre-civil rights South. The Journal of Economic History, 67(2), 410–444.
- Tolnay, S., & Beck, E. M. (1995). A festival of violence: An analysis of Southern lynchings, 1882–1930. University of Illinois Press.
- Williams, J. (2022). Historical lynchings and the contemporary voting behavior of Blacks. American Economic Journal: Applied Economics, 14(3), 224–253.
- Wright, G. (1986). Old South, new South: Revolutions in the Southern economy since the Civil War. Basic Books.
- Wright, G. (1999). The civil rights revolution as economic history. The Journal of Economic History, 59(2), 267–289.
1. There are many important studies on Brown and the role of 1960s civil rights legislation, including Hanushek et al. (2009), Heckman and Payner (1989), Donohue and Heckman (1991), Wright (1999), and Almond et al. (2006).
2. See Kellogg (1973), Meier and Bracey (1993), and especially Sullivan (2009) for detailed histories of the NAACP.
3. See Founding and Early Years–NAACP: A Century in the Fight for Freedom | Exhibitions–Library of Congress (loc.gov).
4. Data on WWI is available from a concurrent effort by researchers at the University of Washington, which is available at NAACP History and Geography 1909–1980–Mapping American Social Movements (washington.edu).
5. County-level population is linearly imputed between Census years.
6. Figure 4 represents 10-year survival rates for branches, by year of entry. For example, the 1919 cohort of new branches are included in the survival line where the percent surviving is equal to 1 in 1919.
7. An economically important variable (i.e., a causal force that governs an important share of the outcome) may be dropped, or badly misestimated, in the LASSO procedure if other correlated covariates are included. For example, the LASSO model drops Black churches from our final specification. However, it is possible that other selected variables, such as Black population and education, are strongly correlated with church activities and cannot be separated in the regression given our available data.
8. This category includes American Indian/Alaska Native, Chinese, Japanese, Other Asian or Pacific Islander, and other race, not elsewhere classified.
9. For example, Dee (2004), Milligan et al. (2004), Sondheimer and Green (2010), Green et al. (2011), and Henderson and Chatfield (2011) find positive effects of education on political behavioral outcomes, whereas Berinsky and Lenz (2010), Kam and Palmer (2011), Pelkonen (2012), Persson et al. (2016), and Bömmel and Heineck (2020) largely find null effects.
10. Aaronson et al. (2021) find this to be the case in estimating the effects of education and school quality on mortality for Southern Black residents exposed to the Rosenwald schools. Since education also induced greater migration of Black people and migration leads to worse health, isolating the direct effects of education on health is challenging.
11. Income here is calculated by using the average wage and salary income in an occupation.
12. Historical accounts of racial conflict during WWII include Sitkoff (1971) and Mormino (1994).