Sustaining Democracy with Force: Black Representation During Reconstruction

In a recent Broadstreet post, Volha Charnysh discussed the reasons that scholars increasingly engage in historical political economy (HPE) work. Is it a way of understanding the past? Of understanding how we got here now? Or is it a way of using new evidence to test core theories in political economy?

In a paper that is forthcoming in an issue of Journal of Historical Political Economy on the “Slavery and Its Legacies”, Mario Chacon, Sidak Yntiso, and I provide an example that speaks to each goal. We study the importance of enforcement of the political rights of the formerly enslaved during Congressional Reconstruction (1868-1877). A key pillar of Congressional Republican Reconstruction was that suffrage would be extended to Black adult males in the defeated Confederate states, and that this would be enforced by the U.S. military.[1]

We test whether the presence of local troops predicts the election of Black politicians to Southern state legislative bodies. We do so by exploiting the unwillingness of Congress to provide a sufficient occupational force to protect the largely rural Black population, which was sparsely distributed across the vast South. Specifically, we use recently digitized information from Downs and Nesbitt (2015) on the precise locations of U.S. Army garrisons between 1868 and 1878 to predict the election of Black candidates to the mandated state constitutional conventions of 1868 and 1869 and in the state legislatures between 1868 and 1878. In this period, more than a thousand Black candidates were elected either as a delegate to these critical constitutional conventions or as a member of the state legislature.

We use the spatial and temporal variation in both the location of U.S. Army troops and the county from which Black legislators were elected to estimate the relationship between federal enforcement of political rights and the incidence of Black representation. This spatial and temporal variation in Black legislators and US Army troops is shown in Figures 1 and 2, respectively. Since members of each state’s lower house were elected on 2-year cycles (either in even or odd years), each panel is a state legislative electoral cycle. Figure 1 shows the number of Black state legislators elected in each county in each electoral cycle. Figure 2 similarly shows the location of federal Army posts during each cycle.[2]

Figure 1: Spatial and Temporal Variation in Black legislators
Figure 2: Spatial and Temporal Variation in US Army troops

Using panel data, as well as several additional strategies to address the problem of the selection on troop locations, we find strong support that local federal troop presence increased the incidence of the election of Black legislators. Of particular note, we exploit the rapidly declining size of the federal occupational force, which is evident in Figure 2. Most posts needed to be closed simply because Congress shrank the size of the Southern occupation (the number of posts in the South fell from roughly 200 in 1871 to 70 by 1876). We also provide evidence that the presence of troops amplified the importance of other factors, such as the local presence of the Freedmen’s Bureau (Rogowski 2018) and the presence of a large pre-war free Black population (Logan 2020), on increasing Black representation.

We think that these results speak to each of the aims of HPE. First, they help us understand this critical episode in American history. Our findings in many ways support the conventional wisdom of historians that one of the primary, if not the primary, reason for Reconstruction’s failure lies in the federal government’s inability to enforce the effective participation of Black citizens in Southern politics (e.g., Chalmers 1987, Downs 2015, Foner 2014). While many historical accounts have recognized the importance of the federal occupation for sustaining black political mobilization in this period, our work is the first demonstrate this using rigorous identification strategies and a complete dataset of black politicians at the state legislative level

Second, this work clearly also speaks to how the U.S. came to be as it is today, especially in terms of racial inequality. The failure of Reconstruction, which was followed by nearly 75 years of segregation and limited rights for the formerly enslaved and their ancestors, is a primary reason why American slavery continues to cast such a long shadow.

Yet, we were primarily motivated by the third aim – using new evidence to test core theories in political economy. In the large literature studying the consequences of democratization (e.g., on representation, public policy, inequality), many theorists emphasize that newly granted de jure political rights can be offset by the increased use of de facto power and methods by existing elites (e.g., Acemoglu and Robison 2008). Yet directly measuring de facto power is extremely difficult and proxies, such as land inequality, are typically employed (e.g., Acemoglu et al. 2015).

Several features of Reconstruction give us leverage to study the importance of de facto power during democratization. The extension of the franchise to the formerly enslaved was imposed by the victorious Northern states, and an occupation force was provided to enforce these rights. At the same time, the occupation force was never of sufficient size and decreased rapidly during Reconstruction. So, while not directly measuring variation across Southern counties in the de facto power of existing elites, we argue that the local presence of federal troops limited the ability of these elites to use their resources to offset newly gained Black de jure power. Given the necessarily patchwork nature of the federal occupation (with sharply decreasing coverage over time), we capture spatial and temporal variation across the South in the ability of white elites to offset Black political power. We think our findings provide strong evidence in support of the theoretical PE literature emphasizing the importance of de facto power on outcomes during democratization. Furthermore, it suggests that grappling with and controlling for elites’ de facto power is critical for social scientists investigating the consequences of democratization.

[1] Congressman (and future President), James Garfield, said the Republican plan should “[p]lace civil Governments before these people of the rebel States, and a cordon of bayonets behind them” (as cited by Downs 2015: 167).

[2] Specifically, we take the median number of troops across each month of the electoral cycle.


  • Jeffrey Jensen is a Lecturer of Political Science at NYU Abu Dhabi. He studies comparative political economy through the lens of comparative politics and quantitative methods. He holds a PhD in Political Science from NYU and completed a post-doctoral fellowship at Duke University’s Program in Political Institutions and Public Choice (PIPC).

Leave a Reply