Race, Segregation, and Incarceration in the States, 1920-2010


The incarceration rate in the United States today is so high that it can only be described as a pattern of mass incarceration. Worse, this system of mass incarceration operates through structures of gendered and racial discrimination which unfairly target black men. Societal racism in the forms of federal policy and residential segregation produce and normalize this discrimination. However, racial ideologies and practices, as well as segregation, have differed geographically throughout United States history. I examine the racial dynamics of incarceration on a state level, asking how racism and segregation have intersected with the demographics of incarceration in the United States over the last 90 years. Since women make up such a small proportion of the prison population, I narrow my focus to the mass incarceration of men, specifically the disproportionate representation of black men in American prisons.


Figures 1-5 visualize incarceration trends of American men, ages 15 to 70, on national and state scales from 1920 to 2010. Figure 1 shows the change in the prison population by racial proportions over time. The width of the bars represents the total prison population and the height represents the proportion of that population that is White, Black, or Other race. Figures 2-5 are a series of animated maps of the United States representing various proportions of prison and total populations: Figure 2 shows black men as a fraction of all men in each state; Figure 3 shows black men as a fraction of all prisoners in each state; Figure 4 represents the fraction of men in prison of all men in each state; Figure 5 shows the fraction of black men in prison of all black men in each state. Figure 6 shows Black-White segregation indices by state for 2005-9. The District of Columbia is excluded for all Figures because its abnormally high black-white population ratio skews state trends.

The data for Figures 1-5 were drawn from the Integrated Public-Use Microdata Series (IPUMS) decennial census data. I used 1% samples for all years except for 1970, 1980, and 2010, which were extracted from 1% state fm1, 1% metro, and American Community Survey (ACS), respectively. All data were weighted by PERWT, the sample weight of each individual. Alaska and Hawaii are excluded prior to 1960.

Although Figures 2 and 3 and Figures 4 and 5 compare related proportions of populations, I had to give them different scales because of the huge racial gap in incarceration. Black men make up such a greater fraction of prison population than total population that it is impossible to measure the proportions meaningfully on the same scale.

I used a variable created by IPUMS, RACESING, to establish Black, White, or Other racial categories for prisoners. RACESING translates multiple-race responses into their most likely single-race categories (White, Black, Asian, Native American, or Other) to make race consistent and comparable across US history. In my analysis, Other is comprised of Asian, Native American, and Other RACESING responses.

Prior to 1990, the census differentiates between correctional, mental, and elderly/handicapped institutions. However, the 1990, 2000, and 2010 censuses lump all of these institutions into one category. Consequently, it is impossible to know exactly how many individuals were in prison starting in 1990. I establish a correction factor, modeled after the 1980 ratios of individuals in correctional institutions to all institutions, as a conservative estimate of the incarcerated populations in 1990, 2000, and 2010. In reality, the proportion of black prisoners are most likely higher than my numbers will show because the incarceration rates for black men have risen faster than the rates at which black men are going to mental and elderly/handicapped institutions. Nonetheless, I apply these conservative factors to Figures 1, 3, 4, and 5 to avoid overestimation. Table 1 and 2 (below) show how I develop the correctional factor using the prison to institution ratios from 1980. I apply a correctional factor of 0.72 when analyzing incarceration rates of just black men and a correction factor of 0.50 when analyzing incarceration rates of all men.

Table 1: Correction Factor for Black

Year Prison Institution Correctional Factor
1980 183500 252300 0.72

Table 2: Correction Factor for All Men

Year Prison Institution Correctional Factor
1980 430700 863300 0.50

I assume that all prisoners are incarcerated in their state of residence. However, the census does not report the home state of prisoners. Since prisoners who commit federal crimes must go to a federal prison, and not all states have a federal prison, those states without a federal prison will be underrepresented, and those states with federal prisons will be overrepresented because they will necessarily contain prisoners from other states. To reduce the cost of over-crowding, state prisons sometimes transfer a small number of their inmates to prisons in other states. However, Mawhirt points out that the number of inter-state prison transfers is so small—only 0.2%—that it is reasonable to treat state prisoners as if all are imprisoned in their state of residence (Mawhirt 2012).

Since the census does not provide a variable that quantifies segregation, the data for Figure 6 are drawn from the University of Michigan Social Science Data Analysis Network’s analysis of the 2005-9 ACS and reflect segregation indices for those years. The indices are a measure of dissimilarity. A score of 0 represents complete integration and a score of 100 represents complete segregation. It can be also understood as the percent of blacks that would need to be relocated to another neighborhood in order to be distributed in the same way as whites. Unfortunately, state-level segregation indices are not available prior to 2000.

Code to generate Figures.

Code to generate Tables.


Figure 1

The width of the bars in Figure 1 shows that the nation’s male prison population has more than doubled since 1920, with noticeable increases in inmates from 1980 through 2000. The height shows that a large and increasing number of this population is black, peaking in 2000. The proportion of male Other race prisoners has also risen since 1920, but not at the alarming rate of black men.

Figure 2

Figure 2 represents how a large proportion of the Southern male population has been black since 1920. The male African American prison population increases in the Rust Belt and especially New York during the Great Migration. Few black men have ever made up the male population in Idaho, Montana, Wyoming, North Dakota, South Dakota, Maine, New Hampshire, and Vermont.

Figure 3

Figure 3 shows that the proportion of black men of the total male prison population has varied geographically at different points in time, but that it has increased in general across the United States from 1920 to 2010. The Rust Belt has consistently had a large  proportion of black inmates, and this proportion was especially large in 1980 and 1990. Proportions increased in the South from 1960 to 1980 and then decreased. Midwest and West proportions have increased over time.

Figure 4

Figure 4 shows how the percent of men in prison has increased in all states from 1920 to 2010. The proportion of men in prison jumps from 1980 through 2000. The South has always had a slightly bigger proportion of its men in prison than more Northern states.

Figure 5

Figure 5 illustrates that the proportion of black men in prison of the total black male population has risen in all states since 1920, increasing remarkably from 1980 through 2000. The Southern states have slightly lower proportions of their black men in jail in from 1980 to 2010. A few states in the Midwest–Iowa, Nebraska, Minnesota, Wisconsin Kansas, and Wyoming–rank high in this statistic from 1980 to 2010.

Figure 6

Figure 6 shows that states in the Rust Belt are the most segregated. The South and Southwest have low levels of segregation. States in the Midwest are moderately segregated.


Collectively, Figures 1-5 prove that incarceration, and especially the incarceration of black men, is a growing problem in the United States. The male prison population has multiplied from 1920 to 2010. Most of this expansion occurred in the past half century, rising from 319,500 male prisoners in 1970 to 1,260,868 in 2010. Figure 1 illustrates that incarceration follows an increasingly racial pattern. In 1920, 35.2% of male prisoners were black, although they only made up 9.2% of the male population. In 2010, 53.6% of male prisoners were black, although they only made up 10.4% of the male population. The overrepresentation of black men in America’s prisons suggests that the US criminal justice system has a history of discriminating against this subset of the population.

The mass incarceration of black men is the result of an historically embedded racial caste system in American political, cultural, and social institutions that can be traced back to slavery. Ava DuVernay’s documentary 13th points out that because white-dominated society depended on free black labor for economic prosperity during slavery, white elites had to find a new institution through which to exploit the black work force when slavery was abolished. That institution was mass incarceration. Although the 13th amendment abolished slavery, it made an exception for the punishment of crime. As a result, blacks were incarcerated at extreme rates and then taken advantage of for free labor. “Black criminality” was constructed as a convention for the unfair roundup, political control, and treatment of African Americans (DuVernay 2016). Alexander Michelle’s book The New Jim Crow suggests that black criminality and institutional racism still exist and are illustrated through the incarceration trends we see today. Black imprisonment is merely a changing of the rules, so that white society can legally oppress black men for political, social, and economic domination (Alexander 2010).

Figures 1, 4, and 5 show that the imprisonment of black men did not increase at a constant rate. Instead, it skyrocketed from 1980 through 2000. In The Prison and the Gallows: The Politics of Mass Incarceration in America, Marie Gottscalk attributes this sharp rise to the law-and-order campaigns of Presidents Richard Nixon, Ronald Reagan, and Bill Clinton. These three presidents implemented rigid prison reform in the 70s, 80s, and early 90s that put more men and black men in prison. They exaggerated crime and especially the illegal drug trade through the War on Drugs to scare the public into voting for them and then supporting their draconian policies (Gottschalk 2006). Meanwhile, the media reinforced the idea of black criminality through reports and images that legitimized the racial policies which led to the unfair arrest and imprisonment of black men (DuVernay 2016).

DuVernay and Alexander suggest that War on Drugs was a response to the Civil Rights movement, which undermined the white status quo and system of racial hierarchy. The incarceration of black men was a way to get society back under white control (DuVernay 2016 and Alexander 2010). Figure 3 supports this theory: the fraction of black men in prison of the total male prison population in the South increases significantly from 1960 to 1980 during the Civil Rights movement.

President Nixon was the first president to declare a “War on Drugs” and a more general war on crime. He increased the power of federal drug control agencies, approved no-knock warrants, and magnified punishments for marijuana possession. Nixon also increased mandatory minimum sentencing for possession of “crack” cocaine, a drug primarily associated with inner-city black communities. The penalties for crack cocaine were 100 times more severe than powder cocaine—more commonly associated with whites—even though their chemical effects are the same. This difference put disproportionate numbers of black men in prison (DuVernay 2016).

The War on Drugs took off with the election of President Reagan in the 1980s. He created the Office of National Drug Control Policy to tighten drug legislation and his wife, Nancy Reagan, started a “Just Say No” campaign (DuVernay 2016). Together, the Reagans instilled fear in the public that America was a dangerous place full of gangs, violence, and drugs. They shifted public opinion to embrace a disciplinary and unforgiving culture, resulting in the legitimation of higher arrest and incarceration rates across the nation, shown in the sharp rise in prisoners in Figures 1, 4, and 5 from 1980 to 1990.

Apprehension about crime and punishment carried over into the Bush and Clinton Administrations. President Clinton’s 1994 Violent Crime Control Act increased federal spending on policemen and investigative lawyers, imposed tougher prison sentences, and opened new prisons (Gottschalk 2006). Once again, these policies materialized in the huge increase in prison populations in from 1990 to 2000 in Figures 1, 4, and 5. Today, around half of all federal inmates are in prison for drug convictions, attributed to the anti-drug policies of the 70s, 80s, and 90s (The Sentencing Project).

These federal influences set the conditions for the expansion of the carceral state across the nation. As we can see in Figures 4 and 5, federal policy and political elites had an important impact on incarceration trends in the states. Racialized drug policies that unfairly target black men exacerbated the disparity between black and white male incarceration rates to worsen across the states over time, visible in Figures 3 and 4. The proportion of black male prisoners increased by 17% in just 30 years, from 1970 to 2000. However, together, Figures 2, 3, 4, and 5 show that these trends were not distributed evenly on state and regional levels. They suggest that incarceration rates are dependent on geographic factors, such as segregation and local and regional racism.

Figure 2 shows how the black-white male demographic has changed geographically from 1920 to 2010. Figure 2 lays the foundation for Figure 3–the proportion of black male of the total male prison population is only meaningful in its relationship to the proportion of black men of the total male population. In a fair and equal society, there should be a larger proportion of black men in prison where black men make up a larger proportion of the male population. Therefore, Figures 2 and 3 should show that black men made up a larger percentage of the prison population in the South, since more black men lived there. Yet, Figures 2 and 3 show that this is not the case. In fact, the South has the lowest black male prisoner population from 1990 to 2010, the years when the incarceration of black men increased the most. Instead, proportions are the highest in the Rust Belt–Illinois, Indiana, Michigan, Ohio, Pennsylvania, and New York–and Midwest even though they have lower proportions of black men of the total male population.

Figure 2 shows that the Great Migration, a huge movement of Americans from Southern to Northern and Western states from 1910 to 1970, had important effects on racial demographics in the US. During the Great Migration, masses of black men in the South, and most likely their families, moved to industrial cities in the Rust Belt, seeking economic and social opportunities before, during, and after World War II (Gregory 2005).

Together, Figures 2 and 3 show that as the proportion of black males of the total black male population rose in the Rust Belt from 1920 to 1970, so did the proportion of blacks of the male prison population. While this rise would make sense, the proportion of black men of the male prison population was still overwhelmingly greater than the proportion of black males of the total male population. Furthermore, Figure 4 shows us that there were not a lot of men in prison at this time, so the fact that many of those men were black suggests that black men experienced discrimination in the Rust Belt. In their book American Apartheid: Segregation and the Making of the Underclass, Massey and Denton explain that black men in the Rust Belt were subject to societal racism through residential segregation during the Great Migration, which Mawhirt proves in her article “Segregation and Incarceration: How Life in the Ghetto Leads to Life in Prisons for Young Black Men” is linked to higher rates of incarceration.

Figure 6 shows that the Rust Belt contains the most segregated states in the United States today. Massey and Denton explain that it has been this way every since the Great Migration. When blacks settled in Northern industrial cities during the Great Migration, they experienced harsh job and housing discrimination upon arrival. Until the late 1960s, the Federal Housing Administration and Home Owners Loan Corporation restricted black families from white neighborhoods and enacted federal policies that made it extremely difficult for them to get loans. “Red lining” undervalued neighborhoods of color, so that white people abandoned and disinvested in black neighborhoods. These policies resulted in extreme segregation and deterioration of black neighborhoods through economic and social isolation (Jackson 1985). Indeed, the Rust Belt contains some of the country’s most segregated cities—Chicago, Detroit, Cleveland, Philadelphia, and New York City. These cities were the primary destinations of black families during the Great Migration. Even after discrimination in the housing market was outlawed with the Fair Housing Act of 1968, de facto discrimination persisted (Jackson 1985).

The consequences of discriminatory housing policies during the Great Migration indirectly attributed to the rise in the incarceration rate of black men over time. Massey and Denton explain that once segregation existed, it worsened, as more and more white people left poor areas while black people could not afford to leave. Certain cities in the 70s and 80s reached levels of “hypersegregation,” or patterns of separation that crossed multiple dimensions of geographic variation (Massey and Denton 1993, 74). Mawhirt draws on Massey to explain how segregation is linked to incarceration through poor education and high dropout rates. Disinvestment in schools and education because of white flight led to less job opportunity and poverty, increasing the incentives for crime, especially in neighborhoods where less stigma is attached to criminals (Mawhirt 2010). Although in Figure 3 we see that the proportion of black male prisoners to all male prisoners increases across the country from 1970 to 1980, the Rust Belt and industrial North increase significantly. This jump suggests that the residential segregation of blacks in the Rust Belt contributed to a larger proportion of its black males in jail. Massey and Denton reject the theory that black culture is a “culture of poverty” (Massey and Denton 1993, 5). Instead, he blames a history of segregation and oppression for the poverty, crime, and incarceration in black communities today. It is no surprise that the incarceration rate jumped from 1980 to 1990 and 1990 to 2000, following the patterns of hypersegregation experienced in the 70s and 80s.

Figures 4 and 5 compare the proportions of black men in prison to the proportion of all men in prison. Figure 4 shows that the percent of all men in prison has increased across the nation, with slightly higher proportions in the South. It makes sense that more men might be imprisoned in the South, which is known to be a more conservative region.

I use Figure 4 to set the foundation for Figure 5. In a just society, Figures 4 and 5 should show a higher proportion of black men in prison of the total black male population where there are higher proportions of men in prison of the total male population. Yet, Figures 4 and 5 show that the fraction of black men in prison of all black men is not correlated with the percent of men in prison of all men. In fact, the South tends to have the lowest proportions of its black male population in jail, when it has the highest proportion of its men in prison. Even though the South is stereotyped to be a culturally racist and conservative region, they do not lock up more black men relative to other regions. Instead, it appears that the highest proportions of black men in prison of all black men actually occurs in the states with the smallest proportions of black men—states like Iowa, Colorado, Nebraska, Kansas, Wyoming, Maine, Vermont, and New Hampshire. For instance, in Iowa in 2010, blacks make up 24.3% of the male prison population, but only 2.13% of the total male population.

It is interesting to explore this high black-to-prison, low black-to-white population pattern through the lens of segregation. These states with low proportions of black men but with high proportions of black men are moderately segregated in Figure 6. They are less segregated than the Rust Belt, but more segregated than the South and Southwest. However, these measures of segregation are based on dissimilarity measurements, meaning that, in these states, blacks are moderately overrepresented in some areas and underrepresented in others. However, dissimilarity is only one dimension of the geographic variation of segregation, and it may not capture the whole picture.

Massey and Denton point out how segregation is multidimensional, and it can also manifest in patterns of isolation, concentration, centrality, and cluster (Massey and Denton 1988). When taken into account, these dimensions of segregation might speak to how states like Iowa, Colorado, Nebraska, Kansas, Wyoming, Maine, Vermont, and New Hampshire have such high proportions of their black men in prison. In particular, isolation could play an important role. Especially since these states have low proportions of black men, isolation–a measure of how exposed the minority is to the majority population–would be very high. It would make sense that a higher percentage of black men are incarcerated in areas with more exposure, where blacks are always subject to, reminded of, and possibly oppressed by the white majority. It is also easier for the white majority to ignore, distance, and marginalize a minority it does not regularly interact with. Without data that approaches segregation across multiple dimensions, it is difficult to analyze how geographies of segregation intersect with incarceration. There has been little research that attempts to quantify segregation across multiple dimensions, and even less on a state scale, since segregation studies tend to focus on big cities and their surrounding suburbs. Further investigation into the experience of race and segregation in rural areas could speak more to its high rates of incarceration.


In conclusion, my analysis suggests that the mass incarceration of black men is a complex reality, rooted in a discriminatory network of politics, ideologies, and geographies that have reinforced, justified, and normalized the marginalization of black men across national and state scales in American society. Federal policies and sentiments during the 70s, 80s, and 90s permeated regional and state criminal justice systems. Residential segregation has exacerbated crime and incarceration rates in poor, black communities. However, the degree to which black men are unfairly targeted and oppressed is also contingent upon state and time. At different times, the Rust Belt, South, and Midwest have displayed high proportions of their black men in prison. Recently, the mass incarceration of black men has increased in the rural Midwest, where there are few blacks but a large proportion of them in prison. For some reason, structural or cultural, it is easier for black men to be arrested, convicted, and locked up than their white counterparts in these states. I call for a closer look at the experience of race in rural geographies at the intersection of incarceration rates.

It is important to understand and investigate the complex history of mass incarceration in the United States in order to fix it. A segregated and racist society is a broken one. Police brutality, political unrest, and racial division are growing in the United States. As we transition from our first black president to our next leader, it is important to remember the role of political elites in shaping federal policy and public attitudes. Our next president has the potential to dramatically improve or worsen racism within the US criminal justice system and American institutions in general.

Works Cited

Alexander, Michelle. The New Jim Crow: Mass Incarceration in the Age of Colorblindness. New York: New, 2010. Print.

13th. Dir. Ava DuVernay. Netflix. N.p., 30 Sept. 2016. Web. 7 Oct. 2016.

Gottschalk, Marie. The Prison and the Gallows: The Politics of Mass Incarceration in America. New York: Cambridge UP, 2006. Print.

Gregory, James N. “A Century of Migration.” The Southern Diaspora: How the Great Migrations of Black and White Southerners Transformed America. Chapel Hill: U of North Carolina, 2005. N. pag. Print.

Jackson, Kenneth T. “Crabgrass Frontier: The Suburbanization of the United States.” Oxford University Press 8.2 (n.d.): 191-245. Web.

Massey, Douglas S., and Nancy A. Denton. American Apartheid: Segregation and the Making of the Underclass. Cambridge, MA: Harvard UP, 1993. Print.

Massey, Douglas S., and Nancy A. Denton. “The Dimensions of Residential Segregation.” Social Forces 67.2 (1988): 281-315. JSTOR [JSTOR]. Web. 30 Oct. 2016.

Mawhirt, Jess. “Segregation and Incarceration: How Life in the Ghetto Leads to Life in Prisons for Young Black Men.” Colgate Academic Review 7 (2010): n. pag. Digital Commons @ Colgate, 20 July 2012. Web.

“The Sentencing Project.” The Sentencing Project. N.p., n.d. Web. 30 Oct. 2016. <http://www.sentencingproject.org/>.