The Surprising Geography of Police Killings: Back-of-the-Napkin Calculations on Race, Region, and Violence
In the United States, the police kill African Americans at a rate that is about 100 percent greater, or two times, 200 percent, their proportion of the national population. In 2016, black people were 24 percent of those killed by cops, in 2015 they were 27 percent of such victims, but in both years black people were only 13 percent of the national population.1 These outrageous disparities have very correctly triggered a nationwide rebellion.
But where do these racial disparities actually take place?
Amidst this moment of reckoning the South, cast as the cradle of racism, seems to come in for special criticism. Antebellum Southern slave patrols are regularly name-checked as an origin of American policing. Confederate monuments are toppling, as they should. NASCAR banned the Confederate flag. A Nation writer decried “stupid” Southerners for flouting social distancing at a bacchanalian redneck vehicle jamboree on the beaches of Galveston, Texas. A Washington Post columnist asked rhetorically if Donald Trump wasn’t actually the last president of the Confederacy. And, let’s admit it, most of the country thinks of the South as profoundly backward.
Given this vibe one might be surprised by the actual regional demographics of police killings. What follows is a very preliminary, incomplete, back-of-the-napkin sketch of data on police killings. My main source on police killings is the Guardian’s Counted Project. Economic and demographic data come from the U.S. Census, the Bureau of Labor Statistics (BLS), and the Kaiser Family Foundation. I am rounding numbers with decimals up and down. For a discussion of the sources used see the first two footnotes.2
Let’s start with Tennessee, the state that gave birth to the Ku Klux Klan. It seems reasonable to assume that the cops in Tennessee kill African Americans at a disproportionately high rate.
In 2016, police in Tennessee killed 25 people. Of these, nineteen, or 76 percent of the total, were white. Meanwhile, whites were 78 percent of the state’s total population. Tennessee police killed three black people, which was 12 percent of the total. However, African Americans were 17 percent of the state’s total population.
In other words, African Americans were, relative to their proportion of the state’s total population, actually 29 percent “underrepresented” in the stats on police killings. White people were 2 percent underrepresented in the police homicide stats.
Thus, Tennessee cops actually killed whites at a higher rate than they killed black people even as both whites and blacks were “underrepresented” in the police homicide stats. Latinos and Pacific Islanders each suffered one police homicide, and because they make up small percentages of the state’s population, were “overrepresented” in the police stats.
In Kentucky the cops killed 22 people. Of this total nineteen, or 86 percent, were white. The state’s population, as a whole, was 87 percent white. Two of the Kentuckians killed by cops that year were black, meaning they were 9 percent of the casualties. Meanwhile, African Americans were 8 percent of the state’s total population. One of the victims of Kentucky police homicide was Latino.
This means white people were slightly underrepresented among those killed by police while Latinos and African Americans were overrepresented relative to their proportion of the state’s population. But the black victims of police homicide in Kentucky were 12 percent overrepresented, not 100 percent overrepresented as they are in the national stats.
What about the Deep South where a greater percentage of the population is black? For example, take Mississippi—it doesn’t get any more “Deep South” than Mississippi.
In 2016, cops in Mississippi killed eleven people: six, or 55 percent, of these were white and five, or 45 percent, were black. The state’s population was 59 percent white and 37 percent black. This means Mississippi cops killed black people at a rate 49 percent higher than their prevalence in the state’s total population. Thus, we can say Mississippi displays a racist pattern as regards police killings. But it is only half as racist as the national numbers.
In Louisiana, cops also killed black people at disproportionately higher rates than they kill white people. African Americans were twelve of the 22 people killed by police. They were 32 percent of Louisiana’s population but were 54 percent of those killed by police in 2016. That ratio gets closer, but is not all the way, to the national aggregate numbers.
Florida is also closer to, but not at, the national average. In the Sunshine State African Americans were 16 percent of the population yet constituted 25 percent of those killed by cops in 2016. Cops in Florida thus killed African Americans at a rate that was 56 percent greater than the African-American percentage of the state population.
In Georgia cops killed thirty people in 2016. African Americans, being 17 of these victims but only 31 percent of the population, were 19 percent overrepresented. Latinos were 17 percent of police homicide victims but only 9 percent of the population and were thus almost 100 percent overrepresented. Whites on the other hand were 28 percent underrepresented in such stats, being only 43 percent of those killed by cops despite constituting 60 percent of the state population.
However, if we cross the Savannah River into South Carolina, the state that started the Civil War, the patterns change. In 2016 Palmetto State cops killed eighteen people. Of this total, four (or 22 percent) were African American even as they constituted 28 percent of the state population. This meant black people were 27 percent underrepresented in the police homicide stats. White victims of police homicide numbered fourteen (or 78 percent of the total) even as whites were only 67 percent of South Carolina’s population.
In other words, white South Carolinians were 16 percent overrepresented in the police homicide stats and they were significantly more likely to be killed by cops than were black South Carolinians.
I could go on with similarly weird and counterintuitive Southern examples but I will spare readers the jumble of numbers.
So then, where do cops kill black people most disproportionately?
One of the worst offenders as regards the disproportionate killing of black people—that is to say, the state with some of the most anti-black cops in the country—is liberal Massachusetts.
The Bay State—which during the Civil War produced the ultra-heroic, all-black 54th Regiment about which the fantastic film Glory was made—has police that kill black people at five times, or 500 percent the rate at which black people appear in the state’s total population. No wonder people joke about “up South in Boston.”
In 2016, police in Massachusetts killed fourteen people: five were white, five were black, and four were Latino.
White people are 79 percent of the population but only 35 percent of those killed by cops, and were thus 56 percent underrepresented in the police homicide stats.
Massachusetts police also kill Latinos at a very high rate. Latinos were 11 percent of the state population in 2016, but they were 28 percent of those killed by police. Thus, Massachusetts Latinos showed up in the police killing stats at a rate of 254 percent their proportion of the state’s total population, or 154 percent greater than the Latino share of the population.
The key number, however, is this: Only 7 percent of Massachusetts’s residents are black, yet they constituted 35 percent of people killed by cops. African Americans therefore appear in Massachusetts police homicide stats at five times the rate, or with 400 percent greater frequency, than do they appear in the state’s total population count. Now we are beginning to see where the national average comes from.
Illinois has a similar profile. In 2016 Illinois cops killed 29 people: nine of them (or 31 percent of the total) were white, while 61 percent of the state’s total population was white. Latinos were 27 percent of those killed by cops despite being only 17 percent of the state’s population.
Illinois cops also killed seventeen black people, (or 58 percent of the total) even as black people were only 14 percent of the state’s total population. In other words, during 2016 Illinois cops killed African Americans at a rate four times (or 314 percent greater than) the black percentage of the population.
Similarly, in Minnesota, cops kill black people at three times their prevalence in the state’s total population: 6 percent of the population versus 21 percent of those killed by cops. In New York police kill black people at three times their proportion of the population: they are only 16 percent of the population but constitute 48 percent of those killed by cops. In Michigan police kill African Americans at a rate about 2.5 times their share of the state population; they are 14 percent of the population but 37 percent of those killed by cops.
Moving west, the cops show anti-black racism in their patterns of killing but not to the level of what we see in the Northeast and Midwest. California fits the northern pattern. Cops killed black people at more than three times their share of the population. But Western police racism, expressed as lethal violence, falls most heavily on Latinos and Native Americans. Measured on a per capita basis no other racial or ethnic group comes near experiencing the appalling level of police violence meted out to Native Americans.
To be fair to the police of Greater Yankeedom, in general, they kill less often than do Southern or Western cops.
The Racialization of Poverty North and South
Why is Northern policing so disproportionately racist? In 1831 Tocqueville noted the peculiar vehemence of Yankee racism: “slavery recedes, but the prejudice to which it has given birth remains stationary…. prejudice of the race appears to be stronger in the States which have abolished slavery, than in those where it still exists; and nowhere is it so intolerant as in those States where servitude has never been known.”3
This Northern prejudice often produced state constitutions that simultaneously outlawed slavery and prohibited African Americans settlement. Ohio, for example, outlawed slavery in its original 1802 constitution. But it also aggressively barred black immigration and enforced the ban with mob violence.
Northern tier states were also the first to pass eugenic forced-sterilization laws. By 1926 most Northern states had such laws but none of the Southern states did.
I suspect that modern patterns of “racialized” poverty, which is to say the racial demographics of poverty, does much to explain Northern police racism. Keep in mind, much of what police do is harass the visibly and “disorderly” poor. Disorderly frequently comes down to doing things in public that, if you had more money, you would do in private: drinking, smoking, buying and selling, yelling, arguing, disrobing, sitting down, and sleeping.4
The racism of Northern police also has something to do with the more “racialized” nature of poverty in the North as compared to the South. In the North, people of color tend to be heavily overrepresented in the ranks of the poor, whereas in the South there are higher rates of poverty and more of the white population is very poor. One crude way we see this is comparing the relative gap between white and black poverty rates in the North and South.
In the South the black poverty rate is typically about twice as high as the white poverty rate. But in most of the northern-tier states the black poverty rate is three times as high as the white poverty rate.5 This is not because black people are necessarily wealthier in the South, though the highest black poverty rates do cluster in the north, but rather because there are more poor white people in the South.6
The U.S. Census defines four major regions: Northeast, Midwest, South, and West. Of these, the South has the lowest median household income; it also has “the largest share of counties with high income inequality.” The South remains the region with the lowest median wages,7 has “maintained the highest rates of poverty over the past 40 years,” and has “the largest share of Americans living in poverty of all regions.” Food insecurity is highest in the South. It has the highest adult and infant mortality rates and the greatest prevalence of illnesses like cardiovascular disease, obesity, and HIV/AIDS. Southerners suffer higher occurrences of occupational fatalities, and the South has many of the highest rates of incarceration.8
The Political Economy of North and South
The South, from the settling of Jamestown onward, has always been home to a large population of poor whites. The South was intentionally designed to be a land of gentlemen and servants. This plan, if you will, shaped southern land distribution. Huge lots were given to rich men, while very little was made available to the common classes. The Yankee north, despite its many faults, pursued an intentionally more equal distribution of land. These divergent sectional settlement patterns had profound and long-term consequences for later economic development.
This sectional difference in land disposal patterns meant that the South never developed a large class of independent small farmers, whereas that class predominated in the North. As Charles Post has shown in his book The American Road to Capitalism, it was from this stratum of family farmers that American industrial capitalism emerged. During the nineteenth century, these small farmers, increasingly subject to market competition and price signals, began specializing and mechanizing. As subsistence production declined, production for sale increased. As it did, consumption increasingly depended on purchasing commodities with money in markets. Through it all the capitalist division of labor deepened, commodification and what Marx called “real subsumption” spread. With class struggle, in the form of growing unionization and then with the New Deal, the wealth produced by Northern industrialization, even as it made robber barons rich, also helped reenforce older Northern patterns of a more widespread, if modest, prosperity.
In the Slave South, several factors blunted this process. The extremely uneven land holding of the South limited the rise of a class of innovating, increasingly market-oriented small farmers. Uneven land distribution also translated into a lower population density and fewer cities, which meant smaller, less competitive markets. And as John Majewski explains in Modernizing a Slave Economy, weak and acidic soils, which are easily depleted by mono-cropping, encouraged the use of “shifting cultivation,” which in turn further reenforced the pattern of large land holdings, low population density, and class inequality.
Slavery also hindered economic development and industrialization because slaves were a fixed cost that had to be utilized even when not working on the cash crops. Because slaves could not be fired like free workers, slave owners needed to maximize their use of slave labor. This disincentivized and undermined the use of labor-saving equipment, resupply through markets, and the outsourcing of tasks to commercial specialists (like blacksmiths or carpenters). Put simply, instead of buying cheap, well produced bacon on emerging commercial markets supplied by small innovating farmers, slaveowners were incentivized to make their slaves raise hogs when they were not raising cotton. Thus even as slaves produced cash crops for export and plantations ran with capitalistic tools of efficiency, like modern account books, the fixed costs of slavery also encouraged nonmonetized production for use. This meant that in the South a smaller portion of production was governed by the law of value, and what Joseph Schumpeter called capitalism’s “gales of creative destruction.”
With large parts of the population (slaves) consuming little and producing much of what they consumed in a nonmonetized, production-for-use fashion, even small yeoman farmers who might have innovated and mechanized along capitalist lines, were for lack of markets effectively held back and stuck in a twilight economy that was capitalist but still heavily governed by the slow logic of production for use. Thus southern industrialization and capitalist “expanded reproduction” were thwarted.
In Slavery and Freedom, James Oakes summarized how slavery underdeveloped the South as follows: “Slavery hindered technological innovation even where its profitability depended on the latest techniques for processing and transportation. It slowed the growth of cities and industry, hampered the growth of a consumer market, reduced the flow of savings, and promoted soil exhaustion and demographic instability by dampening interest in long-term improvements on the land.”9
In the South the pattern of economic development was about cash-crop exports and later also resource extraction. This pattern of economic development reenforced the region’s tremendous class inequality. In the words of the Southern chronicler J.W. Cash, this made the South a society of “Big Men and Little Men, with strict reference to property, power, and the claim to gentility.”10
As a result of the South’s tremendous class inequality, the region’s demographics of poverty have long been less racially skewed than in the North.
In her fine book Masterless Men: Poor Whites and Slavery in the Antebellum South, Keri Leigh Merritt shows that poor, landless whites constituted a full third (some have said one half) of the population of the U.S. South!11 To be clear these were not the hardscrabble small farmers. Rather these were a semi-itinerate, rural Lumpenproletariat, who owned no land and instead lived by occasional day labor, grazing hogs, gathering herbs, cutting wood for sale, stealing, poaching, making and selling liquor, fencing stolen goods, and prostitution. Prone to binge drinking, violence and cavorting with both free and enslaved African Americans (even as they were known for their loudly professed hostility to black people), these poor whites were by most accounts often genuinely dangerous. The planter class hated them. So too, it seems, did much of the smallholding yeomanry.
Prior to emancipation, slavery being the system that controlled most African Americans in the South, both extrajudicial mob violence and formal criminal justice were largely targeted at controlling this class of poor white Southerners.
Even today, in most Southern states the demographic distribution of poverty more closely tracks the overall demographic profile of the state than do poverty rates in the North.12 Of the states with the top ten highest white poverty rates all except for Idaho and New Mexico had been part of the Confederacy.
Making of the Yankee Ghetto
Concentrations of black poverty in the Northern states that once banned black settlement is the result of the racist articulation of deindustrialization and urban renewal. The Great Migration, that is the large-scale relocation of African Americans from the rural South to the urban Northeast, Midwest, and West, began with the industrial boom of World War I. Pulled north by the lure of jobs, higher wages, and greater freedom, the migrants were also pushed north by the hard times brought on by the increasing mechanization of Southern agriculture, and by the despotism of Jim Crow segregation and lynch-law terror. Roughly six million black people moved north before the migration subsided around 1970.
The greatest part of this wave happened from World War II until 1970. But African Americans arrived in the land of industrial democracy and upward mobility just as that political economy began a process of radical restructuring driven by automation and then industrial relocation. Almost as soon as African Americans established themselves in Northern industrial occupations and cities, deindustrialization and racist slum clearance began.
As Thomas Sugrue shows in The Origins of the Urban Crisis: Race and Inequality in Postwar Detroit, industrial employment in Motor City actually peaked in 1950, a full three decades before “deindustrialization” became a sociological watchword. As unionized industrial employment shrank, so too did the service sectors. According to Sugrue, black workers actually continued to move up the wage and skill ladder even as deindustrialization took hold. But this hardly made up for a shrinking regional economy and rising class inequality at a national scale.
Just as industrial employment was peaking, federally subsidized “slum clearance” and highway construction programs began reshaping Northern and Western cities. Coupled with suburbanization along racist lines, these developments increasingly forced black people into de facto segregated and underinvested communities. As businesses and middle-class whites left the urban core, municipal tax bases shrank, services and employment suffered, and concentrations of black poverty became defining features of the Northern-tier rustbelt.
The rustbelt geography became that of the doughnut city, with the African Americans’ deindustrialized core surrounded by autonomous, and for a long time de facto white, segregated suburbs.
The Modern Low-Wage South
Meanwhile, poverty in the U.S. South remained and remains widespread. This is revealed in the disproportionally high percentage of its population working for low wages. In 2016 the Bureau of Labor Statistics reports that the “states with the highest percentages of hourly paid workers earning at or below the federal minimum wage” were: Kentucky, Louisiana, Mississippi, South Carolina, and (one western outlier) Idaho.13
That year saw 2.2 million Americans working for wages at or below the federal minimum. Fully 49 percent of these workers worked in the South. These low-wage workers were predominately women. White women and people of color of all genders were (and still are) disproportionately represented in low-wage work. However, in absolute numbers, low-paid workers as a whole, were and are predominately white. The BLS reports that 74 percent of workers making wages “at or below the federal minimum” are white. (About 10 percent of that number are likely white Latinos but the BLS does not disaggregate in that fashion.) Thus white people work low-wage jobs in proportion to their share of the population. African Americans, at 18 percent of this workforce, are overrepresented. Latinos and Asians are each slightly underrepresented among low-wage workers. Thus for every African American working for the federal minimum wage or less there are four white workers in the same position, and, although the BLS does not offer numbers for race and region together, we can assume that most of both groups are in the South.14 In other words there are lots of poor white people in the South, and this probably helps explain why white people are killed at a higher rate in the south than in the north, and that, in turn, helps explain why black people so disproportionally show up in the northern police killing stats.
One clear takeaway from all these numbers is that Northern liberals—after all, they run most Northern city governments—should not feel too terribly smug when surveying the South, or applauding symbolic victories over racism, because very material forms of racism unfold up North on their watch and these are rooted not only in police prejudice but regional political economy and industrial policy. Transforming those “root causes” would be a massive though not impossible task. It would require challenging the prerogatives of capital; that is, confronting actual capitalists, i.e., campaign donors. That is a daunting prospect. And so, the liberal political class prefers progressive cultural change, renaming and redecorating, to the harder job of progressive economic change. Because, in the grand scheme of things, symbols are cheap.