George Floyd was not an outlier; a Black American is arrested every 15 seconds and murdered every 71 minutes
Is recent violence against George Floyd, Ahmad Arbery, and Breonna Taylor an outlier amplified by social media or emblematic of systematic racism?
This question should not only be answered with polemics, however passionate, but smart analysis of best available data. Bringing empirical analysis into the public square is essential for reasonable discourse; it guards against recency bias, flawed extrapolation from anecdotal evidence, and provides perspective to put individual events in national and historical context.
To that end, I analyze data from the 2018 FBI Uniform Crime Reporting program (the most recent year for which data is publicly available) and identify eight findings that illustrate the social and economic costs of pervasive personal and structural racism on black communities.
1. In 2018, over 2.1 million (2,115,381) Black Americans were arrested. This combined population would be the fifth largest city in the United States, just behind Houston and larger than Philadelphia, Phoenix, and San Antonio.
2. On average, a Black American is arrested every 15 seconds; 4 are arrested every minute, 241 per hour, and 5,796 every day. The number of Black Americans arrested daily is more than twice as many as the total number of NFL players.
3. While fewer Blacks Americans were arrested than Black Americans, when accounting for relative population size, Black Americans are more than twice as likely to be arrested (2.3x) than Black Americans.
4. At the household level, more than 1 in 8 (13%) of Black households are likely to be affected by an arrest.
5. In 2018, at least 7,407 Black Americans were murdered, more than double the number of fatalities of the September 11th terrorist attacks.
6. On average, one Black American is murdered every 71 minutes, about 20 a day.
7. Though a smaller portion of the population, more Black Americans were murdered in 2018 than White Americans (7407 versus 6088). When accounting for relative population size, Black Americans are 7 times more likely to be murdered than White Americans.
8. The economic cost of murder is equivalent to 41% of total Black income. Accounting for wealth inequalities, the economic cost of homicide in Black communities is more than 11 times (11.4x) greater than in White communities.
While the full costs of arrests and homicides cannot be fully quantified by these measures, this data provides a snapshot of how extensively Black communities are affected by arrests and murders. These findings do not explain why the rate of black arrests and murder is so high, but do suggest disparities are systematic. Moreover, they show that recent attention, action, and anger in response to recent high-profile incidents are not much ado about nothing.
Technical Appendix
1. Number of Black American arrests is sourced here. City population from the US Census is sourced here.
2. Estimate of daily arrests (5796) is calculated by dividing total number of arrests (2115381) by 365. Arrests by hour (241) is calculated by taking the daily estimate (5796) and dividing by 24. Arrests by minute (4) is calculated by taking the hourly estimate (241) and dividing by 60. And estimate of seconds per arrest (15) is calculated by dividing 60 by the number of arrests per minute (4). The estimate of NFL players used to compare daily arrests is sourced here.
3. I first calculate the size of Black (43,984,096) and White population (251,103,235) based on Census data here. I then calculate the likelihood of arrest by dividing number of arrests by size of population — the likelihood of being arrested among White population is 2.1% (5,319,654 arrests / 251,103,235 White population), and the likelihood of being arrested among Black population is 4.8% (2,115,381 arrests / 43,984,096 Black population). To calculate the relative risk of arrest, I divide likelihood of arrest among Black Americans by likelihood of arrest among White Americans (4.8% / 2.1%), yielding a relative risk of 2.3x.
4. I estimate the number of Black American households (16,411,976) by dividing the Black American population (43,984,096) by median household size (2.68) as reported by the Census here. I then divide the number of arrests (2,115,381) by number of Black households to identify the incidence rate (12.9%).
5. Black American homicide data is sourced here. Fatalities from the September 11th terrorist attacks is sourced here.
6. The estimate of daily Black murders (20.3) is calculated by dividing the number of murders (7407) by 365. To calculate the minutes per murder (71) I calculate the total number of minutes per day (60 x 24 = 1440) and divide that by the daily estimate of murders.
7. I first calculate the size of Black (43,984,096) and White populations (251,103,235) based on Census data here. I then calculate the likelihood of murder by dividing number of murder by size of population — the likelihood of being murdered among Whites is .0024% (6,088 murders / 251,103,235 white population), and the likelihood of being murdered among Blacks is .0168% (7,407 murders / 43,984,096 black population). To calculate the relative risk of murder, I divide likelihood of murder among Black Americans by likelihood of murder among White Americans, yielding a relative risk of 6.95x.
8. To calculate the economic cost of murder for Blacks($740.7 billion), I use the Value of Statistical Life estimate of $10 million (detailed here and here) and multiply that by the number of murders (7,407). To calculate total Black income, I multiply population (43,984,096) by median income ($41,511; from US Census here); total amount is $1.826 trillion. I repeat the calculation with the White American population; the economic cost of murder is $608.8 billion and total white income is $17.1 trillion (251,103,235 population x $67,937 median income). Dividing costs by total income returns cost as a percent of total income — 40.6% for Black American population and 3.6% for White population. The magnitude of difference is then calculated by dividing Black cost ratio by White cost ratio to yield a 11.4x greater effect among African American community.