Maybe it Was the Lockdowns that Killed 8,682 Black People

Recalibrating a prior HWFO opinion

On November 14, 2024, HWFO published a mathematical analysis showing that the homicide spike of 2020 and subsequent fall back to historically low rates caused the deaths of approximately 8,682 black Americans. That article provided solid mathematical reasoning and extrapolation showing that the spike killed between eight thousand and nine thousand black folks in excess of the number we should have expected with no murder spike. The article further opined that the cause of the spike was the Black Lives Matter protest, making BLM not only the least effective protest ever but likely the most counterproductive one. The article stated, in spicy language:

We had nationwide riots and protests against the police in the wake of the George Floyd killing, with 5000+ demonstrations, 20 million protesters, 20 deaths, 14,000 arrested, $2 billion in insured damages, an average of $100 of insured damage per protester, at least one burned police station, the founding of an anarchist autonomous zone, and a several month long siege at a court building. This included a number of protesters larger than the entire population of New York State chanting “All Cops Are Bastards” and “Defund The Police.” This in turn caused the total number of people willing to be police officers to shrink for two straight years, only stabilizing in 2022.

So that could have been it.

There is a known scientific correlation between increased police presence and decreased crime. It is impossible to prove causation for a murder spike like this because murder is a multivariate problem, but there are no other reasonable causes I can identify, and if you can find a cause that’s more compelling than unwashed hordes of angry protesters burning police stations down domestically while the rest of planet Earth didn’t, please post it in the comments.

I have recently been made aware of another theory by the Brookings Institute, and I think it’s well supported enough that it has forced me to challenge my priors. The lockdowns themselves were to blame, not the ACABs and BLMs. We will cover that study in a moment, but first let’s recap how to calculate 8,682 dead black people.

8,682 Black Lives Ended

First, we calculate the baseline murder rate in the United States across a long span of fairly regular fairly low (for us) murder rate. This is the expected murder rate for the country.

Then we figure out how many people died in excess of that murder rate, by analyzing the four year spike.

Then we take a look at the fact that black folks disproportionately bore the brunt of the spike, at victimization rates that varied by year but were always exceeding “normal,” and we multiply that ratio times the excess deaths to determine the excess number of dead black people within the spike itself.

8,682 is not the number of black people that died in homicide across that time frame. It is the extra dead black people exceeding what normally would have died without a spike. This is a pretty inarguable way to calculate excess dead black people. In fact, the number might be higher if the excess murders were concentrated in black areas.

The only question, really, is who is to blame for the excess dead black people. Is it BLM, as my prior article speculated? The Brookings Institute thinks not. They think it was the lockdowns.

The Lockdowns Killed Them

A month after the HWFO article was published, the Brookings Institute did a very interesting and very robust job of presenting an alternative analysis, in rebuttal to articles like mine, and I think they did a good job. Here’s the article.

Their opener specifically calls out other publications which were saying the same thing I was.

Some commentators have suggested the increase in homicides during 2020 was a response to the murder of George Floyd by a Minneapolis police officer in May of that year. Others hypothesized that it was caused by a police “pull back,” in which officers chose to do less work in reaction to the protests that followed Floyd’s death.

I made the second case prior. It made sense to me. But Brookings identified something very curious when they looked at the data with a finer grain.

As more information has become available, these theories appear to be less supported by evidence than some initially thought.1 The evidence indicates that the national homicide rate was already on track to reach a peak far above the previous year even before Floyd was killed. (See Figure 1.)

Let’s look carefully at that graph. It shows 2019 murder and 2020 murder trends by day and month. The lines were very closely overlapped until the lockdowns began. Thereafter, the gap between the two lines grows, both before Floyd and after Floyd, until it establishes a new flat trend at a higher daily murder rate. The area between these two lines represents “excess murders,” which we know from the table above are 46% to 59% dead black people.

The Brookings article goes on to establish that (1) the excess deaths were concentrated in areas with higher poverty, and (2) teen boys and young men not in school or work were much more likely to engage in violence. They show some example maps of their finer grained geographical analysis:

And from these maps we can easily see why black people bore the brunt of the murder spike, by simply cross referencing them with demographic maps, like this:

Narrator: It’s the same map.

Brookings goes on to compare Baton Rouge murder trends to other historical events such as floods or Hurricane Katrina that caused school and work closings, and shows how those also tie to murder spikes. It’s a robust analysis, and one that has me reconsidering elements of my prior position.

Could Be Both? Who to Blame?

One argument against the Brookings case is that murder went down sharply in places like Rio de Janeiro and Cali Colombia as a result of Covid-19 lockdowns, so it’s possible that our lockdown was unique or different in some way, perhaps owing to all the national police hate. But Brookings analysis defenders might point to how murder rates here were rising before Floyd as a solid counter argument.

The blame for the 8,682 dead black people then is either the people screaming on Facebook to burn police stations and All Cops are Bastards and frantically sharing any instance of police misdeed to feed their clickbait engine, or it was the people screaming on Facebook about how “leaving the house is murder” and “reopening the economy is murder” and “opening the schools is murder” and so forth and so on, leading to a protracted economic and educational lockdown of the country which profoundly and directly damaged black communities.

Funny thing, that. I don’t know what your feed was like in 2020, but in my feed these two groups of people were the same people.