This week, the FBI released grim crime statistics for 2020. While crime overall went down, the murder rate rose 29 percent — a substantial increase over 2019, though still well below the historic highs of the early 1990s.

For months, criminologists, politicians and pundits have been dishing blame and speculation about what’s behind the rise in homicides. Explanations have included declining numbers of police officers due to the coronavirus, resignations and poor recruitment; “de-policing” due to low morale among cops in the wake of the George Floyd protests; an increase in domestic violence due to social isolation; a rise in gun purchases and gun stockpiling; reforms in sentencing and cash bail policies; increased tension, anger and shorter tempers fueled by the pandemic; and a drop in public trust for and willingness to cooperate with law enforcement, especially in high-crime areas.

There are arguments for and against all of these causes, and our political affiliation tends to color which of them we find persuasive. But it’s worth noting that the murder spike affected the entire country, urban and rural, regardless of political leadership. Traditionally, economic hardship brings more property crime but has little effect on violent crime. On the other hand, the social indicators associated with a rise in violent crime — social unrest and upheaval, distrust in government, and anger among marginalized communities — all were in ample supply last year.

But there’s another, much simpler explanation that might explain at least part of the surge in murders last year: There were fewer witnesses.

A large body of academic research has found a strong correlation between the crime rate and the number of police officers on the street. There are still some gaps in the research, most notably whether the conclusions are scalable. Because these studies are based on observing real-world policies, they focus on what happens to crime in specific neighborhoods when cities increase or decrease police presence.

These studies also tend to ignore the costs of over-policing — multiple cops on every corner in the United States would probably reduce official crime to near zero, but most of us wouldn’t want to live in that society. A recent study published by the Niskanen Center found that although an increased police presence does correlate with fewer murders, it also comes with additional arrests, encounters, and presumably, harassment and animosity between police and marginalized neighborhoods.

Criminologists such as John Pfaff have argued that the “more cops, less crime” academic research might not be measuring the effects of police officers so much as a “sentinel effect.” Pfaff points out that other research suggests that unofficial “sentinels” such as private security and neighborhood watch programs might have a similar or greater effect on crime rates. It may not be that we need more cops — we just need someone watching.

But last year, we had a massive, once-in-a-century pandemic that emptied the streets. We stopped commuting. We stopped going to bars, concerts and other late-night events. We stopped visiting one another. We retreated from public spaces.

This effect explains other crime statistics from 2020. Crimes such as theft, robbery, assault and non-domestic rape typically occur when people interact with another. We didn’t interact much in 2020, and as a result those crimes were less frequent. The other major category to increase last year was vehicle theft, a crime you’d expect to go up with fewer people on the streets. Presumably, a lack of witnesses would also stimulate more burglaries of empty houses, but that effect would likely be offset by more of us staying home more often. Sure enough, burglaries dropped in 2020, but not as much as person-on-person crimes such as rape or assault.

But murder is different. The three biggest drivers of murder are domestic violence, gang violence and the illicit drug trade. It’s pretty clear how the pandemic might have exacerbated domestic violence. And neither gangs nor the illegal drug trade is going to close shop because of a pandemic or government-ordered lockdown. Addicts don’t stop needing fixes. Indeed, the United States saw a massive increase in overdose deaths in 2020, suggesting a spike in demand for illicit drugs. Meanwhile, as cops fell ill with covid-19, cities also saw reduced enforcement and police presence in high-drug-trafficking areas. It also seems safe to say the supply chain disruptions that affected just about every other market in 2020 also impacted the illegal drug market.

All of these factors would create mass uncertainty in illicit drug markets, which would send dealers and gangs battling for control of the new landscape. Officials in cities that saw some of the largest murder surges in 2020 and 2021 — including Albuquerque, Indianapolis, Chicago, Los Angeles and others — all have said much of the spike could be attributed to drug- and gang-related murders.

So the events of 2020 produced conditions that both encouraged turf wars and gang rivalries, and at the same time undermined the single biggest deterrent to public violence — the presence of witnesses, whether they were police or bystanders.

Of course, it would be foolish to pin the murder surge on a single cause. But the good news is that if the sentinel effect was a major contributor, we should see the numbers come down as the country returns to normal. And indeed, the homicide surge appears to be slowing, and even reversing in some cities that have opened up, such as New York and Boston.

The lack-of-witnesses explanation isn’t particularly sexy or divisive. It doesn’t easily lend itself to partisan finger-pointing, culture war posturing or a rush to legislation. Perhaps that’s why it’s rarely mentioned.