During the prime crack years of the 1980s and early 90s, crime, and especially media coverage of crime, was almost inescapable in America. After all, as panic over smoked cocaine began to escalate, murder and other violent and property offenses were, in fact, happening more often. The same dynamic had emerged in the 70s, with heroin. But despite an unprecedented epidemic of opioid overdose deaths that began way back in the late 90s, US crime rates have been declining for decades, and are now less than half their 1991 peak.
When the murder rate jumped by 5 percent nationally last year, helping Donald Trump stoke up fear of crime during his angry presidential campaign, it raised the prospect that the spiraling opioid epidemic might be birthing its own crime wave. But that surge was driven largely by cities like Chicago, even as rates of overdose death are 50 percent higher in rural areas than urban ones. And property crime—which includes offenses typically committed by drug users seeking money to support their addictions like burglary and theft—continued to fall.
Past epidemics were also often presaged or accompanied by a major rise in arrestees testing positive for the drug of the moment. But that hasn't been true this go around either, according to Eric Wish, director of the Center for Substance Abuse Research at the University of Maryland, who has run the longest study monitoring this data in Washington, DC. In America's peak crack years, as many as two thirds of people who got arrested in DC—regardless of the charge—tested positive for that drug. These days, only six percent are positive for opioids, Wish told me.
These facts do not fit at all with the usual narrative Americans tend to hear about the link between criminal behavior and addiction. While the drugs/crime relationship has always been far more nuanced than the way it's portrayed, there wasn't such a glaring disconnect between reality and mythology during the heroin epidemic of the 1970s, nor during the crack era. The mystery of the missing opioid crime explosion offers unique insight into the myths and realities of addiction—and lessons for a Trump administration that is rabidly anti-crime even as it promises to address the opioid problem.
"It's a bit of puzzle: The homicide spikes we are seeing don't seem to have much to do with the opioid epidemic," Harold Pollack, co-director of the University of Chicago Crime Lab, told me of the disconnect.
Watch our segment about beating fentanyl, a deadly opioid way stronger than heroin:
"The idea that addiction leads to crime was always at best partial truth," added Mark Kleiman, professor of public policy at the New York University Marron Institute of Urban Management.
Most of us have a general idea about how drug use might feed criminal activity. People start using drugs, and if they're illegal, drugs can be expensive. Continued, frequent use of drugs tends, sooner or later, to boost tolerance, which means regular users often crave higher doses over time. Once someone is addicted, substances are said to "hijack" their brains—and they may start committing crimes in order to pay for what they now believe they need.
In reality, the correlation between addiction and crime is heavily dependent on economic circumstances and childhood experience, according to criminologists I interviewed. Rich people with addiction rarely start committing non-drug crimes after they get hooked: they can obviously afford to pay, so why risk it?
For the rest of America, it's more complicated.
"When you look at when criminal behaviour begins, people usually commit crimes prior to using drugs," said Khary Rigg, assistant professor of mental health law and policy at the University of South Florida. This has been found to apply both to violent crime and to sex work—though what comes even earlier for violent criminals who have addiction tends to be a violent home in a violent community. (For addicted sex workers, there is often a history of childhood sexual abuse.) In other words, addiction tends to exacerbate pre-existing crime patterns, but does not create them, at least in many cases.
Most of the data here comes from studies of poor populations, but less is known about what happens in the working and middle classes—groups especially hard hit these days by opioids. It seems obvious that some people with addiction do turn to crime after they lose jobs and other sources of support that previously allowed them to sustain themselves. But this crime probably occurs in context: i.e., an addicted office worker is more likely to turn to white-collar crime than she is to start mugging grandmothers.
In this epidemic particularly, there's very little information so far about the criminal trajectory of users—other than that if they have been turning to predatory crime after getting hooked (or if their addictions are worsening their offending), it's not showing up as a corresponding spike in rates of victimization. Indeed, during most of the last ten years, the correlation seemed to be opposite: as opioid overdose and addiction rates rose, predatory crime often fell.
Experts I consulted agreed on some of the key elements that have contributed to this low-crime divergence from American drug history. One important factor is that this epidemic began with medical opioids: While most people who get addicted to prescription opioids are not themselves pain patients (around three quarters of users get them from sources other than doctors), these drugs originate in big pharma factories—not cartel-controlled, illicit poppy fields. Doctors running pill mills are not likely to have shootouts or turf wars. Nor is big pharma going to go around gunning down rivals. Perhaps just as important: Dealers who source their drugs from doctors are likely to sell only to a small network of people they know, where, again, turf violence is not common.
In contrast, crack was a new product introduced to a street market in which people frequently bought from strangers on street corners. Reasonably stable turf that had been established for powder cocaine and heroin sales on these streets was up for grabs, and the result was a great deal of violence. Both cocaine and alcohol are also linked to increased violence among those who are already predisposed to it because they increase paranoia and anger; heroin and other opioids, however, tend to be calming—which may be another reason this epidemic has not raised violent crime.
"The pharmacological properties of opioids matter," said Pollack, the University of Chicago scholar. "Cocaine, methamphetamine and alcohol are more crimogenic, and that matters."
So why was the 70s heroin wave so violent and this one less so? One influence may be age and gender: according to an authoritative study culling nationwide data, the heroin-addicted population now is older and includes more women than in the past. Since the overwhelming majority of violent crime is committed by young men, this may help explain the relative paucity of opioid-related violence. And the presence of fentanyl in the drug supply is killing larger numbers of users than past epidemics, making today's addiction problem much more visible.
Technology has also been a critical factor in reducing the link between addiction and street crime. As Ric Curtis, a professor of anthropology at John Jay College of Criminal Justice in New York who's studied drug markets here for decades, put it, "The biggest difference today versus the 80s is that most of the business in the 80s was conducted in the streets with encounters between dealers and users who were pretty anonymous to each other, frequently. In the 90s, however, Curtis said the business "started to shift toward more private venues and a much more network-based approach, facilitated by technology like cell phones and the internet."
When deliveries are arranged via cellphone or online, it's more difficult for anyone to use violence to monopolize a market—and street encounters where a dealer could step on a rival's toes are less common.
Finally, as with any drug scare in America, no analysis is complete without considering the question of race and racism. This current epidemic skews largely white: since 2010, according to the previously cited heroin addiction study, 90 percent of new users have been white. But even if a black person and a white person use exactly the same amount of a particular drug, "Whites get arrested less," according to Rigg, the University of South Florida scholar. "And also, once they do have contact with law enforcement, often times, they are not sent to prison and just get let off or go to drug court or don't face charges at all." This reduces the harm done to whites facing addiction—if they don't get put into the system, they aren't branded with a criminal record that reduces their odds of employment and they don't get immersed in the criminal subculture of being incarcerated.
So if there is an increase in opioid dealing or other nonviolent crimes by white people that we're missing, it's being hidden by biased enforcement and the overall trend towards treatment rather than incarceration. That trend—towards viewing addiction as a medical, rather than a moral problem—has also been given a tremendous boost by the media focus on opioid addiction as a white problem. The reasons for this, of course, are awful: straight-up racism. But the goal is worthy, and one that at least some elements of the Trump administration might actually be on board with.
Criminalization only makes addiction worse. When addiction—and the supply of drugs—are medicalized, on the other hand, their connection with crime is greatly reduced. Once we stop seeing people with addiction through the racist lens of the drug war, we can start to treat them more humanely and simultaneously improve and save many, many more lives. And even if the next drug epidemic isn't as heavily white as this one, maybe the fresh memory of the opioid problem will force America to react a bit differently—or at least in a less egregiously racist way.
Follow Maia Szalavitz on Twitter.