The Economics of Addiction

In 1999, fewer than 10,000 people died from causes related to drug overdoses in the United States. In 2021, the number was over 106,000.

Professor of Economics Jeremy Greenwood is working with Nezih Guner of Universitat Autonoma de Barcelona, Catalan Institution for Research and Advanced Studies, and Barcelona School of Economics; and Karen A. Kopecky of the Federal Reserve Board–Atlanta and Emory University, to study the numbers behind these numbers.

One of their working papers, “The Downward Spiral,” tries to answer a question that stretches beyond drug addiction: Why do rational people make risky choices?

“Why would they rationally elevate themselves to misusers—someone who doesn't follow the prescription or who doesn't have a prescription in the first place—recognizing that they could become addicted, and then if they become addicted, they could die?” Greenwood asks.

Greenwood says there are three major hypotheses about addiction postulated by economists in the last 10 years. In 2015, Anne Case and Angus Deaton found that working-age white men and women without four-year college degrees were dying of suicide, drug overdoses, and alcohol-related liver disease; these have come to be called “deaths of despair.” Then in 2021, Casey Mulligan identified “lethal unemployment bonuses,” which refers to the idea that financial support from the government subsidizes drug use for some. Others, such as David Cutler and Edward Glaeser, suggest that the crisis was driven by supply: cheap drugs becoming widely available. Greenwood believes a combination of the three has contributed to the current crisis.

In “The Downward Spiral,” which covers the period from 2000 to 2019, the authors used a computer model to follow the choices people make. “Before 2010, Purdue Pharma was selling Oxycontin and claiming that it wasn't very addictive,” Greenwood says. “They cited a study about opioid addiction, but the study actually said if you used opioids in the hospital under a doctor's directions, they wouldn't be addictive. So before 2010, I think it's unclear if people knew whether it was addictive or not.”

In 2007, Perdue lost a lawsuit. “After that, I think you can assume that people actually knew what the odds of addiction and the odds of death were,” says Greenwood. In fact, the team’s research showed that during that period, people underestimated the odds of dying from opioid addiction by just 13 percent. So why did people continue to become addicted?

Prices dropped, says Greenwood. “The government and various insurance programs paid for opioids. And then we had fentanyl, a synthetic opioid, that really lowered the price. Prices fell legally maybe 350 percent, and illegally maybe 150 percent. So that's a big stimulus to people taking opioids.”

The researchers’ second paper, “Substance Abuse during the Pandemic: Implications for Labor-Force Participation,” is already being cited and has received attention from the White House Office of National Drug Control Policy. There are multiple reasons for the drop in labor force participation during the pandemic, says Greenwood, including businesses shutting down and people not wanting to go out to work. He, Guner, and Kopecky wondered how much might have been caused by an increase in substance abuse.

“If you have an opioid use disorder, your labor force participation rate is 73 percent of a non-user. And if you are a meth user, it's even worse: 67 percent of a non-user,” Greenwood says. “Our hypothesis was that, if the pandemic caused an increase in substance abuse, then you should see a decrease in labor force participation.”

The researchers found that between 9 to 20 percent of the fall in labor force participation was due to an increase in substance abuse. They plan to continue their research into the current “normal times.” Across U.S. states, the number of people who state they are out of the workforce in the government’s Current Population Survey is highly correlated with deaths from substance abuse.

As for solutions, unsurprisingly, it’s complicated. A man interviewed in San Francisco who was content to live on the street and do drugs makes Greenwood ask, “If you enjoy drugs, what are we going to do? A lot of people think that is stupid. But I think it's a misperception. They just have different tastes.”

The researchers’ computer simulation also showed that easier availability of Narcan would reduce the number of deaths by overdose but would also increase opioid use. “If the odds are dying become lower because Narcan's available, then people are more likely to do it,” says Greenwood. Empirical evidence also shows that states where Narcan is widely available have a higher rate of opioid use, he says.

Likewise, legalizing drugs would reduce crime but create more substance abusers, many of whom will need assistance of some kind. “The interventions are kind of tricky because you have to take into account that people change their behavior,” Greenwood says.

Banning opium dens in the early 20th century may have led to a switch to other drugs. Before Prohibition, most people drank beer, but during Prohibition they switched to drinks with a higher alcohol content. “There's always been a fight going on,” Greenwood says. “The government's trying to regulate things. Then the black market switches to something else, the government tries to regulate that. And so it's like a cat and mouse game going back and forth.”

Greenwood says that, in the end, it may come down to demand. “There was no problem getting heroin and cocaine and all these things in the U.S. in earlier years and getting bootlegged alcohol from Canada in Prohibition. So I think we really have to kill the demand somehow. It’s difficult to do this and the options are limited. You can legalize drugs and regulate prices, quantities, and qualities. You can mount public information campaigns making people aware of the dangers of drugs. You can criminalize drug use, levy fines, mandate rehabilitation, and impose prison sentences,” he says. “Of course combinations of these policies can be used.”