Trendspotting: Is More Substance Abuse Pulling Down LFP?

NewsWire: 5/9/22

• Has substance abuse contributed to the decline in labor force participation? According to a new study, increased substance abuse accounts for as much as a quarter of the decline in prime-age LFP. (Bloomberg)
  
• NH: We have written numerous NewsWires on America's stubbornly low LFP. And we have offered multiple explanations: early retirements, fear of illness, long Covid, increased child care responsibilities, and a decreased desire to work. (See "Explaining Steep Drop in Americans' Desire to Work" and "Is Labor Force Participation Down for Good?") Now a new paper co-authored by the Atlanta Fed offers a further contributing factor: increased substance abuse.
• The paper argues that between February 2020 and June 2021, increased substance abuse contributed from 8.7% to 25.8% of the -1.5M person decline in LFP among prime-age workers (ages 25-54). Those are some significant percentages, so let's explore how the researchers came to that conclusion. 

• First, they needed to estimate the number of new substance abusers since the pandemic began. A substance abuser is anyone who uses opioids without a prescription or uses them in excess of their prescribed dosage. And all heroin and meth users are automatically classified as misusers. While the CDC provides preliminary data on overdose deaths (see “OD Deaths Continue to Skyrocket”), there are no recent surveys on the number of substance abusers. The last National Survey on Drug Use and Health (NSDUH) was released in 2019. 
• To get around this absence of recent data, the researchers averaged annual death rates of substance abusers before the pandemic began (deaths divided by the total number of substance abusers). They derived these average death rates for several pre-pandemic years (2015-18) and found that they were reasonably stable over time. Assuming that the rates remained constant into the future, they then took the number of overdose deaths between April 2020 and June 2021 above the pre-pandemic trend and divided them by their assumed death rates. They concluded that, over these 15 months, 27M prime-age people had a substance abuse disorder and 5M were new abusers. 

• The researchers then used previous NSDUH results to calculate the LFP rates of prime-age substance abusers. For example, the LFP rate for those with an opioid use disorder is 69.7%, -13.2 percentage points lower than an individual with no disorder. For each substance, they multiplied the number of abusers by the "difference between the LFPR of abusers and non-abusers." 
• The results showed that 2.7M prime-age individuals were out of the labor market due to substance abuse. 391K of those individuals developed abuse disorders during the pandemic. The authors then applied this cumulative total to the absolute decline in LFP from February 2020 to June 2021, which accounts for 25.8% of the -1.5M person decline. 

• Of course, this paper makes a few big assumptions. First, the authors assume that the number of "additional non-participants due to elevated substance abuse" in June 2021 was equal to the cumulative number in those 15 months. This was necessary because total deaths are reported annually, not monthly. So a June 2021 death rate could not be derived. The authors are probably OK in assuming that the number of new substance abusers by June 2021 would be at least as high as the 15-month average. 
• The more questionable assumption was that the death rate of substance abusers stayed the same from February 2020 to June 2021. We have frequently discussed the recent rise in fentanyl. The drug is so deadly it takes only 2-3mg to overdose, and its prevalence has undoubtedly increased the death rate. (See "Fentanyl Epidemic Accelerates.") The researchers do acknowledge this reality: It’s possible that the number of opioid users didn’t change and the increase in overdose deaths was from users switching to fentanyl. In that case, “the only impact on LFPRs is from the increase in the number of meth users,” which would explain only 8.7% of the decline in LFP.  
• IMO, both the death rate and the number of substance abusers increased. So the real impact of substance abuse is probably somewhere in the middle, perhaps around 200K people. Is this the biggest factor keeping LFP down? No. In November, we looked at survey data on reasons why people of all ages weren't working. (See "Reasons Americans Aren't Working.") We found that in October 2021, 3.7M Americans were home because of Covid-19 symptoms, and 5.5M simply "didn't want to be employed." The Dallas Fed also estimated there were +1.5M more retirements than expected between February 2020 and March 2021.
• Recently, LFP has been edging closer to pre-pandemic levels. As of April 2021, the LFP rate for all workers is -1.2 percentage points below February 2020, and the rate for prime-age workers is down only -0.6 percentage points. We expect this is mainly due to a combination of people unretiring, ebbing Covid-19, and schools reopening. As these pandemic-era drivers subside, drug abuse could actually become a larger share of the total drag on labor force growth in future months.

To view and search all NewsWires, reports, videos, and podcasts, visit Demography World. For help making full use of our archives, see this short tutorial.