Photographer: Bill Oxford/iStockphoto

Harvard-Educated Doctors Prescribe Far Fewer Opioids

Physicians from the lowest ranked schools hand out three times the pills

Economists digging into the data behind America’s opioid addiction have uncovered a key trend in prescribing practices: doctors who went to top-ranked schools are dramatically less likely to hand out the pills. 

The study, by Princeton University economists Molly Schnell and Janet Currie, leads this week’s economic research roundup. It’s followed by a related study that suggests we may be underestimating how lethal America’s opioid crisis has become. Then we take a peek around the world, where major labor inflows are fading, and end on a wonky look at how different flavors of central bank quantitative easing affect growth. Check this column each week for new and pertinent research. 

Doctor education and opioids

Doctors who graduated from top schools prescribe way fewer opioids, according to this new National Bureau of Economic Research working paper. Looking at data spanning 2006 to 2014, they find that physicians trained at the lowest ranked U.S. medical schools prescribe nearly three times as many opioids per year as those educated at Harvard Medical School, which is the top-ranked program. Looking specifically at general practitioners, those trained at Harvard wrote 180 prescriptions a year, compared to 550 per year for the lowest-ranked. 

Why is this happening? The authors say their results suggest that it’s probably medical training: the gap exists between prescribers from different schools who work at the same hospital, for instance, arguing against patient selection. Why do we care? While prescribing is slowly leveling off, pill addiction remains a widespread problem. Stemming its spread is important to public health and workforce readiness. 

Addressing the Opioid Epidemic: Is There a Role for Physician Education?
Published August 2017
Available at the NBER website

Worse than we thought

Speaking of opioids, we may be significantly underestimating how many people they’re killing in the U.S. Up to a quarter of death certificates don’t note the specific drug that caused an overdose, and Christopher Ruhm at the University of Virginia is trying to correct the data to account for that, in part by looking at the characteristics of the non-classified deceased and estimating the statistical chances that they were using opioid pills or heroin.

Click here for more on the heroin and opioids epidemic.

He finds that nationally, pill and heroin overdose deaths were probably 24 percent and 22 percent higher than reported rates in 2014. The rates were understated in states including Pennsylvania and Indiana and overstated in places like South Carolina and New Mexico. 

This piece of research isn’t economic, per se, but it’s getting a nod here because it has economic implications. If the epidemic is worse in reality than it looks on paper, it could be locking more people out of the labor market than we’d otherwise expect. What’s more, it could be fueling discontent: other research has showed a correlation between so-called “deaths of despair” and populist voting in the last election. 

Geographic Variation in Opioid and Heroin Involved Drug Poisoning Mortality Rates
Published Aug. 7

Forthcoming in the American Journal of Preventative Medicine

Is the world’s labor boom souring? 

The global economy is on the cusp of a major reversal, London School of Economics’ Charles Goodhart and Talking Heads Macro’s Manoj Pradhan write in this Bank for International Settlements paper.  Between the 1980s and 2000s, developed nations saw an unprecedented increase in labor supply thanks to demographic trends and China and Eastern Europe’s inclusion in the World Trade Organization. It also experienced what the authors call a demographic “sweet spot”: dependency ratios fell as people lived longer and had fewer children. Together, the trends allowed a huge amount of people to work, pushing down wages, slowing gains in marginal labor productivity and driving up inequality.

Now that’s changing as the world’s working-age population shifts into slow-growth mode. That could boost inflation and cut income disparities. It could also push up interest rates as savings fall faster than investment, though the authors admit that this proposition is controversial – many economists think that population aging will slow growth and drag on interest rates. What’s more, debt overhang could make a global decline in interest rates hard to reverse. 

Demographics will reverse three multi-decade global trends
Published August 2017
Available on the BIS website

QE isn't created equal

Central banks’ large-scale bond purchases vary in effectiveness, depending on what type of asset is purchased. Fed economist Robert Kurtzman and University of Southern California economist David Zeke argue that purchases on non-financial corporate bonds might create distortions in the cost of capital among companies, as those whose bonds are bought see lower financing costs. That could cause capital mis-allocation.  With government bond purchases, on the other hand, cheaper capital costs are spread between both small and large firms more evenly. 

The end result? A quantitative easing program that buys large-firm corporate debt can come with mis-allocation that’s large enough to make the government bond purchases a more effective tool for increasing output, at least when rates are higher than zero. 

Misallocation Costs of Digging Deeper into the Central Bank Toolkit
Published August 2017
Available on the Federal Reserve website


    Before it's here, it's on the Bloomberg Terminal.