A few months back, The Economist had an article on the drug problem in Scotland. This statistic caught my eye:

Scots in the poorest areas are 16 times more likely to die a drug-related death than those in the richest places.

The difference between the overdose death rates for the rich and poor is too large to be attributed to mere coincidence.  This discrepancy cries out for some sort of  explanation.  

One possibility is that a lack of money makes people depressed, and depressed people are more likely to abuse drugs in order to ease the pain of poverty.  Another possibility is that both poverty and drug abuse are caused by a third factor.  How could we distinguish between these alternative theories?

The same article in the Economist has a graph which shows some equally startling differences in drug overdoes rates across countries:

There seems to be little or no correlation between the wealth of nations and the rate of drug abuse. Scotland’s per capita GDP is right at the average of the European Union, the US is far richer than the European average, while Portugal is much poorer than average.   And yet it is Portugal that has the lowest level of drug overdose deaths.

So how can we reconcile the fact that drug abuse within a country like Scotland is highly correlated with income, while at the international level one sees little evidence of poverty causing drug abuse?  One possibility is that both drug abuse and poverty are caused by a third factor.

Suppose that both poverty and drug use are correlated with some aspect of personality.  Also assume that all countries have a mix of personalities, some more susceptible to drug use than others.   And finally, assume that international difference in per capita income are not caused by difference in national personalities.  I’m not suggesting that any of these assumptions are completely true, rather I’m claiming that they allow us to understand why drug use with a country might strongly correlate with poverty, but not across countries.

If we assume that in both poor and rich countries a fairly stable percentage of the population has a personality that makes them susceptible to drug abuse, then we would not necessarily expect more drug abuse in poor countries than in rich countries. If we also assume that the same aspect of personality that leads to drug abuse also leads to poverty (say a lack of self control), then we can explain both the within country correlation and the lack of international correlation.

Case closed?  Not quite.  This doesn’t explain why there is such a large difference in drug overdose deaths between different countries (and even between different regions, as we see in the UK.)  Something else is going on, but poverty alone doesn’t seem to be an explanation. 

One possibility is that some countries have legal systems that are more tolerant of drug use.  But that also doesn’t seem to fit the data, as both Switzerland and Portugal rely more on a softer “harm reduction” approach, whereas the US and Scotland have much stricter enforcement of laws against drug use.  Indeed, it may be the case that strict enforcement of drug laws actually increases overdose deaths, by creating an underground market where drug quality is highly unreliable.

Instead, I suspect that drug use is correlated with two factors—personality and regional culture.  Regions of the world where people lack community support systems may have higher rates of drug abuse, and people with certain personality types within any given area are more likely to abuse drugs.  The worst situation of all would be people who lack self control and live in relative isolation in countries without strong support systems for individuals struggling with life.

In the US, the two states with the lowest rates of drug overdose deaths are Nebraska and South Dakota.  Both states have many small rural towns with a strong sense of community.  North Dakota used to be quite similar in a cultural sense, but the fracking boom in the Bakken region has brought in many younger workers who are isolated from their family and community.  North Dakota still has a fairly low drug overdose rate, but it’s now significantly worse than its two neighbors to the south.  And yet no one would argue that the oil boom has made North Dakota poorer—indeed just the opposite is true.  

To conclude, poverty probably does not directly cause drug overdose deaths.  But within any given region, the factors that cause poverty likely do correlate with the factors that cause drug abuse.