Moschion and Powdthavee (2018) on Drug Use and Happiness

Julie Moschion and Nattavudh Powdthavee, “The Welfare Implications of Addictive Substances: A Longitudinal Study of Life Satisfaction of Drug Users.” Journal of Economic Behavior & Organization 146: 206-221, February 2018. [This outline is based on the pre-publication version, pdf available here.]

• People in general seem to have difficulty in predicting what will make them happy.

• It may be that drug users mispredict the utility that they will get from drugs, in contrast to the rational addiction approach. So, the authors look to see if people who start to use drugs sometime in the previous six months are happier or less happy. If they are less happy, then this suggests (but by no means proves) that drug use is not rational, but results from mispredictions of the hedonic payoff from drug use.

• Mispredictions of future utilities might stem from focusing illusions or projection biases. If you are feeling blue today, focusing illusions and projection biases might tilt you, excessively, in the direction of drug use.

• A higher future price of the addictive good lowers current utility for rational addicts, but increases it for mispredicting addicts, even though both “types” of addicts respond to the future price increase by decreasing current consumption.

• Some (but not all) previous work has found that higher cigarette taxes or public smoking bans have been associated with increased subjective well-being for (predicted) smokers. [Even sure-fire evidence that ex-smokers are happier than they were when they were smoking does not undermine rational addiction theory: rational addicts know that they will be happier if they give up smoking, they just find the costs of quitting to be too high to justify the transition.]

• The longitudinal, bi-annual data comes from a housing insecure Australian population, 2011-2014, with n exceeding 1,100. The life-satisfaction question is of the 0-to-10-scale, “all things considered” variety. This is the dependent variable, while the main independent variables concern past, current, and future substance use. This sample is more than four times as likely to smoke as the average Australian – but is less likely to engage in heavy drinking.

• The data reveal that daily smoking is quite persistent over six-month periods, and relatively easy to initiate. Weekly illegal drug use is not very persistent, though cannabis consumption is quite persistent.

• Lower life satisfaction seems to precede daily smoking, daily pot use, and weekly illegal drug use, with the illegal drug use result being most robust. Current (high) consumption of any of the substances, including alcohol, seems to lower current happiness. Weekly street drug use is similar to getting a divorce, in terms of life satisfaction. The negative effects of illegal drugs tend to be worse for women and for the less educated.