Two empirical studies in the economics of alcohol abuse
Date
1988
Authors
Jones, Daryl Ross
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The issue of alcohol abuse has been studied by all the social sciences. This thesis examines two areas studied by economists. These are production losses caused by abuse and production returns to treatment.
There are three ways that alcohol abuse can have an adverse impact on the level of production. First, studies have shown that abuse has a detrimental impact on attendance and productivity. Secondly, drinking problems can become so acute that they render the abuser unemployable. Finally, alcohol abuse can cause premature mortality.
Chapter 2 estimates lost production due to alcohol abuse in British Columbia in 1979. The estimates are prepared according to alternative assumptions regarding unemployment; a full employment assumption and a replacement worker assumption. With the latter assumption, costs are restricted to losses from employed abusers (i.e., reduced efficiency and increased absenteeism). These factors are estimated to reduce the production of 50,000 workers by approximately 25 percent. This is estimated to have decreased 1979 production by almost $200 million. Under the full employment assumption, it is estimated that the production cost is about $329 million. The increased cost results from 2,652 unemployable abusers (estimated cost, $34 million) and the premature death of 800 individuals (estimated cost, $95 million).
Chapter 3 develops two cost-benefit models to estimate production returns from treatment (i.e., one for each of the foregoing assumptions). Presumably, if the abuser quits drinking the production problems and their ensuing costs are eliminated. If a clinic is responsible for abuser sobriety, the elimination of future costs is a benefit of the treatment program.
As abusers, employers, and the government share the benefits of successful treatment, the models are adapted to distribute the costs and benefits among these three groups. In addition, chapter 3 presents an empirical application of the model. It examines 1985 production returns to in-patient treatment using data from the Dallas House program (Victoria, British Columbia).
Each cost-benefit equation is estimated using sensitivity analysis and benefits are assessed with 108 variable combinations. The results indicate that, with both models, treatment produced a net production benefit for all three groups. This result was achieved because the government provides the majority of funding for the Dallas House treatment program. The benefits to employers, under many variable combinations, were not sufficient to offset total treatment costs.
This result differs from several previous studies which found that occupational treatment programs offer significant benefits to employers. The discrepancy is caused by the fact that previous studies implicitly assumed that all the production benefits of
treatment accrue to employers.