Under the influence of the market: an applied study of illicitly selling and consuming heroin

Illicit market activities are associated with considerable financial and psychological harms for drug users and third parties, and impose significant demands on public services. Since policy interventions interact with the market, evaluations of cost-effectiveness necessitate knowledge of how the market operates. There seems to be emerging consensus in the economic literature that economic factors, such as prices, can influence consumption patterns of addictive commodities. However, the descriptive validity of the notion of ‘maintenance doses’ have provided the rationale for predictions about the shape of the demand curve, and particularly for changes in slope, or ‘kinks’. Data collected from over 500 heroin users contacted outside a needle exchange service in Oslo are analysed in this paper. Wide variation in levels of consumption, income and prices are reported. Using a switching regression model of heroin consumption with endogenous switching between dealing and non-dealing, the influences if economic factors on the decision to deal and the level of consumption are estimated. Drug users in employment are estimated to be almost half as likely to sell drugs and a 10% increase in non-dealing income is estimated to reduce the probability of dealing by 3%. Non-dealing users’ consumption patterns are found to be price-elastic (elasticity = 1.23) and significantly influenced by the level of personal income (elasticity = 0.47). Influences on the extent of drug sales are also considered. Whilst appearing less responsive, dealers’ market activities have predictable relationships with economic factors and market conditions. For example, a 10% increase in the proportion of active dealers arrested in the previous month is found to reduce the size of the dealing population by over 3%. There is, however, some evidence that the remaining dealers are more active, such that attempts by enforcement agencies to reduce the size of the overall market are offset. No evidence of kinks in the demand curve for heroin are identified for the range of prices quoted on the Oslo market over a 13-month period. Given the multitude of influences on both the costs and benefits of consumption and the observed heterogeneity of drug users and their patterns of behaviour, this may not be surprising. The high price responsiveness of non-dealers’ heroin consumption and the importance of income levels in predicting behaviour emphasises the importance of policies which succeed in manipulating the economic conditions faced by illicit heroin users.