Rational Addiction to Heroin and Other Drugs

In Summary

A popular approach that economists take to study the use of illegal drugs is the concept of rational addiction. The idea is that individuals willingly choose to consume heroin and other drugs, as they would do with food, goods or services. While this idea is controversial, it allows both researchers and policymakers to view the issue related to drugs (declining physical and mental health, social stigma, dependence) beyond common moral barriers. Policymakers are usually hesitant to act on these issues without the cover of people’s safety (not the ones consuming) – not usually in sync with the well being of the drug users.

The article explains mathematically how drug consumption relies on both one’s history (individual capital) and one’s social environment (social capital), in the same way as any other good one would consume (yes, including bread, donuts and whatnot). The article also covers the two current strategies in place to prevent and cope with drug use: the war on drugs and harm minimization.


A rational addiction – why do people consume heroin?

Drugs are a great example of a good that is associated with economic “bads” – i.e. goods or outcomes that produce a negative effect on the consumer and are undesirable (Varian, 2010, p. 41). Consuming an opioid like heroin may lead to several biological adverse effects like nausea, vomiting, loss of memory, overdose but also to social negative events such as marginalization and imprisonment – effects that are usually known by the consumer. On the marketplace, “bads” have a negative price, i.e. one would need to be paid or receive something in exchange of a “bad”. When one consumes heroin, he/she makes the choice to accept some level of the adverse effects in exchange for some level of the benefits procured by the consumption of the drug (cf. Figure 1.1 below). This conception of the trade-off that a drug user is ready to make lead us to believe that he/she makes a rational choice.

Rational addiction
Figure 1.1: Utility curve for the consumption of heroin, comparing the bads and the good it procures to the consumer. A consumer only agrees to get quantity Qbad of bad if he/she gets a quantity Qgood of good.

When Becker and Murphy proposed the model of rational addiction, they choose to assume that an individual has good information about the drug and about its addictive property. Allowing this condition to take place, they can assume that consuming the drug is the product of the individual’s rational choice and thus that the drug is bought and consumed on a market just like any other product. Rational addiction is a dynamic model that builds on past consumption – identified as “stock of capital” – to determine present consumption of the drug. Such modelling allows economists to introduce the strong biological effects specific to addictive drugs as endogenous variables, such as reinforcement, when higher current consumption increases the future consumption of a good, and tolerance, when, for a given level of consumption, the same level provides a lower utility in the future – i.e., the consumer must continually increase her/his consumption of the good (Becker & Murphy, 1988, pp. 681–682). This “rational” consumption could well be applied to other goods that produce a pleasant experience to the consumer that may reinforce any further consumption in a future period of time.

This dynamic model was at the root of Becker’s broader approach on individuals’ preferences. Expanding from the idea that a stock of capital inherited from the past was key to determine one’s motivation to further consume drugs, Becker further developed the notions of “personal capital” and “social capital” in an attempt to conceptualize how each individual’s preferences are shaped by their past consumption and by their peers’ consumption (Becker, 1996). “Social capital” as conceived in this model helps us understand the political interest for a strong policy against drugs. It also shows that this issue extends beyond the individual to her/his social environment.

In Becker’s approach on individual preferences, utility at time t is:

(1.1)     u=u(x_{t},y_{t},P_{t},S_{t})

With x_{t}, y_{t} different goods on the market, P_{t} the stock of personal capital at time t, and S_{t} the stock of social capital at time t. The personal capital (that corresponds to the “stock of capital” in Becker and Murphy’s article) evolves over time, subject to one’s consumption of the good (x_{t}, Becker speaks of “amount invested in personal capital”) and to a constant depreciation rate of personal capital (d_{p}) – equation (1.2).

(1.2)     P_{t+1}=x_{t}+(1-d_{p}) \times P_{t}

“Social capital”, or the influence that the preferences of one’s social network have on one’s preferences, is another important notion that help understand the rational consumption of drugs. In Becker’s model, social capital (S_{t+1}) is form of the sum of the consumption (X_{i}) of every member (i) of one’s social network added to the social capital of the previous time period (t).

(1.3)     S_{t+1}^{i}=X_{t}+(1-d_{s}) \times S_{t}^i     with X_i=\sum x_i

Individuals’ preferences are intrinsically (endogenously) influenced by their peers’ preferences – preferences that are exposed in their consumption. A heroin addict will certainly be encouraged to consume the drug regularly if he/she hangs out with other heroin addicts. Inversely, he/she may also choose the company of other heroin consumers for their common interest in that good. Globally, the utility of the good increases in part because it produces the positive externality of belonging to the group (Becker, 1996).

“But while individuals do not have much direct influence over their social capital, they often have an enormous indirect influence over it, since they try to become part of social networks that benefit rather than hurt them. […] This endogeneity of social networks creates a tendency for social capital to raise rather than lower utility” (Becker, 1996, p. 13).

Now, this model can only provide a rationale for the addiction of a drug user, but it cannot explain how the consumer chose to use heroin in the first place. In Becker and Murphy’s model, this first choice to consume is the result of an exogenous event that it cannot account for, nor explain. In their article, they use the example of veterans of the Vietnam War and the trauma they went through to identify such exogenous “temporary” event (Becker & Murphy, 1988, p. 691). Do individuals who become drug users see their future differently or have more preferences towards risky behaviors than those who are not interested to use drugs? Theoretical and empirical studies applying the rational addiction model looked into potential divergences in time preferences and in risk aversion between drug users and non-users (Blondel, Lohéac, & Rinaudo, 2007; Orphanides & Zervos, 1998). A difference in time preferences could suggest that drug users have a significantly different discount rate of future consumption than non-users. If future utility is strongly discounted, the drug user would experience a sort of “myopia” and disregard the effects of her/his present consumption of drugs (Orphanides & Zervos, 1998). Testing this hypothesis empirically, Blondel et al. (2007) found that while drug users do not seem to discount the future more heavily, they do seem to display more risk-seeking behaviors.

With this conception of drug addiction, drug users seem to make an informed and rational choice that maximizes their utility accounting for their preferences (time and risk): one is willing to accept the bads in exchange of some of the good and since there is reinforcement and tolerance, the consumer may increase her/his consumption of heroin over time – building on her/his personal capital. This situation implies that if one cannot pursue this increasing consumption due to an unaffordable high incremental cost, he/she will get a diminishing marginal utility over time of their constant level of consumption – potentially leading the individual to stop consuming the drug completely. The consumption pattern of the people an individual chooses to relate with (social capital) will encourage her/his consumption of the same good in the future. A social network of heroin addicts will increase the utility of consuming heroin, but a social network of ex-addicts who decided to quit consuming may stimulate the utility for one to also quit.


Reducing the burden of drug use and abuse

Economists usually refrain from intervening in people’s choice of consumption: assuming an individual makes a rational choice to consume, their taste and preferences are not disputable. After all, “de gustibus non est disputandum”. However, economists do recognize the existence of market failures for which government intervention may be advised. The existence of costly negative externalities to the consumption of drugs are a cost that the consumer does not face alone when choosing to consume. Those costs affect both the individual her/himself and his social environment through declining health, impaired driving, rape or violence, but also the society as a whole as every process involved in the supply of the drug contributes to other unlawful activities and trafficking (such as the guerrilla of the armed group FARC in Columbia, cartels in Mexico). While he/she may be making a rational choice to consume, the drug addict does not necessarily take in account these externalities. To respond to these issues, governments employ two different strategies: a “war on drugs” and “harm minimization”.

The former relates to the policy of prohibition of drugs, severely enforcing the laws against production, dissemination and use of drugs like heroin or marijuana. In economic terms, this strategy is an attempt to eliminate the demand and the supply of drugs to such level that the price at which supply meets demand is discouraging individuals to pursue their consumption of heroin. In a city like Baltimore, this strategy implies the maintenance of a strong police force with surveillance and intervention capabilities. It means that people are not allowed to consume this good, despite a consumption pattern that seems to follow “rational” choice. By discouraging the development and the use of drugs, the individual’s social network may become the main influence in the individual’s further consumption of the drug. Not only the social capital increases the utility of the drug to the individual, but it also allows easier (thus cheaper) access to its supply. Relatively to the “war on drugs”, this implies that if the police is not able to reach into some social networks, these networks will keep their drug use and attract more drug addicts. This makes the overall trafficking and use of drugs, particularly heroin, concentrate in the city’s criminal circles which may drive drug users to also engage into other unlawful activities in exchange of the participation to the group. In this case, the social capital theorized in Becker’s preference model may explain how smaller groups of people with very high preference for heroin will exist and prevent the drug from being totally absent of the market. In this respect, while the “war on drugs” is necessary to control the dissemination of harder and more addictive drugs, like heroin or cocaine, it is not completely capable of ensuring the safety of others – the primary negative externality we aim to prevent.

Meanwhile, “harm minimization” relates to a “softer” approach which objective is to provide relief, support and social reinsertion to drug addicts to limit the extent of harm of the drug and to encourage drug users to quit. Programs like prescription of methadone or other medical substitutes, safe needle site injection centers, counseling in missions and community centers are examples of such strategy. The first program provides drug users with substitutes to heroin that are less damaging (with less “bads”) like methadone, buprenorphine or naltrexone. The second program actually provides a complementary good to heroin by providing clean syringes and a confidential safe site to use them. Contrarily to accessible substitute goods, access to complementary goods usually increases the use of the initial good. Many people argued that providing such centers legitimized and promoted the use of heroin and other injectable drugs; there was no evidence of such increase in use, while researchers assessed lower criminality among drug users (Hathaway & Tousaw, 2008). The third program usually acts directly on the personal and social capital of a drug user, provided support and guidance on how to change her/his life habits that drive the use of heroin.

The “war on drugs” may be inevitable for harder and more biologically addictive drugs to lower their supply, but it may lead to the formation of strong cohesive groups of users who have to resolve to criminal ways to provide for their consumption and protect the group. In the light of past events, this may be a barrier to further social promotion in the poorer strata of the society. “Harm minimization” strategies may help individuals escape from these social group and, with the right support, support them in their reinsertion in society. The city is constantly working to create new opportunities of community work to not let the success of social reinsertion at the whim of luck.



I would like to thank Dr. John Bridges for his pedagogical feedback in the development of this essay.



Becker, G. S. (1996). Accounting for taste. London, UK: Havard University Press.

Becker, G. S., & Murphy, K. M. (1988). A Theory of Rational Addiction. Journal of Political Economy, 96(4), 675. http://doi.org/10.1086/261558

Blondel, S., Lohéac, Y., & Rinaudo, S. (2007). Rationality and drug use: An experimental approach. Journal of Health Economics, 26(3), 643–658. http://doi.org/10.1016/j.jhealeco.2006.11.001

Hathaway, A. D., & Tousaw, K. I. (2008). Harm reduction headway and continuing resistance: insights from safe injection in the city of Vancouver. The International Journal on Drug Policy, 19(1), 11–6. http://doi.org/10.1016/j.drugpo.2007.11.006

Orphanides, A., & Zervos, D. (1998). Myopia and Addictive Behaviour. Economic Journal, 108(446), 75–91. Retrieved from http://onlinelibrary.wiley.com/journal/10.1111/%28ISSN%291468-0297/issues

Varian, H. R. (2010). Intermediate Microeconomics (8th ed.). New York, USA: W. W. Norton & Company.

Print Friendly, PDF & Email
The following two tabs change content below.

Gatien de Broucker

Health economist / Économiste de la santé at Johns Hopkins University

Latest posts by Gatien de Broucker (see all)

Leave a Reply