High and dry?

The market for illicit drugs is seen as the cause of many social ills. The trade in illicit drugs gives rise to an underground economy that generates addiction, crime, and violence. In less affluent and minority communities, the drug economy crowds out the incentives to join the formal sector and raises incarceration rates. In an effort to counter these trends, massive amounts of resources are devoted to interfering with the drugs market – the so-called 'war on drugs'. The emphasis of the war on drugs is on tougher penalties and more law enforcement. In fact, there is little evidence that recent efforts to increase penalties and law enforcement have measurably reduced the availability of drugs. On the other hand, the prison population convicted of drug-related crimes has skyrocketed. (In the US, for example, the prices per pure gram of cocaine and heroin have declined substantially during the periods when budgets on law enforcement rose and penalties increased. At the same time, in the period 1981-2003 the prison population convicted of drug-related crimes has shot up ninefold.)

Partly in response to these twin failures, some quarters (including, prominently, the Economist magazine) have advocated liberalisation. Becker et al. (2006), for instance, state the argument in favour of liberalisation as follows. Penalties and enforcement aim to decrease consumption by raising the unit cost of producing/distributing the drugs, and thus ultimately the price faced by consumers. Whereas such interdiction is costly to enact and to evade, taxes could be levied at low administrative cost. Therefore, drugs should be regulated by taxes rather than interdiction. Of course, this argument works to the extent that the government has the power to tax 'legalised' drug sellers without driving them underground. Realistically, even in a world with legalised drugs there would be 'illegal' (i.e., tax-evading) drug sellers, thus limiting the efficacy of taxes as regulatory instruments. (The case of tobacco in Britain is instructive. In response to high duties on tobacco in the UK, a large underground market developed reaching 20 per cent of total tobacco expenditure between 1995 and 1999. See Cullum and Pissarides (2004).)

California Governor Arnold Schwarzenegger recently kicked off a debate on the legalisation of marijuana – though, admittedly, as a measure to deal with the California budget crisis. The position of the current US administration has not yet fully emerged. On the one hand, President Obama pledges 'full partnership' in Mexico's war against the drug cartels. On the other hand, the background of the recently nominated 'drug czar' portends an emphasis on decriminalisation (Johnson and Goldstein 2009).

In a recent working paper (Galenianos et. al. 2009), we argue that the dichotomy between war on drugs and decriminalisation may be a false one – a carefully designed scheme of penalty reduction can, surprisingly, interfere with the functioning of the drugs market.

Understanding the drugs market

Currently, we think of illicit drug markets as centralised markets, in which the usual demand and supply curves meet and form a market-clearing price. But this view fails to account for some key economic forces that shape the retail drugs market. The first, and most important, is that retail transactions for illegal drugs are subject to significant moral hazard. What we mean is that the seller can covertly dilute ('cut') the product, and this dilution is largely unobservable to buyers until after they consume. The following table, which is based on data from undercover Drug Enforcement Administration (DEA) purchases, shows that moral hazard is indeed present in this market. The table documents an extreme instance of the moral hazard – the rip-off, a transaction in which the buyer is sold essentially zero-purity drugs. A significant fraction of "street-level” transactions are seen to be total rip-offs. Most important, the price paid in a rip-off is not appreciably different from that of a non-rip-off transaction, suggesting that buyers cannot observe dilution.

The practice of selling drugs in branded bags ('dope stamps') is further corroborating evidence of a quality problem in the illegal drugs market. Dope stamps could be boasts of quality ('America's Choice', 'Dynamite'), status brands, and even corporate names. The purported effect of a dope stamp is quality certification. However, because the stamps can be faked by "unscrupulous” competitors, the certification value of a dope stamp is limited and often very short-lived (a couple of days, often) (Wendel and Curtis 2000).

If this opportunistic behaviour is possible, why is it not more prevalent? And, indeed, why doesn't moral hazard foreclose the possibility of trade? The answer lies in the possibility of long-term relationships between buyers and sellers. A seller who wants to keep a customer will not rip him or her off. Long-term relationships are the second key feature of the drugs market.

The third basic fact is the presence of considerable dispersion in the price/quality ratio. As an example, let's look at the distribution of pure quantity of crack cocaine traded for $20 in Washington, DC in the period 1989-1991. This empirical distribution shows a great deal of dispersion in the quality obtained for $20. This is further evidence against the view of a centralised market because in such a market the "law of one price” ought to hold. Based on the facts laid out above, we propose a new perspective on illicit drugs markets.

In our view, moral hazard and search frictions (the costs incurred by a buyer in searching for a new seller) are key to these markets and put them at risk of collapse. We therefore propose a new model in our paper that properly recognises the role played by these two economic forces. We think it works because it predicts dispersions that are similar in shape as those evidenced in the above example of the crack cocaine market.

Potential drug policies

This new view of the drugs market leads us to evaluate policy in a more nuanced, sometimes counter-intuitive way. For example, the analysis suggests alternative channels to suppress the market. If it's true that the market is undermined by moral hazard, then economic theory suggests policy-makers could leverage the moral hazard, i.e., induce sellers to dilute more. This could be achieved through a policy of reducing the sentences of sellers who 'cheat' and sell low-purity drugs. This policy would have a similar effect as increasing the wholesale price of drugs – a key objective of the war on drugs, and one that is pursued at great cost. However, the policy we propose can be implemented for free. In fact, the direct effect of this policy is to save money by reducing incarceration rates. This is a goal that is very important in its own right, of course, quite apart from any monetary savings, and that the current war on drugs definitely fails to achieve. Finally, the policy could easily be implemented with minor changes to the existing sentencing guidelines.

Ideology and politics, of course, plays a large role in the debate on drugs policy. While the strategy of shortening sentences for 'high diluters' looks a little like decriminalisation, it is the kind of decriminalisation that conservatives might find more acceptable, because it would undermine the drug trade. Therefore, the dichotomy between the war on drugs and decriminalisation may indeed be a false one. Certain kinds of 'smart decriminalisations' could be found which, rather than interfere with the war on drugs, might help achieve its very objective.

There are, of course, potential unknowns associated with the policy we are suggesting, such as the possibility of increased violence in these markets, or the possibility of an increase in overdoses. (Whether higher rates of dilution increase or decrease the health risks of users depends on the dilutant. Overdoses occur most frequently when a user consumes drugs of unexpectedly high purity.)

Still, the likely advantages are sufficiently important, we think, that this policy is worth considering seriously.

Manolis Galenianos is an Assistant Professor of Economics, Pennsylvania State University.

Nicola Persico is a Professor of Economics, New York University

Originally published on www.VoxEU.org. Reproduced with permission.