Price Theory and the War on Drugs

Lessons from taking price theory and transaction costs seriously

You are reading Economic Forces, a free weekly newsletter on economics, especially price theory, without the politics. Economic Forces arrives weekly in the inboxes of over 14,000 subscribers. You can support our newsletter by sharing this free post or becoming a paid subscriber:

---

When I was in grad school, I had a professor named Li Way Lee. He was a pure price theorist and a great teacher. His teaching always seemed spontaneous. He would show up to class with The Wall Street Journal and he’d hold it up and say, “every article I read in here is about market share” and then he’d hold up the textbook and say “nothing in here mentions market share as an objective. Isn’t that odd?” From there, we were off, exploring all aspects of firm decisions to determine how to reconcile the differing emphasis in the textbook and the popular press. In his Industrial Organization class, he made a bunch of would-be economists read anti-trust cases and debate the underlying economic arguments.

He also wrote papers on seemingly random topics that would only cite maybe five other papers — and yet he published very well. He wrote a paper on the allocation of time when people face transaction costs, which is capable of explaining a lot of otherwise puzzling behavior we observe in labor markets. He also wrote a paper on car pooling. And another on regulation as a form of exchange (something that Brian wrote about previously). Today, I want to write about one of his other papers, on the market for illegal drugs.

A common argument that you hear from economists is that government policies designed to reduce drug use are aimed at the wrong side of the market. These policies tend to be designed to punish drug dealers, producers, and criminal organizations. In other words, they target the supply-side of the market. A reduction in supply will reduce the equilibrium quantity and raise the equilibrium price. Critics argue that this is a bad policy because it raises the relative price of illegal drugs and as a result will encourage violent competition between competing suppliers. Instead, these critics argue that policy should be aimed at the demand-side since a reduction in the demand for drugs will not only lead to a reduction in the equilibrium quantity of drugs sold, but will also reduce the relative price of drugs.

This view, however, oversimplifies matters. The market for illegal drugs features some transaction costs that are absent from other, legal markets. In addition, some of the transaction costs faced by sellers are actually contingent on the behavior of the consumers. As a result, we need to modify our typical supply and demand analysis to incorporate this behavior and these costs. And by doing so, we can evaluate policies and potentially explain otherwise paradoxical behavior.

In thinking about the market for illegal drugs, consumers broadly face three distinct costs: (1) the cost of purchasing the drug, (2) the expected cost of being caught purchasing the drug, and (3) the cost of being caught in possession of the drug. It is therefore important to dig into the incentives created by these various costs. The law of demand tells us that if any of these costs rise, the quantity demanded will decline. However, these latter two costs require a bit more thought.

Consumers have an incentive to minimize costs. The expected costs from purchasing a drug can be thought of as the probability that one is caught in the act of purchasing the drugs multiplied by the punishment. Consumers face this expected cost whenever they engage in a transaction. One way to minimize this cost would be to reduce the number of transactions. Nonetheless, for a given level of desired consumption, fewer transactions means larger transactions. Consumers who engage is larger transactions are likely to then hold larger inventories of the drug, making it more likely that they will be caught in possession of drugs and subject to those penalties. This reveals an additional tradeoff that is absent from an analysis of legal markets.

On the supply side, there are also additional costs that are absent from legal markets. For simplicity, let’s assume that the supply-side is entirely vertically integrated from production to sales (which doesn’t seem totally unreasonable) and that the only transaction cost is the expected cost of being caught selling drugs. The marginal cost to the supplier not only includes the standard marginal cost of production and distribution that any firm would face, but also this additional expected cost of being caught and punished. The latter could occur from selling drugs to an undercover police officer of simply that the sale was observed by law enforcement. The expected cost is the probability that this occurs multiplied by the punishment.

Again, the supplier faces this cost whenever a transaction takes place. The more transactions that take place within a particular period, the higher the marginal cost. Nonetheless, there is a crucial difference here. What the criminal organization primarily cares about is the total quantity that it sells. At the same time, the consumers largely determine the number of transactions and the size of purchases. What this means is that not only do suppliers face additional costs, but also that the magnitude of those costs is, in part, determined by the choices made by consumers. The typical relationship between the supplier’s marginal cost curve and its supply curve does not hold. In fact, the supply curve is steeper than the marginal cost curve.

All of these additional details are important. Nevertheless, the reader might wonder to what extent any of these things (admittedly real world costs) make a difference in terms of supply and demand. To illustrate why these details are important, I will provide some examples from the paper.

In his paper, Lee brings up the experience with the 1972 National Commission on Marijuana and Drug Abuse. The commission’s report suggested that going after drug users was ineffective and recommended decriminalization of users (but not suppliers). Eleven states followed this recommendation. However, when researchers examined the effects of decriminalization, they found some puzzling results. A standard supply and demand analysis would suggest that decriminalization of users would reduce the cost of drug use and increase demand. The net result would be greater consumption and a higher price. In contrast, researchers found that drug use actually declined.

Some people attributed the decline to a “forbidden fruit” argument, in which people place greater value on goods that are illegal or otherwise hard to obtain. This claim is difficult to justify. In addition to the evidence of a modest decline in consumption, corresponding survey evidence found that there was little change in attitudes about drug use following decriminalization.

Lee’s model provides a better explanation. It is true that decriminalization, by lowering the cost of purchasing drugs, would increase demand. However, that is not the end of the story. It is also true that users facing a smaller (or zero) penalty for engaging in transactions would encourage consumers to engage in more transactions. Since decriminalization only applied to users and not suppliers, a greater frequency of transactions increases the marginal cost to suppliers. Evidently, the corresponding reduction in supply was larger than the increase in demand and the quantity of drugs consumed declined.

A similar analysis can be used to understand the effects of the policies brought about by the War on Drugs of the 1980s. Despite the organized effort to reduce drug use, there is actually  evidence of an increased supply. In fact, even surveys of high school students in the late 1980s revealed remarkable answers about how much easier it was to obtain heroin and cocaine. The simple model cannot explain this. In fact, it is precisely the opposite as what the simple model implies. Greater punishment should mean a lower equilibrium quantity.

By considering these additional transactions costs, one can make sense of this result. For example, there were two changes with respect to users. The first is that they were more likely to be arrested, but over time faced lower penalties for drug possession. The main effect of this policy was therefore to increase the cost of transactions while reducing the cost of holding inventories. Consumers were therefore more likely to engage in less frequent, but larger transactions to minimize costs.

All else equal, drug dealers faced greater expected punishments and this would tend to reduce supply. Yet, all else wasn’t equal. The fact that drug users were incentivized to significantly reduce the frequency of transactions in favor of larger transactions meant that dealers were less likely to get caught and face punishment. All else equal, this lowers the marginal cost faced by suppliers and encourages the supplier to produce more. These two effects work in opposite directions. Determining which effect dominates is an empirical question. Evidently, the latter effect dominated, thereby increasing supply. The paradoxical result is thus explained by the model.

This paper is now 30 years old. It would be interesting to see how it’s conclusions hold up to subsequent changes in drug policy and enforcement. Regardless, there are some broader lessons I would like to draw.

Policymaking is difficult. Even policies that seem like obvious solutions to a problem might not produce the desired result. Brian and I love to say “it’s all just supply and demand.” When people see us at conferences and other events, they refer to us as the “the supply and demand guys.” That’s great. However, it is important to keep in mind that one of the central tenets of price theory (at least as we present it here each week) is that the proper analysis requires thinking through all of the relevant costs. Each cost introduces a new margin upon which people will choose to adjust. Thus, when a simple supply and demand model yields paradoxical results, it is often because the analyst has left out a relevant cost and therefore a relevant margin for analysis. It is the job of the researcher to identify the proper margin. Li Way Lee was great at this and we should all strive to be as well.