I recently presented a series of DIY calculations for determining the size of the U.S. marijuana market. This post got a lot of play around the web, most notably from Andrew Sullivan, who again did me the kindness of adding his millions of regular readers to my regular readers (both of them). A number of commenters at his site, RBC and other web outlets (e.g., Washington Monthly) jumped on my estimate that the average price of marijuana was $120-$144/ounce (i.e., $120 with an upward sensitivity of 20%). Across the comments I read of those people who reported their marijuana purchasing experiences, the average price quoted was $425/ounce.
How does one account for the price estimate generated in this fashion being so much higher than, for example, reports from Oregon that indoor-grown marijuana prices are as low as $2000/pound ($125 per ounce) and prices for outdoor grown weed are as low as $700/pound ($44 per ounce)? Or the reports of law enforcement officials (who are often accused of overstating the cash value of drug seizures) from South Texas showing that marijuana prices have plummeted to $540/kilogram ($15/ounce)? Are some people fibbing or simply wildly off-base? In my view, no. All have won and all must have prizes because of the bifurcated nature of the markets for many drugs, including marijuana.
In the early 1980s, I was working with drug addicted people on the mean streets of Detroit, which were very mean indeed in those days. The mainstream media at the time and my middle class friends saw cocaine as a glamorous drug consumed by bankers and movie stars at prices upwards of $100/dose. My daily experience however was that cocaine was becoming an unglamorous drug (crack) purchased at $5/dose or less by streetwalkers, homeless people, gangbangers, and housing projects residents. Had the Internet of today existed then and I had posted that the average price of a dose of cocaine in the U.S. was very low, I would no doubt have been inundated with people saying (to paraphrase a commenter on my marijuana market post) “if Humphreys knew a place where cocaine sold for $5/dose, the world would beat a path to his door”.
But what I was seeing in Detroit was real, namely the emergence of a “downmarket” in cocaine. The “upmarket” was also real and was mainly composed of casual users who bought high-quality product. They paid high prices and often purchased their drug with bundled services (e.g., home delivery). The “downmarket” was mainly composed of heavy users who bought low-quality product at cheap prices. The upmarket was mostly Caucasian in racial makeup, with good social capital (e.g. college education, employment). The downmarket was mainly composed of lower-income people of color. The upmarket was not just rich people; it included many middle class and some working class people. But the downmarket was overwhelmingly composed of people living in poverty (with a few cross-overs now and then).
All of what I have said about the cocaine market in the 1980s could be said about the bifurcated market in marijuana today. And that makes assessment of average marijuana prices a challenging task.
At first blush, determining the true average price of a drug across a bifurcated market might seem as simple as summing the upmarket and downmarket price and dividing by two. But this is not correct because the frequency of use and drug purchasing are much higher in the downmarket, meaning that the cheap price must be weighted much more heavily than the high-end price when determining the average national price. For marijuana, a good survey for determining average national price would require a sample in which about 80% of the marijuana purchasers live in or near poverty, and the other 20% of purchasers were financially better off.
This of course is not the sort of sample breakdown you will get on the Internet. Internet users clearly “skew upmarket“: They are more educated, have higher incomes, are more likely to be employed and are more likely to be Caucasian than is the U.S. population as whole. Although I don’t have any data on the characteristics of Internet users who read Andrew Sullivan and debate public policy with Stanford University professors, I feel safe in assuming that if anything they are even higher in education and income than is the average Internet user. Finally, to the extent that online contributions to marijuana price estimates come from websites and groups advocating marijuana legalization, this is yet another layer of upmarket bias as those sentiments are strongest among white, highly-educated and affluent people. All of the above factors combined would led us to expect that the comments about my original post would very much reflect an upmarket perspective, as indeed they do.
By and large, the various comments quote prices for an upmarket product, namely sensimilla, which is only about a fifth of the marijuana market (the rest is much cheaper commercial grade product, most of which comes from Mexico). The comments often reflect the upmarket phenomenon of service bundling, e.g., “At my dispensary in San Francisco”. If you live in a public housing project in Dallas and the local drug dealer comes to his door when you knock and hands you a half pound of marijuana in a garbage bag for $300, the price truly reflects the price of the drug. But at a dispensary, with a physical location that you enter, staff on hand and an elaborate system to reduce your perceived risk of arrest (i.e., the doctor’s recommendation), you are buying marijuana bundled with other services and the “price of weed” you pay is in fact only partly the weed.
Let me illustrate the upmarket perspective concretely using two of the many posted comments:
From one of Andrew Sullivan’s readers: I have never encountered decent quality marijuana for less than $200/ounce in the western US.
From a Washington Monthly reader: I’ve been a user since ’68 and nearly everyone I know is a user. Not abusers but users. From my experience, the avg. couple uses daily, usually after work in the evening aka a glass of wine etc. Most couples will average 1-2 oz/month, especially upper income users of which there are many.
The world these two comments describe — the upmarket — is real, but is a small part of the total marijuana market. Sullivan’s reader insists on “decent quality” which is not generally a consideration for marijuana smokers in the downmarket. And the pot smokers described by the Washington Monthly commenter are not the people who do most of the purchasing; lower income people who smoke all day almost every day complete far more transactions and thus influence average prices far more. These two comments and others like them reflect a reality as surely as did Pauline Kael’s alleged remark that she didn’t know anyone who voted for President Nixon in 1972. But Nixon won a landslide for the same reason that the average price of pot in the U.S. is far cheaper than $425/ounce: The experience of a selected part of the population can be extremely different from the experience of the whole.