What America's Users Spend on
Illegal Drugs
1988–1998
December 2000
Appendix C
Drug Prices
Several sources report prices paid for illegal drugs.1 The problem with those sources, for present purposes, is that they report prices as broad ranges, unsuitable for the calculations used in this report.
Recent studies provide a method for estimating prices within a narrower range.2 Basically, this methodology is to estimate the price paid during market transactions (completed by police as undercover agents, and hence reported to a data source) using regression analysis to control for the quantity and quality (purity is the measure of quality) of drugs sold. Results from the regression are then used to estimate the price paid on average at a given time and place for a given quantity and quality of drugs.
For this report, we analyzed data from the System to Retrieve Drug Evidence (STRIDE), which were available from January 1981 through September 1996. The data and our basic approach are described elsewhere.3 We have updated that method for present purposes, and we will report full results at a later date.
One problem when using regression analysis to estimate illicit drug prices is that the typical quantity and quality of drug entering a retail transaction is unknown. As others have noted,4 retail transactions take so many forms that an average retail price is hard to identify. Nevertheless, the calculations used in this report required one.
Estimating the street price for heroin was relatively straightforward. Rocheleau and Boyum5 reported that hardcore heroin users bought about 1.7 bags of heroin per purchase in New York, 1.9 bags per purchase in Chicago, and 2.0 bags per purchase in San Diego. They also reported that a bag contained about 25 milligrams of pure heroin—suggesting that a typical purchase comprised roughly 50 milligrams of pure heroin. They do not explain how they determined that a typical bag contained 25 milligrams of pure heroin, but Rhodes and Pittayathikhun6 report a similar figure—80 pure milligrams per purchased—based on an analysis of data from the domestic monitoring system (1987–1992). They also reported that a typical purchase was 20 percent pure over this period. More recent estimates, also based on the domestic monitoring program data, suggest that purity was closer to 40 percent in 1996.7 For purposes of estimating the regression, then, we assumed that a typical purchase was of two to four bags, containing a total of 80 milligrams of pure heroin from 1988 to 1992, and a total of 100 milligrams of pure heroin from 1993 to 1995. These assumptions were used to estimate the average price paid for heroin, based on regression predictions, for each year 1988 through 1995. Results are reported in the text.
Because similar information is not available for cocaine purchases, we use a different approach: Let P = F(Am, Pu) be a functional representation of the relationship between price paid (P) and the amount (Am) and purity (Pu) of drugs purchased. This functional relationship was determined by regression analysis as explained earlier.
Let $D represent the average dollar amount that a hardcore cocaine user spends per week on cocaine. This number was reported in Table 2 of the main report. Setting Pu equal to the average purity of drugs sold at the retail level, and assuming that the user buys drugs once per week, the typical amount of drugs in a weekly purchase must be the solution to the equation:
If Am* is the solution to this equation, then one estimate of retail prices is F(Am*, Pu).
Similarly, assuming that the user buys drugs at T separate times during the week, the purchase amount must be the solution to the equation
If Am** is the solution to this equation, another estimate of retail price is F(Am**, Pu).
Now, if few hardcore users buy drugs less frequently than once per week, and if few heavy users buy drugs more frequently than T times per week, then F(Am*, Pu) and F(Am**, Pu) provide low and high prices, respectively. The average of these is reported in the text.
This price range does not encompass all prices paid at retail. Many hardcore drug users undoubtedly pay much more. Others probably pay much less. These limits are intended to encompass the price that is typically paid at retail. That is, it is a range that seems likely to include the price that hardcore drug users pay on average for retail-level drug transactions. Prices are reported in Table 3 of the main report.
Endnotes
1 For example: Drug Enforcement Administration, Illegal Drug Price/Purity Report United States: January 1992–December 1995, August 1996; National Institute on Drug Abuse, Epidemiologic Trends in Drug Abuse, Volume II Proceedings, December 1995.
2 G. Brown and L. Silvermen, "The Retail Price of Heroin: Estimation and Applications," Journal of the American Statistical Association, 69, no. 347 (1974):595–606; J. Caulkins and R. Padman, "Quantity Discounts and Quality Premia for Illicit Drugs," Journal of the American Statistical Association,88, no. 423 (1994):748–57; W. Rhodes, R. Hyatt and P. Scheiman, "The Price of Cocaine, Heroin and Marijuana, 1981–1993," The Journal of Drug Issues, 24, no. 3 (1994):383–402; J. Caulkins, Developing Price Series for Cocaine, MR–317–DPRC (Santa Monica, CA, Rand).
3 W. Rhodes, R. Hyatt and P. Scheiman, "The Price of Cocaine, Heroin and Marijuana, 1981–1993," The Journal of Drug Issues, 24 no. 3 (1994):383–402.
4 J. Caulkins, "What is the Average Price of An Illicit Drug?" Addiction, 89, no. 7 (July 1994): pp. 815–19.
5 A. Rocheleau and D. Boyum, Measuring Heroin Availability in Three Cities (Office of National Drug Control Policy, Washington, D.C.: 1994): 49.
6 W. Rhodes and T. Pittayathikhun, Heroin Prices and New Addicts: 1981–1994 (report submitted to the National Institute of Justice, Washington, D.C.: April, 1995).
7 Drug Enforcement Administration, DEA Briefing Book (Drug Enforcement Administration, Washington, D.C.: October, 1996).
