What America's Users Spend on Illegal Drugs, 2000–2010

What America's Users Spend on Illegal Drugs, 2000–2010: Technical Report

B. Kilmer
S. Everingham
J. Caulkins
G. Midgette
R. Pacula
P. Reuter
R. Burns
B. Han
R. Lundberg
Copyright Date: 2014
Published by: RAND Corporation
Pages: 71
https://www.jstor.org/stable/10.7249/j.ctt6wq7r9
  • Cite this Item
  • Book Info
    What America's Users Spend on Illegal Drugs, 2000–2010
    Book Description:

    In January 2012, the U.S. White House Office of National Drug Control Policy (ONDCP) asked RAND to generate national estimates of the total number of users, total expenditures, and total consumption for four illicit drugs from 2000 to 2010: cocaine (including crack), heroin, marijuana, and methamphetamine. Drug users in the United States spend on the order of $100 billion annually on these drugs (in 2010 dollars). While this total figure has been stable over the decade, there have been important compositional shifts. From 2006 to 2010, the amount of marijuana consumed in the United States likely increased more than 30 percent, while the amount of cocaine consumed in the United States decreased by approximately 50 percent. These figures are consistent with supply-side indicators, such as seizures and production estimates. Methamphetamine consumption rose sharply from 2000 through the middle of the decade, and this was followed by a large decline through 2008. Heroin consumption remained fairly stable throughout the decade, although there is some evidence of an increase in the later years. For all of the drugs, total consumption and expenditures are driven by the minority of users who consume on 21 or more days each month.

    eISBN: 978-0-8330-8571-9
    Subjects: Psychology, Political Science, Health Sciences

Table of Contents

  1. Front Matter
    (pp. 1-1)
  2. Table of Contents
    (pp. 2-4)
  3. Section 1. Introduction
    (pp. 5-5)

    In January 2012, the White House Office of National Drug Control Policy asked RAND to generate estimates of the total number of users, total expenditures, and total consumption for four illicit drugs from 2000 to 2010: cocaine (including crack), heroin, marijuana, and methamphetamine (or meth). The Main Report provides an overview of our methodology and presents our results.

    This Technical Report presents additional details about our methods and calculations. Similar to the Main Report, it initially focuses on the analyses for cocaine, heroin, and meth. Section 2 provides a detailed account of the regression models which helped us understand which...

  4. Section 2. Predicting Positive Drug Tests in ADAM
    (pp. 6-16)

    Drug use among adult male arrestees is a function of both individual and market-level demand factors, and the relationships can differ by substance. For example, Dave (2008) finds that race, ethnicity, age, the drug possession arrest rate, and STRIDE-derived street-level cocaine prices are statistically associated with the probability that an arrestee in the ADAM program tests positive for cocaine; for heroin, age did not matter, but the street-level price of heroin, possession arrest rate, race, and employment status did. Our challenge was to find state- and substate-level variables that both predicted the share of arrestees testing positive for each drug...

  5. Section 3. Estimating the Number of Hard-Drug Users
    (pp. 17-25)

    ONDCP (2012c) established estimates for 2000–2003 using ADAM, TEDS, and an assumption of a constant proportionality between chronic drug users in a county and treatment admissions nationally. The basic model is:\[C={{C}_{a}}*(V/{{V}_{a}}),\caption{(3.1)}\]whereCarepresents the number of chronic users in ADAM counties,Varepresents the number of treatment admissions in ADAM counties,Vrepresents the number of treatment admissions in the country, andCrepresents the number of chronic users in the country. In essence, the number of CDUs in ADAM counties is scaled up by the ratio of treatment admissions in the country relative to those counties....

  6. Section 4. Estimating Expenditures on Hard-Drug Users
    (pp. 26-31)

    The approach for generating monthly drug expenditures in the previous report is fairly straightforward, but impossible for us to replicate with the available information.14Using data from those who paid cash as well as those who bartered for at least some of the product, the report regressed monthly expenditure (value of last purchase multiplied by the number of times purchased on that day, multiplied by the number of days purchased in the past 30) on past-month use days for those who reported using everything they obtained (i.e., they did not sell or give away any). The report uses the estimated...

  7. Section 5. Estimating Hard-Drug Consumption
    (pp. 32-39)

    The previous report followed a six-step process to develop price series for user-level cocaine, heroin, and methamphetamine purchases based on STRIDE undercover purchase data between 1998 and March 2007 (ONDCP, 2012c). They first limited the dataset to include only purchases between $20 and $200 in nominal terms to mitigate the risk of nonretail-level purchases entering the data. Next, they excluded purchase of greater than 10 pure grams for the same reason. Finally, they defined a price-purity metric as pure grams per dollar spent and exclude purchases with values greater than 0.1 as outliers—to rephrase, they excluded purchases with values...

  8. Section 6. Details of Marijuana Analyses
    (pp. 40-56)

    The earlier versions of this report employed a straightforward approach based principally on NSDUH data (as in this report). Users were counted as those who reported using marijuana or hashish in the last 30 days. A chronic user was defined as someone who consumes marijuana four or more days per month (e.g., weekly), and an occasional user as someone who consumes three or fewer days per month.

    Then, the counts of total, chronic, and occasional users were adjusted upward to “reduce reporting bias” that results from that fact that “respondents to NSDUH frequently understate their drug use” (ONDCP, 2012c). The...

  9. Section 7. Sinsemilla Consumption in the United States
    (pp. 57-68)

    It has become standard practice over the last 20 years to work with purity-adjusted prices for cocaine, heroin, and methamphetamine, not just the “raw” price per gram, unadjusted for purity. Marijuana is more complicated, because it contains many different psychoactive chemicals, but tetrahydrocannabinol (THC) is the most important, so there is likewise an argument for working with THC potency-adjusted prices for marijuana. That has not been standard practice in the past, but we attempted to factor potency trends into our estimates here.

    Potency affects the amount consumed per unit of intoxication. Roughly speaking, one-third of a gram of sinsemilla containing...

  10. Section 8. Contents of the Spreadsheets
    (pp. 69-71)

    The Excel workbook “RAND WAUSID Calculations (Delivered September 16 2013).xlsx” contains 30 spreadsheets used to generate the results, figures, and tables in the Main Report and the previous sections of this Technical Report. Tabs pertaining to marijuana are shaded green.

    The first four sheets (“Cocaine”, “Heroin”, “Meth”, and “Marijuana”) present the main calculations for our estimates of CDUs, expenditures, and consumption. The first three begin with the total number of adult male arrest events involving a positive test for each respective substance (calculations described in Section 2). A series of adjustments are made to convert these figures to total CDUs,...