Endnotes

¹ Money is not the only form of payment for illicit drugs. Dealers often keep drugs for personal use, users help dealers in exchange for drugs, and users perform sex for drugs (especially crack cocaine). When such "income in kind" is valued at current retail prices, an additional $4 billion to $7 billion must be added to the total for cocaine and an additional $2 billion to $4 billion to the total for heroin. In this report, all expenditures are in 1998 dollar equivalents. These expenditure estimates do not include income in kind.

² By comparison, Americans spent about $43 billion on tobacco in 1993. The Tax Burden on Tobacco (Washington, D.C.: The Tobacco Institute, 1993).

³ The NHSDA excludes military personnel, those incarcerated in jails and prisons, and those who are residents of treatment facilities. Military personnel, whose consumption of illicit substances is monitored through urinalysis, do not have the opportunity to be heavy drug users. Those incarcerated in jails and lockups may use drugs, but that consumption must necessarily be limited by restricted availability. A Bureau of Justice Statistics study reports AIn State correctional facilities, 3.6 percent of the tests for cocaine, 1.3 percent for heroin, 2.0 percent for methamphetamine, and 6.3 percent for marijuana found evidence of drug use. In Federal prisons, 0.4 percent of the tests for cocaine, 0.4 percent for heroin, 0.1 percent for methamphetamine, and 1.1 percent for marijuana were positive." C. Harlow, Drug Enforcement and Treatment in Prison, 1990 (NCJ–134724, July 1992). These percentages are probably high because tests are most likely to be conducted when drug use is suspected. In any case, drug use in prisons cannot account for much of the drug use that occurs in America. Sources at the National Institute on Drug Abuse consider drug use by those in residential treatment facilities to be minimal.

⁴ Evidence that a large segment of the drug-using population is excluded from the NHSDA comes from a number of sources. According to the 1991 NHSDA, drug use is twice as high among respondents who lived in households considered unstable than it is among those who lived in more stable environments, indicating that the NHSDA's bias toward reporting on stable households is likely to miss many heavy drug users. Additional evidence also comes from interviews with nearly 35,000 intravenous drug users who were contacted by National Institute on Drug Abuse-sponsored researchers as part of an AIDS outreach project. Abt Associates' tabulations show that of these drug users, an estimated 40 percent lived in unstable households and about 10 percent could be considered homeless.

Available evidence indicates that NHSDA's respondents understate heavy drug use. A. Harrell, K. Kapsak, I. Caisson, and P. Wirtz, "The Validity of Self-Reported Drug Use Data: The Accuracy of Responses on Confidential Self-Administered Answer Sheets," paper prepared for the National Institute on Drug Abuse, Contract Number 271–85–8305, December 1986. M. Fendrich, T. Johnson, S. Sudman, J. Wislar and V. Spiehler, "Validity of Drug Use Reporting in a High-Risk Community Sample: A Comparison of Cocaine and Heroin Survey Reports with Hair Tests," American Journal of Epidemiology 149(10): 955:62, 1999. Consistent with these observations, the Substance Abuse Mental Health Services Administration reports that virtually no heroin addicts answer the National Household Survey on Drug Abuse. Substance Abuse Mental Health Services Administration, Preliminary Estimates from the 1993 National Household Survey on Drug Abuse (June 1994).

A comparison of the demographic characteristics of the heavy cocaine users in the NHSDA with those of heavy cocaine users based on other sources (the Drug Use Forecasting program, the Drug Abuse Warning Network, and the National AIDS Demonstration Research project) shows a marked difference between those populations and the one represented in the NHSDA. Incomes are greater, unemployment is lower, and there are fewer respondents using more than one drug in the NHSDA. D. Hunt and W. Rhodes, "Characteristics of Heavy Cocaine Users Including Polydrug Use, Criminal Behavior, and Health Risks," paper prepared for Office of National Drug Control Policy (ONDCP), December 14, 1992.

Finally, estimates of heavy drug use reported in the NHSDA are difficult to reconcile with other data sources maintained by the Substance Abuse Mental Health Services Administration, especially with reports of the treatment for cocaine or heroin. These incompatibilities are discussed later in this report.

⁵ A large percentage of heavy drug users are arrested at some time in their drug-using "careers," so the criminal justice system provides valuable supplemental data when counting heavy drug users. For example, in the 1993 Household Survey, about 58 percent of weekly cocaine users surveyed had been arrested and booked at some time, 39 percent during the year prior to the survey. In the National AIDS Demonstration Research data, 81 percent of heavy cocaine users had been arrested at some time in their lives, and one-third had been in jail or prison during the six months prior to the interview.

⁶ The population of hardcore users is not identical to the population of users who need substance abuse treatment. Still, using the 10 days per month threshold, the DUF data show that 57 percent of hardcore cocaine users and 77 percent of hardcore heroin users deemed themselves to be in need of treatment. These self-reports probably understate the need for treatment, because denial of the need for treatment is high among hardcore users.

⁷ Because urinalysis will detect cocaine and heroin use within two to three days of its consumption, it is unlikely that urinalysis will fail to identify an individual who uses cocaine on at least a weekly basis. (Most weekly users use it more frequently than once a week.) However, an occasional user is likely not to have used cocaine or heroin within two to three days of his or her arrest. Consequently, DUF would frequently fail to identify occasional users. Arguably, the EMIT test used by DUF understates drugs in the urine of arrestees. C. Visher and K. McFadden, A Comparison of Urinalysis Technologies for Drug Testing in Criminal Justice, NCJ–129292, June 1991. However, it seems reasonable that occasional users are more likely than hardcore users to have an erroneous negative urine test, so we have not adjusted the DUF urine test results to reflect the EMIT tests false negative rate of about 20 percent. For evidence supporting this decision, see T. Mieczkowski, "Immunochemical Hair Assays, Urinalysis, Self Reported Use and the Measurement of Arrestee Cocaine and Marijuana Exposure in a Large Sample," paper presented at the Annual Meetings, American Society of Criminology, New Orleans, November 7–22, 1992.

⁸ S. Everingham, C. Rydell and J. Caulkins, "Cocaine Consumption in the United States: Estimating Past Trends and Future Scenarios," Socio-Economic Planning Sciences, Vol. 29 (4) December 1995: 305–314. The authors report that heavy users of cocaine use 70 percent of all cocaine. Estimates based on retail sales expenditure, reported later, are consistent, but also show that hardcore heroin users account for a larger fraction of heroin sales than hardcore cocaine users account for cocaine sales.

⁹ Drugs are sometimes received as income-in-kind, especially by drug-using dealers who keep part of what they otherwise would deal, and also those who exchange drugs for sex. Income-in-kind is not included in the retail sales dollar amounts, but it is factored into the measures of metric tons of drugs consumed.

¹⁰ To project hardcore user estimates from the DUF data, we estimated the number of hardcore users in 1998 as a linear projection of estimates from 1995, 1996 and 1997. We set estimates for 1999 and 2000 equal to the 1998 projection. Finally, we applied a three-year moving average to all the estimates from 1989 through 2000. The three-year moving average is reported in the text. Statistics for 1998 had already been reported for the NHSDA, so we used a linear projection (using data from 1988 through 1998) to estimate comparable figures for 1999 and 2000. The final hardcore users estimates equal the smoothed estimates from DUF data plus one-half the estimate of hardcore use from the NHSDA.

¹¹ A large number of drug users use both heroin and cocaine. For example, of the hardcore drug users in the 1995 DUF sample: 70 percent are hardcore users of cocaine only, 16 percent are hardcore users of heroin only, and the other 14 percent are hardcore users of both.

¹² W. Rhodes, S. Langenbahn, R. Kling, and P. Scheiman. What America's Users Spend on Illegal Drugs: 1988–1995 (Washington, D.C.: Office of National Drug Control Policy, Fall 1997). See Appendix A.

¹³ D. Hamill and P. Cooley, National Estimates of Heroin Prevalence 1980–1987: Results from Analyses of DAWN Emergency Room Data, RTI Technical Report, (Triangle Park, N.C.: Research Triangle Institute, 1990).

¹⁴ R. Simeone, W. Rhodes, and D. Hunt, Methodology for Estimating the Number of Hardcore Drug Users, report submitted to the Office of National Drug Control Policy, March 1997.

¹⁵ SAMHSA estimates that 7.1 million people needed treatment in 1994. Persons needing treatment are divided into two categories, Level 1 and Level 2. The Level 2 category is a more severe category of need and contains about 3.6 million people. We have used this 3.6 million figure in our calculations under the assumption that Level 2 users are similar to the hardcore drug users described in our report. See: Substance Abuse and Mental Health Services Administration, "The Need for and Delivery of Drug Abuse Services: Recent Estimates," February 22, 1996.

¹⁶ SAMHSA defines those who are severely in need of drug treatment using four criteria. NHSDA respondents were classified as in need of treatment if they reported any of the following in the past 12 months:

• Been dependent on any drug other than marijuana;
• Reported injecting cocaine, heroin or stimulants;
• Received drug abuse treatment at a specialty facility; and
• Used drugs frequently.

To account for the underestimation of hard-core drug use in the NHSDA, SAMHSA adjusted the number of people needing treatment using a ratio estimation technique that links NHSDA data to data from the Uniform Crime Reports and the National Drug and Alcohol Treatment Unit Survey. This ratio estimation technique inflated estimates of treatment need by 20% in 1991 and 1992 and 30% in 1993. Although we did not have figures for the ratio estimation in 1994, we assumed a similar adjustment of 20 to 30%. See: Substance Abuse and Mental Health Services Administration, "The Need for and Delivery of Drug Abuse Services: Recent Estimates," February 22, 1996 and "Estimating Substance Abuse Treatment Need for a National Household Survey," by Joan Epstein and Joseph Gfoerer, OAS Working Paper, presented at the 37th International Congress on Alcohol and Drug Dependence, August 20–25, 1995, UCSD Campus, La Jolla, California.

¹⁷ Using SAMHSA's description of their technique for estimating the number of persons needing treatment, we developed the following algorithm using the NHSDA. Persons were classified as severely needing treatment if they met at least one of the following criteria:

• Dependence on any drug other than marijuana in the past 12 months. Six question types from the 1994 revised NHSDA were used to approximate the DSM-III-R criteria for drug dependence. Respondents were classified as dependent if they answered at least three of these six questions positively for any drug except marijuana. We originally defined dependence using positive answers to at least two of the six questions, since the DSM-III-R uses three of nine questions to determine dependence. However, this procedure yielded estimates that were too high.
• Reported using needles to inject cocaine, heroin or stimulants at least once during the last year.
• Reported receiving drug treatment at a hospital (as an inpatient), a drug treatment facility (as an inpatient), or at a mental health facility over the past year.
• In the past year, reported using marijuana daily and met the criteria for marijuana dependence described above, reported any heroin use, reported using cocaine at least weekly, or reported daily use of other drugs, including inhalants, hallucinogens, stimulants, sedatives, analgesics, and tranquilizers.

We inflated the estimate obtained through this method by 25% to approximate the ratio estimation technique used by SAMHSA.

¹⁸ National Institute on Drug Abuse, Epidemiological Trends in Drug Abuse, Volume I: Highlights and Executive Summary, Community Epidemiological Work Group, December 1996: Exhibit 5, page 18. We excluded Minneapolis/St. Paul from this summary, because that site did not exclude alcohol—only from its treatment statistics.

¹⁹ Treatment Episode Data Set (TEDS): 1992–1997. SAMHSA, August 26, 1999. Downloaded from the Internet 11/18/1999: www.samhsa.gov/teds9297.htm

²⁰ Center for Disease Control and Prevention, HIV/AIDS Surveillance Report 1998, Vol. 10 (No. 2).

²¹ Trends in lifetime prevalence of heroin use among 12th graders rose from 1993 to 1997, but leveled or dropped from 1997 to 1998. Table 5–1, National Survey Results on Drug Use from the Monitoring the Future Study, 1975–1998 (Bethesda, Maryland: National Institute on Drug Abuse, 1999).

²² Treatment data are difficult to interpret. From the Treatment Episode Data, we observe that treatment admissions for heroin increased from 167,000 in 1992 to 218,000 in 1997; furthermore, while 77 percent of heroin users injected in 1992, only 68 percent injected in 1997. Perhaps these trends imply more heroin users in the late 1990s. It certainly implies a larger prevalence on non-injection drug use. Substance Abuse Mental Health Services Administration, Treatment Episode Data Set (TEDS): 1992–1997.

²³ Table 2.10 Downloaded from the Internet on 11/15/99: www.samsha.gov/oas/p0000018.htm

²⁴ R. Simeone, W. Rhodes, and D. Hunt. Methodology for Estimating the Number of Hardcore Drug Users. Report submitted to the Office of National Drug Control Policy by Abt Associates Inc., March 1997.

²⁵ Weekly expenditures on cocaine and heroin have decreased over time, but this change results from using the CPI to convert expenditures to 1998 dollar equivalents. Many hardcore users spend two-thirds of their incomes on drugs, but they probably do not see themselves as spending less over time because the price of cocaine and heroin has fallen in real terms since 1988. The CPI is not a good reflection of a hardcore drug users' market basket.

²⁶ K.J. Riley, Crack, Powder Cocaine, and Heroin: Drug Purchase and Use Patterns in Six U.S. Cities, joint report of the National Institute of Justice and the Office of National Drug Control Policy (Washington, D.C., December 1997).

²⁷ We are indebted to Linda Truitt for these calculations.

²⁸ On this point, see J. Caulkins, B. Johnson, A. Taylor and L. Taylor, "What Drug Dealers Tell Us About Their Costs of Doing Business," Journal of Drug Issues 29(2), Spring 1999. This study was about the distribution of crack, but a similar marketing scheme is likely to pertain to heroin.

²⁹ Two factors make the assumption of higher spending questionable. First, incomes of most drug users cannot support a higher level of drug use. Second, heavy drug users have a high level of unemployment and underemployment. D. Hunt and W. Rhodes, "Characteristics of Heavy Cocaine Users, Including Polydrug Use, Criminal Activity and Health Risks," paper prepared for ONDCP, December 14, 1992. As discussed in Appendix B, illegal income from property crimes and prostitution accounts for much of the expenditure on drug use. However, illegal income cannot account for higher expenditures than are reported in this study. Drug dealing is often advanced as a way to support hardcore drug use, but in total, street-level dealing cannot generate the dollars that ultimately must go to satisfy the cash demands of middle-level and upper-level dealers. If expenditures are much greater than reported here, the income source for supporting that level of consumption is suspect.

³⁰ Reuter and Kleiman estimated that the market for cocaine was about $8 billion in 1982. This is about $14 billion in 1998 dollars. Because of the accelerating use of cocaine from that time until the mid-1980s, and after accounting for inflation, it is not surprising that their estimate is less than the figure reported here. Their $8 billion estimate for heroin expenditures equals about $14 billion in 1998 dollars. That is considerably less than our 1989 estimate. P. Reuter and M. Kleiman, "Risks and Prices: An Economic Analysis of Drug Enforcement," in Crime and Justice: An Annual Review of Research, volume 7, ed. M. Tonry and N. Morris (Chicago: University of Chicago Press, 1986), 194. Carlson, who conducted a study of the underground economy for the Internal Revenue Service, reported that an estimated $11 billion was spent on cocaine in 1982. K. Carlson et al., "Unreported Taxable Income for Selected Illegal Activities: Volume I: Consensual Crimes," paper prepared for the Internal Revenue Service under contract number TIR–81.57, September 1984. In an update of his study, Carlson estimated that cocaine expenditures increased from $5.8 to $6.6 billion between 1988 and 1991. K. Carlson, "Unreported Illegal Source Income 1983-1995," paper prepared for the Internal Revenue Service under order number 89-11565, May 15, 1990. Since he relied heavily on the NHSDA, and because his estimates are not adjusted for inflation, it is not surprising that his estimate is much lower than the one reported here. Carlson's estimate of heroin expenditures, based on the National Narcotics Intelligence Consumers Committee estimates for 1982, was in keeping with Reuter and Kleiman's $8 billion figure. His updated study, based on NHSDA data, put that figure at roughly $7 billion a year between 1988 and 1991.

³¹ Heroin distribution seemed to change toward the end of the 1980s and 1990s. As discussed later in this report, there was a marked decrease in the cost of heroin and an equally marked increase in the purity of heroin available to American consumers. At least as of 1995, Colombia had replaced Southeast and Southwest Asia as the principal source of heroin sold in the United States, and distribution practices changed as a consequence. As Appendix B argues, ethnographers increasingly reported that drugs were being distributed by profit dealers instead of users.

³² Using the CPI to inflate expenditure on drugs is arguable. The Federal government computes the CPI from a weighted average of prices paid by consumers for what is deemed to be a typical market basket. The problem when applying this CPI to hardcore users is that their market basket is grossly atypical—two-thirds to three-quarters of their income may be spent on illicit drugs. (See J. Fagan, "Drug Selling and Illicit Income in Distressed Neighborhoods: The Economic Lives of Street-Level Drug Users and Dealers," in Drugs, Crime and Social Isolation, edited by A. Harrell and G. Peterson, (Washington, D.C.: The Urban Institute Press, November 1994). Because the nominal prices of cocaine and heroin have fallen over much of the period examined through the retail sales calculations, hardcore users have seen a deflation, not an inflation, in how much they spend on their typical market basket, most of which may be for illicit drugs. Thus, when asked about drug expenditures, hardcore users may well say they spend about the same amount in 1998 as they spent in 1988.

³³ Recent reports by the Community Epidemiological Work Group have told of increasing numbers of heroin users: "In the most recent reporting period (1997–1998), heroin indicators continued to increase in 12 CEWG cities. In some cities, heroin use indicators have been trending upward for more than three years." December 1998 Advance Report. Downloaded from the Internet 11/15/99: www.cdmgroup.com/cewg/docs/1298-miami/1298adv.ntm#heroin

³⁴ M. Childress, B. Dombey, and S. Resetor. A Systems Description of the Cocaine Trade (Santa Monica, CA: Rand, 1994).

³⁵ M. Childress, et al. A Systems Description of the Cocaine Trade (Santa Monica, CA: Rand, 1994).

³⁶ W. Rhodes, P. Johnson, S. Han, Q. McMullen, and Lynne Hozik. Illicit Drugs: Price Elasticity of Demand and Supply. Report submitted to the National Institute of Justice by Abt Associates Inc., February 17, 2000.

³⁷ National Narcotics Intelligence Consumers Committee, The NNICC Report 1993: The Supply of Illicit Drugs to the United States (Washington, D.C., August 1994): 61.

³⁸ The estimate of 0.0136 ounces is equivalent to 0.39 grams. The 1997 NNICC report says that a joint contains one-half gram on average, and that a ". . . blunt may contain as much as 6 times this amount." If the NNICC estimate is correct, our estimates would be about 25 percent too low, but the source of the NNICC estimate is unknown. The NNICC Report 1997: The Supply of Illicit Drugs to the United States (Washington, DC: DEA, November 1998).

³⁹ Researchers disagree about trends in reporting practices, but they agree that self-reported tobacco use is only about three-quarters as large as reports based on foreign imports and tobacco sales resulting in state and federal excise taxes. K.E. Warner, "Possible Increases in the Under reporting of Cigarette Consumption," Journal of the American Statistical Association, 73 (1978):314–317. E.J. Hatziadreu, J.P. Pierce, M.C. Fiore, et. al, "The Reliability of Self-Reported Cigarette Consumption in the United States," American Journal of Public Health, 79, (1989): 1020–1023.

⁴⁰ In 1993, about 74 percent of arrestees who tested positive for marijuana use at the time of booking reported some marijuana use during the month before the survey.

⁴¹ Using several self-report surveys, BOTEC Analysis Corporation estimated that marijuana costs $222 an ounce and that an ounce could be divided into 60 joints, yielding a unit price of $3.70 per joint. Based on these assumptions, BOTEC estimated that Americans spent $13.1 billion on 1,599 tons of marijuana in 1992. After adjusting for inflation, BOTEC's estimate is greater than the estimate presented in this report. The difference can be accounted for by three factors: methodological differences in estimating the number of users based on the NHSDA; BOTE's inclusion of criminally active user estimates; and BOTEC's higher price estimates. A..L. Chalsma and D. Boyum, "Marijuana Situation Assessment," (Washington, D.C.: Office of National Drug Control Policy, September 1994).

⁴² We noted previously that heavy cocaine users and heavy heroin users frequently appear in the DUF data, but infrequently appear in the NHSDA data. The reverse occurs for other illicit substances. With few exceptions, which are specific to cities, other illicit substances have relatively low prevalence among arrestees.

⁴³ Their answers, which were in ranges of days per year, were converted to a fixed number. For instance, the range three to five days became four days.

⁴⁴ Estimates of frequency of use from the 1991 NHSDA were applied to earlier years.

⁴⁵ Drug Enforcement Administration, Illegal Drug Price/Purity Report United States: January 1990–December 1993, April 1994. Community Epidemiology Work Group, Epidemiologic Trends in Drug Abuse, (Rockville, MD: National Institute on Drug Abuse, June 1994).

⁴⁶ M. Layne, P. Johnston, W. Rhodes, Following the Flow of Cocaine: The Sequential Transition and Reduction (STAR) Model, 1996–1999, May 2000.

ⁱⁱDefense Intelligence Agency, 1999. Interagency Assessment of Cocaine Movement: August 1999 Eighteenth Edition, Mid-Year Review, p. 2

ⁱⁱⁱWe used movement events from the CCDB for our calculations, and they differ slightly from figures published in the IACM. See Cala, 1999.

⁴⁷ W. Rhodes, M. Layne and P. Johnston, Estimating Heroin Availability. Report submitted to the Office of National Drug Control Policy by Abt Associates Inc., May 2000.

⁴⁸ Rhodes, W., Truitt, L., Kling, R. and Nelson, A. The Domestic Monitor Program and the Heroin Signature Program: Recommendations for Change (Cambridge MA, Abt Associates Inc., June 30, 1998).

⁴⁹ Coomber argues that this dilution of imported heroin is a product of the heroin production process. Thus it probably varies from source to source. South American heroin appears to be the most pure; Mexican is typically the least pure. R. Coomber, "The Cutting of Heroin," Journal of Drug Issues, 29 (1), 1999: 17–35.

⁵⁰ W. Rhodes, M. Layne, and P. Johnston. Estimating Heroin Availability. Report submitted to the Office of National Drug Control Policy by Abt Associates Inc., May 2000.

⁵¹ Calculations began with all the seizure reports contained in the Heroin Signature Program data file. These reports are not comprehensive of all seizures at ports of entry. From this file we selected all reports where: (1) the seizure occurred at an airport, at the border, or through the mail; (2) the seizure happened in 1995 or later; and (3) the seizure involved less than ten kilograms. Each report was characterized by the amount of pure heroin seized, and then the sample was weighted so that the distribution by source country for the seizure data matched the distribution by source country for the consumption data. For example, if 10 percent of the seizures came from South America while 15 percent of consumption came from South America, we weighted the seizures from South America by 15/10 or 1.5. By source area, the weights were:

0.73 for unknown
2.67 for Mexico
0.87 for Southeast Asia
1.32 for Southwest Asia
1.67 for South America

As a practical matter, then, this weighting gives greater emphasis to Mexican and South American heroin.

⁵² The Canadian Center on Substance Abuse reports that 5.9 percent of Canadians tried heroin at some time; 1.1 percent of the population used heroin during 1994. Canadian Center on Substance Abuse, Canadian Profile 1999 Illicit Drugs, downloaded from the Internet www.ccsa.ca/cp99.11.htm, November 11, 1999.

⁵³ Personal communication with Bill Wolf, Drug Enforcement Administration; November 12th, 1999.

⁵⁴ Drug Enforcement Administration Memo: "International Chemical Conference on the Multilateral Chemical Reporting Initiative."

⁵⁵ http://www.usdoj.gov/dea/programs/diversion/divpub/substanc/methamph.htm. G. Haislip, Methamphetamine Precursor Chemical Control in the 1990s.

⁵⁶ http://www.usdoj.gov/dea/pubs/meth/threat.htm. Methamphetamine: A Growing Domestic Threat—The Methamphetamine Problem.

⁵⁷ Personal communication with Bill Wolf, Drug Enforcement Administration; November 12th, 1999; Drug Enforcement Administration Memo: "Shifts in Predominance of Precursors."

⁵⁸ http://www.usdoj.gov/dea/programs/diversion/divpub/substanc/methamph.htm. G. Haislip, Methamphetamine Precursor Chemical Control in the 1990s.

⁵⁹ Personal communication with Bill Wolf, Drug Enforcement Administration; November 12th, 1999.

⁶⁰ Drug Enforcement Administration Memo—April 9, 1997.

⁶¹ Drug Enforcement Administration Memo—W.J. Wolf Jr., July 27, 1999.

⁶² G. Haislip, Methamphetamine Precursor Chemical Control in the 1990s, Drug Enforcement Administration, January 1996, downloaded from the Internet www.usdoj.gov/dea/programs/diversion/divpub/substance/methamph.htm.

⁶³ The DEA no longer estimates the amount of marijuana under cultivation outdoors in the United States. The DEA also notes that indoor cultivation continues and that there is no way to estimate the extent of this practice. The NNICC Report, 1995: The Supply of Illicit Drugs to the United States (Washington, D.C.: National Narcotics Intelligence Consumers Committee, August 1996).

⁶⁴ Drug Enforcement Administration, Intelligence Division, U.S. Drug Threat Assessment (Washington, D.C.: U.S. Department of Justice, 1993).

⁶⁵ Details of the statistical model can be found in P. Johnston, W. Rhodes, K. Carrigan and E. Moe, "The Price of Illicit Drugs: 1981 Through the Second Quarter of 1998." Paper prepared for the Office of National Drug Control Policy by Abt Associates Inc., February 1999.

⁶⁶ A standardized retail cocaine purchase consists of 0.35 pure grams of cocaine at 67 percent purity. By assumption, retail cocaine purchases involve transactions of 0.01 to 1.0 pure grams.

⁶⁷ A standardized middle level cocaine sale involves 30 pure grams (37.5 bulk grams) of cocaine at 80 percent purity. Middle level cocaine transactions are estimated to range from 15 to 140 grams, costing between $10 and $1000 per gram.

⁶⁸ A standardized importation level purchase is 358 pure grams at 73 percent purity. Importation level purchases were 0.1 metric tons and larger.

⁶⁹ A standardized purchase level for injection drug users is 40 milligrams at 13 percent purity. Purchases of 100 pure milligrams or less were considered to be purchases by injectors.

⁷⁰ A standardized purchase level for those who sniff heroin is about one-third pure gram at 39 percent purity. Purchases between 0.1 and 1.0 pure grams fit this category.

⁷¹ A street-level purchase is 2.94 pure grams at 41% purity. This includes purchases of between 0.001 and 10 pure grams.

⁷² An importation-level purchase is 321 pure grams at 71 percent purity. A purchase was considered to be at the importation level if it exceeded 100 pure grams.

⁷³ These estimates reflect retail level sales ranging from 0.001 to 10 grams; the retail price is evaluated at 3.1 grams. The importation level is for purchases of 1 metric ton and more. The prices are evaluated at 1.8 metric tons.