THE RTI STUDY FOR DIMINISHED PRODUCTIVITY

The Research Triangle Institute ("RTI") of North Carolina issued a report in 1984 that attempted to calculate the economic cost to society of alcohol and drug abuse and mental illness. (7) It appears to be the source of the claim that drug abuse costs the United States thirty-four billion dollars a year in lost productivity. (8)

This claim seems to indicate that somehow individual worker productivity was assessed by some quantitative measure and was noted to be seriously diminished. I often ask my students how they would design a study which might quantify the effect of drug use or abuse on productivity and they generate interesting ideas and designs. None, however, have ever come up with anything as creative as the RTI report.

The National Institute on Drug Abuse ("NIDA") Household Survey is the actual source of the RTI statistics. (9) The RTI analysis itself does not include any studies about productivity, Nor does it document that drug use was associated with any diminished productivity. The RTI report relied on the 1982 version of the Household Survey, which included a question to the adult (older than eighteen years of age) respondents about daily marijuana use. If the respondent stated that he or she had ever used marijuana daily for twenty or thirty days in one month, this respondent's household income was compared to another household without a daily user. The user households were estimated to have a household income approximately twenty-eight percent less than those households that did not have such an interviewee. According to the RTI report: "These result have been used to calculate the reduced productivity due to daily marijuana use. Once again, the prevalence rates of ever using marijuana daily in the labor force by age and sex groups....were applied to the number of persons in the labor force from those groups that experienced reduced productivity. This number was multiplied by the expected productivity in the workplace and in the household and then by the impairment rate (27.9 percent) to produce the final value: The loss due to marijuana abuse was estimated at $34.2 billion for 1980." (10)

This somewhat opaque quotation means that the $34 billion "lost" to drug-induced diminished productivity and the one-third reduction is productivity of the drug user come from a very unusual calculation. The homes with daily users had a reduced household income. Because the daily users all were between eighteen and thirty-four years of age, the proportion of the work force represented by them was multiplied by total income expected and this was multiplied by the reduction figure to generate the $34 billion guess. I have never seen a published analysis of the RTI study. It is a complex and detailed document of one hundred thirty-seven pages whose appendices number nearly one hundred thirty more. Although many of its assumptions regarding the cost of alcohol and drug abuse are questionable, I have focused only on the loss of productivity formulation.

The reduced income figure itself may be incorrect. It relies upon reports from interviewees that they used drugs and relies upon their estimates of household income. The income data were grouped in seven broad categories with intervals as broad as $10,000. (11) To generate the reduction figure with categories like this amounts to measuring centimeters using a device graduated in meters. The control households were generated by selecting a household whose respondent resembled the marijuana use respondent. This meant that the two respondents resembled each other in regard to age, education, marital status, and occupation. (12) The similarities did not include race, location, or non-children resident(s) in the household. Therefore, a twenty-four year-old single black, high school graduate brick-layer who smokes marijuana and lives with his grandmother and uncle in Harlem, could have his family income compared to a twenty-four year-old single white, high school graduate, brick-layer living with his parents in an affluent New Jersey suburb. If the household in Harlem had a lower income, it would be attributed to diminished productivity caused by marijuana use.

As the RTI authors noted, however, the association between marijuana use and reduced income is not necessarily causal. (13) Although age and occupation were similar, the user may not have concerned himself with income and may have made other job choices. (14) Moreover, the heavy use of marijuana could have been secondary to job loss in a poor economy.

This manipulation, which assigned lost productivity to sometime past heavy use of marijuana, did not show any association with current use of marijuana (or any other drug), nor could it show any productivity decrease due to past use of any other drug. It also could not show any effect of any drug use in any person greater than age thirty-five. (15) Thus, according to the study, regular use of marijuana in the past on the part of females aged twenty-five to thirty-four cost the nation $2,396,000,000. (16) However, marijuana use by females greater than thirty-four, cost the nation nothing. (17) In males aged twenty to twenty-four, regular marijuana use in the past would have cost the economy $6,723,000,000. (18) On this basis, however, any use of cocaine, prescription drugs, or phencyclidine (PCP) by this group cost the nation nothing. Of course, the study really does not indicate this because its inadequacy prevented any such calculations. Its inadequacy regarding the marijuana calculations should be viewed in the same fashion. To stand in front of an audience and say that drug use cost the nation $34 billion dollars in decreased productivity is foolish and improper.

Readers may have noted that the past paragraphs followed the mode of the RTI report and used income and productivity as equivalents. I do not believe they are equivalents. Are those who choose to work in positions where they earn less than the maximum possible less productive? If so, a judge is less productive than a practicing lawyer, a medical school professor is less productive than a practicing physician, a farmer is less productive than a florist, and an elementary school teacher is less productive than an owner of a day-care center.

Click here for Part IV - Illegal Drug Use Declining.