Drug Use and Race Philip Shropshire brings up an issue that I've heard about before but never investigated: the rates of drug use among blacks, whites, and Asians. In response to my request for statistical sources for his assertion that the "majority of drug users in this country are white and middle class", Philip returned with an assortment of links on the topic. I had heard the statistic before, but I had thought it akin to the disingenuous claim that "most welfare users are white", which ignores the crucial fact that blacks are welfare users at higher rates. The study that Philip cites (link in PDF) that uses government statistics indicates that:

"Data from the 1997 National Household Survey on Drug Abuse (NHSDA) showed that whites had the highest lifetime rates of use for most illicit drugs, including marijuana (36%), cocaine (12%), LSD (9%), inhalants (7%), stimulants (5%), and PCP (3%). Blacks had slightly higher rates of lifetime crack use (3%) compared to whites (2%) or Hispanics (2%). The lifetime rate of heroin use across racial/ethnic categories was less than 1 percent,and the data suggested no reliable differences between blacks, Hispanics and whites for this drug." ... Across all age groups, lifetime illicit drug use was highest among whites (38%), followed by blacks (31%) and Hispanics (26%). Differences among whites, blacks, and Hispanics in any illicit drug use were highest among persons aged 18 to 34 years of age. Among 18- to 25-year-olds, lifetime illicit drug use was highest for whites (50%), followed by blacks (35%) and Hispanics (32%). Among 26- to 34-year-olds, lifetime illicit drug use was highest for whites (58%), followed by blacks (40%) and Hispanics (29%)

Intrigued, I did a little bit of probing myself and found that the site was quite dispassionate in its analysis and seemingly nonpartisan. I looked around for other sources to corroborate this data and found this link from the CDC that has a time series of drug use rates among whites, blacks, and hispanics. The statistics here are a bit more mixed; whites have higher rates of alcohol use across the board, but they're in a statistical dead heat with blacks in terms of marijuana use. In any case, the point is that the marijuna rates are not statistically different, but alcohol rates are. Perhaps more importantly, I noticed that the study above cited the 1997 NHSDA rather than the current one. I looked for the latest version (which was 2000 rather than 2001 or 2002) and found the following key paragraph:

-The rates of current illicit drug use for major racial/ethnic groups in 2000 were 6.4 percent for whites, 5.3 percent for Hispanics, and 6.4 percent for blacks. The rate was highest among the American Indian/Alaska Native population (12.6 percent) and among persons reporting more than one race (14.8 percent). Asians had the lowest rate (2.7 percent). -Although Asians as a group had the lowest rate of current illicit drug use, there were variations among the various specific Asian subgroups. For persons aged 12 and older, the rates ranged from 1.0 percent of Chinese and 2.1 percent of Asian Indians to 6.9 percent of Koreans, 5.0 percent of Japanese, and 4.3 percent of Vietnamese. These estimates are based on combined 1999 and 2000 NHSDA data, to ensure adequate sample sizes for these population subgroups (Figure 2.12). -Based on combined 1999 and 2000 data, rates of past month illicit drug use in the population aged 12 and older were 10.1 percent for Puerto Ricans, 5.5 percent for Mexicans, 4.1 percent for Central or South Americans, and 3.7 percent for Cubans (Figure 2.12). -Among youths aged 12 to 17 years, the rate of current illicit drug use was highest among American Indian/Alaska Natives (22.2 percent for combined 1999 and 2000 data).

Now, what does this all mean? It seems that blacks and whites are in a statistical dead heat with respect to rates of illicit drug use. On the basis of other data (such as savings rates or number of sexual partners), we would expect the "time-value-of-pleasure" tradeoff to be negotiated in favor of the immediate by blacks and the long-term by asians, with whites intermediate. In other words, we would expect blacks to be less likely to trade off pleasure today for (more) pleasure tomorrow, and asians more likely to do so. The drug use rates seem to be anomalous in this regard. It may be that there is a countervailing factor at work that differentially enables drug use among whites rather than blacks, such as a surplus of discretionary funds. It may also be that there are genetic factors at work (e.g. alcoholism predilection) that are specific to the drug use case and again impact whites more than blacks. I don't know what the answer is, but I wanted to point out that drug abuse rates for the Chinese are lower than that of any other group. Although the sample size of the asians in the survey is uncertain, it seems clear that lumping all Asians together is a (fairly crude) statistical approximation that must be reevaluated in each domain to which it's applied. A couple of other points: 1. I think I have been fair in admitting when I've been wrong on the facts. But will Mr. Shropshire or ProgCal ever admit that (say) blacks commit crimes at much higher rates than whites in both the reports of the police (Uniform Crime Reports) and the reports of the victims (National Criminal Victimisation Survey)? 2. I want to make clear that I support decriminalization of drugs that do not cause violence. In fact, I expect that decriminalization would result in a booming market in designer drugs that bring pleasure with few side effects.