An A.P. story notes a recent reduction in methamphetamine use detected by workplace drug testing and attributes it to the federal crackdown on pseudoephedrine, a meth precursor:

Methamphetamine use continued to decline in nearly every part of the country last year as the government sharpened its crackdown on precursor chemicals used to make the illegal drug.

Overall, the number of workplace employees who tested positive for meth dropped 22 percent last year, according to a study released Wednesday by New Jersey-based Quest Diagnostics Inc., the nation's largest drug-testing company. Meth use in the Northeast, however, remained steady.

At the same time, the Drug Enforcement Administration issued a report showing the number of illegal meth lab seizures plunged 31 percent last year, from 7,347 to 5,080.

White House drug policy director John Walters said laws restricting the sale of cold medicines containing pseudoephedrine, a key ingredient used to cook meth, and efforts to thwart drug trafficking from Mexico have disrupted the market for meth...

As the number of meth labs began shrinking in the United States, they have been replaced by "superlabs" in Mexico and Mexican-run labs in some U.S. border states. DEA Acting Administrator Michele Leonhart said interdiction efforts, coupled with U.S. pressure on the Mexican government to reduce imports of pseudoephedrine into that country, have helped cut down meth trafficking across the border.

The thing is, methamphetamine use, as measured by Quest's data, the National Survey on Drug Use and Health, and the Monitoring the Future Study, has been declining at least since 2004. The Combat Meth Act, which restricted retail access to pseudoephedrine and made it such a hassle to buy effective cold and allergy remedies, did not become law until March 2006 and did not take full effect until September 2006. Several states passed their own laws restricting pseudoephedrine before Congress acted. But the only effect on the methamphetamine supply seemed to be a shift from mom-and-pop domestic producers to those Mexican superlabs. In January 2006, for example, The New York Times reported that "the drop in home-cooked methamphetamine has been met by a new flood of crystal methamphetamine coming largely from Mexico." That May, CBS News noted the same phenomenon. Since then, maybe interdiction and pressure on the Mexican government has, as John Walters claims, made a dent in the foreign supply. But the Quest data do not show that.

In fact, a closer look at the national survey data indicates that the decline in meth use probably started around 2000, years before the states passed their anti-pseudoephedrine laws. In the Monitoring the Future Study, both lifetime and past-year meth use by high school seniors fell steadily from 1999 through 2003. Past-year use rose slightly in 2004 before resuming the downward trend. The trend for past-month use was similar, with a more gradual decline. In the National Survey on Drug Use and Health, past-year and past-month meth use have been pretty much flat since the survey began in 2002. Lifetime use fell steadily from then until 2006 (the most recent year for which data are available), when it went up.

We are probably seeing the usual flow and ebb of drug fashions, in which a drug gains popularity and then loses it after enough people run into problems with it to discourage others. It's doubtful that the government's recent supply control effort has had much of an impact on these trends, and it certainly can't be credited with reductions in drug use that predated it.

In the March issue of reason, Greg Beato wondered how Americans learned to stop worrying and love workplace drug testing. Back in 2002, I took a deeper look at the subject.