Economic Scene: Less Marijuana, More Alcohol?

WHAT do teen-agers do when they are priced out of the market for
marijuana?

Some, presumably, take oboe lessons or join the 4-H Club. But others
look for solace in less wholesome pursuits. And, surprisingly,
economists may have more to say on the subject than toilers in the
fields of psychology or criminology.

Drug policy is grounded on the premise that illicit drugs are birds of
a feather -- that reducing the availability of one decreases the
consumption of others. But economists who measure the demand for
illicit substances the way, say, Exxon analyzes the demand for grades
of gasoline, challenge this conventional wisdom. Their identification
of a strong substitution effect between marijuana and alcohol suggests
that the full court press against the weed is partly responsible for
stubbornly high levels of binge drinking by teen-agers.

According to the University of Michigan's Institute for Social
Research, the proportion of high school seniors regularly using
marijuana fell to 14 percent last year, barely one-third the rate
reported in 1978. Their use of alcohol has been on the wane, too,
slipping by a fourth since the late 1970's.

This might seem proof that alcohol and marijuana drugs are complements
-- more like bread and jam than cake and pie. But simple correlation
cannot account for the slew of factors that influence drug consumption
over time and place.

That is where a yet-to-be-published study by John DiNardo of the
University of California at Irvine and Thomas Lemieux of Princeton fits
in. Their work, supported by the Rand Corporation and the National
Institute of Alcoholism and Alcohol Abuse, focuses on the mid-1980's,
when the threat of losing Federal highway aid forced states to adopt a
uniform minimum drinking age. In 1980, 1/2 the states had minimum
drinking ages ranging from 18 to 20. Eight years later, all states
were up to age 21.

The two economists estimated demand curves for marijuana and alcohol,
using a variety of data that might influence consumption -- everything
from parents' education to unemployment rates -- to isolate the effect
of drinking sanctions.

The good news is that the higher "price" for alcohol -- that is, the
greater difficulty of obtaining it -- reduced drinking. The bad news:
Other factors being equal, raising the drinking age from 18 to 21
increased the proportion of high school seniors who smoked marijuana by
an estimated 10 percent.

To Peter Reuter, an economist at Rand, this conclusion is most
interesting for what it implies about marijuana policy in the 1980's.
If marijuana is a substitute for alcohol, he notes, alcohol is, by
definition, a substitute for marijuana. Thus tough marijuana
enforcement must increase drinking. And, indeed, another new study for
the National Bureau of Economic Research by Karen Model suggests Mr.
Reuter is on the mark.

Ms. Model, a Ph.D. candidate at Harvard, examined the impact of
marijuana decriminalization on hospital emergency room admissions for
drug abuse reported to the Federal Drug Abuse Warning Network in the
mid-1970's. And as the substitution hypothesis would suggest, Ms. Model
found that emergency room episodes related to drugs other than
marijuana were 12 percent lower in the states that had decriminalized
the weed. Lowering the effective "price" of marijuana, she concluded,
reduced the abuse of other substances.

The data did not allow Ms. Model to isolate alcohol emergencies from
those caused by the use of heroin, cocaine or prescription chemicals.
But Ms. Model believes alcohol is far and away the most likely drug
replaced by marijuana. Both alcohol and marijuana were widely seen by
users as "soft" recreational drugs, in contrast to, say, cocaine,
heroin or LSD.

Marijuana and alcohol use are both down; why, then, worry? Because the
level of teen-age alcohol abuse remains remarkably high.

In 1991, some 30 percent of high school seniors reported having had
five or more drinks in a row sometime in the previous two weeks. The
comparable figure for college students (almost all of whom had to break
the law to obtain alcohol is 43 percent -- and there is no downward
trend.

To those who focus on the risk of accidental injury and other medical
crises, heavy drinking seems a more serious worry than marijuana. Ms.
Model found that other factors equal, states decriminalizing marijuana
reported lower overall rates of drug- and alcohol-related emergencies.

And while both substances have been implicated in auto accidents, Frank
Chaloupka, an economist at the Chicago campus of the University of
Illinois, believes that substitution toward marijuana is, on balance, a
life saver.

In a statistical analysis that parallels Ms. Model's, he found that
states without criminal sanctions against marijuana possession suffered
fewer auto fatalities.

"If the choice is more marijuana use or more dead teen-agers," Mr.
Reuter concludes, "the choice is easy."