By:LINDA SEEBACH
Ms. Seebach is an editorial writer at the Rocky Mountain News in Denver

The conventional wisdom has it that the United States imprisons a
disgracefully large proportion of its population "more than any other
industrialized country" is the way it's commonly stated.

People who talk like that seem to think that it matters whether people work
in farms of factories, but it doesn't matter at all how many crimes they
commit. Because by one crucial measure, the probability that someone will
be imprisoned for a serious crime, the U.S. incarceration rate is far lower
now than it was in the 1950s, when the acrime rate was much lower.

No coincidence, that.

The National Center for Policy Analysis in Dallas recently issued "Crime
and Punishment in America: 1997 Update," by Morgan Reynolds. It documents
the commonsense conclusion that when the probability of punishment goes
up, as it has been doing in recent years, the number of crimes goes down.

Just since 1993, the probability of going to prison for murder is up 17
percent, incarceration rate up 5 percent.

Burglary: down 15 percent, incarceration rate up 14 percent.

And as the history of the past halfcentury shows, the relationship holds
in the opposite direction as well. When the probability of punishment goes
down, crime goes up.

The probability is calculated as the number of prison sentences divided by
the number of such crimes. If it seems low, that's because the number of
less seroius crimes such as burglaries, which are rarely solved, greatly
exceeds the number of more serious crimes such as murder, which have a
comparatively high rate of conviction. But the pattern is clear and
consistent.

Probability of prison for serious crime: 1950, 5.27 percent, 1960, 3.63
percent, 1970, 1.33 percent, 1980, 1.57 percent, 1990,2.39 percent, 1995,
2.68 percent

And moving in concert (with a certain lag time while potential criminals
figure out that the rules are achanging):

Serious crimes reported to the police, per 1,000 population: 1950, 5.0,
1960, 5.9, 1970, 14.3, 1980, 22.8, 1990, 19.7, 1995, 16.7

The crime rate is continuing to move down, but it's still three times what
it was in 1950, and the likelihood that someone will be convicted and sent
to prison is still only about half what was then.

Because states' policies have changed at different rates, it's possible to
observe the pattern on a smaller scale.

California and Texas, the report notes, have followed opposite paths. In
1980, the California state prison population was 30 percent below the
national average and its rate of violent crime was 40 percent higher than
average. A decade later, California's prison population was three times
higher and its crime rate a third lower.

Texas started the decade with a prison population 50 percent above the
national average and crime 5 percent higher; by 1989, the prison population
in Texas had fallen 5 percent below the national average and its rate of
serious crime was 38 percent higher.

The conventional wisdom complains that it's too expensive to put all these
people in prison "We're spending more on prisons than on colleges."
That's like complaining that Californians spend a lot more on earthquake
preparation than Minnesotans. They have more earthquakes, and you spend as
much as you have to protect yourself.

Criminals unlike earthquakes, choose their opportunities. If prison
spending is inadequate to punish crime, the result is more crime.

As victims and their families can testify, that's even more
expensive.

Ms. Seeback is an editorial writer at the Rocky Mountain News in Denver.