STUDY SHOWS BUILDING PRISONS DID NOT PREVENT REPEAT CRIMES

The rate at which inmates released from state prisons commit new crimes
rose from 1983 to 1994, a time when the number of people behind bars
doubled, according to a Justice Department study released yesterday.

The report found that 67 percent of inmates released from state prisons in
1994 committed at least one serious new crime within three years. That is 5
percent higher than among inmates released in 1983.

Criminologists generally agree that the prison-building binge of the last
25 years, in which the number of Americans incarcerated quadrupled to
almost two million, has helped reduce the crime rate simply by keeping
criminals off the streets. There has been more debate about whether longer
sentences and the increase in the number of prisoners have also helped to
deter people from committing crimes. The new report, some crime experts
say, suggests that the answer is no.

"The main thing this report shows is that our experiment with building lots
more prisons as a deterrent to crime has not worked," said Joan Petersilia,
a professor of criminology at the University of California at Irvine and an
expert on parole.

A likely reason for the increase in recidivism, Professor Petersilia said,
is that state governments, to save money and to be seen as tough on crime,
cut back on rehabilitation programs, like drug treatment, vocational
education and classes to prepare prisoners for life at home.

Only about 15 percent of state prison inmates are enrolled in academic or
rehabilitation classes, she said.

The report was prepared for the Bureau of Justice Statistics, the
statistical branch of the Justice Department, by Patrick Langan and David
Levin. It is the first major study of recidivism in more than a decade. The
report examined 272,111 former inmates in 15 states during the first three
years after their release.

The report indicates that the first year after an inmate is released is
critical to his or her success in returning to civilian life. For example,
the study found that two-thirds of the inmates who were rearrested were
rearrested within 12 months of their release.

In addition, the report found that the number of times prisoners had been
arrested was the best predictor of whether they would commit more crimes
after being released and how quickly they would return to their criminal ways.

Prisoners who have one prior arrest have a 40.6 percent recidivism rate
three years after being released, the study said. With two prior arrests,
the rearrest rate within three years climbs to 47.5 percent. With three
prior arrests it rises again, to 55.2 percent within three years. Then,
with each additional prior arrest, it continues to rise, reaching a
rearrest rate of 82.1 percent for inmates with more than 15 prior arrests.

Prisoners with a greater history of previous arrests are likely to be
rearrested faster. Prisoners with one prior arrest have a 21 percent
rearrest rate within a year of their release. But inmates with 16 or more
prior arrests have a 74 percent recidivism rate within the first year after
their release, the study found.

Professor Petersilia said the finding about the recidivism rate reaching
67.5 percent in three years was particularly striking because earlier
studies, dating to the 1960's, have all found that rearrest rates of
American prisoners tend to hover at about two-thirds within the first three
years after release.

These findings have occurred even as the prison philosophy of the day has
shifted from rehabilitation to getting tough on crime to deterrence, with
seemingly little difference in the outcome.

Jeremy Travis, a senior fellow at the Urban Institute, said the report
showed that "the population getting out of prison continues to pose a high
risk of reoffending." This is true, at least in part, Mr. Travis said,
because state prison departments, with a few exceptions, have done a poor
job of managing the critical period when an inmate is released.

Often, Mr. Travis said, inmates are released having received little or no
job training, drug treatment or education in how to be a better parent.
Many are unable to find jobs and are barred by law from living in public
housing projects, so they quickly return to crime, said Mr. Travis, a
former director of the National Institute of Justice and deputy police
commissioner in New York City.

The report found that people who commit crimes involving money are more
likely to be rearrested than those committing homicide or rape, with a
recidivism rate of 70 percent for robbers and 74 percent for burglars
compared with 41 percent for those whose previous crime was homicide and 46
percent for those whose previous crime was rape.