Percent of Released Prisoners Returning to Incarceration


Crime in America.Net

We have a reader request; she asks about recidivism of people released from prison (arrests, convictions, returns to prison).

Possibly one of the most significant studies on recidivism (defined for this report as a return to prison) was offered by the Pew Center on the States, Pew Safety Performance Project and released in April, 2011.

The study analyzed returns to prison for 33 states for those released in 1999 and 41 states for those released in 2004 making the study the most comprehensive analysis of returns to prison ever done.

The report compared earlier studies on recidivism conducted by the US Department of Justice for 15 states for releases in 1983 and 1994 and concluded that recidivism rates “have consistently remained around 40 percent.” California was excluded from this finding due to that state’s large percentage of the nation’s prison population and the fact that California’s rate of return to prison is traditionally high.

The report singles out three states, Oregon, Michigan and Missouri for lowering rates of return to prison.


Before the Pew report there was nothing of a national and substantive nature except for a US Department of Justice study from the Bureau of Justice Statistics that was released in June of 2002 based on inmates released from prison in 1994.

Note that there are a variety of older studies from the Bureau of Justice Statistics that remains definitive for many years. This study involved a very large set of data from 15 states and the methodology (quality of the research) was very good. This is the most quoted recidivism study in the country.

It followed offenders for three years and tracked results.

Principle Findings:

The study (see summary below) found that:

Two-thirds (sixty-seven) of offenders were arrested for “serious” crimes.

Fifty-two percent of the offenders were returned to prison for “serious” crimes and technical violations (they didn’t follow the rules of their release after prison).

Current Studies:

There are a wide variety of state and local studies that also examined recidivism but none with the same large data set (numbers of people followed) and none with the same rigorous methodology. We will address these studies in a future post.

The Bureau of Justice Statistics is redoing the study discussed here and will have results in 2011.

Main Characteristics of Recidivism:

The study found two primary variables in recidivism which seem to be present in virtually all past and current studies:

Younger prisoners and those with longer records were more likely to be rearrested.

Post-prison recidivism was strongly related to arrest history.

Can these Results be Changed?

Yes. Interventions providing programs for offenders both in and out of prison can lower recidivism from 10 to 20 percent. See:

Previous Study:

A previous study titled “Recidivism of Prisoners Released in 1983” was released in April of 1989.  It was an analysis of the criminal records of more than 16,000 men and women, representing the almost 109,000 offenders who were released from prisons in 11 States during 1983.

The study links correctional data with federal and state criminal history records to provide a complete portrait of criminal careers for more than a half of the State prisoners released.

About 47% of the former prisoners were convicted of a new crime and 41 percent were sent back to prison or jail.


2002 Study:


WASHINGTON, D.C.—Sixty-seven percent of former inmates released from state prisons in 1994 committed at least one serious new crime within the following three years.

This was a rearrest rate 5 percent higher than that among prisoners released during 1983.

State prisoners with the highest rearrest rates were those who had been incarcerated for stealing motor vehicles (79 percent), possessing or selling stolen property (77 percent), larceny (75 percent), burglary (74 percent), robbery (70 percent) or those using, possessing or trafficking in illegal weapons (70 percent).

Those with the lowest rearrest rates were former inmates who had been in prison for homicide (41 percent), sexual assault (41 percent), rape (46 percent) or driving under the influence of drugs or alcohol (51 percent).

About 1 percent of the released prisoners who had served time for murder were arrested for another homicide within three years, and about 2 percent of the rapists were arrested for another rape within that period.

Within three years, 52 percent of the 272,111 released prisoners were back in prison either because of a new crime or because they had violated their parole conditions (e.g., failed a drug test, missed a parole office appointment).

Men were more likely to be rearrested than were women (68 percent, compared to 58 percent), blacks more likely than whites (73 percent vs. 63 percent) and non-Hispanics more than Hispanics (71 percent vs. 65 percent).

Younger prisoners and those with longer records were also more likely to be rearrested.

Post-prison recidivism was strongly related to arrest history.

Among prisoners with one arrest prior to their release, 41 percent were rearrested. Of those with two prior arrests, 47 percent were rearrested. Of those with three earlier arrests, 55 percent were rearrested. Among those with more than 15 prior arrests, that is about 18 percent of all released prisoners, 82 percent were rearrested within the three-year period.

The 272,111 inmates had accumulated more than 4.1 million arrest charges prior to their current imprisonment and acquired an additional 744,000 arrest charges in the 3 years following their discharge in 1994 – an average of about 18 criminal arrest charges per offender during their criminal careers.

These charges included almost 21,000 homicides, 200,000 robberies, 50,000 rapes and sexual assaults and almost 300,000 assaults.

Almost 8 percent of all released prisoners were rearrested for a new crime in a state other than the one that released them. These alleged offenders were charged with committing 55,760 new crimes in states other than the imprisoning state within the three-year period. New York, Arizona and California had the most arrests of out-of-state offenders in this study.

The data were from the largest recidivism study ever conducted in the United States, which tracked prisoners discharged in 15 states representing two-thirds of all state prisoners released in 1994.

They were 91 percent male, 50 percent white, 48 percent black, 24 percent Hispanic (of any race) and 44 percent were younger than 30 years old.

Most of them had been in prison for felonies: 22 percent for a violent offense (such as murder, rape, sexual assault or robbery), 33 percent for a serious property offense (mostly burglary, motor vehicle theft or fraud), 33 percent for a drug offense (primarily drug trafficking or possession) and 10 percent for public order offenses (mainly drunk driving or weapons crimes).

Most former convicts were rearrested shortly after getting out of prison: 30 percent within six months, 44 percent within a year, 59 percent within two years and 67 percent by the end of three years.

The study findings are based upon the prison and criminal records of an estimated 272,111 discharged prisoners in 15 states who were tracked through fingerprints records made at various points of contact with the justice system, both within the state in which they had served time and other states to which they traveled.

The BJS special report, “Recidivism of Prisoners Released in 1994″  was written by BJS statisticians Patrick A. Langan and David J. Levin.






  2. Ken Vanderploeg says:

    I am a bit confused by the data presented: first it states that 67% of released prisoners in 1994 committed at least one new serious crime within 3 years but then it says only 52% were back in jail for either a new crime or for not complying with conditions of parole. That means that a lot of those that committed another serious crime never saw more jail time. I don’t know what qualifies as a “Serious crime” but does not result in jail time especially when it is not a first time offence.

  3. stanton wilson says:

    Michigan released thousands of inmates back into society, only to fail to arrest them and return them to prison after they violated their parole or committed new crimes. Currently there is a study to show how many murders have been committed by early released prisoners. Many parolees are not being supervised and therefore no one knows the actual status of these parolees. 11,000 rape kits went untested for years in Michigan, resulting in 21 serial rapist still remaining on the streets after only 600 kits being tested. Michigan needs to be excluded from this study as they have done little to protect the public. As I recall “protecting the public” is a part of

  4. Prisoners returning o prison


  1. […] in and out of prison because he simply refuses to admit there is a problem and turn his life around? 52% of offenders return to prison because they don’t follow the rules after prison. Admittedly, intervention programs can reduce this risk  – but here must be something […]

  2. drug treatment houston

    52 Percent of Released Prisoners Return to Prison-67 Percent Rearrested

  3. […] So, Harry, did you know that the people around you shape who you are? Of course you do. If you don’t check out this article on “recidivism”. […]

  4. […] have to pay for his stay is discouraging, especially when he’ll have nearly a 50% chance of ending up back in prison. The type of control they are looking for is to prevent these types of people from being able to […]

  5. […] whopping 52% of released prisoners become repeat offenders. If they have the skills and experience to get a job after getting released, they’ll be less […]

  6. […] reforming after prison sentences is not great. A study by the Pew Center on the States found that half of inmates are back in prison within three years of being released from jail. It is a little better in sports […]

  7. […] of those people put themselves at risk of being incarcerated, in which case, they would have had a 52% chance of returning to prison with in 3 years, and a strong possibility of not finding a steady […]

Leave a Reply