The Contribution of Prisons to Public Safety


The following memo was sent to Oregon legislators in response to a hearing on Senate Bill 856, the so-called "Smart on Crime" bill.

A correction was made on April 28 and clarifications were added on May 5 and May 12.


Date:          April 9, 2007 (correction added April 28, clarifications added May 5, May 12)

From:        Howard Rodstein, Policy Analyst and Director

To:             Senate Judiciary Committee

CC:            House Judiciary Committee, Public Safety Ways and Means Committee

                  President Peter Courtney, Senator Brown, Senator Ferrioli

                  Speaker Jeff Merkley, Representative Dave Hunt, Representative Wayne Scott

Subject:     What have we received for all this prison construction?


During the recent hearing on the “Smart on Crime” bill, Senator Walker raised the question of what Oregonians have received for all of our prison construction.[1] I was surprised to hear the question and cannot let it pass without answering.


The short answer is that Oregon’s criminal justice system has gained vastly-increased credibility with criminals, victims as well as with law-abiding citizens; we have prevented many people from engaging in crime; we have saved taxpayers’ dollars; and, most importantly, we have prevented tens of thousands of victimizations. This has been by far the most effective state government initiative in recent memory.


The Oregon Criminal Justice Commission (CJC) 2007 Report to the Legislature[2] says that, for each inmate added in 1995, about 28 crimes were avoided. In 2005 about 11 crimes were avoided for each additional inmate. The average over that period is about 20 crimes avoided each year for each additional inmate. From July of 1995 to July of 2005, Oregon added about 5,500 inmates. According to the CJC analysis, this should translate to about 110,000 avoided crimes per year.


[At the end of April, the Oregon Criminal Justice Commission estimated that 98,786 crimes were prevented in 2005 by the increase in incarceration from 1995 to 2005. Accounting for the effects of Senate Bill 1145, the 1997 bill that moved inmates serving less than 12 months from prisons to county jails or to community supervision, would raise their number very close to my estimate of 110,000. ]


The Oregon Prison Population Forecast[3] says that 3,700 additional prison beds are required because of Measure 11. That's 3,700 armed robbers, kidnappers, child molesters, rapists, drunk drivers who maimed and killed people, attempted murders, murderers and other violent criminals who are in prison instead of on our streets because of Measure 11. It is impossible to say exactly how many crimes don't occur because these criminals are in prison. But we can say that in 1995, Oregon had 16,408 reported violent crimes including 129 murders and manslaughters while in 2002, despite increased state population, we had 10,208 reported violent crimes including 72 murders and manslaughters[4].


The CJC 2007 Report also says that we can expect a 3.4 percent decrease in the violent crime rate and a 2.6 percent decrease in the overall crime rate for every 10 percent increase in incarceration.


[ The preceding paragraph was revised on April 28 from my original memo. In the original memo, I said we had ten 10-percent increases in incarceration rate and that this translated to 34 and 26 percent decreases in violent and overall crime respectively. After further research, I determined that my original calculation of ten 10-percent increases was wrong and it is not clear what the right number is. I regret the error. See Addendum below for details. ]


I find it disturbing that many in the Capitol building are unaware of or dismissive of these facts.


Opponents of incarceration are fond of telling us that Oregon ranked 3rd among states in prison growth[5]. You will never hear them tell you that, despite this growth, Oregon is still a low-incarceration state.


In 1995, Oregon ranked 42nd among states in incarceration rate[6]. After all the prison construction, we now rank 34th. In 2006, only 21 percent of convicted felons received prison sentences[7].


[My original memo said, “In 2005, only 16 percent of convicted felons received prison sentences.” This statement was based on a letter that the Oregon Criminal Justice Commission sent us in early 2007. We accurately quoted the CJC letter. On May 10th the CJC informed us that the figure they provided was incorrect. They were unable to give us a corrected figure for 2005 but did report that in 2006, 21 percent of convicted felons received prison sentences.]


There is not a single inmate in an Oregon prison solely for drug use. 70 percent are in for violent crimes, 15 percent are repeat and repeat-repeat property criminals, 10 percent are drug manufacturers and/or dealers and the remaining 5 percent are classified as "statute" or "other"[8]. Despite what you hear in the halls and hearing rooms of the Capitol, this is hardly a state that that is “over-incarcerating”.


Prison population growth in Oregon is slowing. It is projected at 4 percent for the next biennium. Less than half of the increase in the Governor’s DOC budget is due to prison population growth.


In the current biennium, the cost of Measure 11 is $223 million, according to the Legislative Fiscal Office. This amounts to $31 per Oregonian per year. This is less than the monthly cost of cable television for most families. And it pales into insignificance compared to the “crime tax” extracted from Oregonians by criminals each year[9].


Presented with these facts I think that the vast majority of Oregonians would agree that what Oregon has received for “all this prison construction” is well worth the price – in fact, it’s the best value they receive for their tax dollar.


Now prisoner advocates and some legislators are attempting to dismantle a system that has restored the credibility of Oregon’s criminal justice system, protected all Oregonians and spared countless families from tragedy. Their main argument is cost. To dramatize it, they claim we are spending more on corrections than higher education. This is wrong. Figures from the Legislative Fiscal Office show that the Governor’s 2007-2009 all-funds budget for higher education, excluding tuition funds, is $3.9 billion while the all-funds budget for corrections, excluding parole and probation, is $1.1 billion.


May 5, 2007 Update: After rechecking with the Legislative Fiscal Office, we have learned that the governor’s 2007-2009 budget calls for spending approximately $1.95 billion of taxpayers’ money on higher education. Here is a breakdown we got from LFO:





Total Higher Ed Budget

$4.9 billion


Federal Funds

$1.1 billion

Mostly federal income tax money for research and student aid

General Funds

$827 million

Mostly state income tax money

Charges and Fees

$670 million

For housing, meals, etc.

Capital Expenditures

$600 million

From bonds and other sources

Nongovernmental grants and donations

$325 million



Not only are their figures wrong, but the entire premise of their comparison is misguided. Parents can and do put their children through college without the help of the state through years of hard work and sacrifice – countless thousands have done so for generations. But parents cannot make sure that drug manufacturers who poison their children, predatory child molesters, drunk drivers who maim and kill, and murderers are incapacitated. Only the state can do that by fulfilling its first responsibility – to protect law-abiding innocent people.


Judging from the bills they have filed[10], their idea of “modest sentencing reform” includes cutting sentences for violent crimes by up to 50 percent and increasing earned time for drug manufacturers and repeat property offenders from 20 percent to 33 percent.


Instead of lobbying the legislature to put more criminals on the street, they should be working to change a culture that tolerates, excuses, and in some cases even promotes criminality.


A brief look at our history is needed to put these issues in context.


From 1960 through 1985, violent crime in Oregon increased by nearly 700 percent[11]. The response of state government during this 25-year period was to build one new prison with a capacity of 400 beds. By 1988, crime in Oregon was so bad that Department of Corrections Director Michael Francke admitted that the state’s corrections system had “lost its credibility - with law abiding residents and with criminals themselves. ‘We don't have a deterrent to the guys on the street,’ Michael Francke said. ‘We've lost control of it.’"[12]


In 1994, voters took matters into their own hands by passing Measure 11 with a two-thirds majority. Measure 11 required minimum mandatory sentences for serious violent crimes – aggravated assaults, armed robbery, kidnapping, child molestation, forcible rape, manslaughter, attempted murder, and murder. It went into effect on April 1, 1995.


In 2000, voters reaffirmed their decision by rejecting the repeal of Measure 11 by a nearly three-fourths majority, despite a campaign of well-documented misrepresentations[13] from repeal supporters.


In 1995, Oregon ranked 42nd among states in incarceration rate and had 522 reported violent crimes per 100,000 residents. Our state then experienced seven consecutive years of decreasing violent crime. Our violent crime rate dropped to 292 per 100,000 residents. The total decrease was 44 percent, compared to 28 percent nationally. Oregon ranked first among states in reducing the violent crime rate over that period[14].


Relative to the 1995 violent crime rate, we now have a savings of approximately 8,000 reported violent crimes per year. Since 1995 this savings adds up to 66,000 fewer violent crimes[15].


We do not claim that Measure 11 was solely responsible for this decrease in violent crime. But there is ample evidence to show that increased incarceration made a significant contribution. For example, the Oregon Criminal Justice Commission (CJC) 2007 Report to the Legislature says that violent crime decreases by 3.4 percent for every 10 percent increase in incarceration2. Since 1995, Oregon has experienced ten such 10 percent increases in incarceration.


Opponents of incarceration dismiss the connection between declining violent crime rates and increased incarceration in Oregon on the grounds that violent crime rates decreased all over America. A prominent Oregon prisoner advocate was recently quoted in the Register-Guard as saying "There is recognition that our current state strategy hasn't made us any safer."[16] This view cannot explain why Oregon led the country in violent crime decrease from 1995 to 2002. It also ignores the fact that nearly every state reacted to soaring crime with increased incarceration. In the 1990's, thirty-three states increased time served for violent crimes[17]. In addition, twenty-three states implemented three-strikes laws[18]. And according to the RAND Corporation report[19] on Measure 11, by 1994 all 50 states implemented some form of minimum mandatory sentencing, making Oregon the very last state to do so.


The simple fact is that keeping violent criminals off the street prevents violent crime.


While a strong stand against violent crime contributed to a stunning decrease in violent crime, a tepid response to drug and property crime has allowed these problems to fester and created generations of drug-addled criminals who prey on law-abiding Oregonians. As of 2005, Oregon ranked 4th among states in property crime[20]. To a large extent we brought this upon ourselves through our increasing tolerance of drug abuse and criminality.


We support treatment for drug addiction so long as it is rigorously tested for effectiveness. But we cannot treat ourselves out of our drug problem when drug dealers, serving large-scale manufacturers, are working day and night to hook new users. It takes far longer to detoxify an addict than to create a new one. This will be a losing battle until we change broad societal attitudes about drug abuse and restore the value of personal responsibility. Minimizing the seriousness of drug manufacturing and dealing and tolerating drug use will only perpetuate our problems.


A lot of attention is paid to recidivism of released prisoners. We are told that, if only we spend more money on treatment, we can reduce the recidivism rate. The implication is that we are wasting money locking up criminals instead of spending more money on treatment programs which are portrayed as panaceas. A closer examination reveals that “evidence-based” treatment programs are estimated to produce modest decreases in recidivism – 6.3 percent for highly-touted evidence-based cognitive/behavioral programs, for example[21].


During the period of dramatically increased prison population and dramatically lower spending per inmate on treatment, we have seen decreases in recidivism that surpass those projected for evidence-based cognitive/behavioral programs. The average recidivism for the 8 years prior to Measure 11 was 33.5 percent while the average recidivism for the 8 years after Measure 11 was 30.8 percent[22]. This represents an 8 percent decrease in recidivism despite dramatically reduced treatment program spending[23].


If you believe that evidence-based programs will dramatically reduce crime, try it out on the 34,000 adult[24] criminals who are on parole and probation in our communities. If you are right, you will dry up the stream of criminals going into prison without sacrificing additional innocent people through major cuts in sentences for the most serious criminals.


A credible criminal justice system not only keeps criminals in prison – it keeps others out of prison. Deterrence is a well-known mechanism though quantifying it is difficult. But there is another, almost universally-ignored mechanism. A paper[25] by Flores, Latessa and others lists the "big four" criminogenic risk factors as criminal history, antisocial attitudes, personality and associates. In other words, who you hang out with has a significant impact on whether you engage in crime. The 5,500 inmates added from 1995 to 2005 represent 5,500 criminogenic risk factors removed from our society. There is no escaping the conclusion that keeping these people in prison contributes to keeping others out of prison.


A story involving one walking criminogenic risk factor who was not sentenced under Measure 11 is instructive. James Daniel Nelson was released from prison in 2003 after serving 11 years for murder under sentencing guidelines. Though his behavior in prison was atrocious and he was supposedly on post-prison supervision, he wasted no time in recruiting a new “street family”. Within two months, this family, at Nelson’s direction, stabbed, burned and murdered 22-year-old Jessica Kate Williams[26]. Thirteen family members were convicted of crimes ranging from assault to murder. The cost to taxpayers of releasing James Daniel Nelson is in the multiple millions. If Measure 11 had been in effect when Nelson was convicted of his first murder, he would have been in prison until 2017 – a costly proposition to be sure, but far less costly than releasing him. Jessica Kate Williams would be alive and Oregon would have 12 fewer prisoners.


Incarceration is not cheap but it is very cost-effective, especially incarceration of violent criminals. A Washington State study found a savings of $8.20 in 1995 and $4.30 in 2005 for each dollar spent incarcerating an additional violent criminal. The CJC report says "there are many similarities between Oregon and Washington that make these estimates seem reasonable for Oregon. "[27]


A recent Multnomah County case[28] illustrates how a supposedly enlightened disdain for incarceration can lead to human tragedy and squander taxpayers’ money. In the name of “the best interests of the child”, two teenagers who had engaged in serious crimes involving drugs and guns were allowed to remain “on probation” in the community instead of being sent to a juvenile facility where they would have gone to school, learned discipline and received intensive counseling. Supposedly under close watch, they were uncontrolled and undeterred. According to police, one of them, Joseph Eugene Allen, 16, shot the other, Davonte Garland Lightfoot, 14. Now one teenager is dead, another is facing a long stay in a juvenile facility, and taxpayers are left to pay the bill - all in the name of “the best interests of the child.”


As reported in the New York Times[29], after more than a decade of decreases, violent crime in cities is now showing “a sharp surge”. In city after city, rising violent crime is front-page news[30]. This trend has not reached Oregon – yet.


Our best tool to prevent this from happening here is the tool that has worked effectively over the last 12 years – putting serious violent criminals and repeat property offenders in prison. Now is no time to flinch. Calls to do so represent a false economy which will boomerang as they have in the past. If we go down that road, we will pay many times over in both human and economic terms.


The voters of Oregon want, deserve and are willing to pay for a no-nonsense criminal justice system that puts public safety first. Oregon legislators should heed this voter mandate.


[1] Senate Judiciary Committee, March 26, 2007, minute 24.


[2] Oregon Criminal Justice Commission 2007 Report to the Legislature,


[3] Oregon Prison Population Forecast, October 2006,


[4] FBI Uniform Crime Reports


[5] Eugene Register-Guard, January 28, 2007, “Changes in Prisons Urged”


[6] Bureau of Justice Statistics, Data Online


[7] Oregon Criminal Justice Commission


[8] Oregon Criminal Justice Commission 2007 Report to the Legislature


[9] According to a landmark 1996 study by the National Institute of Justice, crime cost "$105 billion annually in property and productivity losses and outlays for medical expenses. This amounts to an annual 'crime tax' of roughly $425 per man, woman, and child in the United States. When the values of pain, long-term emotional trauma, disability, and risk of death are put in dollar terms, the costs rise to $450 billion annually (or $1,800 per person)." Accounting for inflation would make these figures grow to about $621 and $2484 respectively. These figures represent the cost to each citizen outside of tax payments for law enforcement, incarceration, and other criminal justice system costs. (

[10] SB 631, SB 2904, SB 1014, HB 2880


[11] Bureau of Justice Statistics, Data Online


[12] Salem Statesman-Journal, February 21, 1988


[13] The Oregonian, “Campaign Mixes Fact, Fiction”, October 28, 2000


[14] Bureau of Justice Statistics, Data Online


[15] violent_crime_savings.pdf


[16] Eugene Register-Guard, January 28, 2007, “Changes in Prisons Urged”


[17] Bureau of Justice Statistics


[18] Justice Policy Institute,


[19] RAND Corporation Report on Measure 11,


[20] Bureau of Justice Statistics Data Online


[21] Washington State Institute for Public Policy, “Evidence-Based Public Policy Options To Reduce Future Prison Construction, Criminal Justice Costs, And Crime Rates”, October 2006. This paper predicts that the use of evidence-based cognitive-behavioral programs will result in a decrease in recidivism from 63 percent to 59 percent – a change which is characterized as a 6.3 percent decrease in the recidivism rate.


[22] Oregon Department of Corrections Recidivism Report, November 1, 2006,


[23] Oregon Criminal Justice Commission 2007 Report to the Legislature, Graph 8


[24] October 2006 Prison Population Report


[25] "Evidence of Professionalism or Quackery: Measuring Practitioner Awareness of Risk/Need Factors and Effective Treatment Strategies",


[26] The Oregonian June 30, 2003


[27] Oregon Criminal Justice Commission 2007 Report to the Legislature


[28] The Oregonian, January 10, 14 and 16 and March 7, 2007


[29] New York Times, March 9, 2007, “Violent Crime in Cities Shows Sharp Surge”


[30] Minneapolis Star Tribune, March 8, 2007; Wisonsin State Journal, March 17, 2007; Seattle Times, March 21, 2007; Palm Beach Post, April 1, 2007; Philadelphia Inquirer, April 2, 2007 . . .




The following explains my error in the original version of this memo which said Oregon had ten 10-percent increases in incarceration from 1995 to 2005.


My error was a result of mixing total (prison and jail) incarceration from 2005 with prison incarceration from 1995. This mixup was a result of a confusing spreadsheet from the Bureau of Justice Statistics (see Bureau of Justice Statistics Spreadsheet below).


On reviewing this with Michael Wilson of the Oregon Criminal Justice System, we came up with five ways of comparing the 2005 incarceration rate to the 1995 rate.


Increase in Incarceration Rate

Number of 10-percent increases

Data Used

41 percent


Prison and jail inmate population based on Bureau of Justice Statistics data

50 percent


Yearly average prison population calculated from DOC data

56 percent


Mid-year prison population from Bureau of Justice Statistics data

62 percent


Start-of-year prison population from DOC data

77 percent


Prisoners sentenced to more than 12 months as reported by the Bureau of Justice Statistics


In 1997, the Oregon Legislature passed Senate Bill 1145 which moved prisoners serving less than one year from DOC prisons to county jails or to community supervision. Because of this change the 50, 56 and 62 percent figures are comparing populations that changed character between 1995 and 2005. Therefore these figures should not be used.


The 41 percent figure (increase in prison and jail incarceration rate) and 77 percent figure (incarceration of prisoners serving more than 12 months) arise from comparing like quantities and thus are more meaningful.


It is not clear which of these figures is pertinent in the context of the Oregon Criminal Justice Commission’s finding of a reduction of 2.6 percent in overall crime rate and 3.4 percent in violent crime rate for each 10-percent increase in incarceration rate. This is a question for future study.



Bureau of Justice Statistics Spreadsheet


This spreadsheet, from the Bureau of Justice Statistics, was the source of my confusion which resulting in my mixing of total (prison and jail) incarceration rates with prison incarceration rates:


PDF version of this memo