A Decision Making Framework For Reducing State Correctional Populations

As discussed earlier in this report, one of the principal factors influencing the state’s correctional costs is the number of inmates in prison and parolees under state supervision. For 2008–09 and 2009–10, the Governor proposes to generate hundreds of millions of dollars in state savings in corrections by reducing the inmate and parolee populations. Below, we discuss the historical causes of the increased prison population and costs, provide a framework for considering population reduction options, critique the Governor’s proposals, and offer an alternative that we believe could better balance the important trade–offs inherent in any proposal to reduce state correctional populations.

Corrections in a Historical Context— Rising Caseloads and Spending

Significant Prison Growth Driven by Several Factors. As previously noted, the prison population has increased significantly over the past 20 years. The factors contributing to this increase are the (1) number of new admissions sent to prison by criminal courts, (2) amount of time served by non–lifer inmates, (3) number of inmates in prison with life sentences, (4) number of parolees returned to prison by criminal courts for new felony offenses, and (5) number of parolees returned to prison by the state’s administrative revocation process. As Figure 5 shows, most of these factors have increased significantly between 1987 and 2007.

Based on the above findings, we estimate that about two–thirds of the total increase in the prison population since 1987—67,000 of the 104,000 total inmates—can be attributed to the increase in court admissions, including both new admissions and parole violators returned to prison by the courts. (We explore the reasons for the growth in court admissions in the box below.) The increase in the lifer population has contributed to an additional 26 percent of the population growth. We estimate that the increase in parole violators returned to prison by CDCR and the average time served in prison combined contributed to only about 9 percent of the growth. Figure 6 summarizes the share of the prison population increase that can be attributed to each of these explanatory factors.

Increases in Inmate Population and Incarceration Costs Drive Prison Expenditures. As previously discussed, General Fund spending on corrections (including all formerly separate departments that are currently included in CDCR as a result of a 2005 state reorganization of correctional agencies) has significantly increased over time. For instance, from 1987–88 to 2007–08, corrections spending increased from $1.7 billion to $10.1 billion, an average annual increase of 9 percent. By comparison, total General Fund spending has grown at a slower average annual rate of 6 percent. As a consequence, spending on corrections takes up about twice as much of the state budget than it did 20 years ago, increasing from 5 percent to 10 percent of total General Fund spending.

The growth in corrections expenditures is in effect a result of (1) having substantially more inmates and parolees in the state correctional system and (2) the increased costs to incarcerate and supervise those offenders. As shown in Figure 7, the average cost to incarcerate an inmate has more than doubled over the past 20 years from about $20,000 in 1987–88 to about $46,000 in 2007–08, an average annual increase of about 4 percent. One of the main reasons for this is the growth in inmate health care costs, which have increased by over $1.5 billion since 2000 and have been largely due to the outcome of litigation in federal court over inmate health care. Increases in security–related expenses (primarily for correctional officer salaries and benefits) and other incarceration expenses (such as for transportation, reception and diagnosis, education programs, and administration) have also driven up the average incarceration cost. As shown in Figure 7, the average cost to supervise as parolee has also significantly increased.

Reducing Inmate and Parolee Populations Key to Reducing Costs

Since corrections expenditures make up 10 percent of the state’s total General Fund budget, it is reasonable for the Legislature to consider reducing CDCR’s budget to help address the state’s current massive General Fund shortfall. As discussed above, the three principal drivers of corrections cost increases over the past 20 years are inmate health care, employee compensation for security staff, and prison and parole populations. Of these three factors, reductions to the prison and parole populations have the most potential to achieve large–scale budget savings. For example, reductions to inmate health care expenditures would be difficult to realize given the requirements of various federal court orders and settlements. Therefore, while we recommend that the Legislature consider all options available to reduce corrections costs (including employee compensation costs), we believe that reductions to the inmate and parole populations (such as those proposed by the Governor) have the most potential for significant General Fund savings in both the short– and the long–term. (As discussed in the box below, federal courts are currently considering whether to order a reduction in California’s inmate population to reduce overcrowding as a means to provide a constitutionally adequate level of inmate medical care.)

Criteria for Evaluating Population Reduction Options. In general, the Legislature can choose from a number of policy options to reduce the state’s prison and parole populations and to generate General Fund savings. As with any type of budget cut—whether in corrections or another state program—we recommend that the Legislature carefully weigh the trade–offs inherent in the various options to reduce state correctional populations. Each option carries different benefits, particularly in regard to the magnitude of state savings that can be achieved, as well as differing potential negative consequences or challenges to implementing the reduction. Below is a brief discussion of the five primary criteria we think the Legislature should consider when evaluating various population reduction options:

• Budget Savings. What is the magnitude of savings that will be achieved? To what extent is the actual level of savings dependent on changes to department operations? How quickly will the savings level be achieved?
• Public Safety. How will the option affect public safety? Can any negative impacts to public safety be mitigated by the use of evidence–based correctional practices, such as risk assessments, community–based sanctions, and substance abuse and other treatment programs? Will the option help to reduce recidivism rates of offenders?
• Prison Overcrowding. To what extent will the option reduce prison overcrowding? To what extent does the particular population reduction option result in ancillary benefits, such as avoiding the need to build additional prison bed capacity? To what extent does a particular option facilitate improved prison operations for inmate health care services and other programs?
• Ease of Implementation. Does the option require only simple actions (like statutory changes) or something more complicated (like implementing a new program)? Will savings be delayed because of implementation requirements, such as to conduct reviews of inmates’ case files or to lay off state workers?
• Shift of Responsibilities to Local Governments. Will the option increase local costs to incarcerate more offenders in county jails or supervise offenders on county probation? What impact will the option have on jail overcrowding? Will the option affect local law enforcement or court–related workload?

There are unlikely to be any options to reduce the state corrections populations that produce only positive benefits with no trade–offs. In other words, there are no “perfect” solutions. Instead, we recommend that the Legislature review various options with an eye towards identifying those options that (1) best meet legislative policy goals, including achieving state savings and assuring public safety, and (2) mitigate the potential negative trade–offs.

Three General Strategies for Reducing State Corrections Populations. We would categorize the available policy options for reducing inmate and parole populations into three general strategies:

• Reducing prison admissions from the criminal courts.
• Reducing the amount of time inmates serve when sent to prison.
• Reducing parole caseloads and returns to prison.

Figure 8 lists the various specific population reduction options within each of the above general strategies, which we describe in detail below.

Strategy 1: Reducing Prison Admissions From Criminal Courts

In general, reducing prison admissions from the courts has the potential for the greatest state savings. This is because for each person not sent to prison, the state saves the costs to (1) incarcerate that individual for their crime (an average of two years), (2) supervise that offender on parole for three years, and (3) incarcerate the offender if he is returned to prison on a parole violation. As a consequence, each person diverted from state prison saves the state an average of at least $100,000 over several years.

Make Certain Offenses Ineligible for Prison. One specific option for reducing court admissions is to change sentencing laws to make certain crimes ineligible for state prison. For example, as we discussed in our Analysis of the 2008–09 Budget Bill (pages D–112 through D–117), the Legislature could change “wobblers” to misdemeanors. Wobblers are crimes that current law allows to be prosecuted in the criminal courts as either felonies or misdemeanors. These include a variety of property and drug offenses, such as vehicle theft, forgery and fraud, and drug possession. Currently, there are about 28,000 inmates who are in state prison for wobbler offenses. Changing these crimes to misdemeanors—making them ineligible for prison—would save the state approximately $750 million annually in both incarceration and parole costs. The offenders would instead be incarcerated in local jails and/or supervised by county probation departments, mitigating the effect on public safety. (As discussed later in this report, we propose changing drug possession and select other crimes to misdemeanors and realigning the affected offenders to local jail and probation. Counties would be provided additional resources to supervise these offenders and provide them with intensive substance abuse treatment services.)

Although changing wobblers to misdemeanors would be relatively simple to implement through only statutory changes, this option would shift a large number of offenders to the county correctional system. Moreover, this particular option could increase jail overcrowding and already–sizable adult probation caseloads. Such burdens could be lessened by changing only select wobblers to misdemeanors, thereby reducing the number of offenders affected (as well as the level of state savings), or by identifying a funding source to help offset the new county correctional costs. Another potential trade–off is that the sentencing change would reduce the incarceration time, and thus the severity of punishment, for the affected offenders compared to current law. This is because the maximum jail sentence for a misdemeanor is one year, which is about six months less than the actual time these offenders would serve in prison on average.

Change Property Crime Thresholds. Current law uses dollar thresholds to distinguish between some crimes and assign penalties accordingly. For example, a theft is considered petty theft (a misdemeanor not eligible for punishment with a state prison sentence) when the value of the property taken does not exceed $400. Otherwise the theft is considered grand theft—a felony potentially punishable with prison. For some crimes, the dollar threshold has not been updated in a number of years. One option to reduce the prison population would be to update these thresholds for inflation, which would in turn reduce the number of offenders eligible for felony–level penalties. The actual level of savings from this option is unknown but could potentially be in the tens of millions of dollars annually within a couple of years. Updating property crime thresholds would be relatively simple to implement, requiring only a statutory change. The affected offenders—probably no more than a couple thousand—would be incarcerated and supervised by local correctional agencies.

Divert Offenders to Community–Based Programs. Another option is to expand the availability of local community–based punishment and treatment programs and require the courts to divert certain offenders to these programs. These programs could include intensive probation supervision, substance abuse and mental health treatment, and day reporting centers. Proposition 36 (enacted by the voters in November 2000) and drug courts (now a discretionary choice for judges during sentencing) are two examples of this type of approach that are currently being used. If, for example, 10 percent of total court admissions were diverted to community–based programs, the state inmate population would be reduced by about 13,000 offenders, saving the state more than $300 million annually. Ideally, the offenders would be diverted to programs that are evaluated and proven to be effective at reducing recidivism, thereby having a positive effect on public safety.

There are, however, a couple of trade–offs with this option. First, the affected offenders would be supervised in the community instead of incarcerated in state prison and, thus, county corrections costs would be likely to increase. In addition, some of the affected offenders could potentially commit new crimes while in the community and jeopardize public safety. Such impacts could be mitigated by targeting only lower–risk felons who would otherwise go to prison. It would take significant financial investments—offsetting, perhaps, one–third or more of the total savings—to expand the availability of community–based programs. These activities would likely also delay the date when savings could be achieved.

Strategy 2: Reducing the Time Served by Prison Inmates

Another strategy for the Legislature to consider is reducing the amount of time inmates serve in prison for their offenses. Such a strategy would reduce the prison population and place the affected offenders in the community under state parole supervision sooner than would occur under current law. Under the current revocation process, parole agents would have the option to send these offenders back to prison, which would help mitigate the potential public safety impact of this strategy. In addition, there is little conclusive evidence that the length of time someone serves in prison affects his recidivism rate. However, it is certainly true that early release means that the affected offenders would be in a position to commit crimes that they could not commit if they remained incarcerated. Finally, these options would have little direct impact on local governments because there would be no direct shift of responsibilities from the state to county probation or jails.

Reduce Sentences for Certain Crimes. One option for reducing time served in prison is to change current sentencing laws. For most felonies, current law provides criminal court judges with a choice of three prison terms—the sentencing triad. For example, first–degree burglary is punishable by two, four, or six years in prison, and the judge can choose which of these sentences is most appropriate given the circumstances of the crime and offender’s criminal history. In addition, current law provides for a number of sentence enhancements—additional time that can be added to an offender’s sentence—based on such factors as prior offenses or possession of a weapon during the commission of the crime. The “Three Strikes and You’re Out” law is one example of a sentence enhancement. The Legislature could choose to reduce the triad sentences for certain crimes, or it could reduce or eliminate particular sentence enhancements. The reduction in the prison population and commensurate savings level would depend on the specific statutory changes made, but savings could certainly reach hundreds of millions of dollars annually.

The ease of implementing this option, as well as when the savings would be achieved, would depend on how the option was adopted. Specifically, if the option was adopted on a retroactive basis—meaning it would apply to existing inmates as well as future convicts—it would generate savings more quickly because current inmates would be released earlier than they otherwise would. However, such a retroactive policy would be somewhat difficult to implement because release dates would have to be recalculated, either by resentencing offenders in court or by CDCR recalculating these dates. On the other hand, if the option was adopted on a strictly prospective basis—meaning the reduced sentences would only apply to those convicted after implementation of the law change—it would be much easier to implement because the release dates of existing offenders would not need to be recalculated. However, a prospective policy of reducing state prison sentences would delay any budget savings.

Release Certain Inmates Early. Another option is to release selected inmates from prison early. For example, the Governor proposed last year to release all inmates with no history of serious, violent, or sex offenses from prison 20 months early. This proposal, which was not adopted, would have reduced the prison population by 26,000 inmates and saved the state over $500 million annually when fully implemented. Early release can be targeted towards particular types of inmates—based on factors like age, infirmity, or offense history. In crafting an early release policy, the Legislature also can adjust the periods of time by which sentences are reduced. For example, it could allow specified inmates to be released one day early or months in advance. The reduction in the state prison population and the state savings achieved would depend on the specific eligibility criteria and how early the inmates were released.

A major concern with releasing inmates early is that they could very well commit crimes while in the community when they otherwise would be in prison. While this possibility can never be eliminated, one way to reduce the possibility is to make the individual offender’s risk level the primary criteria for releasing offenders early. The CDCR currently uses a validated risk assessment tool to determine the likelihood that an offender will commit new offenses after being released from prison. The department’s assessments have found that roughly 30 percent of inmates released are of low risk to reoffend, 40 percent are of moderate risk, and 30 percent are of high risk.

Increase the Early Release Credits That Inmates Can Earn. Most inmates are eligible to earn credits towards reducing time off of their sentence, such as by participating in a prison work assignment or education program and refraining from disciplinary problems. Most inmates earn “day–for–day” credits, one day off their sentence for each day that they participate in work or a program. In order to achieve budgetary savings, the Legislature could increase the amount of credits inmates are eligible to earn. One current example of increased credit–earning is that inmates who work in the state’s fire camps currently earn twice the usual credits. The level of savings that could be achieved from increasing credits for other inmates would depend on how many inmates were eligible for the credits and the additional amount of credits these inmates could earn. This option would require some effort by CDCR to implement because it would have to change the way that the department calculates these credits, which is already a complicated task.

Reduce Time Served by Parole Violators. About 70,000 parole violators are returned to prison through an administrative process by CDCR parole agents and the Board of Parole Hearings (BPH) each year, serving an average of about four months. A policy to reduce this prison time by one month, for example, could reduce the inmate population by about 5,000 inmates and save the state over $100 million annually. However, this option could be difficult to implement. Currently, BPH deputy commissioners have broad discretion for determining appropriate revocation sentences. So, legislative action designed to reduce the revocation time served would probably require standardizing in statute the period of time that offenders are returned to prison for various parole violations.

Strategy 3: Reducing Parole Caseloads And Returns to Prison

As discussed above, about 70,000 parolees are returned to state prison by BPH each year, with another 20,000 parolees returned by the courts for new felony convictions. Consequently, at any given time, there are about 60,000 inmates in prison for a parole violation—20,000 who were returned by BPH and 40,000 returned by the courts. Therefore, strategies to reduce the number of parole violators returned to prison can have a significant impact on reducing the state’s prison population and associated costs. In addition, reducing parole returnees can yield ancillary fiscal benefits to the state. These are related to reducing the number of parole revocation hearings, processing of inmates in and out of prisons at reception centers, and local assistance payments to county jails for holding parole violators awaiting their hearings.

Implement Programs to Reduce Reoffending. According to the research, a variety of in–prison and community–based programs have been found to be effective at reducing the likelihood that offenders will commit new offenses when back in the community. These include substance abuse and mental health treatment, education, and employment training programs. The research also finds that many of these programs are cost–effective, meaning that the savings achieved through reduced reincarceration and other factors are greater overall than the cost to implement the program. While the Legislature has invested significant additional funding for rehabilitation programs in recent years, CDCR still does not have sufficient program capacity to provide these programs to all inmates and parolees who would benefit from them. Therefore, the Legislature could choose to expand these programs (or to make other changes designed to make them work more effectively). The key benefit of this option is that it is designed to improve public safety by reducing rates of reoffending.

The impact of this option on the prison population and costs would depend on the number and types of programs expanded. Research shows that even the most effective correctional programs will not eliminate all reoffending. In fact, a 10 percent reduction in reoffending is often considered a notable achievement for such programs. A 10 percent reduction in parolee returns to prison would reduce the inmate population by about 6,000 inmates and reduce state incarceration costs by about $150 million annually. However, most of these savings would be offset by the costs to implement and expand the programs. In addition, expanding these types of programs would take time and, especially in prison settings, could be difficult given current overcrowded conditions at many prison facilities. Locating acceptable sites for parole programs in the community can also be challenging. Therefore, significant savings would be unlikely in the first few years, though savings in the longer term are certainly achievable if the programs are well–designed and implemented.

Make Certain Parole Violations Ineligible for Prison. As shown in Figure 9, about 44 percent of the 72,000 parolees returned to prison for parole violations in 2007 had a technical violation as their principal offense (such as for failing a drug test or not attending a required meeting). Another 37 percent were returned primarily for misdemeanors, and 19 percent were returned primarily for new felonies. One option to reduce state costs would be to prohibit parole returns for certain types of less serious violations. For example, prohibiting returns for technical and misdemeanor violations could reduce the state prison population by about 16,000 inmates and save the state $400 million annually.

The most significant trade–off with this option is the potential impact on public safety. Though the affected offenders would remain under state parole supervision, they would receive no prison or jail sanction for their violation. Moreover, this option does not account for the fact that, while the new violation is a potentially low–level technical or misdemeanor violation, the offender may have a more serious and violent criminal history that nonetheless makes him a risk to public safety if retained in the community. One approach for mitigating these public safety concerns would be to only allow parole violators who are evaluated as low–risk for violence or to reoffend to remain in the community when a less serious parole violation has occurred. In addition, this option could be implemented in conjunction with another option we discuss below to ensure that those offenders who are no longer eligible for a prison return for their violation are diverted to community–based sanctions and treatment programs.

Divert Parole Violators to Community–Based Programs. Just as the courts could divert certain new felons to community–based sanction and treatment programs, CDCR could divert parole violators to such programs. In fact, as part of the 2008–09 budget package, the Governor directed the department to implement the Parole Violations Decision Making Instrument (PVDMI), a system designed to guide parole agents in determining whether a parole violator should be directed to community–based programs (rather than returned to prison). However, the 2008–09 budget did not include additional funding to expand the availability of such programs, making it unclear whether significant savings can be achieved at this time from implementation of the PVDMI. The Legislature could choose to invest in additional community–based programs. If they were successful at diverting 25 percent of parole violators from prison, for example, the prison population would decline by 5,000 inmates and state incarceration costs would decline by about $125 million annually. However, a third or more of these savings would be offset by the costs to implement the new community–based programs. The exact costs would depend on the number and types of programs implemented.

Expanding community–based programs for parole violators would likely take a couple of years to implement if developed statewide on a large scale. Therefore, the estimated total savings would likely not be achieved in the first couple years. The other trade–off with this option is that some of the affected offenders could commit additional offenses while in the community. However, such an outcome could be avoided somewhat if certain high–risk and violent offenders were deemed ineligible for diversion.

Discharge Certain Parolees Early. Nearly all inmates are placed on parole supervision after they are released from prison, with most serving a standard three–year parole term. Under current law, parolees are eligible to be discharged from parole as early as one year after release if they have not been returned to prison for a parole violation and their parole officer recommends early discharge. One option would be to allow parolees to be discharged even earlier than after one year. Qualification for early discharge could be based on the offense history or risk level of the parolee. The policy could also be based on whether the parolee meets certain criteria associated with successful community reintegration (such as steady employment, stable housing, and no evidence of drug use)—a concept referred to as earned discharge. Early and earned discharge would reduce the parole population, and to a lesser extent would reduce the inmate population because there would be fewer parolees eligible for revocation to state prison. The level of potential savings from implementing early discharge could reach a few hundred million dollars annually, depending upon which parolees were eligible and how much the time that they were required to serve on parole was reduced.

The primary trade–off with this option is the potential impact on public safety. Affected offenders would no longer be under state supervision, and if they committed new offenses, they would not be eligible for return to prison through the state’s revocation process. However, they could still be convicted in court for any new crimes. In addition, early discharge would take some additional effort by CDCR staff to implement. That is because the department would need to identify which inmates met the criteria for early discharge and verify that they meet any relevant criteria before discharging them. (In our 2008–09 Budget: Perspectives and Issues publication, we proposed the realignment of parole to counties. Our proposal has many of the same state benefits as early discharge but provides better public safety outcomes by keeping the offenders under local probation supervision. We continue to recommend consideration of this approach.)

Governor’s Budget Includes Three Population Reduction Proposals

Summary of the Governor’s Proposals. The Governor’s budget includes three proposals intended to reduce the inmate and parole populations and generate combined savings that would reach $800 million annually at full implementation (2010–11). Figure 10 shows the savings level for the three proposals, as well as other related savings from implementing the proposals.

The population reduction proposals included in the Governor’s 2009–10 budget plan are identical to those the administration proposed as part of the November and December special sessions to address the serious state budget shortfall. In total, this set of proposals would reduce the state corrections populations by about 16,000 inmates and 66,000 parolees at full implementation. We describe each of the Governor’s proposals below.

• Early Discharge. The Governor proposes direct discharge (no parole) for all inmates released from prison who have no current or prior serious, violent, or sex offenses. This proposal would be responsible for a majority of the estimated reduction in the inmate and parole populations under the Governor’s plan.
• Credit–Earning Enhancements. The Governor also proposes to allow inmates to earn additional credits that would reduce the time they would otherwise serve in prison. Most of the estimated savings would result from two specific changes: (1) allowing many inmates to earn the same level of credits while in local jails, awaiting transfer to state prison, that they can already earn in state prison (one day of credit for each day served) and (2) allowing many inmates to earn additional credits for completing rehabilitation programs, including academic or vocational education and substance abuse treatment.
• Update Property Crime Thresholds. The administration proposes to update certain property crime thresholds for inflation. Under this proposal, many property crime thresholds would more than double. For example, the threshold for grand theft would increase from $400 to $950.
• Ancillary Savings. The Governor’s budget plan identifies other areas of the department’s budget that could be reduced if the inmate and parole populations were reduced substantially. This includes savings for revocation hearings, local assistance payments to county jails, and various community–based programs for parolees.
• Implementation and Other Costs. The administration identified about $7 million in ongoing costs for the department to implement the above changes, particularly to do the casework necessary to recalculate release dates and identify which inmates are eligible for early discharge. The proposed budget also includes adjustments to take into account actions already scored in the department’s budget, such as savings related to the implementation of the PVDMI, so as not to double–count the savings.

Proposals and Estimated Savings Seem Reasonable. We find that the administration’s estimates of the savings that could be achieved from these proposals are reasonable, although the actual savings would depend on various factors. These factors include (1) the number of discharged offenders who commit new crimes for which they are sent to prison, (2) the number who are no longer eligible for state prison because the value of their property offense no longer meets the statutory threshold, and (3) the number who successfully complete prison programs. The administration’s estimated savings for the current and budget year assume implementation by February 1, 2009. The administration has acknowledged that actual savings would be reduced in the short–term if implementation was delayed.

The Governor’s proposed package of reforms generally provides a reasonable starting point for the Legislature to consider making correctional population reductions a part of its approach to balancing the state budget. For example, the proposed increase in credit–earning would result in only a modest reduction in time served for most inmates. In addition, the program completion credits may actually help reduce the likelihood that offenders commit new crimes after release by providing an incentive for them to participate in programs that are linked to reduced recidivism. Our analysis also indicates that the change in the property crime thresholds would result in only a small shift of property offenders from state prison to the local corrections system.

Direct Discharge Proposal Raises Concerns. The Governor’s early discharge proposal to release tens of thousands of offenders from prison without any community supervision (either through state parole or county probation) raises some concerns because of its potential impact on public safety. Moreover, as now proposed by the administration, the release of particular offenders would not be based upon any assessment of their risk to reoffend. In order to help alleviate some of these public safety concerns, we propose below an alternative package for the Legislature to consider.

LAO’s Alternative Proposals for Correctional Population Reduction

In developing an alternative package of correctional population proposals, we attempt a better balance between the need to achieve budgetary savings and the goal of also protecting public safety. Our alternative essentially builds upon the Governor’s proposals by modifying the direct discharge proposal and replacing it with an earned discharge strategy. We believe earned discharge would provide a better balance with public safety by tying the discharge of offenders from parole to factors such as their risk to reoffend and actual behavior in the community. Our alternative would retain the Governor’s proposals to enhance credit–earning and changing the thresholds for various property crimes. Our alternative package results in roughly the same magnitude of savings as the Governor’s plan when fully implemented.

The primary trade–off of implementing earned discharge instead of direct discharge is that our approach would achieve less state savings than the Governor’s budget proposal. Our alternative makes up for this loss of savings by including three additional policy changes: (1) changing petty theft with a prior, currently a wobbler, to a misdemeanor, (2) expanding the use of alternative community–based sanctions for parole violators, and (3) allowing early release (by two months) for those inmates who are evaluated as being low–risk to reoffend and who have no current or prior serious, violent, or sex offenses. Figure 11 summarizes the components of our alternative package.

LAO Alternative Has Several Advantages. At full implementation, we estimate our alternative package would reduce the prison population by about 23,000 inmates, and save the state about the same amount as the Governor’s plan ($800 million). (Over time, it is possible that the total savings could be even greater as the department modifies other areas of its operations to take into account the decrease in the prison population.) Most of these savings could begin to be realized immediately or within a few months. Only the implementation of alternative sanction programs would take more time in order to expand the availability of these programs. The level of savings that would be achieved in 2009–10 would depend on when the policies are implemented. A later implementation date would reduce budget–year savings because of the time it takes to change department policies and procedures, as well as to eliminate the affected staff positions. For example, we estimate that about one–half of the potential $800 million savings could be achieved in 2009–10 if the proposals were implemented by March 1, 2009. Implementation on July 1, 2009, on the other hand, would reduce budget–year savings, possibly to as little as one–third of the total savings achievable at full implementation.

Importantly, our proposed alternative would minimize the risk to public safety by focusing on lower–level offenders, utilizing risk assessments, and expanding the use of best practices like alternative sanctions and earned discharge. Finally, we estimate that this plan would have only minimal direct impact on local governments. Although the change of petty theft with a prior to a misdemeanor would shift about 4,400 inmates (and about 5,600 parolees) to the local corrections system, this would represent only a 2 percent increase in the local corrections population.

While we think the above package has advantages compared to the Governor’s proposals, it is certainly not the only set of options available to the Legislature. So, the Legislature could certainly remove, replace, or modify any of these components to reflect its own choices about the trade–offs between achieving state savings and protecting the safety of the public, as well as consideration of the other criteria.

Impacts on Prison Facilities From Reducing the Inmate Population

We would note that before the Legislature adopts any policies designed to reduce the inmate population, it should require CDCR to present a plan during budget hearings that would identify how it would implement these reductions. The plan should take into consideration how the projected reduction in the inmate population would be spread among inmates in reception centers, prisons of various security levels, and female prisons. Further, the plan should present how the administration would prioritize the reduction of different types of beds, such as those in gyms and dayrooms, contracted facilities, and reception centers. This plan should also assess the extent to which population reductions would eventually delay or eliminate the future need to construct prison facilities. This above plan would help the Legislature evaluate the full impact of significant population reductions on both operational and capital costs.