Analysis of the 2008-09 Budget Bill: Criminal Justice

Adult Corrections (5225)

Who Is in Prison?

There were 173,312 inmates in the prison population as of June 30, 2007. About 93 percent of the population is male. Other demographics of the inmate population include the following:

• About 49 percent of inmates are incarcerated for nonviolent offenses.
• About 62 percent of all inmates were committed to prison from Southern California, with about 33 percent from Los Angeles County alone and 8 percent from San Diego County. The San Francisco Bay Area is the source of about 11 percent of prison commitments.
• About 46 percent of all inmates are between 20 and 34 years of age, with the number of inmates falling dramatically starting at age 50.
• Of the total inmate population, about 38 percent are Hispanic, 29 percent are black, and 27 percent are white.
• About 63 percent of the inmates are new admissions from the courts, 25 percent are offenders returned by the courts for a new offense while on parole status, and 12 percent are parolees returned to prison by administrative actions for violation of their conditions of parole (see Figure 1).

Inmate and Parole Population Management Issues

Inmate Population Projected to Increase

The California Department of Corrections and Rehabilitation (CDCR) is projecting the inmate and parole populations to increase in the current and budget years. The inmate population is projected to grow more slowly than was projected in the spring, while the parole population is projected to grow more quickly. These projections do not take into account a separate proposal by the administration for the early release of some inmates from prison and to not actively supervise certain offenders on parole which would significantly reduce the inmate and parole populations.

Inmate Population Projected to Increase. As of June 30, 2007, CDCR housed 173,312 inmates in prisons, fire and conservation camps, community correctional facilities, and out–of–state facilities. The CDCR forecasts the inmate population will increase to 179,105 by June 30, 2009, a projected two–year increase of 5,793 inmates, or about 3 percent, compared to the beginning of the current fiscal year. The projected increase in the inmate population is the result of a recent trend of increasing admissions to prison from county courts, as well as more parole violators returned to prison through the state’s administrative returns process. Figure 2 shows the year–end inmate and parole populations for the period 1998 through 2009, with and without the Governor’s population reduction proposals.

Parole Population Projected to Increase. As of June 30, 2007, CDCR supervised 126,330 persons on parole. As shown in Figure 2, CDCR projects the parole population to increase to 134,092 by the end of the budget year, an increase of 7,762, or 6 percent. This increase is primarily a result of the increase in the number of inmates released to parole after serving their prison sentence.

Impact of Governor’s Policy Proposals Affecting CDCR Populations. The administration’s budget includes two policy changes—20–month early release and summary parole—that would significantly reduce the inmate and parole populations by a combined 47,000 offenders in 2008–09. (See our discussion of the Governor’s population reduction proposals later in this chapter.) While the Governor’s budget includes estimates of the direct fiscal impact of these policies on the costs of prison and parole operations, the reduction in the inmate population is not reflected in the department’s bed plan. That means it is unclear whether new contracted beds that are proposed are, in fact, necessary given the number and security level of offenders that will remain in the prison system.

Similarly, the budget plan does not reflect the Governor’s budget proposals (discussed in more detail in our analysis of the “Department of Alcohol and Drug Programs” in the “Health and Social Services” chapter) to reduce funding for Proposition 36 and drug court allocations to counties. Because Proposition 36, an initiative approved by voters in 2000, and drug court programs allow offenders to receive drug treatment in lieu of their incarceration in prison (or county jail), the proposed reduction in spending for these programs could begin to increase the prison population during the budget year.

Costs of Projected Population Growth. The CDCR is requesting additional funds of about $18 million in the current year, growing to $111 million in the budget year related to changes in adult inmate and parole populations. This does not include the fiscal impact of the early release and summary parole proposal which is accounted for in Control Section 4.44 of the budget act. These proposals largely reflect the additional costs to incarcerate the projected increase in the inmate population, in particular by expanding the use of contracted out–of–state and female community correctional facilities.

Housing the Projected Growth in Inmate Population. The Governor’s budget proposes a housing plan to accommodate the additional inmates that CDCR expects to receive by the end of the budget year. The plan has the following major elements:

• Out–of–State Beds. Over the course of the current and budget years, the department’s housing plan includes the activation of about 7,000 contracted beds operated in facilities in other states, reaching a total of 8,000 by the end of the budget year. State inmates began filling those beds in late 2006. As of December 31, 2007, CDCR had about 2,100 inmates housed in out–of–state facilities, an increase of about 1,000 from the beginning of the fiscal year.
• Female Rehabilitation Community Correctional Centers (FRCCCs). The department proposes to contract for a total of 1,300 FRCCC beds by the end of the budget year, growing to 2,000 beds in future years. The FRCCCs would be secure facilities designed to house low–security, female offenders and provide programs and services designed to reduce the likelihood that these offenders reoffend after their release to the community.
• Overcrowding of Existing Prison Space. The housing plan assumes that the activation of additional contracted community beds in the current and budget years will allow the department to reduce the number of inmates placed in temporary housing by about 2,200. Historically, CDCR prisons utilize gymnasiums, dayrooms, and other space as temporary housing to accommodate overcrowding.

Reentry Facilities Not Included in Bed Plan. The department’s budget includes some funding in the current and budget years related to opening two reentry facilities—a 600–bed facility in Stockton (the conversion of the Northern California Women’s Facility) and a 48–bed facility in San Francisco. The funding provided for these facilities is for a pre–activation team in Stockton and program operations at the San Francisco facility. However, the additional beds provided in these facilities are not incorporated into the department’s bed plan described above. If these new reentry facilities are activated as proposed, the department’s bed plan will have to be revised to account for the additional beds available elsewhere in the prison system.

Potential Risks to Accuracy of Projections. As we have indicated in past years, the accuracy of the department’s latest projections depends upon a number of factors, changes to any of which could result in significantly higher or lower populations. These factors include sentencing laws, crime rates, and local criminal justice practices. For example, these projections do not take into account a recent court case alleging that CDCR has tens of thousands of inmates who have incorrectly calculated release dates. If this challenge is upheld by the courts, it could result in some number of inmates being released from prison earlier than they otherwise would, thereby reducing the inmate population relative to the department’s current projections.

Caseload Will Likely Require Further Adjustment

We recommend a $55 million reduction in the 2007–08 and 2008–09 budget requests for caseload funding because recent data indicate that the population is trending lower than the department’s projections. We will continue to monitor the caseload and recommend further changes, if necessary, following review of the May Revision. (Reduce Item 5225–001–0001 by $55 million.)

Actual Inmate Population Has Declined. The fall 2007 projections, which are the basis for the Governor’s budget proposal, anticipated that the inmate population would grow by about 400 inmates during the first half of 2007–08. Instead, the inmate population has declined by almost 1,900 inmates during the past six months. The fall 2007 projections anticipated that the parole population would grow by about 3,100 parolees during the first half of 2007–08. However, the parole population grew by less than 600 over this period. If these trends hold, population–related spending in the budget plan could be overstated by as much as $55 million in both the current and budget years.

Analyst’s Recommendation. Given the recent inmate population trend, we recommend the Legislature act to reduce the budget in the current and budget years by $55 million. The administration should also provide fiscal estimates of the impacts on corrections of proposals to reduce funding for Proposition 36 and drug courts. We would note that CDCR will issue population projections in spring 2008 which form the basis of its May Revision proposal.

Budget Process for Population Changes Needs More Work

Last year, the Legislature enacted budget bill language directing the California Department of Corrections and Rehabilitation to improve its current population budget request in order to make it a more transparent document for legislative oversight. While the department has taken initial steps to comply with legislative direction, there is additional work still needed. Accordingly, we recommend several steps that would significantly improve the process the department uses to budget for changes in the inmate population that will further enhance transparency, as well as provide for a more accurate budget request and a more efficient budgeting process.

Current Steps in the Population Budgeting Process

Prison and Parole Budgets Adjusted for Changes in Population. The budget for a prison is determined by several factors, including its mission, security level, programs, facility design, and size. Each prison has a budget that includes a base complement of funding and staff necessary to manage and operate the facility, and includes support for administrative, security, and other functions. In addition, like many other state departments, CDCR receives additional budget adjustments to account for changes in caseload, in particular changes in the number of inmates and parolees housed and supervised by the department. These adjustments generally include resources for food, clothing, inmate health care, administration, and security staffing. Figure 3 shows the amount of funding provided in the Governor’s budget for the caseload–related changes. As shown in the figure, the Governor’s population budget request (as it is known) also includes funding for other issues, including inmate mental health caseloads, contracted facilities, and the state’s juvenile ward and parolee population.

The CDCR’s process for creating the population budget request is one that takes several months and is completed twice each year as part of the state’s standard budget process. The first time is as part of the Governor’s budget request submitted January 10 of each year, and the second is as part of the May Revision. We describe the basic steps of the process in more detail below.

Population Projections. The process of identifying necessary budgetary changes begins with the identification of what change in the inmate and parolee populations is likely to occur. To this end, in the summer of each year, CDCR staff analyze data on recent and historical trends that affect inmate and parolee populations, including numbers of court admissions, parole revocations, average time served by offenders in prison, and discharges from parole. Using this data, CDCR projects the inmate and parolee populations over the next several years. Department staff update their projections in the winter to serve as the basis of the May Revision adjustment. Figure 2 (earlier in the “Adult Corrections” analysis) shows the actual inmate and parolee populations in recent years, as well as the department’s most recent population projections.

Institution Activation Schedule (IAS). Using the population projections, the department then creates the IAS for the prisons. The IAS takes the inmate population projections, as broken down by gender and security level, and specifies which housing units at each prison will have to activate or deactivate beds (in the case where the inmate population is projected to decrease) each month in order to accommodate the change in population in both the current and budget years. For example, if the projections show that the population of Level III inmates is going to be higher by 20 inmates in March, the IAS might specify that Ironwood State Prison (Blythe) would “double–cell” (placing a second inmate in one cell) at 20 additional cells in its Facility A. Alternatively, the IAS might show that the prison would add 20 more beds to its gymnasium to accommodate the additional inmates.

Staffing Packages Requested for Institutions. Once staff at each institution know how many inmates are projected to be sent to them at various points in the year based on the IAS, they identify how many and what type of positions they would need to provide security and operate other services. Similarly, if the IAS shows that there will be fewer inmates sent to the prison in a year than they now hold, the staffing packages identify what positions at that prison will be cut from the budget. Historically, the department provides for changes in staffing levels based on a ratio of about one staff position for each six inmates. Department policy requires that at least 6 percent of those positions included in each staffing package be for health care staff. Most of the remaining positions are for custody staff, particularly correctional officers, though institutions have flexibility to request other classifications if those would better meet their operational needs. As shown in Figure 4, approximately three–quarters of positions requested by the department in the January 2007–08 budget request were for custody staff. (We would note that this example only includes positions requested for staffing packages related to new activations at adult institutions, and does not include deactivations or other ratio adjustments that are usually made as part of the budget request.)

Staffing Packages Reviewed by Headquarters. After these staffing packages are developed by prisons, they are sent to CDCR headquarters, where they are reviewed by managers. Once approved, headquarters staff identify the budget costs associated with each staffing package and assemble them into the total population budget request. The 2008–09 population budget request includes about 300 separate estimates for activations and deactivations.

Population Budget Request Sent to Legislature. The department’s population budget request includes, in addition to costs based on approved staffing packages, resources for other caseload–related issues, including parole agent caseloads, case records and other ratio–driven staffing needs, juvenile offender populations, and contracted bed usage. As with all budget proposals, the population budget request must be approved by the Department of Finance (DOF) and is then sent to the Legislature for consideration. The entire population budget request generally fills about four, four–inch binders, but the department provides the Legislature an abridged version that was included in two binders this year.

Continuous Adjustments Made to IAS as Actual Population Changes Materialize. As should be expected, the actual population that arrives in state prisons is almost never exactly the same in number or breakout by gender and security level as is projected by the department. Therefore, the IAS has to be updated on an ongoing and continuous basis in order to determine where to assign inmates and what beds need to be activated or deactivated at each institution to accommodate the change in the inmate population that actually occurs. These adjustments, as well as the revised population projections, are the basis of the revised IAS that is included in the May Revision population budget request.

State’s Approach Is Ineffective, Inefficient, and Lacks Transparency

Our analysis of the department’s process of creating its population budget request finds that it is an ineffective approach to identifying the actual budgetary needs of the department, is an inefficient use of department staff time, and fails to provide a transparent budget document for legislative review. We discuss each of these shortcomings in more detail below.

Ineffective Approach to Identify Actual Budgetary Needs

Population Projections Done Too Early to Be Accurate Basis for Budget Request. Because the department’s process is complicated and requires many steps to complete, the department is forced to start its population projections—the fundamental basis of the population budget request—very early. The January budget is based on population trends tracked through the prior June and, thus, do not take into account the most recent seven months of data. Consequently, the projections are often inaccurate. For example, the department’s Fall 2007 projections were already too high by more than 2,000 inmates by the time the January 10 budget plan was released. In this example, a 2,000–inmate discrepancy represents only a difference of about 1 percent in the overall size of the inmate population. Certainly, projections can never be perfect predictors because of the many complex factors affecting the department’s actual caseloads, including crime rates, actual arrests, demographics, court activities, and parole actions. However, a less complicated and more streamlined population budgeting process might allow the department to gather several more months of trend data before completing its projections, thereby improving the likelihood of more accurate projections and, therefore, budget requests that are closer to the funding level the department really needs.

IAS Inaccurate and Potentially Unnecessary. The population budget request includes the IAS, which, as discussed earlier, indicates how many beds will be activated or deactivated at specific facilities at each prison in order to accommodate the projected change in the population. However, prison beds are rarely opened on the schedule laid out in this document. Instead, the activation of new beds is done on a weekly and monthly basis based on the actual population changes that occur, the security and special needs of the incoming population, and changes in availability of beds (for example, due to the repair of cell doors). Therefore, the IAS usually provides little useful information about how the funding provided under the budget would actually be distributed among institutions while making the budget request unnecessarily complicated.

Notably, after completing the IAS and calculating the corresponding changes in staffing costs, the department makes a “below–the–line” budget adjustment to tie its total funding request to a separately calculated aggregate estimate of the change in spending that will result from the projected changes in the inmate population. This aggregate estimate is based on CDCR’s marginal cost to incarcerate an inmate. For example, in the Governor’s January 10 budget plan for 2008–09, this below–the–line adjustment is a $78 million increase for the budget year. This means that the exercise of doing the IAS is largely irrelevant to the amount of funding actually requested in the budget. Moreover, it is not clear that the department’s aggregate estimate of necessary spending changes, which is based on an assumed marginal cost of incarcerating additional inmates, are accurate. For example, the current estimate includes outdated assumptions of costs for certain operating expenses and equipment, such as pharmaceuticals for inmates.

Fixed Staffing Ratio Unresponsive to Operational Needs. In developing the staffing packages that tie to its population budget request, the CDCR has for more than 20 years utilized a fixed ratio that assumes that, for about every six additional inmates projected to come to a prison, that prison will get one additional staff position. However, it is no longer clear that this ratio is appropriate for the modern CDCR. Based on discussions with department staff, the existing fixed ratio methodology appears to date back to a time when corrections, prisons, and population growth were very different from today, particularly given the makeup of the existing population, overcrowding, and expanding health care and programming missions.

One of the key potential problems with the methodology is that it is not responsive to the variation in missions and operations among the state’s 33 prisons. For example, some prisons specialize in housing violent and dangerous inmates; some are reception centers that assess and classify inmates when they first arrive at prison; some focus on providing rehabilitative programs; some are for female offenders; and one is the department’s primary health care facility for inmates. The one–size–fits–all fixed staffing ratio currently employed does not recognize these differences in missions among prisons, perhaps resulting in some prisons being overstaffed, while others are comparatively understaffed. To the extent that prisons are overstaffed compared to what they need to operate effectively, the state is incurring unnecessary costs. Even understaffed prisons could result in unintended costs. For example, they may have to rely more heavily on overtime (a more expensive option than regular time) to make up for having fewer positions than their missions might necessitate.

Many States Do Not Make Population Adjustments at All. Interestingly, it appears that California’s population budgeting process may be somewhat unique nationally. Based on our review of the available literature, as well as conversations with practitioners in other state correctional agencies, many other states do not change staffing levels based on incremental changes in population. Instead, most other states base staffing levels on regular assessments of what staff is necessary to operate housing facilities and programs. It is not clear whether this unique approach to budgeting for CDCR is justified.

Inefficient Use of Staff Resources

The population budget request—produced twice annually—consists of a document that is literally thousands of pages long and requires many hours of CDCR staff time to produce. This includes staff in headquarters and at each institution to develop the IAS, generate and review staffing packages, and produce fiscal estimates. While it is certainly necessary for departments to dedicate staff resources to developing any budget proposal, it is unclear that the amount of staff time dedicated to the current process is really the most efficient use of these resources, particularly given that the final product, as described above, is largely ineffective. A simpler and more streamlined process might allow the department to reprioritize some of these staff resources for better use, such as providing more time to dedicate to the development and analytical review of policy–driven budget change proposals (BCPs).

Lack of Transparency

Length and Complexity Inhibits Careful Review by Administration and Legislature. The length and complexity of the population budget request make it difficult to understand how individual components of the total request tie back to the population projections upon which they are ultimately based. In part, this difficulty is due to the several intermediate steps the current process requires—in particular the creation of the IAS and staffing packages. In addition, complexity is created because there is frequently little detail provided on how individual components of the population budget were calculated or the underlying assumptions and methodologies. Moreover, it is apparent from our discussions with CDCR budget staff that the current methodology is not always well understood even by them. This may be due in part to the complexity of the process as well as the amount of staff turnover that has historically occurred in that office.

Importantly, there is evidence that the complexity of the current process contributes to significant errors. For example, in reviewing the May Revision last year, the Legislature discovered that the department had inadvertently been overestimating its staffing need for mentally ill inmates, a $22 million General Fund mistake in the 2007–08 budget proposal that had regularly occurred since 2005–06. Whatever the specific reasons for these errors, this particular example raises concerns that other large mistakes could be embedded in the population requests that might not be easily identified by DOF or legislative staff in the short time periods that they have to review them.

Population Budget Has Historically Included Non–Caseload Funding Requests. Legislative staff have long been concerned that the department has sometimes included funding requests in the population budget that were not directly a result of caseload changes, but rather policy decisions made by decision makers in CDCR headquarters or institutions. Legislative staff have voiced concern that the population budget request be strictly a technical funding request reflecting only caseload–driven costs.

Separating out technical caseload–related changes from policy decisions is, admittedly, not a simple task in a department in which nearly every change in CDCR is to some degree “population–related.” However, many population–related issues are also driven by policy decisions. For example, funding needs for the correctional officer academy, administrative segregation housing, and medical guarding, while legitimately issues related to population, also are areas greatly affected by department policies and operations, such as staff vacancies, disciplinary procedures, and the quality of health care inside prisons. As such, it may be appropriate that these types of requests be presented to the Legislature separately as policy BCPs.

Status of Recent Efforts to Improve the Process

Recognizing that CDCR’s population budget proposals needed to be improved, the Legislature adopted budget bill language in the 2007–08 Budget Act that requires CDCR to work with legislative staff and DOF to improve the population budget request by defining what issues are allowable in the population request and make it a more transparent document to allow for improved legislative oversight. The department, legislative staff, and DOF staff met several times during the summer and fall to discuss ways to improve the population request, and the department committed to making improvements.

The Governor’s 2008–09 budget proposal includes some initial changes in the way that components of the population budget request are presented. While the length, underlying methodologies, and issues included in the 2008–09 request are generally unchanged from prior years, the request is improved in how it displays some of this information. In particular, the revised document more often than before provides concise, accurate narrative descriptions, as well as the underlying assumptions and methodologies for its calculations. According to the department, the new display that it has begun to use is modeled on caseload budget requests for social services programs. The department also reports that it has dedicated a staff position in its budget office to continue to work on improving its own population budget request. Further, CDCR reports that it intends to make some additional improvements by the May Revision and in the budget year in order to complete the steps necessary to improve its population budgeting process.

Further Steps Necessary to Improve Population Budget Methodology

We recommend that the Legislature adopt language in the 2008–09 budget bill directing the department to continue its current efforts to improve the transparency of the population budget request. The language would also require the department to significantly change its population budget methodology by developing new funding formulas for changes in the inmate population to replace the current reliance on the Institution Activation Schedule and staffing packages.

CDCR Should Continue Current Efforts to Improve Transparency

The department should continue its current efforts to improve the transparency of the population budget request—meaning, its ability to be easily reviewed and understood. In particular, this should include utilizing its new display for additional components of the budget request, such as parole revocation caseload. Also, CDCR should continue its work with legislative staff and DOF to determine which components might be more appropriate to be presented separately as BCPs rather than being included in the caseload request.

In addition, the department should continue to look at whether some of the underlying caseload methodologies could be improved. For example, the calculations for changes in the parolee caseload currently do not include adjustments for each of the various types of caseloads that parole agents actually carry. Specifically, the budget generally assumes a caseload of 70 parolees for each parole agent. However, most parolees are actually on caseloads that range from about 50 parolees to 150 parolees per parole agent, depending on their risk factors. Making these changes would likely make the department’s request more transparent and would more easily allow for legislative review and oversight.

State Should Develop New Method for Changes in Prison Population

While the department’s current population budget request includes some improvements that make it somewhat more transparent, the changes made do not address the underlying problems associated with the complex and ineffective process the department uses to estimate the fiscal needs associated with inmate population changes. The department reports that there was insufficient time during the fall to significantly reform the central components of the budget request.

We agree that making such changes is a significant undertaking and should be done thoughtfully. However, reforms to the process for budgeting for inmate population change could yield significant benefits. Therefore, we offer our recommended approach below for moving ahead with further changes to the process. It is our intent that these recommendations serve as a starting point for further discussions among CDCR, legislative staff, and DOF so that, with additional analysis, the population budget request can be improved. Therefore, we recommend that the Legislature adopt budget bill language directing CDCR, legislative staff, and DOF to again work collaboratively, this time to redesign the methodology for budgeting for inmate population changes.

This new methodology will include several steps, including streamlining the process to complete the population projections later, eliminating the IAS and staffing packages, and using more precise budget formulas that take into account the different populations and missions of individual prisons.

Entire Process Must Be Streamlined. As discussed earlier in this analysis, the inmate population projections that form the underlying basis of the population budget request often prove inaccurate. These inaccuracies are probably at least in part due to the fact that, under the current process, work on the projections must begin many months in advance of the release of the budget request.

To address this key problem, we offer recommendations in more detail below that would streamline the entire population budgeting process, thereby allowing the department to begin and complete its projections later. This would allow the department to incorporate more recent data into its projections and, in all likelihood, improve the accuracy of the population budget request. In addition, we would note that the department has contracted with two academic experts to provide recommendations on how to improve the department’s projections model. If incorporated into their projections, these recommendations may also help to yield more accurate projections.

Eliminate the IAS for Budgeting Purposes. The IAS serves an important ongoing operational purpose for the department by helping to determine into which institutions new inmates will be placed. However, as we have explained, the IAS is both burdensome and largely irrelevant to the population budgeting process. Therefore, we recommend that the IAS no longer be used as part of the population budget request, and that it be replaced with a new and simpler budgeting methodology.

Elimination of the IAS would allow CDCR to halt the time–intensive process of producing, reviewing, and costing out staffing packages—the bulk of the documentation in the department’s multivolume population budget request. As we have noted, the end product does not reflect the resources that will actually be utilized at institutions.

We would anticipate that in lieu of the IAS, the Legislature would receive a higher level summary of the department’s plan to house inmates. For example, this aspect of the budget estimate could display how many inmates the CDCR plans to house through overcrowding of existing prison facilities, in contract beds, in out–of–state beds, and through activation of new prison facilities. This summary information would enable the Legislature to more easily determine whether the budget request is based on cost–effective and reasonable choices as to how inmates would be housed.

Use Formulas for More Precise Budgeting of Resources. Instead of calculating extensive staffing packages to identify the staffing and costs necessary to meet projected changes in population, we recommend that the department—working together with legislative staff and DOF—develop funding formulas to estimate in the aggregate the fiscal resources required for the projected change in the inmate population. These funding formulas would apply a distinctly different staffing ratio for each gender and security level in the prisons rather than rely on the one–size–fits–all historical ratio of one added staff member for every six inmates. For example, an increase in high–security inmates would provide for more staff and funds being added to the budget request than a similar increase in low–security or female inmates.

Number and Mix of Staffing Funded Could Change. The specific funding formulas to be used under our approach would be developed by CDCR based on historical staffing patterns and its expertise in correctional operations. These formulas might result in a different number of staff being added at a particular institution when the population increased instead of the number they would receive under the standard six inmate–to–one staff ratio described earlier.

The revised formula may also be based on a different mix of staffing classifications than before. Clearly, based upon our review of the January population budget request for 2007–08, most new positions funded through the new formula would continue to be custody staff. However, the mix of staff could vary due to the security levels, missions, programs, and priorities of institutions. As shown in Figure 5, the custody positions sought for prisons based on their 2007–08 budget requests varied significantly across institutions, including significant variation between prisons with similar security levels and missions. This suggests that it may be beneficial for the proposed new funding formulas to allow for some flexibility in the classification of positions provided through the population budget requests.

Aim for Changes to Be Cost–Neutral. In designing new funding formulas, the state should aim to make the reallocation of staffing be cost–neutral, so as to not result in additional costs to the state or fewer resources for the department. In the future, when the state’s overall fiscal condition has improved, these formulas could easily be adjusted if, for example, CDCR were able to demonstrate that the formulas significantly underfunded a particular security level of inmates.

Allocate Funding Through Headquarters. This approach would require that these resources be allocated to CDCR headquarters rather than to individual institutions, with CDCR headquarters, in turn, allocating funding to individual institutions as actual population changes unfolded. Just as is done now, budget adjustments would be made in the January and May Revision population requests to reflect the difference between the actual and projected population in the prisons, as well as to reflect changes in the projected population.

Adopt Budget Bill Language to Ensure Continued Progress. To date, CDCR staff has worked collaboratively with legislative staff and DOF to improve its population budget request. Moreover, department officials have continued to express commitment to making further changes. To ensure that these efforts continue along the lines we have proposed, we recommend the Legislature adopt the following budget bill language:

Item 5225–001–0001—California Department of Corrections and Rehabilitation. The California Department of Corrections and Rehabilitation (CDCR) shall continue its efforts in consultation with legislative staff and the Department of Finance to create a more accurate and transparent population budget request for caseload–related funding. In particular, CDCR shall identify appropriate funding formulas to use to estimate staffing levels and funding associated with changes in the projected inmate population. These formulas shall be presented to the Legislature no later than January 10, 2009 so as to be considered during budget deliberations. If approved, these formulas shall be incorporated into CDCR’s budget request for the following year.

Resources and Operations Would Be Better Aligned. While there would clearly be challenges in moving CDCR to a new population budgeting system, we believe our recommended approach would better align resources with actual operations and provide a much more transparent budgeting process. In particular, we believe our proposed changes would provide a more effective way to estimate the fiscal demands of changes in the inmate population because (1) the population projections could incorporate more recent data trends, (2) the IAS and below–the–line adjustments would no longer be necessary, and (3) the budget funding provided would more accurately align with the actual operation needs of different missions and security levels rather than being based on the fixed six–to–one ratio that is currently used. The CDCR staff workload would be reduced, potentially resulting in savings on administrative costs, because the staffing formulas would be far simpler than the current extensive processes involving the creation of the IAS and staffing packages. Finally, the use of funding formulas would provide for a more transparent budget document because it would be far simpler for the Legislature to review these funding formulas, than to sort through the multiple volumes that currently make up the population budget request. In so doing, the Legislature will better be able to fulfill its oversight role, as well as ensure that the document budgets funding for CDCR operations consistent with its actual and projected inmate and parole populations.

Administration’s Population Reduction Proposals Not Best Options for Public Safety

In order to achieve significant budget savings in corrections, the administration proposes to release certain inmates from prison early and place them under minimal parole supervision, a policy that is termed “summary parole.” We note significant public safety risks and operational concerns with the current proposals, and recommend alternatives that we believe offer a better tradeoff between public safety and the achievement of budget savings. In particular, we believe it would be more appropriate to change crimes currently classified as ”wobblers” to misdemeanors and to substitute an earned discharge program for the Governor’s summary parole proposal.

Budget’s Population Reduction Proposals

The administration proposes two policies—20–month early release and summary parole—to significantly reduce the state’s inmate and parolee caseloads and reduce operational costs in CDCR by an estimated $354 million in the budget year with $758 million in ongoing savings in subsequent years. We describe these two proposals in more detail below.

Twenty–Month Early Release. The administration proposes to release specified inmates from state prison up to 20 months prior to their scheduled release date. If an offender who had been sent to prison had less than 20 months to serve, they would be released if a screening of their record determined they were not excluded from early release. The proposal excludes inmates from early release if they have a current or prior conviction for (1) serious or violent offenses; (2) sex offenses requiring registration under Penal Code 290; or (3) other specified crimes, including offenses related to weapons, child and elder abuse, and manufacturing certain drugs. The proposal also excludes inmates who have been found to have committed one of these crimes while in prison, even if they were not convicted for the offense in court.

Summary Parole. The administration proposes to place generally the same offenders who would qualify for early release on “summary parole.” Currently, almost all offenders released from prison are placed onto parole and assigned to a parole agent’s caseload. The frequency of contact between the parolee and his parole agent depends on certain risk factors of the offender, particularly his criminal history and mental health. Under summary parole, the specified offenders would technically remain parolees but would not be actively supervised by a parole agent or subject to return to prison through the state’s administrative revocation process operated by the Board of Parole Hearings (BPH) unless they were also convicted of a new crime in the criminal courts. Offenders on summary parole would be subject to search and seizure by local law enforcement, as well as drug testing, just as they are now.

Projected Caseload and Fiscal Impacts of Proposals. The administration estimates that these two proposals will significantly reduce the inmate and parolee populations over time, beginning with relatively small reductions in the current year, but with much more significant reductions in the budget year and thereafter. Figure 6 shows CDCR’s estimate of the population reductions likely to occur under these proposals, including a combined reduction of 63,000 inmates and parolees by 2009–10. This is approximately a 20 percent reduction in the state prison and parolee populations. Figure 7 shows the department’s estimate of the net savings that are likely to occur as a result of the reduction in inmate and parolee caseloads, and includes offsetting administrative costs in the current year ($16 million) and ongoing ($5 million) to review the case files of offenders to determine which are eligible for early release and summary parole.

Major Problems With Administration Proposals

We have identified significant public safety and operational concerns with the administration’s early release and summary parole proposals. We acknowledge that any option that reduces the number of offenders in state prison or that reduces the level of parole supervision is likely to involve some tradeoff in terms of public safety. But, as the experiences of other states have demonstrated, it is possible to make reasonable choices about the deployment of the state’s prison and parole resources that minimize the practical risks to public safety while achieving significant reductions in prison and parole caseloads and state costs. We are concerned that the particular way in which the administration proposes to implement early release and summary parole does not offer the best such tradeoff for public safety. We conclude that better alternatives are available (including other forms of early release) that also would result in significant budget savings.

In particular, we are concerned that the administration’s proposal would negatively affect public safety by creating a gap in the state’s criminal justice system and by reducing incentives for low–level offenders to take advantage of diversion programs. In addition, we found that the administration’s proposal is incomplete. Many aspects of how these proposals would be implemented have not yet been fully developed by the administration, making it difficult to fully evaluate the operational and fiscal impacts of these proposals. We discuss each of these concerns in more detail below.

Proposal Creates a Gap in Criminal Justice System. Under current state law, there are three basic kinds of crimes: felonies, misdemeanors, and infractions. A felony is the most serious classification of crimes and is defined as those crimes for which offenders can be sentenced to state prison. Felonies can also result in a sentence to county jail, a fine, supervision on county probation in the community, or some combination of the available punishments. Some felonies are designated in statute as violent or serious crimes that can result in additional punishment, such as a longer term in state prison. Misdemeanors are considered less serious and can result in a jail term, probation, a fine, or release to the community without probation but with certain conditions imposed by the court. Infractions, which include violations of certain traffic laws, do not result in a prison or jail sentence.

The administration’s proposals target certain offenders—those who have not been convicted of specified offenses, including serious or violent felonies—for early release from prison and summary parole. The appeal of this approach is that it targets those inmates and parolees who are lower level offenders compared to other state inmates and parolees who have a more serious criminal history. Yet, by targeting this particular group of offenders in the way proposed, the administration’s proposals create a significant gap in the state’s range of criminal sanctions. Specifically, local corrections—jails and probation, primarily—will continue to supervise misdemeanants and many low–level offenders, while prisons will continue to house the most serious felons. However, under these proposals, about 63,000 mid–level offenders—those with no serious, violent, or sex offenses in their criminal history—would effectively go unpunished, serving little or no time in prison and not actively supervised on parole. This is because many of these offenders normally serve prison terms of up to 20 months and would thus be immediately eligible for release and summary parole. Consequently, these mid–level offenders would, on the whole, receive less punishment and supervision than lower–level offenders in county jails and on probation who are serving longer terms. Figure 8 illustrates the gap created in the criminal justice system by the combination of 20–month early release and summary parole.

This gap in the criminal justice system means that some offenders will essentially go unpunished for their crimes because they will serve only a minimal amount of time in prison before they are released to summary parole. It also increases the likelihood that offenders will go unpunished for any additional crimes they commit while on summary parole. This is because the proposed summary parole policy would require local prosecutors to bring new charges in court for new criminal activity before a parolee could be returned to prison. However, if that new criminal activity is another low–level drug or property crime, the offender would serve little or no time in prison for the new criminal court conviction because of the administration’s 20–month early release policy. Moreover, it is unclear to what degree local prosecutors will spend their resources to prosecute these offenders in the first place if it will not necessarily result in a prison sentence. Under the administration proposal, the BPH could revoke these offenders if there was a new criminal conviction. However, the department has not provided the Legislature with details as to how it would implement this provision.

This situation is likely to reduce the incentive for offenders on summary parole to abstain from committing new nonserious, nonviolent felonies. This concern is particularly relevant because there would be no mechanism under this proposal to encourage or require offenders on summary parole to participate in programs that could reduce their likelihood to reoffend.

Reduces Incentives for Offenders to Participate in Prison Diversion Programs. These proposals would also have the unintended consequence of reducing the incentive for some low–level offenders, at the time of their sentencing, to elect to participate in diversion programs such as Proposition 36, drug courts, and mental health courts. Currently, a major incentive for many felons to participate in these programs is that the alternative to participation would be incarceration in state prison followed by parole supervision. However, under the administration’s early release and summary parole proposals, many of these offenders might view a prison sentence as a better alternative than a diversion program because they would serve little or no time in prison and be unsupervised on parole.

Research on these diversion programs generally show them to be effective programs at reducing the recidivism rates of participants. If fewer offenders participate in them because they have lost the incentive to do so, their involvement in crime will be greater and public safety will be harmed.

Implementation Plan Not Developed. We are also concerned that several important aspects of how these new policies would be implemented have not been developed by the administration. These details are important for the Legislature to have in order to evaluate the full implications of the major policy changes proposed by the administration, as well as to compare those proposed changes with any alternative proposal that would significantly reduce the inmate and/or parolee populations, such as those alternatives discussed later in this analysis.

First, the department has not provided the Legislature with an estimate of how the population reduction would affect the number of beds needed at each security level, at reception centers, or for female offenders. Similarly, the department has not estimated how the policies would affect the number of remaining inmates to be housed outside of the state’s 33 prisons, such as in conservation camps, minimum support facilities, or contracted facilities within and outside of California. In fact, the magnitude of the inmate population reduction that would occur could potentially allow the state to close prisons, and the department has suggested that it might consider closing down some prisons or parts of prisons that are particularly old or decrepit. However, CDCR has provided no details as to which facilities these would be or how much in additional savings these closures would generate to the state. These questions are particularly important because the department is simultaneously moving forward with the development of plans to construct more prison beds as authorized by Chapter 7, Statutes of 2007 (AB 900, Solorio), commonly referred to as “AB 900,” which provided $6.1 billion to construct as many as 40,000 additional prison beds. We discuss the potential impact of these policy proposals on AB 900 construction plans in the “Capital Outlay” section of the CDCR analysis.

Second, the department has acknowledged that these policy changes would have a significant impact on the number and type of rehabilitation programs that are needed in state prisons and for parolees in the future. Currently, many programs, particularly parole programs, are oriented toward lower–level state offenders, exactly the population that would be largely removed from the system under these proposals. As a result, many programs might need to be redesigned or relocated to meet the needs of the remaining population, and some programs may need to be eliminated or replaced with different rehabilitation strategies altogether. In addition, the sudden shift of a large group of offenders from prison to summary parole raises the question as to the extent resources would continue to be needed in the community to assist those released on summary parole with drug treatment, education, assistance in finding a job, or mental health treatment. To date, the department has not provided the Legislature with an outline of what program changes would be necessary, an estimate of how long it would take to make these changes, or an analysis of their fiscal implications.

Third, the administration reports that these policy changes would result in a reduction of about 5,900 positions, thereby requiring staff layoffs, most of which would be of custody staff such as correctional officers and parole agents. The department’s estimates of fiscal savings assume that it takes approximately nine months to layoff the affected staff. However, the department has provided little information to substantiate the exact extent of these layoffs or how these layoffs would occur. For example, to what extent would the department propose to close beds it contracts for in the community, potentially reducing the number of layoffs necessary for state employees? How would the layoffs affect the need for resources for recruitment, training, and the correctional officer academy?

Fourth, we would note, and the department has acknowledged, that there exist significant inconsistencies between this proposal and other funding requests contained within the Governor’s budget. Presumably, significant inmate and parolee population reductions such as those proposed would result in a reduced need for resources for activities related to training, parole revocation caseloads, and administration. However, the department’s budget requests significant new funding for many of these activities, including correctional officer recruitment and training, BPH caseloads, and human resources administration staffing. If the Legislature were to move forward with the administration plan, or any alternative approach to reduce the inmate and parole populations, a number of additional budget adjustments would be warranted. Our analysis suggests that these additional savings could exceed $100 million annually.

Some Fiscal Impacts Not Included. As discussed above, there are a number of operational questions that are unaddressed by the administration, many of which have fiscal implications that could significantly increase or decrease the level of savings that would actually be achieved under its proposals.

In addition, other factors could increase the actual savings level achieved from adopting these proposals. For example, the population reductions could result in cost–avoidance related to delaying or eliminating the need for future construction of prison facilities. In addition, the administration’s fiscal estimate of its early release proposal assumes the department would achieve no savings related to inmate medical care from releasing tens of thousands of inmates early. However, there would be savings if a population reduction resulted in even a small reduction in medical costs for inmates, such as for nurse registries or specialized treatment provided outside of prison walls. Notwithstanding the potential for medical care savings, we recognize that the department’s inmate medical program—including its budget—is under the control of a federal court–appointed Receiver who may not agree to a reduction in his medical budget even if a significant number of inmates were released from the prison system early.

On the other hand, there could be some factors that result in reduced savings relative to the administration’s estimates. For example, their fiscal estimates do not assume any costs related to parole violators who return to prison upon a conviction for a new crime. To the degree that occurs, the department’s fiscal estimates could be overstated. The magnitude of these offsetting fiscal impacts is unknown.

Better Options Available to Reduce Inmate and Parole Populations

Should the Legislature conclude that reductions in state correctional populations should be part of the package of actions necessary to address the state’s budget shortfall, our analysis indicates that there are better options available to achieve significant budget savings that minimize the impact to public safety. The option we think best meets these criteria is to change sentencing law to make “wobblers”—crimes that can be sentenced as either felonies or misdemeanors—punishable as misdemeanors only. We describe this option, as well as identify other possible approaches, in more detail below. We also recommend substituting a policy of earned discharge from parole in lieu of the administration’s summary parole approach.

LAO Alternative: Change Wobblers to Misdemeanors

Definition of Wobblers. Current law makes some crimes punishable as misdemeanors—where jail, probation, and/or fines are the criminal punishments allowed—or as felonies—where a sentence to state prison is defined. These crimes are sometimes referred to as wobblers. The sentencing decision on wobblers is left up to the criminal court, with the court‘s decision generally based on the specific circumstances of the crime and the criminal history of the offender. Most wobblers are property or drug offenses and include crimes such as forgery, petty theft with a prior theft, and some drug possession offenses.

Recommended Approach. Should the Legislature choose to identify inmate population reductions as a strategy to achieve budget savings, we would recommend changing criminal penalties to make wobblers misdemeanors instead of the administration’s 20–month early release proposal. While this approach would limit the discretion judges now have to send persons convicted of wobblers to prison, it would prioritize the use of expensive state prison beds for violent and serious offenders. Which specific wobblers to choose would ultimately depend on the policy preferences of the Legislature, as well as the budget savings target to be achieved. Figure 9 lists the criminal offense categories that include wobblers, the number of inmates in prison for those crimes, and the annual cost to incarcerate those offenders. As shown in Figure 9, we estimate that about and 31,000 inmates are in state prison for wobbler offenses at any given time. Thus, the adoption of our recommended option would result in state savings of about $200 million to $300 million in the budget year growing to about $700 million annually within a few years. Annual savings could be even higher, by $100 million or more, if the resulting population reduction resulted in CDCR closing entire prisons or selected older facilities within existing institutions.

The LAO alternative would also reduce the parole population by a comparable number of offenders, resulting in additional savings of several tens of millions of dollars annually within a few years. In the longer term, these population reductions could also result in significant cost avoidance related to capital outlay to the extent that new prison facilities would not have to be constructed. However, we would also note that there could be some factors that could reduce the estimated level of savings from our alternative. For example, it is possible that local prosecutors would change their charging and plea bargaining practices such that a significant number of offenders would continue to be sentenced to state prison even under our proposed change in sentencing law. The degree to which this would occur, as well as the net fiscal impact of these offsetting factors on the state judicial system and local criminal justice agencies, is unclear at this time.

Advantages of Changing Wobblers to Misdemeanors. In our view, the conversion of wobblers to misdemeanors has some distinct advantages over the administration’s proposal, although it also has some aspects in common with the administration’s proposed approach as well. These are summarized in Figure 10.

The LAO alternative has several distinct advantages compared to the administration’s proposal. First, our approach maintains a continuum of criminal justice sanctions by eliminating the gap in the criminal justice system created by the administration’s proposal. Under our approach, offenders diverted from prison would still be subject to criminal sanctions for their crimes at the local level, typically in jail and on probation.

Second, our approach would not create as much of a disincentive for offenders to participate in Proposition 36 and other diversion programs as would occur under the Governor’s proposal. Local court officials could use the prospect of significant jail time as an incentive for these offenders to participate in these programs.

Third, our approach would not require tens of millions of dollars in administrative costs to implement, as would the administration’s proposal. Our approach would not require CDCR to complete tens of thousands of case file reviews. Instead, our proposal would simply eliminate prison as a sentencing option for these offenders, resulting in no additional workload for CDCR or the courts.

Fourth, our approach would have the additional advantage of further reducing the population in CDCR’s overcrowded reception centers. Under the administration’s proposal, offenders would still be transported and housed in reception centers for some period until a case file review determined their eligibility for early release. However, under our approach, inmates would be sentenced to local corrections and never arrive at the reception center at all, thereby reducing reception center overcrowding and costs relative to the Governor’s approach.

Like the Governor’s proposal, the LAO alternative would provide a large and ongoing solution to help address the state’s budget shortfall. The savings would begin to accrue immediately upon implementation, and reach the full estimated savings within a couple of years. (At full implementation, the administration’s proposal results in somewhat lower savings than the LAO alternative—about $527 million compared to $700 million.) Both approaches would target relatively low–level offenders in the prison system whose current offense is not a serious, violent, or sex offense. In so doing, as noted earlier, our proposal as well as the Governor’s would effectively prioritize the use of expensive prison beds for the most serious and violent offenders in the state’s criminal justice system. Also, both approaches to reducing the inmate population might affect the federal court’s consideration of a cap on the inmate population. By acting independently, the Legislature would be able to set its priorities in determining which offenders would no longer be in state prisons, as well as how those offenders were handled in the criminal justice system in lieu of being in prison.

Addressing Potential Trade–offs With Changing Wobblers to Misdemeanors. While the LAO alternative has some advantages, it also involves a couple of tradeoffs that are worth noting.

First, our proposed change in sentencing law would likely result in some property and drug offenders serving somewhat shorter terms of incarceration than under current law. Under current law, the maximum jail sentence for a misdemeanor is one year, while the average time that most of these offenders would serve in prison is now about a year and a half. However, a year and a half is the average time served. Many of these offenders serve a year or less in prison under current law for their crimes. Therefore, the reduction in time served would generally be a few months at most, compared to as much as 20 months less time served under the Governor’s proposal.

Second, our proposed approach would result in affected offenders being a local, rather than state, responsibility for housing and supervision. This would certainly result in additional operating costs to local governments, primarily to counties, for probation and jail operations. It would also potentially add to existing overcrowding levels in some local jail systems, perhaps resulting in additional early releases of lower–level jail inmates. However, we would note that when state sentencing laws were changed in the past to require that many offenders be sent to prison and not to county jails or probation, funding was not shifted from the counties to the state to adjust for the increase in state responsibilities. Moreover, the Legislature recently enacted AB 900 which authorized $1.2 billion in state funding in two phases to construct 13,000 additional jail beds. The Legislature could consider accelerating Phase II of the jail bed funding to make more space available for local jails sooner.

Third, while our proposal targets offenders whose current offense is a relatively low–level property or drug offense, some of these offenders would have committed more serious crimes in their past, potentially including violent and sex offenses. However, based on a review of the available data, less than one–quarter of these offenders are “third strikers” (having been convicted as such under the state’s “Three Strikes and You’re Out” law) or have any prior offenses for a violent crime or a sex offense that would require registration. However, one option available to the Legislature would to exclude offenders with prior violent or sex offenses from being sentenced as misdemeanants. Doing so would reduce the estimated savings by about $150 million to $200 million annually upon full implementation. Even with such a change, we estimate that our approach would still result in annual savings of $500 million to $550 million

Other Prison Population Reduction Alternatives Could Be Considered

While our analysis indicates that changing wobblers to misdemeanors is the best policy option available that would result in significant caseload savings in corrections in the near term, it is certainly not the only alternative. For example, the state could provide a shorter period of early release to a broader range of offenders or provide increased early release credits to inmates. Our 2004–05 Budget: Perspectives and Issues (P&I) identifies several such options (see page 228).

LAO Alternative: Adopt Earned Discharge Instead of Summary Parole

Our analysis suggests that a better alternative to summary parole is earned discharge, which could be adopted in combination with, or independent of, our proposed alternative to change wobblers to misdemeanors.

In our view, there is some merit in the administration’s proposed summary parole policy. It is similar to the practice already in place in many other states where certain low–level offenders do not serve any time on parole after completing their prison term. After serving their prison term, these offenders can be prosecuted if they commit new crimes, but they are not actively supervised on parole or subject to return to prison through an administrative revocation process.

However, we are concerned that adoption of a summary parole policy may not adequately take into account the propensity of some of these offenders to reoffend after their release from prison. Our analysis indicates that a better alternative would be for the Legislature to direct CDCR to implement a policy called “earned discharge” which would permit parolees to be discharged from parole early if they had demonstrated that they had successfully reintegrated into the community. Earned discharge requirements would be set in statute and could be limited to parolees who met specified criteria such as having been involved in no new criminal activity, had no evidence of drug use, were employed, and had stable housing. Certain offenders who met these statutory criteria continuously in the first six months after being released from prison, for example, would be discharged from parole.

The period at which offenders could be discharged early would depend on legislative preferences, but could vary for different groups of offenders. One approach would be to vary that period based on the likelihood that offenders will reoffend. The CDCR is currently implementing an assessment tool that identifies the risk that an inmate will reoffend after release to parole. Outcomes from these assessments could determine each offender’s discharge period. For example, a parolee assessed as being at low risk to reoffend could be discharged at three months if he met the statutory criteria, while an offender assessed as medium risk to reoffend might have to demonstrate six months of continuous compliance with the statutory criteria before early discharge. The Legislature may want to exclude the highest risk offenders and/or those convicted of violent or sex offenses from earned discharge.

The exact level of savings that would be achieved through an earned discharge policy would depend on the number of months parolees would have to serve on parole before they could earn early discharge, how many parolees were eligible to participate in the earned discharge program, the specific statutory requirements to earn early discharge, as well as how many parolees met those statutory requirements. However, if earned discharge was available to most parolees—except those with current or prior violent or sex offenses, for example—the savings could exceed $100 million annually at full implementation.

Even if applied to a broader segment of the parolee population than the administration proposes for summary parole, an earned discharge policy might only result in savings of one–third to one–half of the amount that could be generated under summary parole. However, earned discharge would provide more accountability for offenders and, consequently, considerably less risk to public safety, because the offenders released from supervision would be the very ones who had demonstrated evidence of success in the community. Earned discharge has been recommended by the group of academic and professional experts commissioned to study CDCR operations (commonly referred to as the “Expert Panel”). The department has developed plans to implement earned discharge on a pilot basis during the current year. However, the department reports that implementation of the pilot programs has been delayed.

There are other alternatives the Legislature could consider for parole reform. In the 2008–09 P&I, we propose that the state realign responsibility for supervising certain parolees convicted of drug and property crimes. Under this approach, counties would receive new revenue to offset the additional costs they would incur under this realignment. We estimate that the state would save about $500 million annually from our parole realignment proposal.

CDCR Should Report on Implementation Issues

Finally, we would note that before the Legislature adopts any policy changes that would significantly reduce inmate and parolee caseloads, it should require CDCR to report on the operational implications of various alternatives. In particular, the department should be required to identify the likely impact to inmate housing needs, staffing, offender programs, and other corresponding adjustments to the budget plan, so that the Legislature can weigh the full range of policy and fiscal impacts before making a final decision.

Other Correctional Program Issues

Offset Funding for Community Work Crews

We recommend modifying the department’s request for funding related to inmate community work crews to (1) eliminate a General Fund augmentation for these new positions, and (2) reflect funding of the new positions from reimbursements from local jurisdictions and institutional savings that will occur from the increase worktime credits earned by participating inmates. (Reduce Item 5225–001–0001 by $2.4 million. Increase Item 5225–001–0995 by $1.2 million.)

Budget Proposal. The spending plan proposes an augmentation of $2.4 million in the budget year, growing to $5.8 million in future years, to create inmate community work crews at most prison institutions. These work crews would provide services to local jurisdictions, such as litter removal, weed abatement, and minor repairs. The department would not be reimbursed by the local jurisdictions for the services performed by the work crews. The funding would be used to establish almost 29 new correctional officer positions in the budget year (growing to 60 positions in future years) that would supervise the work crews while off prison grounds, as well as some one–time costs for equipment. The department reports that similar work crews used to exist prior to budget reductions in the 2003–04 fiscal year. Currently, six prisons have community work crews, but local jurisdictions reimburse the state for its costs of providing the services.

Additional General Fund Spending Not Warranted. We have several concerns with this proposal.

First, while we concur with the department that the new inmate work crews could help to promote more positive relationships between institutions and the local jurisdictions in which they reside, General Fund spending for this work is not critical for the maintenance and operation of state prisons. As we noted in the Analysis of the 2007–08 Budget Bill, the prison system faces serious maintenance problems that we believe could be partly addressed through the more effective use of inmate labor. Second, it is unclear whether the department would have sufficient inmates available to work on these crews if the Legislature approves the administration’s proposals to reduce the inmate population (discussed earlier in this chapter). Third, we would note that the department’s request for funding may be understated because it does not account for the costs of paying the inmates who would work on the crews. These costs are likely to be as much as several hundreds of thousands of dollars, depending on how many inmates were paid for this work and what share of the funds were paid out of the General Fund versus the Inmate Welfare Fund.

Analyst’s Recommendations. For the reasons cited above, we recommend rejection of the proposed General Fund augmentation for these new activities. We propose instead that the Legislature approve the proposed new correctional officer positions on a limited–term basis and that the full costs of inmate work crews be funded through two other means: (1) partial reimbursement from local jurisdictions that request the services and benefit from the community work projects, and (2) the redirection of General Fund savings that would be achieved in the prison system due to increased worktime credits earned by inmates.

Our recommended approach stems, in part, from our view that it would be appropriate for the department to require local jurisdictions to pay for half of the costs of the inmate work crews. This cost–sharing arrangement could allow local jurisdictions to have the proposed community beautification work done at a lower cost than they might otherwise pay, while resulting (together with the offsetting bed savings) in no net increase in state General Fund costs. The cost–sharing arrangement should take into account any costs necessary to pay inmates who work on the community crews.

The remainder of the costs, our analysis indicates, could be offset by redirecting prison savings from inmate work credits. Participating inmates would earn greater worktime credits while on the work crew than if they did not have a job assignment in the prison. This would shorten their stay in prison, thereby reducing prison operating costs. We estimate that at full implementation, these additional worktime credits could result in about $3 million to $4 million in state savings annually, about one–half of the amount the department is requesting on an ongoing basis.

Approval of the new correctional officer positions on a part–time basis would enable the Legislature to determine in the future if the program was successful and self–sustaining.

Finally, we again recommend that the department explore strategies for using inmates to supplement its maintenance workforce, as it was directed to do in language adopted as part of the Supplemental Report of the 2007–08 Budget Act.

Reprioritizing Workforce Investment Act Discretionary Funds

We recommend the Legislature redirect a total of $3.9 million in Workforce Investment Act funding to parolee employment programs in the California Department of Corrections and Rehabilitation, resulting in an equal amount of General Fund savings. (Reduce Item 5225–001–0001 by $3.9 million. Increase Item 5225–001–0995 by $3.9 million.)

We discuss our recommendation to modify the administration’s proposed spending plan for federal Workforce Investment funds in the “Employment Development Department” section of the “General Government” chapter.