Analysis of the 2004-05 Budget Bill Legislative Analyst's Office February 2004

Department of Corrections (5240)

The California Department of Corrections (CDC) is responsible for the incarceration, training, education, and care of adult felons and nonfelon narcotic addicts. It also supervises and treats parolees released to the community.

The department operates 32 institutions, including a central medical facility, a treatment center for narcotic addicts under civil commitment, and a substance abuse treatment facility for incarcerated felons. A new maximum-security institution is scheduled to open in spring 2005 at Delano (Kern County). The CDC system also operates 11 reception centers to process newly committed prisoners; 12 Community Correctional Facilities; 38 fire and conservation camps; the Richard A. McGee Correctional Training Center; 126 community reentry programs; two restitution centers; a drug treatment center; 185 parole offices; and 73 outpatient psychiatric services clinics.

Budget Overview

Budget Proposal

The budget proposes total expenditures of $5.3 billion for CDC in 2004-05. This is $447 million, or about 7.8 percent, below the revised estimate for current-year expenditures. Proposed General Fund expenditures for the budget year total $5.2 billion, an increase of $413 million, or 8.7 percent, above the revised current-year estimate. This increase reflects additional funds to offset $852 million in one-time federal funds received in the current year, as well as increases in department positions, parole population, and health care costs. These General Fund costs are partially offset by projected declines in the inmate population due to policies enacted in the 2003-04 Budget Act.

Current-Year Deficiency

The California Department of Corrections has experienced spending deficiencies in recent years and recently projected a current-year deficiency of $540 million. We report on the department's deficiency below and recommend that the Legislature require the department to submit a plan to address its ongoing deficiencies.

In October 2003, the CDC announced that it anticipated a budget deficiency of over $540 million for the current year. The Governor's budget recognizes $493 million of this deficiency, leaving an unfunded amount of $50 million. This latter amount includes items projected to be underfunded in the current budget—such as overtime and workers' compensation—which are partially offset by salary and program savings. Figure 1 shows the components of the department's deficiency.

Items Included in Governor's Budget—$493 Million. The 2004-05 Governor's Budget document indicates that the administration proposes to fund $493 million of the deficiency identified in October 2003 by CDC. This includes funding salary and retirement costs that were not included in the 2003-04 Budget Act. It also includes higher-than-anticipated population because the current budget overestimated the decline in the in mate population resulting from policy reforms and underestimated the number of new admissions to prison.

• Salary Increases—$174 Million. The Governor's budget proposes $174 million to fund the department's current-year salary increases. It is important to note that the 2003-04 Budget Act did not include salary increases for any state department. This is because the budget assumed that employee compensation contracts would be renegotiated, or departments would absorb the costs of any remaining increased salary costs through cuts in other programs. However, the prior administration was unable to negotiate lower salaries for correctional officers, and neither the prior nor current administration approved sufficient current-year cuts to CDC's budget to absorb the full costs of the salary increases.
• Retirement Costs—$179 Million. Retirement costs were not included for any state department because the 2003-04 Budget Act assumed that a pension obligation bond would fund the state's retirement contributions. However, a state superior court has ruled that pension obligation bonds cannot be issued without voter approval. Therefore, the Governor's budget also proposes $179 million to fund department retirement costs in 2003-04.
• Overestimated Population Savings and Underestimated New Admissions—$139 Million. The current-year budget assumes savings due to policy changes designed to reduce the number of nonviolent inmates, including education and day-for-day credits in reception centers, drug treatment furloughs, and intermediate sanctions for nonviolent parole violators. The department will be unable to achieve most of these savings in the current year primarily because the savings estimates did not account for overlap in the populations affected by the reforms. The department is also experiencing moderately higher-than-projected new admissions to prison.

Items Not Included in the Governor's Budget—$50 Million. The Governor's proposed budget does not include funding for the remainder of the department's October request, a net amount of approximately $50 million. This includes items that have generated deficiencies in past years, offset by savings primarily from department vacancies.

• Ongoing Operational Deficiencies—$240 Million. Approximately $240 million of the department's announced deficiency reflects on-going shortfalls in spending authority for overtime, temporary help, workers' compensation, medical supplies, and contract medical personnel, as well as unfunded merit salary increases for department employees. Each of these items is a source of a recurring structural deficiency, suggesting an inability of the department to fully fund these activities within its allocated budget.
• Offsetting Salary and Program Savings—$190 Million. The department anticipates being able to offset the $240 million in budget shortfalls with expected savings of about $190 million from the institutions, inmate health care, and parole programs. Most of these savings are from the department's approximately 2,700 unfilled vacancies.

Analyst's Recommendation. The Governor's budget summary suggests that it will offer a proposal as part of the May Revision to better control aspects of the department's fiscal management. The Legislature should direct CDC to include as part of its cost control proposal, a plan to address those items which have traditionally driven deficiencies in the department's budget. In particular, the proposal should identify ways to bring spending on overtime, temporary help, worker's compensation, and medical costs in line with the department's budget, as well as to identify other sources of offsetting program savings.

Major Additional Savings in Corrections Budget May Be Difficult to Achieve in Budget Year

The budget proposes a significant unallocated reduction in corrections. The administration has indicated that it will provide details of the proposal as part of the May Revision. Given the size of the proposed reduction, we recommend that the Legislature request the administration to provide the specifics of the plan prior to the May Revision in order to allow the Legislature adequate time for review. In this piece, we identify issues for the Legislature to consider as it reviews the administration's forthcoming proposal.

The Governor's budget proposes a $400 million unallocated reduction in corrections. Given that CDC is by far the largest department in the Youth and Adult Correctional Agency, it is likely that most of this amount will need to be achieved through reductions in CDC's budget. Without prejudice to the administration's forthcoming proposal, we believe there are issues the Legislature should take into consideration as it examines the budget for the department.

Department Costs Are Largely Driven by Population and Labor Costs. The department's budget is primarily determined by the projected number of inmates that will be housed in its prisons, and on the number of persons projected to be on supervised parole in the community. Based on these projections, the department determines how many staff are required to (1) supervise this population and (2) provide other mandated services such as education and health care services. Thus, the most important determinates of the costs to operate CDC are the numbers of inmates and parolees, the level of compensation for correctional officers and other custodial staff, and the types and levels of services provided. Therefore, in order to achieve significant reductions in the budget for corrections—as proposed by the administration—consideration must be given to reducing the number of individuals in state custody, the labor cost of providing security and community supervision, or the types and levels of services provided.

• Population-Related Savings and Public Safety. Population savings can be achieved in two ways: (1) bringing fewer individuals into the system by changing sentencing laws and (2) releasing more individuals out of the system through early or accelerated release programs. In considering reductions to the inmate and parole population, we have recommended in the past that the Legislature focus on nonviolent and short-term inmates to minimize the impact on public safety. The Governor's budget already assumes a significant level of population-related savings from parole reforms that are being implemented in the current year. Because these reforms already target many of the nonviolent inmates in the system, the additional savings that can be achieved through population reduction proposals of nonviolent and short-term inmates has been narrowed.
• Labor Costs Savings Require Successful Contract Renegotiation. The administration has indicated that it plans to renegotiate the state contract with the correctional officers' union. We agree there is potentially a significant level of savings that can be achieved through contract renegotiations. For example, as part of our analysis of the department's overtime problem, we recommend that the Legislature direct the Department of Personnel Administration to renegotiate provisions that increase overtime. In addition, in our evaluation of the administration's request for 1,239 relief positions—discussed later in this analysis—we note that renegotiating training-related provisions of the correctional officer contract could generate significant state savings. While we believe such savings can be achieved if there is a contract renegotiation, we would note that such negotiations need to occur in a timely manner in order to achieve the maximum benefit in 2004-05, and to provide the Legislature sufficient time to review the full fiscal impact of the agreement.

In our The 2004-05 Budget: Perspectives and Issues publication, we offer options for CDC budget savings. These options primarily include sentencing reforms aimed at more rapidly transitioning low-risk offenders into the community.

Administrative Efficiencies. Based upon the Governor's budget, we expect the proposal for achieving $400 million in savings to include specific plans for achieving administrative efficiencies, such as closing institutions, contracting out for certain services, and improving fiscal control and accountability within the department by developing systems to identify costs at the institution level which can be reduced. We agree that there are opportunities for savings from these types of proposals, but the feasibility of implementing some of these proposals will depend on the department's successful implementation of changes that are already assumed in the budget. For example, prison closures would be possible if the current-year parole reforms being implemented actually achieve the population reductions assumed in the budget. Likewise, cost savings through better internal controls are possible if central management, as well as wardens and other prison administrators, have adequate tools to track expenditures. In addition, we agree there may be opportunities to reduce costs for certain services, such as health care, mental health, and pharmacy by contracting with private entities for the delivery of these services. However, the department's ability to contract for some services may be limited under current law, and thus require voter approval of the administration's proposed constitutional amendment to expand the circumstances under which the state can contract with private entities.

Administrative Feasibility of Achieving Additional Major Savings in 2004-05. The department already is in the process of implementing several major reform proposals. Any new proposals submitted by the administration to reduce costs by the level assumed in the budget may experience delays due to the current workload limitations of existing staff.

Historically, the department has had a poor record of accomplishment in implementing changes. In fact, the department is currently experiencing delays in implementing reforms that were adopted as part of the current-year budget. The delay is caused by a number of factors, including the freeze on hiring and contracting and the need to negotiate with employee unions on certain components of the reforms. However, other factors that affect the department's ability to implement change are its sheer size and decentralized nature—including 32 institutions and 185 parole offices spread across the state, as well as the autonomy of individual institutions.

During its review of the administration's proposal, we recommend the Legislature realistically assess CDC's ability to fully implement changes that are proposed as part of the administration's $400 million proposal.

Summary. The details of the proposed $400 million unallocated reduction are unknown. Given the magnitude of the proposed reductions, we recommend the Legislature request the administration to provide the details of the plan before the May Revision. Furthermore, without prejudice to the administration's forthcoming proposal, we have concerns about the department's ability to achieve this level of savings in 2004-05 in light of major changes that are already underway, and the time it takes to implement significant changes in a department the size of CDC. We recommend the Legislature take these issues into consideration when assessing the department's ability to achieve the Governor's proposed savings in 2004-05.

Inmate and Parole Population Management Issues

Inmate Population Projected to Decline

The California Department of Corrections is projecting the inmate population to decline substantially in the second half of the current year and into the budget year. These declines will be accompanied by a commensurate increase in the parole population.

Inmate Population Decrease. As of June 30, 2003, the CDC housed 160,931 inmates in prisons, fire and conservation camps, and community correctional facilities. The CDC projects the inmate population to decline to 148,390 by June 30, 2005, a decrease of over 12,000 inmates. This projected decline by mid-2005 is primarily the result of a series of policy reforms enacted as part of the 2003-04 Budget Act, which includes programs to restructure parole, inmate education, and substance abuse treatment. (See The "New Parole Model" later in this chapter.) In the absence of these reforms, the inmate population would have increased to over 163,000 inmates because of a recent trend of moderately increasing new admissions to prison. Under the policy reforms, the prison population is projected to decline to its mid-1990s levels, creating the first substantial drop in the state's inmate population in at least 20 years. Figure 2 shows the year-end inmate and parole populations for the period 1993 through 2005.

Parole Population Increase. As of June 30, 2003, the CDC supervised 116,173 persons on parole. As shown in Figure 2, the CDC projects the parolee population to increase to 124,224 in 2005. This increase is a result of the current-year policy reforms designed to reduce parole revocations. This means that the affected parolees will stay in the community longer (instead of being returned to prison), thereby having the effect of increasing the total parole population. The projected parole population will be the largest in state history.

Implications of Population Changes. The decline in the inmate population will affect the department's housing plan, allowing CDC to consider options that include closing prisons, moving inmates out of dormitory-style housing units, and reducing the use of Community Correctional Facilities. In addition, the reforms will reduce the number of parolees revoked and housed in local jails and state reception centers.

Potential Risks to Accuracy of Projections. As we have indicated in past years, the accuracy of the department's latest projections remains dependent upon a number of significant factors. These include:

• Changes in sentencing laws and the criminal justice system adopted by the Legislature and the Governor or through the initiative process.
• Delays in the implementation of current-year budget reforms, including ongoing problems with the department in (1) contracting with providers of parolee services such as drug treatment, (2) hiring and training staff due to a hiring freeze, and (3) negotiating with employee unions regarding certain components of the reforms.
• Changes in the local criminal justice system affecting the number of persons arrested, charged, tried, convicted, and ultimately admitted to prison.
• Changes in the crime rates, especially for violent crimes, that could cause growth in the inmate population to differ from the latest CDC projections.

Significant changes in any of these areas could easily result in a prison growth rate higher or lower than the one contained in CDC's projections.

Caseload May Require Further Adjustment

We withhold recommendation on the 2004-05 budget request for caseload funding. Recent delays in the implementation of policy reforms included in the 2003-04 Budget Act may result in increased prison population above what is assumed in the proposed budget. We will continue to monitor the caseload and program implementation and recommend further changes, if necessary, following review of the May Revision.

Due to the delays in implementing current-year policy reforms designed to reduce the inmate population by January 1, 2004 the department will be unlikely to achieve the full current-year and budget-year population reductions assumed in the Governor's budget. The department has not yet released updated population projections that include the impact of these delays nor their corresponding budgetary impact.

Pending release of the policy reform implementation schedule and revised population estimates, we withhold recommendation on the 2004-05 caseload-funding request. We will continue to monitor CDC population, and make recommendations as appropriate at the time of the May Revision.

The "New Parole Model"

Summary

The 2003-04 Budget Act included several policy changes designed to reduce the number of parolees returned to prison. These changes, referred to as the "new parole model," include the provision of prerelease planning, reentry services, and intermediate sanctions for some inmates and parolees. In this piece, we review the policy changes and their estimated effect on the inmate and parole populations and the resulting state fiscal impact. We make several findings about the policy changes, including concerns about implementation delays, and provide several recommendations.

Background

California's Parole System

California operates the largest state parole system in the country with the CDC supervising over 110,000 parolees at any given time. The purpose of parole is twofold, to protect public safety by supervising recently released felons and to assist parolees in successfully reintegrating into communities. When an inmate is released from a California state prison, he/she is required to report to his/her parole agent and abide by the conditions of parole, including following all laws. The parole agent is responsible for monitoring the activities of the parolee, assisting the parolee to access programs and services such as drug treatment and job placement programs, and taking appropriate action if the parolee presents a danger to the public or oneself.

California's Parole Failure Rate Highest in Nation

California parolees have the highest rate of returns to prison in the nation with over 55 percent of all parolees returned to prison within two years of release. California has the second lowest level of parole success, with only 21 percent of parolees completing their term on parole supervision without being returned to prison or absconding.

Approximately 80 percent of the parole violators returned to prison are sent back by way of the administrative revocation process, often for noncriminal violations, often referred to as "technical violations." The remainder of parole violators are convicted of new crimes and sent back to prison by the courts.

Researchers have attributed the high parolee failure rate in California to a number of factors, including few programs and services available to assist parolees with reintegration, few alternatives to revocation for parolees who violate their conditions of parole, a failure by CDC to adequately prepare inmates for release, and an emphasis on parolee punishment rather than reintegration.

Costs of Parolee Failure

While there is a clear public safety benefit associated with incarcerating felons and isolating them from society, there are also significant costs associated with parolee failure and revocation. First, the state incurs average prison costs of over $10,000 for each parolee who is returned to prison, costing the state almost $900 million in incarceration costs each year. In addition, researchers have noted other, often less fiscally tangible, costs associated with parole failure. These costs include local law enforcement and court costs, harm to victims and their property, loss of financial and social support to children and families of the incarcerated, and loss of tax revenues from revoked parolees who otherwise could be working.

2003 Parole Reforms—The "New Parole Model"

As part of the 2003-04 Budget Act, the Legislature required CDC to institute a number of parole-related reforms designed to decrease the number of revocations, reduce parolee recidivism, and save the state money. In response, the department designed a series of prevention and intervention strategies, including the institution of prerelease planning, reentry services, and intermediate sanctions. While not all research on these types of programs is conclusive, many evaluations do suggest that these types of efforts can cost less than returning a parolee to prison, and in some cases can significantly reduce recidivism when combined with services aimed at addressing substance abuse, employment, and other needs of parolees. California's new programs are described in greater detail below.

Prerelease Planning

The CDC will dramatically expand its use of prerelease planning for inmates soon to be released into the community. The first element of the prerelease program involves the use of a needs and risk assessment of inmates before they are released from prison. This assessment, which will be given to approximately 60,000 inmates, will accomplish two tasks: (1) it will use information collected about each inmate's personal and criminal history to identify what programs and services would benefit the inmate after his release and (2) it will project the likely risk that the inmate will reoffend. The results of the needs and risk assessment will be provided to the inmate's parole agent prior to the inmate's release. With this information, the agent will be better prepared to provide the types and levels of services and supervision appropriate for each parolee.

The second element of the prerelease program includes the improvement and expansion of existing prerelease classes to approximately 60,000 inmates released to parole each year. These classes will be mandatory for all inmates nearing the completion of their felony term. Revoked parolees will not be required to participate. In addition, the new prerelease classes will cover a broader range of topics and materials than the program in the past which sometimes has been limited to providing information packets to inmates. In past years, approximately 30,000 inmates participated in voluntary prerelease programming.

Finally, the new prerelease program will be operated by the department's parole division with the assistance of contracted social workers, rather than the department's prison employees as is done currently. The department hopes that parole agents and social workers will be better able to prepare the inmates for success on parole and in the community.

The CDC projects that the new prerelease program, in combination with the use of reentry services (see below), will reduce parolee recidivism by 5 percent statewide. The department estimates that over 3,000 fewer parolees will return to prison, thereby saving the state $17 million annually in incarceration costs. This estimate does not include implementation costs which are likely to be significant.

Reentry Services

The department currently operates Police and Corrections Teams (PACTs) in 15 parole districts. These teams are partnerships between CDC, local law enforcement agencies, and community service providers and are designed to coordinate the supervision of and services for parolees. For example, some parole districts operate a PACT Parolee Orientation where a number of local service providers come to parole offices at one time to educate parolees about services available in the community, such as health, housing, and employment services.

The 2003-04 Budget Act expanded the use of PACTs to all 25 parole districts and provided funding to staff each PACT with a Community Resource Coordinator. The coordinator acts as the liaison between the parole office and community service providers and recruits their participation in the PACT Parolee Orientation. Participation in the orientation will now be mandatory for all released parolees. The department believes that the use of the PACTS and the orientation will provide parolees with an improved opportunity to take advantage of community resources, such as health, housing, and employment services, that can assist in successful reintegration.

Mentally Ill Parolee Case Management

The current budget also funds the expansion of the department's Transitional Case Management Program for mentally ill parolees. This program allows for the continuation of mental health services on parole for all offenders who were diagnosed with a mental illness while in prison. Specifically, these parolees will receive ongoing psychiatric treatment and case management services at the department's 73 parole outpatient clinics (POCs). Research demonstrates that the receipt of mental health services can be critical to the successful continuation and completion of parole for mentally ill offenders. For example, one analysis conducted by CDC staff indicates that mentally ill offenders who received POC and case management services stayed in the community longer than similar offenders. The CDC projects that over 5,000 new parolees will receive POC services under the program expansion and the resulting reduction in recidivism will reduce state incarceration costs by $14 million each year, not including offsetting implementation costs.

Intermediate Sanctions

Intermediate sanctions are punishments administered by parole agents in the community for parolees who violate the conditions of their parole and minor violations of the law. Intermediate sanctions are designed to give parole agents a broader range of punishment options for parolees who commit these violations. In the absence of intermediate sanctions, agents are generally faced with the choice of returning the parolee to prison—sometimes seen as too severe of a punishment for minor violations—or administering no punishment at all.

The administration will redesign and expand some existing programs to be used as intermediate sanctions. The intermediate sanctions adopted include electronic monitoring, Community Correctional Reentry Centers (CCRCs), and Substance Abuse Treatment Control Units (SATCUs). The CDC estimates that over 30,000 parole violators will be assigned to these intermediate sanctions each year, reducing prison operating costs by over $100 million. These sanctions are described in more detail below.

• Expanded Use of Electronic Monitoring. Electronic monitoring refers to the use of a device that alerts the parole agent when the parolee wearing the devise is no longer within a defined distance from his home. Typically, parolees are allowed to leave their home at prescheduled times for work, treatment, and other activities sanctioned by their agent. The department has used electronic monitoring technology for some time. However, it has been used solely as a condition of parole to monitor a small, select group of potentially dangerous parolees, including sex offenders. Under the new intermediate sanctions program, electronic monitoring will be significantly expanded and will be used for nonviolent and nonserious parolees who commit parole violations. The CDC estimates that each year approximately 8,000 parolees will be required to use electronic monitoring devices for up to 45 days. The CDC estimates that use of electronic monitoring will cost an additional $5 per day per parolee.
• Conversion of CCRCs. The CCRCs were originally established as work furlough programs for inmates being released from prison. Under the new parole program, these secure housing facilities will be converted to residential intermediate sanction programs operated by community-based organizations under state contracts. At the CCRCs, parolees will participate in various programs including substance abuse treatment, stress and anger management, victim awareness, computer literacy, life skill, and employment training programs. Parole agents can place nonviolent parole violators in CCRCs for up to 120 days, and the department estimates that approximately 8,000 parolees will be placed in CCRCs for an average of 45 days each year. Because of the programming provided, CCRC beds are more expensive to operate on a per day basis than prison beds. However, CDC projects savings from this reform because parolees will be in CCRCs for a shorter period than had they been revoked to prison and the programs will better prepare parolees for success in the community.
• Expanded Use of SATCUs. Some SATCUs have been used by CDC in the past simply as short-term "dry-out" beds in jails and reception centers for parolees with substance abuse problems. Under the parole reforms, SATCUs will still be for parole violators with substance abuse problems, but their use will be dramatically expanded with an estimated 16,000 parolees assigned each year. Parolees assigned to SATCUs will participate in a 30-day residential drug abuse program at local jails, followed by a 90-day aftercare treatment program in the community.

Findings and Recommendations

The implementation of the new parole policies should significantly reduce the number of nonviolent parolees returned to prison, thereby generating substantial state savings. However, we find that there have been implementation delays that will significantly limit the impacts of these programs in the short term, and that further opportunities exist to expand these parole programs to generate state savings. Based on our findings, we recommend that CDC provide the Legislature with an updated implementation plan at budget hearings. We identify further opportunities for expanding the existing parole reforms. We discuss our findings and recommendations in more detail below.

Department Should Report on Impacts of Implementation Delays

The current budget assumes implementation of the prerelease, reentry services, and intermediate sanctions reforms by January 1, 2004. The mentally ill parolee program was scheduled to phase in implementation beginning on July 1, 2003. However, statewide freezes on hiring and contracting have delayed the implementation of all elements of the parole reforms. Currently, CDC expects to have these programs operating by March 1, 2004. The delays will limit the ability of the department to use these programs to reduce the inmate population, thereby limiting the actual savings that will be achieved from these reforms, particularly in the current year.

At the time this analysis was prepared, the department had been unable to provide an updated estimate of the impact of the implementation delays on the inmate population and its budget. The department currently plans to submit information regarding the impact of the delays on population and budgeted savings as part of the May Revision and after the completion of the spring population projections. However, this will provide a very limited time for legislative review. Therefore, we recommend that CDC provide the Legislature with an updated implementation plan for the parole reforms at budget hearings, including the population and fiscal impacts of delays.

Opportunities for Additional Savings by Expanding Parole Reforms

While the current parole reforms begin to bring the state's parole system more in line with research findings and other states' policies for reducing parolee recidivism, there remain further opportunities to improve the parole system. These additional opportunities include broadening the eligibility of prerelease planning and intermediate sanctions and incorporating additional intermediate sanctions. These program expansions could further reduce the institution population and state costs while continuing to provide public safety. These program opportunities are described in greater detail below.

Expand Eligibility for Prerelease Planning. Under the new parole policies, all inmates approaching the end of a new felony term will be required to participate in prerelease programming. In other words, all new commitments and parole violators with new terms will participate. We believe the prerelease planning services may benefit a broader group of inmates, in particular those who are serving time for a parole violation. While fewer parole violators will be in prison if the intermediate sanctions and other parole reforms work as intended, there will still be as many as 50,000 such inmates in the system. Moreover, those parolees who do return to prison may be a subgroup of inmates most in need of additional programming given their history of parole failure. We estimate that even a modest decrease in returns to custody for this population could save the state several million dollars.

Expand Eligibility for Intermediate Sanctions. Under the parole reforms currently being put in place, intermediate sanctions will be used as an alternative to prison for parolees with nonviolent and nonserious criminal histories. However, research demonstrates that some factors other than criminal history—for example, age or completion of certain prison programs—have been shown to be statistically relevant factors in predicting future criminality. Therefore, there may be some inmates who do not qualify for intermediate sanctions under the current policy for whom incarceration may not be the most cost-effective alternative in the event of a parole violation. However, revocation would remain the appropriate response for many parolees, such as those with a recent history of violent crime or sex offenses.

Expand Intermediate Sanctions and Graduated Sanctions. We surveyed the literature to identify other potential intermediate sanctions that could be implemented in California. We found that many other states are utilizing intermediate sanctions in addition to the ones adopted under the 2003-04 Budget Act, including community service, day fines, intensive supervision, and day reporting centers. Research suggests that alternative sanctions such as these can be less expensive than revoking a parole violator to prison and, when combined with treatment and service programs, can reduce recidivism by as much as 25 percent.

In addition, expanding the types of alternative sanctions available to parole agents and parole violators will allow more opportunities for agents to match an appropriate level of sanction with a parolee given the characteristics of the parolee, his or her criminal history, and the nature of the parole violation. In fact, the current array of intermediate sanctions may not be the most effective or appropriate for all violators. For example, the use of community service or fines might be a more appropriate response for an employed parolee who commits a single technical violation than more restrictive sanctions such as confinement in SATCUs or CCRCs.

Offering a greater array of sanctions will also allow the department to establish a system of graduated sanctions for repeat violators. Graduated sanctions allow agents to increase the severity of punishment in response to increased frequency or severity of parolee infractions with the expectation that some parolees will curtail their violations to avoid the increasing penalties for their behavior. Prison would still remain a punishment option for offenders. The expanded use and/or eligibility of intermediate sanctions to even a few hundred additional parolees would save the state several million dollars.

While there could be a couple million dollars in implementation costs to expand the use of prerelease planning and intermediate sanctions, these reforms could generate net General Fund savings by further reducing the inmate population. In view of the potential to increase General Fund savings while protecting public safety, we recommend that the Legislature direct CDC to provide an estimate of the fiscal impact of these proposals. Based on this information, the Legislature may wish to direct the department to phase in the above described expansions during 2004-05. Because the department is already implementing several new programs, the program expansions recommended here may also experience delays due to workload limitations on existing staff. However, we believe that these expansions merit consideration because of the General Fund savings that would be generated when implemented.

Corrections Overtime Expenditures

Summary

Despite recent oversight and administrative efforts by the Legislature and the Department of Corrections (CDC), the department's overtime expenditures are expected to remain over $230 million in the current year. In this analysis, we provide background information on this issue, then an update on the current status of overtime, and finally recommendations for controlling departmental overtime.

Background

The CDC has over 45,000 positions funded at almost $3.5 billion in the current year. Of those positions, approximately half are "posted" positions in the institutions. Posted positions are those that provide supervision of and services for inmates at state prisons. For example, most correctional officer positions and many medical positions are posted.

Overtime expenditures occur when a posted position is unfilled or when the employee who regularly fills that position is absent. In either case, when another employee fills that position by working hours in excess of his or her own scheduled hours, the department pays that employee at a rate of 1.5 times his or her salary for each hour of overtime worked.

Costs Associated With CDC Overtime

Figure 3 shows the department's overtime expenditures over the past five fiscal years, from 1999-00 through 2003-04 (projected). It shows that annual CDC overtime expenditures rose almost $50 million between 1999-00 and 2002-03. These expenditures reached a high of $263 million in 2001-02 and declined $26 million in 2002-03. The decrease in overtime expenditures occurred primarily in the Institutions and Community Correctional (Parole) programs and is expected to continue—though remaining over $230 million—in the current year.

We recognize that it is not unusual for departments to have some overtime, but the growth in and persistent nature of the overtime expenditures by CDC suggests that the department has a continuing problem. Overtime expenditures are a significant contributor to the department's annual deficiency. In 2002-03, for example, CDC spent approximately $45 million more for overtime payments than it was budgeted. The department projects its current-year overtime deficiency to exceed $50 million.

High overtime usage may also result in increased workers' compensation costs. This is because working overtime can lead to employee fatigue and injuries, resulting in increased workers' compensation claims. The CDC's workers' compensation costs are projected to be approximately $200 million in the current year. However, the department is unable to identify how much of these costs were attributable to employees working overtime.

Previously Identified Drivers of CDC Overtime

Past reports by our office, the Bureau of State Audits, and the Department of Finance identified several drivers of overtime costs within CDC, including correctional officer vacancies, inadequate numbers of relief positions, use of overtime to cover for sick leave and other paid leave programs, and inadequate management controls and oversight. Prior to 2002-03, between 10 percent and 15 percent of officer positions in CDC were vacant. Prison administrators typically had the option of using relief officers—specifically designated to fill otherwise unfilled positions—or overtime to fill vacant positions. Using relief officers is a less costly alternative than overtime because relief officers are paid at their base salary rather than 50 percent above, and many relief officers are newer officers paid at the lower wage range. However, there were too few relief officers available at most institutions to fill all vacancies. Instead, administrators relied more on using overtime.

As with vacant positions generally, posted positions vacant due to sick leave must be filled by another officer. Prison administrators often used overtime to fill positions left vacant by sick leave. Sick leave was a driver of overtime expenditures for the department because correctional officers' average use of sick leave was significantly higher than non-CDC state employees—6.5 hours per month compared to 4 hours. In addition, a report by the Bureau of State Audits in 2000 found that the department failed to effectively manage paid leave programs, such as holiday, vacation, and sick leave, resulting in excessive overtime and other costs. In particular, the report cited the failure of the department to collect and analyze personnel data from institutions, information that could be used to more effectively manage and direct personnel resources.

Actions Taken by Legislature and CDC

The Legislature and CDC have both taken steps to address the issues driving departmental overtime costs. In particular, the Legislature has focused its efforts on filling posted positions. For example, in 2000 and 2001 the Legislature authorized the expansion of CDC's training facility in order to provide the department with more correctional officers to fill vacancies in posted and relief positions. As a result, the statewide correctional officer vacancy rate fell from 12 percent in 2001-02 to less than 2 percent at the end of 2003.

The CDC made changes in recent years in an attempt to better manage overtime. In general, these efforts have focused on collecting and analyzing fiscal and personnel management information. The department now (1) performs annual fiscal reviews of each institution, including overtime expenditures; (2) operates overtime/sick leave management committees in every institution; (3) uses the Watch Office Tracking System at each institution to provide daily, weekly, and monthly expenditure reports to central management, including overtime expenditures; and (4) operates the Resources Review Team within the Financial Services Division to coordinate personnel management practices and gather information. The department has also submitted a Feasibility Study Report to support its proposed Business Information System project which is intended to be a department-wide personnel and budget management data system.

Findings

While the problem of the California Department of Correction's overtime expenditures has been discussed in past reports, our analysis identifies ways that provisions of the correctional officer contract and the missions of different institutions contribute to what remains an ongoing overtime problem. These findings are discussed in greater detail below and are based on information provided by the department, visits to prisons, and discussions with prison administrators and staff.

Vacancies in Many Posted Positions Affect Overtime

As discussed above, progress has been made in some areas, in particular filling correctional officer positions. However, there remain a high number of vacancies in other posted positions that prison administrators often fill with overtime. For example, the vacancy rate for sergeants and lieutenants is approximately 8 percent, and each of these vacancies is more costly than an officer vacancy because of the higher pay earned by correctional supervisors. In addition, over 23 percent of the department's posted medical positions—including registered nurses and medical technical assistants—are vacant. The department believes that these vacancies are the result of inadequate salaries to compete with outside medical facilities. Figure 4 (see next page) shows vacancies and overtime expenditures for selected CDC posted positions in 2002-03. While the single position that is the biggest driver of overtime expenditures is correctional officers, the department spent almost $60 million in overtime costs on the other four classifications of posted positions listed here.

Correctional Officer Contract Contributes to Overtime Costs

Several provisions of the current correctional officer contact result in increased overtime costs. One such provision is the contract requirement that senior officers have the first opportunity to work overtime shifts. This provision results in higher overtime expenditures than if overtime hours were more evenly distributed among all officers because senior correctional officers have higher base salaries than junior officers. Also, the contract limits the number of hours that Permanent Intermittent Correctional Officers (PICOs) can work as well as the number of PICOs that institutions can hire. The PICOs are relief officers that institutions can call to temporarily fill vacant positions. These officers typically have lower salaries than full-time officers and, therefore, are less expensive when addressing overtime requirements. In addition, the annual salary increases required by the contract between 2003 and 2006 will contribute to the department's overtime costs. This is because as the salary of correctional officers rises, the cost of paying overtime—at 1.5 times that salary—will rise proportionately.

According to the department, a significant increase in sick leave use by correctional officers can also be attributed to the current bargaining contract as a result of two factors. First, the current correctional officer contract eliminated the Extraordinary Use of Sick Leave program that institutions previously used to track and identify sick leave abuse. Second, the new contract changed how sick leave counts towards officers' 40 hours of work each week, making it easier for officers to earn overtime after having taken sick leave. Sick leave use by correctional officers has risen approximately 25 percent since 2000 to an average of 8.3 hours per officer per month. The department estimates that it spent over $80 million for overtime in 2002-03 to cover sick leave.

Overtime Spending Reflects Institution Factors

Spending for overtime by each of the individual prisons varies considerably, ranging from approximately $3 million to over $15 million in 2002-03. This wide variation suggests that overtime expenditures may be in part a reflection of institution-specific factors and activities. In order to evaluate this variation, we used a regression analysis which suggests that the variation in overtime expenditures is related to differences in custody level, assault incidents, and special missions at each prison. Below we illustrate how each of these factors contribute to overtime costs.

Custody Level. For the purpose of distinguishing those inmates who are most dangerous, violent, and a threat to escape, CDC classifies inmates on a scale of I through IV with a Level IV inmate being considered the most dangerous or highest threat to escape. The CDC houses inmates in prison housing units with a corresponding ranking designed to ensure the appropriate custody level. The 17 institutions housing Level IV inmates (including San Quentin which houses condemned inmates) average 34 percent more in overtime costs than the other 15 institutions when we control for inmate population. Overtime expenditures are generally related to the security level of institutions because the most dangerous and violent inmates require additional guarding and are the inmates most likely to engage in assaults and other incidents (see below).

Assault Incidents. The five prisons with the highest number of recorded assault incidents in 2002 averaged 72 percent more in overtime costs than those five prisons with the fewest assaults when controlling for differences in population size. Inmate incidents, particularly violent ones, often require officers and other staff to complete additional paperwork and provide additional security coverage as a precaution against an escalation of the incident. To the degree that assaults lead to additional workload, officers are likely to work overtime hours to complete these tasks.

Special Missions. Several prisons have special missions that they are required to carry out in addition to custody and standard services and programming. Nineteen prisons fulfill at least one special mission, including operating reception centers, security housing units, correctional treatment centers, psychiatric services units, and fire camp training. These 19 institutions averaged 22 percent more in overtime expenditures than those institutions not operating under special missions when we control for population size. It is not clear why the institutions with special missions experience more overtime than the other institutions, though it may be that these activities require higher levels of staffing but are not budgeted accordingly.

Some Prison Activities Regularly Result in Overtime Costs

Some prison activities often result in overtime costs. In particular, CDC policies and administrative decisions regarding the staffing for medical guarding, transportation, and "administrative segregation overflow" regularly require the usage of overtime by officers and cause high overtime costs.

• Medical Guarding. Institutions have to transport some prisoners to local hospitals and clinics for specialty health services that the prisons cannot provide. The CDC spent over $26 million in overtime for medical guarding in 2002-03. These expenditures covered the costs of providing a team of correctional officers to transport inmates between the hospital and prison and supervise them while in the hospital. Often, the length of time required to complete the medical guarding and transportation requires officers to work beyond their scheduled eight hour shift, thereby generating overtime costs.
• Transportation. The department transports inmates for a variety of reasons, including assignment from a reception center to a mainline institution, reassignment to a different institution, release from prison to their home county, and attendance at legal proceedings. Many of these trips can last longer than an officer's assigned shift, thereby requiring overtime. In 2002-03, department overtime expenditures for transportation were $2.3 million.
• Administrative Segregation Overflow. All prisons have temporary housing for inmates who, for a variety of reasons, need to be separated from the general population. This housing is referred to as administrative segregation (or "ad seg") and requires more correctional officer staffing than traditional housing in order to supervise and escort these inmates. Ad seg overflow occurs when more inmates are assigned to ad seg than there are cells available, and prison administrators temporarily convert a housing unit designated for another population to ad seg. This additional housing for ad seg is referred to as ad seg overflow. The department experiences overtime expenditures for ad seg overflow because institutions do not receive additional funding for the increased officers necessary to supervise the overflow into additional housing units. The CDC was unable to provide information on the total overtime costs attributable to ad seg overflow in 2002-03. However, anecdotal evidence suggests that the use of ad seg overflow is a regular occurrence at many institutions.

Institution Staffing Not Based on Activities and Population

The department's institution staffing formula does not adequately reflect the staffing requirements associated with various inmate populations and institution activities. Instead, after establishing the base staffing level necessary to operate the institution—based on population, prison design, custody level, and missions—staffing grows by a ratio of one employee for every six inmates above the 100 percent capacity level. While this overcrowding staff can be made up of any mix of personnel to which the institution and administration agree, the fixed one employee to six inmate ratio does not take into account the variation in populations, activities, and prison design at different institutions. For example, similar growth in Level I, General Population inmates and Level IV, mentally ill inmates would result in the same total staffing increases even though the two populations would require very different supervision and programming. We believe that this fixed ratio may work to drive some overtime costs because it may be difficult for those institutions serving high custody inmates and multiple missions to meet all of their staffing needs with the same ratio as other institutions.

Recommendations

Based on the findings described above, we offer a number of recommendations for reducing CDC's overtime expenditures. The recommendations include filling noncorrectional officer posted position vacancies, renegotiating provisions of the correctional officer contract, reexamining how the department staffs prisons, and improving the accountability of individual institutions to control overtime spending. We believe that these recommendations will result in net fiscal savings for the state by significantly reducing the department's overtime costs.

Fill Posted Position Vacancies

While significant progress has been made in filling correctional officer vacancies, we recommend that the Legislature adopt supplemental report language requiring the department to identify cost-effective ways to fill vacancies in its other posted positions, especially in those posted position classifications that have the highest vacancy rates (sergeants, lieutenants, and medical positions). The department already offers some recruitment and retention bonuses and has centralized its efforts to recruit qualified medical personnel. However, the department should identify other means of improving its recruitment efforts in order to fill its vacant positions. In so doing, the department could generate net savings by filling these positions at a regular salary level, rather than paying the overtime premium. In addition, filling the posted medical positions will reduce the costs to the department for contracting for medical personnel. It is important to note, however, that the difficulty the department has at hiring and retaining qualified medical technical assistants and registered nurses is part of a broader trend facing other state departments and local hospitals and clinics. Therefore, statewide action may be necessary to fully address this problem.

Direct State to Renegotiate Unit 6 Contract

The state's current fiscal situation has forced it to revisit employee contracts with most unions, and a number of changes have been made to those agreements. In addition, the new administration has indicated a desire to renegotiate contracts to generate budget savings. At the time this analysis was prepared, no change had been made to the correctional officers contract. We recommend that the DPA renegotiate provisions of the correctional officer contract that have contributed to increased overtime expenditures. Specifically, DPA should negotiate to reinstitute the department's program to monitor and discipline sick leave abuse, eliminate the preference given to senior officers for voluntary overtime, and remove the limits on the use of PICOs. These changes, as well any reduction to or delay in correctional officer salary increases, would generate net savings to the General Fund by significantly reducing department overtime costs.

Examine Staffing Formulas for Prisons and Prison Activities

We recommend the adoption of budget bill language requiring the department to identify a more efficient way of distributing staff resources. Our analysis indicates that some factors of the prison population (for example, custody level and inmate assaults) and prison activities (for example, special missions and administrative segregation overflow) significantly contribute to overtime expenditures. However, the department relies upon a fixed ratio of one staff per six inmates to determine the staffing level in individual prisons. It does not take these other factors into account. This is significant because it suggests that staffing may not be directed in the most efficient way to meet the varying staffing requirements of institutions, thereby contributing to the department's high overtime expenditures. A more effective approach to staffing might be to replace the fixed staffing ratio with variable ratios that reflect the staffing requirements dictated by differences in populations and activities. The department should also present to the Legislature its policies on how it staffs medical guarding, transportation, and administrative segregation overflow and determine a less costly way to staff these activities. The following budget bill language is consistent with this recommendation:

5240-001-0001 Provision X. No later than January 10, 2005, the Department of Corrections shall submit to the Chair and Vice Chair of the Joint Legislative Budget Committee, and the Committee on Budget in both the Assembly and Senate, a report identifying ways to reduce overtime costs caused by the inefficient staffing of state prisons. In particular, this report should identify the fiscal impact of the current staffing ratio used to staff prisons and alternative staffing procedures that would generate state savings, including less costly staffing methods for transportation, medical guarding, and administrative segregation overflow.

Improve Institution Accountability

The administration has indicated its intent to utilize better cost control procedures to manage CDC's spending and improve the fiscal accountability of the department. The relief factor proposal (described later in this chapter) is one part of this plan, and the department indicates that more aspects of the plan will be released as part of the May Revision. We recommend that the administration provide this information to the Legislature prior to the May Revision so that it will have adequate time to review it. The plan should detail how it will address the overtime problem, particularly the variation in overtime expenditures across institutions. While much of the disparity in overtime expenditures at different prisons can be attributed to factors such as special missions and custody level, these factors do not explain all of the variation in overtime costs, suggesting that institution management plays a critical role in controlling overtime spending. For example, the R.J. Donovan (RJD) Correctional Facility and Wasco State Prison (WSP) are similar institutions in that both are male, level III institutions with a reception center and have an above average number of assaults. Yet, while WSP operates with a larger population than RJD, Wasco's overtime expenditures totaled $6.1 million while expenditures at RJD were $9.8 million, a difference of 61 percent. The administration should identify in its forthcoming plans ways to better control overtime spending at individual prisons. This could include, for example, changing the overtime authorization process for prison supervisors, improving the effective utilization of fiscal management systems by institutions, and creating incentives for prison administrators to stay within their overtime budgets.

Relief Factor Proposal Lacks Sufficient Detail

We withhold recommendation on the Governor's proposal to increase funding by $99.5 million and 1,239 relief positions to fill in behind officers who are on leave because the administration has not provided sufficient details about key aspects of the request to allow for legislative review. We recommend that the Legislature require the department to provide more detail on this proposal prior to budget hearings, including the estimated savings from reductions in overtime costs, the distribution of positions to institutions, and the administration's fiscal control proposal.

Background

Posted Positions. The CDC is charged with the responsibility for the custody and care of state prisoners. In order to carry out this mission, the department must have sufficient staffing to supervise and treat inmates 24 hours a day, seven days a week. For this reason, many positions in prisons are "posted" positions, work assignments that must always be filled. In other words, if a posted position is left vacant because the employee normally in that assignment has taken a vacation or sick leave, for example, then another employee would have to fill that position for the day. Most posted positions in institutions are correctional officer assignments, though there are also correctional supervisor, medical, and cook posted positions.

Posted Position Relief. The department is able to estimate the average number of days that officers typically take off for various leaves, including regular days off (RDO), holidays, vacations, and sick leave. For example, a guard tower that needs to be staffed during one shift, seven days a week will have one officer assigned to that position, but he will need other officers to fill in twice each week on his RDOs. Based on statewide averages for correctional officers, the officer will also need relief for another 43 days each year because of holidays, vacation, and sick leave.

Relief Factor. The department uses the information on average relief required for RDOs, holiday, vacation, and sick leave to compute a relief factor. For example, the relief factor for most correctional officer positions is 1.67. This means that in order to staff that posted position seven days a week, the department requires not only the full-time officer assigned to that post, but also the equivalent of 67 percent of another full-time officer for relief. This factor is used by CDC to estimate the total number of employees it requires to offer relief for posted positions. The CDC then requests funding in the budget commensurate with this level of staffing.

Governor's Relief Factor Proposal

The Governor's budget requests 1,239 new relief positions at a General Fund cost of $99.5 million in 2004-05. The proposal has two components.

• Additions Related to Past Reductions—$15.3 Million and 42 Positions. The proposal adjusts for past budget reductions to ensure full funding and staffing of the 1.67 relief factor. These adjustments total $15.3 million and 42 positions. Of this amount, $2.8 million and 42 positions are to establish positions requested to achieve the 1.67 relief factor, but not approved in the 2001-02 Budget Act. The remaining $12.5 million is meant to backfill for relief funding that was redirected to fund vacant positions in the 2003-04 Budget Act.
• Increase Relief Factor for Additional Leave Programs—$84.2 Million, and 1,197 Positions. The department also proposes funding and positions to increase the relief factor to 1.76 for correctional officers, sergeants, and lieutenants. This increase reflects additional leave programs that are sanctioned by the department and used by officers, but that are not included in the current relief factor of 1.67. These include military, bereavement, and family medical leaves, as well as officer training and result in about 10.5 additional days of leave. The department also proposes to make annual adjustments to the relief factor to reflect actual usage patterns.

Figure 5 shows how the proposed relief factor is computed for a full-time correctional officer post. The relief factor is computed for each type of leave by dividing the number of days for each leave category—for example, vacation, 16.51—by the number of days in a personnel-year (PY) (207.72 days). Totaling the resulting relief factors gives the total relief factor for the position.

The department states that the increased relief factor is necessary to provide adequate staffing behind officer positions and will provide two fiscal benefits. First, the staffing of more relief officers will allow officers to take more of their allotted vacation time than they do currently, thereby reducing the bank of accrued leave that the department must pay out when officers leave the department. These payouts can be particularly expensive because the accrued leave balance is paid at the salary level of the employee when he/she leaves a department, which can be a higher pay rate than when the leave was initially earned.

Second, the department argues that the increased relief staffing levels should reduce its reliance on overtime and temporary help. Prison administrators rely on overtime and Permanent Intermittent Officers—temporary, part-time officers—when full-time relief officers are not available. The CDC is currently running deficiencies in both overtime and temporary help totaling $79 million.

LAO Comments

While more relief positions are probably necessitated by the leave usage types and rates identified by the department, the relief factor proposal lacks sufficient detail. To allow for legislative review, we recommend the department be required to provide a more detailed plan prior to budget hearings.

Implementation and Fiscal Details Lacking. The 32 state prisons operated by CDC each have unique characteristics. For example, they vary in their design and mission, the size and makeup of their workforce, and the amount of overtime that is required of staff. Prisons also vary in their leave usage patterns, including average rates of sick leave and vacation. Cumulatively, these factors make a significant difference in the requirements for staffing at each prison, including the number of relief officers.

Despite these differences, the department's plan does not detail how it would distribute the proposed 1,239 positions among its 32 institutions. The department has suggested that at some future time it intends to create separate relief factors for each prison. We agree that this would be an effective approach to accounting for the variation in relief required at different institutions. However, no details of such a plan are provided.

In addition, while the department claims that the relief factor proposal will have certain fiscal benefits to the state, it does not provide an estimate of how much the proposal will reduce spending for leave balances, overtime, and temporary help. The lack of such information makes it difficult for the Legislature to (1) assess the proposal's full fiscal impact and (2) hold the department accountable for achieving the suggested results.

Department May Lack Necessary Fiscal Controls. The administration suggests in the Governor's Budget Summary that it will propose fiscal control measures designed to enable CDC to better manage how positions are used at institutions. We agree that the decentralized nature of CDC, autonomy of individual institutions, and limited fiscal tools by central management has historically limited the ability of CDC headquarters to control institution spending. In light of the department's fiscal control problems, we are concerned that CDC may lack the necessary fiscal controls for institutions to ensure that the proposed relief positions are utilized as intended by each institution.

Training Provision of Correctional Officer Contract Biggest New Cost Driver. The current bargaining contract with the correctional officer union requires that, beginning in 2004-05, the department provide monthly officer training during officers' normal work schedule rather than off-schedule as is done currently. As with leaves, this training requirement will necessitate that the department fill positions with relief officers during those hours. More than half of the total cost of the additional leave programs is a result of this contract provision. The administration has already indicated that it will seek to renegotiate provisions of the correctional officer contract in order to generate state savings. Returning to the former practice of requiring correctional officers to participate in training outside of their normal scheduled hours would reduce the cost of this relief factor proposal by approximately $50 million annually.

Analyst's Recommendation

To enable the Legislature to fully assess this proposal, we recommend that the Legislature require CDC to provide a detailed plan of how it will implement the relief factor proposal and what reductions in spending would be achieved. We also recommend that the department report prior to budget hearings regarding the details of its fiscal control and accountability proposal for CDC. Specifically, the administration should address how provisions of its fiscal control proposal will impact the implementation of the relief factor proposal at state prisons. Finally, in light of the magnitude of the current request and the state's fiscal challenges, the Legislature may wish to direct the department to provide a plan for phasing in the relief factor proposal over time.