Performance-Based Expansion
		Under the current planning process, even if the Legislature were to agree to authorize and fund CDC's entire proposal, the State's prisons would become increasingly crowded.
		The Department of Corrections estimates that in mid-2000 the State will run out of places for additional inmates in existing facilities. Even if the Legislature were to agree immediately to build a traditional prison, CDC maintains that it could not design and construct a new prison by that date.

Facility Planning

Finding 4: The State lacks an adequate process for assessing the needs and options for housing, training and treating felons sentenced to state prison.

During the prison construction boom, the California Department of Corrections developed a process for designing and constructing new facilities that leveraged the efficiencies of the private sector to construct large public facilities while providing for legislative oversight. Unfortunately, as the political consensus for additional prisons has eroded, the experienced team that developed and refined this process has been idled and is being dismantled.

CDC has another process that comes before design and construction -- a facility planning process used to develop proposals for new prisons. Briefly, the department forecasts the future prison population and proposes to build what it believes to be the politically acceptable number of high-security prisons.

The process has a number of shortcomings that at the very least has contributed to the inability of divergent interests to agree on what kind of prisons should be built when. The current process does not allow for an open discussion of the major assumptions underlying the need for new prisons, or the opportunities for the State to chart a different course. Similarly, the process does not allow for an open discussion or analysis of the alternatives for housing prisoners. And perhaps most importantly, the process never results in a plan for housing all of the inmates that are expected to reach state prison -- so even if disagreements were resolved and the proposals were implemented, the prisons would become increasingly crowded.

The Current System

The California Department of Corrections periodically develops a five-year facility plan that shapes the department's legislative proposals for constructing new facilities.

CDC's facility plan begins with the department's inmate population projections. The department estimates population needs twice a year -- in the spring and the fall. The estimates are based on certain assumptions -- but primarily that the immediate future will closely resemble the immediate past. A computer model then estimates the number of new felons that could be expected to be sentenced to prison, when they will be released on parole -- and for a majority of the parolees -- when they will be returned to prison and released again.⁽¹⁰⁷⁾

The department believes that its population projections are as accurate as possible. Understandably, the forecasts almost always under-project or over-project the incoming population. Also as can be expected, the projections are more accurate in the near-term than at the end of the five-year projection horizon.

In the 1995-96 time frame, the projections tended to overestimate the effects of Three-Strikes sentencing, and as a result, overestimate the inmate population. The department attributes this to its assumption that judges would adhere strictly to the law. Because judges found room to disregard some previous felonies in sentencing repeat offenders, CDC believes that not all of those who qualified for the longer sentences under the law received them.

More recently, in 1997 the department underestimated the number of parolees who would violate the terms of their release and return to prison -- further demonstrating the difficulty of predicting the future.⁽¹⁰⁸⁾

Because of the biannual adjustments, the short-term inaccuracies are more important to day-to-day operations of the prisons than the longer-term task of planning and constructing new facilities. And because the opening of new prisons routinely lags behind the increasing inmate population, the department is continually planning for new facilities.

Another important element of the population projections is the estimate concerning the kind of inmates that can be expected. All state prison systems classify inmates in order to place them in appropriate housing. Inmates who are dangerous or likely to attempt escape are placed in the most secure cells; inmates who are not dangerous can be housed in less-restrictive and less-costly facilities, such as dormitories.

The classification system used in California was established in 1986 and places inmates in one of four categories -- Level I through Level IV, with Level IV the most restrictive.

The department's classification system employs more than 20 variables, including gang affiliation and behavior during previous incarcerations. The length of sentence, however, dominates the score, and as a result, most inmates with sentences exceeding 10 years are classified as levels III and IV and so are confined initially to cells. Because the "length of sentence" variable is heavily weighted in the calculation, inmates receiving long sentences as a result of the Three Strikes Law or other sentencing enhancements are more likely to be classified at Level III and Level IV than in the past. The result is an even higher demand for the most expensive, high-security housing.

As the table below shows, the construction costs for facilities varies considerably depending on the facility's level of security. The figures in the chart are based on the construction costs of the recently completed Corcoran II prison. The figures are based on the "high-occupancy" use of the facility. If it were to hold the population originally intended for that prison design, the capital costs per inmate would double.

California's Inmate Classification System
Classification	Type of Security	Construction costs per inmate
Level I	Camps and dormitories with low security level	$35,160
Level II	Open dormitories with secure perimeter, armed coverage possible	$51,302
Level III	Secure perimeter, armed coverage, cells adjacent to exterior walls	$54,478
Level IV	Secure perimeter, internal and external armed coverage, cells not adjacent to exterior walls.	$63,479

Source: Department of Finance

The table below shows CDC's anticipated growth in male felons, which make up more than 90 percent of the prison population, broken down by security level.⁽¹⁰⁹⁾

Projected Increase in Male Felon Prison Population
	Reception	Level I	Level II	Level III	Level IV
Dec. 31, 1997	19,390	34,294	29,060	35,875	21,885
June 30, 2003	25,350	46,746	39,480	50,713	31,197
Numerical change	5,960	12,452	10,420	14,838	9,312
Percent change	30.7	36.3	35.8	41.3	42.5

In recent years, CDC's classification system has come under increasing criticism. The concern is that the classification system "over-classifies" inmates -- committing the State to build more expensive high-security facilities where inmates are less likely to receive work or educational opportunities.

A Department of Finance analysis of the classification system in 1996 found the CDC may be giving too much weight to the length of an inmate's sentence and not enough weight to the inmate's behavior during previous prison terms and to the inmate's age. The Department of Finance also found that Level III inmates were not involved in more incidents in prison than Level II inmates. And Department of Finance investigators found that inmates older than 35 -- no matter what their classification -- were involved in fewer incidents than Level II inmates.⁽¹¹⁰⁾

CDC maintains that length of sentence is an important predictor of an inmate's propensity toward violence and escape. But the Department of Finance argued that research does not support that conclusion.

Also in 1996, the California Correctional Peace Officers recommended that the State alter its classification system. In its assessment of the prison housing crisis, the union argued that many inmates are placed in medium security Level III prisons because they are judged to be escape risks, not because they are violent. The union pointed out that a secure perimeter -- such as the lethal electrified fences around medium and maximum security prisons -- could be used to surround dormitory-based facilities.

CDC responded by hiring two consultants -- one to analyze the system and one to interpret the analysis. The results, described in more detail on the following page, did not resolve the debate. At a minimum, the analysis shows that the department needs to begin collecting information that will allow for the classification system to be refined in the future with greater confidence that it will improve and not disrupt prison operations.

The controversy underscores the importance of the basic assumptions in CDC's facility planning process and the need to reach agreement on the assumptions before agreement can be expected on the proposal the process produces.

Historically, CDC's Planning and Construction Division uses the population projections, broken down by security levels, to estimate the number of new prisons that will be needed.

In preparing the plan, department officials say they internally consider options for housing inmates, including the potential for expanding re-entry work furlough centers. The only alternative proposed by the department was maximum security cells.

Similarly, the facility plan also does not explicitly discuss the role or potential role of the community correctional facilities that are under contract with the department. Those facilities, even though they house low- and medium-security inmates, are monitored by the Parole and Community Services Division. They are not planned by the Design and Construction Division and not managed by the Institutions Division.

Recent facility plans reach one conclusion and make one recommendation: The conclusion is how many prisons are needed to meet the entire anticipated inmate population. The recommendation is how many prisons should be built. The most recently completed plan, dated June 1996, anticipates that 17 new prisons will be needed by 2001. The plan, however, only proposes to build six new prisons. While conceivably the remainder of the prisons could be authorized in future years, the plan itself makes no provision for accommodating the needs even five years out.

The proposal generated by the plan is then included in the Governor's budget. In recent years, since the public rejected a general obligation prison bond measure in 1990, construction has been financed with revenue bonds. No matter which financing mechanism is used, legislative approval is needed before additional prisons can be built.

The facility plan does describe the department's refined process for actually designing and constructing prisons once they are authorized. The process has been lauded by some reviewers and even critical reviewers have suggested modest improvements. Among its attributes: CDC has emphasized standardized designs to minimize design costs and reduce errors. At the same time, it has institutionalized learning by constantly evaluating facilities that are in operation and refining the standardized design. The department also has emphasized life-cycle costs to make sure that a dime saved in construction does not cost the State a dollar in additional operational costs or avoidable maintenance.

Political Stalemate

The current planning process produces a proposal that is presented in whole or in part in the Governor's budget. In recent years, however, the proposals have resulted in political stalemate:⁽¹¹¹⁾

• For the 1995-96 fiscal year, the Governor's budget proposed $2 billion in lease-revenue bonds to build six new prisons. The Legislature did not authorize the bonds or the prisons.



• For the 1996-97 fiscal year, the Governor's budget proposed a Public Safety Bond Act for the 1996 ballot that would seek voter approval for $1.9 billion in general obligation bonds. The bonds would be to build six of 13 new prisons that CDC then said would be needed by 2001.

• For the 1997-198 fiscal year, the Governor's budget proposed using $265 million in federal grants to design three new prisons, and $249 million in lease-revenue bonds to finance construction of one of the three, in anticipation of additional federal funds to cover much of the construction costs. Environmental reviews also would have begun for three additional prisons.

The budget proposals have run into legislative opposition by some lawmakers who believe that budget for the Department of Corrections is expanding too quickly, that more felons should be held in county facilities and that intermediate sanctions between probation and prison should be expanded.

In 1994, the Legislature passed and the Governor signed the Community-Based Punishment Act, which solicited from counties proposals to house and supervise more inmates at the county levels. While many of the community plans outlined the potential for intermediate sanctions, most counties also proposed additional jail construction. In addition, some local officials, leery of two decades of budget crunching, were suspicious about plans to shift responsibility from state prisons to county correctional agencies without assurance of the needed revenues.

Just as no new prisons have been funded, none of the community punishment plans have received state funding.

In addition, the Legislature has proactively entered the facility planning process. During the prison construction boom legislation was passed directing CDC to contract with local governments and private companies willing to build and operate low-security facilities, primarily to house parole violators. More recently, the Legislature has considered and rejected proposals to expand the use of privatized prisons.

Meanwhile, the last new prison authorized by the Legislature, Corcoran II, opened in the autumn of 1997.

Inadequate Planning

There are two indicators that the State's process of planning for additional correctional facilities is inadequate. The first is that even if the proposals were enacted completely, the system would get increasingly crowded. And secondly, the process always produces the same answer -- more state-run maximum security prisons.

Progressively Overcrowded

With no additional facilities authorized, the State's 33 prisons are expected to become increasingly crowded. With no politically acceptable solution in sight, that trend takes on an urgency. But it also is important to recognize that even when the wave of new prison construction was cresting -- and political support was united -- prisons were becoming more crowded.

Under the current planning process, even if the Legislature were to agree to authorize and fund CDC's entire proposal, the State's prisons would become increasingly crowded.

Between 1985 and 1995 the State opened 18 new prisons. During that same time, the prison system went from 150 percent of design capacity to 180 percent of design capacity. The department in recent years has acknowledged that the system can function well far above the design capacity. Nevertheless, that standard provides a benchmark for the State's ability to accommodate the inmate population -- given changes in sentencing laws and parole revocations procedures, and the ineffectiveness of the correctional programs as currently implemented to reduce recidivism.

As discussed in Finding 3, the department believes the purpose of prisons is incapacitation, not rehabilitation. From a facility standpoint, that perspective will require ever increasing prison facilities -- a fact not fully reconciled by the department's planning process or political leaders.

More Maximum Security

A final indicator that the current planning process is inadequate is that despite a diversity of inmates and a changing inmate population, the facility plan has for the most part only produced a single housing solution -- one based on high-security prisons with low-security support facilities.

The department's view is that it is only planning for the anticipated demand and the department maintains that few inmates are in facilities higher than their classifications. But they also feel compelled to build higher security facilities -- in part because they know that the demand will always exceed supply. In a crisis, it is more acceptable to put a low-risk inmate in a medium- or high-security prison than it is to put a high security inmate in a low-security prison. Because the department proposes to meet only a portion of the housing needs, its feels compelled to build only high-security facilities.

The planners also assert that many of the factors that outsiders believe should be considered for expansion in the planning process -- such as intermediate sanctions, expanding halfway houses, or expanding programs known to reduce the demand for prison space -- are fixed in policy or statute. Those issues are simply not on the table. One division head told the Commission's advisory committee that in a "brainstorming" session on the facility crisis, it was suggested that the department propose a greater use of alternative sanctions. The suggestion was summarily dismissed as beyond the realm of possible solutions.

Where the Process Falls Short

The three main shortcomings of the existing planning process can be traced to a number of places where the planning process excludes the public examination of issues and options that could produce a more cost-effective plan that might garner more political support:

• Policy makers never confront the entire need. As stated above, one indicator that the process needs improvement is that prisons are getting more crowded. Planners maintain that the department's ultimate proposal cannot be dictated by meeting all of the department's needs, but rather must consider all of the other capital outlay projects supported by the administration. Putting the need for prisons in context of other public needs is important.

But it also is important for the State to understand and plan for anticipated local jail and state prison populations. Those needs cannot be met unless policy makers are required to reconcile increased sentencing legislation with the infrastructure needed to implement them. Without a realistic proposal, the actual cost of the State's correctional policies are pushed into future budgets, and policy makers are not required to consider the cost-effectiveness of existing policies. Meanwhile, the department is in the awkward position of habitually "under planning."

• Housing alternatives are not explored openly. CDC's plans do not contain any discussion of alternatives, and as a result the alternatives are never openly or collaboratively considered. The master plan assumes that community correctional facilities will continue to exist, but does not consider or propose their expansion. Work furlough or re-entry programs for soon-to-be-released felons could be expanded -- but have not been and are not discussed as a way of reducing the in-prison population.

Similarly, the department's Preventing Parole Failure Program has been shown to more than pay for itself -- just in terms of the reduced demand for housing parole failures. But expanding that program is not considered as part of the facility plan as a way to meet growing inmate populations.

• Assumptions are not discussed openly. The facilities plans are built on a number of important assumptions. As discussed in Finding 1, CDC assumes that incarceration rates will continue to increase. While this assumption is rooted in the department's computer program, which bases future trends on historic trends, that assumption is the bedrock for a multi-billion prison construction proposal.

The classification system also shapes significantly the kinds of facilities the department will need, but refining the classification system is detached from facility planning and is only raised as an issue by policy analysts outside of the department. The California Correctional Peace Officers Association, the Legislative Analyst's Office and the Department of Finance have all suggested that a reassessment of these assumptions could lead to more inmates being housed in lower security and less costly facilities.

• The plan does not consider alternatives for implementation. CDC has a number of contracts with for-profit, non-profit and governmental organizations to house inmates and provide services. This aspect is discussed in greater detail in Finding 5. But it is important to note that the facility plan does not consider the State's options for operating any additional facilities that the department will construct.

The plan only relies on one model for procuring and operating prison space. While there have been contracts with public agencies and private companies to provide some beds, those contracts were never considered part of the department's plans to house inmates, based on analysis of the best alternatives, or even implemented by the department's Institutions Division. Instead, they were agreed to without analysis and implemented by the Parole and Community Services Division, which had little expertise in large contracts or with operating institutions.

The department's view of its facility planning process is clearly more limited than would be commonly expected. Department officials believe the legislative forum is the appropriate venue for debating assumptions and alternatives. CDC officials maintain that their planning process is not intended to publicly explore options or to result in a proposal to change programs or the State's responsibilities for housing inmates. That narrow focus is similar to the narrow focus in CDC's lauded design and construction process.

But another attribute of the design and construction process is the institutionalized learning that requires constant re-evaluation of the product and an exploration of alternatives, with progress measured on verifiable outcomes. The design and construction process also has been successful for using life-cycle analysis, which views expenditures as investments and seeks to minimize costs and maximize benefits.

A facility planning process that was similarly based on an evaluation of the use of existing facilities and an exploration of the alternatives could be expected to produce different results. A facility planning process that accounted for the life-cycle costs of recycling inmates could also be expected to produce different results. And if different results did not build support for the ultimate proposal, a more open planning process might.

Summary

The existing facility planning process does not give the State the best opportunity to examine what is needed in the way of new prisons and how that need can be met. With no process for collaboratively exploring the assumptions underlying the demand for additional prisons, and for exploring the options for meeting those prisons, the department proposals are placed in the often adversarial legislative process. An open process conducted by a panel of correctional officials could be expected to deliver the State a better plan for housing more inmates.

Recommendation 4: The Governor and the Legislature should require the modified Board of Corrections to develop plans for additional correctional facilities.

• A modified Board of Corrections should be the planning body. The responsibility should be placed with a panel comprised of gubernatorial and legislative appointments, one that represents a wide variety of interests. It should hold public meetings to gather information and consider alternative ways to incarcerate felons.

• The board should develop plans for cost-effectively accommodating the entire projected state and local inmate population. An initial step to developing a facilities plan should be a review of the classification system to ensure the State is not over classifying inmates and as a result building too many high-security prisons.

• The facility plan should provide for competitive procurement of additional facilities. The facility plan should whenever feasible provide for the acquisition of services -- including the construction and operation of prisons -- through competitive procedures that allow for proposals by the Department of Corrections, local government agencies, non-profit groups, for-profit companies, or partnerships among those organizations.



• The plans should be submitted to the Governor and the Legislature for enactment and funding.



• The board should help to identify and resolve issues associated with siting correctional facilities. Among the issues the panel should consider are the impacts on school systems and local infrastructure, as well as ways the staff and inmates of facilities can become greater assets to host communities.