Fiscal Impacts of Proposition 47

Introduction

Proposition 47, which was approved by voters in November 2014, made significant changes to the state’s criminal justice system. Specifically, it reduced the penalties for certain non–violent, nonserious drug and property crimes and requires that the resulting state savings be spent on mental health and substance use services, truancy and dropout prevention, and victim services. In this report, we (1) provide an overview of the major effects of Proposition 47 on the state, (2) recommend a framework for satisfying the measure’s requirement that the state savings it creates be deposited into a new special fund, and (3) describe and assess the administration’s approach for meeting the measure’s fiscal requirements, including the allocation of the monies deposited in the new fund.

Back to the Top

Overview of Proposition 47

Measure Impacts Various State Entities. Proposition 47 reduced certain nonserious and non–violent property and drug offenses from wobblers or felonies to misdemeanors. (Wobblers are crimes that can be charged as either a felony or a misdemeanor.) The implementation of the measure has impacted the workload of the following state entities:

• California Department of Corrections and Rehabilitation (CDCR). The measure made two changes that reduced the state prison population and associated costs. First, changing crimes from felonies and wobblers to misdemeanors made fewer offenders eligible for state prison sentences. Second, the measure allowed for the resentencing of inmates currently in state prison for crimes reduced from felonies to misdemeanors under the measure. In addition, the resentencing of individuals who were in state prison has temporarily increased the state parole population. This is because the measure required that resentenced inmates generally be placed on state parole supervision for a year.
• State Courts. The measure creates a temporary increase in court costs resulting from workload associated with (1) resentencing offenders who were serving sentences for crimes reduced from felonies to misdemeanors under the measure and (2) reclassifying the convictions of such offenders who have already completed their sentences. However, the above costs to the courts have been partly offset by savings from the conversion of felonies to misdemeanors. This is because misdemeanors generally take less court time to process than felonies. Once the resentencing and reclassification workload is complete, courts should experience ongoing savings from the conversion of felonies to misdemeanors.
• Department of State Hospitals (DSH). The measure reduces DSH costs by decreasing the number of commitments to state hospitals. This decrease has been primarily due to fewer individuals accused of felonies being committed to state hospitals as a result of being deemed incompetent to stand trial (IST). This is because counties are generally responsible for treating IST individuals accused of misdemeanors, including those whose crimes were reduced from felonies to misdemeanors by the measure.

Amount of Annual Savings Deposited in New Special Fund. Proposition 47 also requires (1) the Department of Finance (DOF) to annually estimate savings to the state resulting from the measure in the preceding fiscal year and (2) the State Controller to deposit an equivalent amount from the General Fund into the Safe Neighborhoods and Schools Fund (SNSF). The measure requires the first deposit to be made in July 2016 based on the level of savings created in 2015–16. The funds deposited in the SNSF are continuously appropriated and are required to be annually allocated as follows: (1) 65 percent for the Board of State and Community Corrections (BSCC) for grants to public agencies to support various recidivism reduction programs, (2) 25 percent for the California Department of Education (CDE) for grants to support truancy and dropout prevention programs, and (3) 10 percent for the Victim Compensation and Government Claims Board (VCGCB) for grants to trauma recovery centers (TRCs). Under the measure, up to 5 percent of the funds allocated for each of the above purposes could be used for administration.

Framework for Meeting Fiscal Requirements

No Net Impact on State Expenditures. Given that Proposition 47 does not lay out a detailed process for how the state must estimate and pay for the deposit into the SNSF, there essentially is no one “correct” way to meet the proposition’s fiscal requirements. In order to assist the Legislature, we provide below a general framework for how we believe (1) the amount of the deposit should be determined and (2) how program budgets should be adjusted to pay for the deposit. Our recommended framework is based on a key principle that the overall intent of the measure was to have no net impact on state expenditures. In other words, after accounting for all of the state savings and costs from the measure (including the SNSF deposit), it should neither increase nor decrease state expenditures.

Account for All Costs and Savings Within Each Fiscal Year. As discussed above, Proposition 47 requires that the annual deposit to the SNSF be based on the level of state savings in the prior fiscal year. We find that the most reasonable method to determine this amount is to estimate the total amount of savings that resulted from the proposition in the prior fiscal year minus any increased costs resulting from implementing the measure in that year. The difference between the total savings and total costs should then be deposited in the SNSF.

Budget Savings and Costs Separately. To pay for the deposit in a way that results in no net impact on the General Fund and provides the greatest transparency, we recommend that the Legislature do the following:

• Reduce Program Budgets to Account for Total Savings. As indicated earlier, CDCR, state courts, and DSH experienced a reduction of workload as a result of the implementation of Proposition 47. We recommend that the Legislature reduce the program budgets of these entities to account for savings associated with the reduced workload. In some cases, the Legislature will not need to take a specific action because the existing budgeting methodology for the program results in adjustments to the program’s budget to account for decreased workload, as is the case with CDCR prison operations. In other cases, the Legislature would need to take action as part of the budget process to reduce specific appropriations.
• Augment Program Budgets to Account for Increased Costs. The implementation of Proposition 47 has also increased some workload for state entities. For example, the measure requires courts to process resentencing and reclassification hearings. We recommend that the Legislature provide specific augmentations to those state entities that experience increased costs. In some cases, the Legislature will not need to take a specific action because the existing budgeting methodology results in automatic adjustments to the program’s budget to account for the increased workload, as is the case with CDCR parole operations. In other cases, the Legislature will want to take action to increase specific appropriations.

Based on this framework, we assess the administration’s plan for 2016–17. As we discuss below, the administration has taken a different approach in several areas than what we recommend above.

Back to the Top

How Much Money Should Be Deposited to SNSF in 2016–17?

Currently, the administration estimates that $29.3 million from the General Fund would be deposited into the SNSF on July 31, 2016 for expenditure in 2016–17, based on its estimates of the savings and costs resulting from the implementation of Proposition 47. These estimates are summarized in Figure 1, which we discuss and assess in detail below. Overall, we find that the administration likely underestimates the savings and overestimates the costs resulting from the proposition. In addition, we find that the administration did not account for all costs and savings from the measure in the fiscal year in which they occur, as we recommend as part of our proposed framework above.

Estimated Total Savings for 2015–16

Administration’s Estimates

The administration estimates a total of $62.7 million in savings from the implementation of Proposition 47 in 2015–16. These savings consist of the following:

• Prison Savings—$52.2 Million. The administration estimates that Proposition 47 will reduce the average number of state prison inmates in 2015–16 by about 4,700 for a total savings of about $52 million. This total consists of $39 million in savings from a reduction in inmates housed in state prisons and $13 million in savings from a reduction in inmates housed in contract beds.
• DSH Savings—$8.7 Million. The administration estimates that Proposition 47 will reduce the average number of patients in its hospitals in 2015–16 by 73, resulting in savings of $8.7 million.
• Court Savings—$1.7 Million. The administration estimates that the courts will have a reduction of 6,000 criminal cases in 2015–16 because of Proposition 47, resulting in savings of $1.7 million.

LAO Comments

Prison Savings Likely Underestimated. The administration’s estimate that Proposition 47 will reduce the number of state inmates by about 4,700 in 2015–16 appears reasonable. However, our analysis indicates that the administration has underestimated the savings associated with this inmate population reduction. One key assumption that the administration needed to make in order to estimate the prison savings from the population reduction is how much CDCR saved in 2015–16 from housing fewer inmates in state prisons and in contract beds. The administration assumes that the department was able to reduce the number of contract beds by about 400 due to Proposition 47 and that the balance of the population reductions from the measure—about 4,300 inmates—resulted in a reduction in the inmate population in the state’s 34 prisons. In other words, the administration’s approach implies that, had Proposition 47 not passed, CDCR would have otherwise accommodated the 4,700 inmates in 2015–16 by placing (1) 400 inmates in contract beds and (2) 4,300 inmates in the state’s prisons.

While we acknowledge that it is impossible to know precisely how many contract beds CDCR would be using in 2015–16 had Proposition 47 not passed, we find that the administration’s assumptions are unrealistic. This is because CDCR is currently housing just 900 inmates less than the federal court–ordered limit on the prison population. (In recent years, the state has been under a federal court order to reduce overcrowding in CDCR’s 34 prisons. Specifically, the court found that prison overcrowding was the primary reason the state was unable to provide inmates with constitutionally adequate health care and ordered the state to reduce its prison population to 137.5 percent of design capacity.) As such, adding 4,300 inmates to the state’s prisons would have resulted in the state exceeding the population cap by 3,400 inmates. In our view, it is more realistic to assume that had Proposition 47 not passed, CDCR would have attempted to maintain a similar population level in the state prisons—along with a similar cushion below the population limit—and thus accommodated the additional inmates by contracting out for an additional 4,700 beds. In other words, we estimate that Proposition 47 allowed CDCR to avoid the need for 4,700 contract beds in 2015–16. This alternative assumption yields a prison savings estimate for 2015–16 that is $83 million higher than the administration’s estimate—bringing total prison savings to $135 million.

Court Savings Likely Underestimated. By reducing certain offenses from felonies to misdemeanors, Proposition 47 results in cases being filed as misdemeanors in the courts rather than as felonies. To estimate how this would affect the courts, the administration made two key assumptions. First, the administration projected that 80,000 fewer felony cases would be filed and 74,000 additional misdemeanor cases would be filed, resulting in an total decrease of 6,000 cases in 2015–16. (Fewer overall cases would be filed to the extent locals choose not to pursue misdemeanor charges for cases that would have been previously charged as felonies.) Second, the administration assumed that these misdemeanor cases would take the same amount of time to process as they would have if they were felony cases. As a result the administration’s estimate of $1.7 million in savings is due solely to their projection that 6,000 fewer cases will be filed and does not reflect any savings in workload from converting a felony case to a misdemeanor case. We note, however, that the judicial branch’s existing workload studies show that misdemeanors take significantly less time on average to process than felonies. Specifically, felony cases require approximately six times more judicial time and approximately three times more staff time than misdemeanor cases. We acknowledge that these averages may not apply directly to Proposition 47 cases as these cases represent the least serious felony cases and the most serious misdemeanor cases. However, the administration has not accounted for any cost difference between these cases. Thus, it is likely that the administration’s court savings are underestimated—potentially by $10 million or more.

Recommend Judicial Branch Provide Updated Calculation of Savings. While we acknowledge that average case processing times for felonies and misdemeanors specific to Proposition 47 may not exist, the judicial branch can use the data underlying their existing workload studies, as well as data collected from trial courts, to provide estimates of average case processing times for these filings. We believe this is a more reasonable method for calculating the savings generated from cases being filed as misdemeanors instead of felonies. Accordingly, in order to obtain a more accurate understanding of Proposition 47 impacts on trial court workload, we recommend the Legislature direct the judicial branch to provide an updated calculation of savings by April 1. Specifically, the judicial branch should recalculate estimated savings by using adjusted average case processing times for felonies and misdemeanors to account for differences in the staffing resources needed to process these different case types. We also recommend that the Legislature direct the judicial branch to report the level of savings experienced due to Proposition 47 separately for 2014–15 and 2015–16.

Estimated Total Costs for 2015–16

Administration’s Estimates

The administration estimates a total of $33.4 million in increased costs from the implementation of Proposition 47 in 2015–16. As we describe below, these costs are related to increased parole caseloads and court workload.

• Parole Costs—$6.5 Million. The administration estimates that the average number of parolees in 2015–16 will increase by about 1,700 due to individuals being resentenced and placed on state parole for a year under the measure, resulting in costs of $6.5 million.
• Court Costs—$26.9 Million. As shown earlier in Figure 1, the administration’s estimates regarding the deposit to the SNSF in 2016–17 assumes $26.9 million in court costs. This amount simply reflects the total amount that was provided to courts in the 2015–16 budget for increased Proposition 47 workload related to resentencing. As we discuss in more detail below, this amount includes costs related to 2014–15, as well as excludes other workload costs in 2015–16 resulting from the measure for which the Governor is proposing to fund as part of his proposed budget for 2016–17.

LAO Comments

Includes Court Costs Incurred in 2014–15. Of the $26.9 million in court costs identified by the administration, our analysis finds that $25 million is for workload associated with resentencing hearings that occurred in 2014–15. As noted previously, we believe that a more reasonable way to calculate net savings in a given fiscal year is to account for all costs and savings incurred within that year. Thus, it is inappropriate to include 2014–15 costs in estimating the costs resulting from the implementation of the proposition in 2015–16.

Excludes Some Court Costs Incurred in 2015–16. We also find that the administration’s $26.9 million cost estimate for the courts does not include costs for some workload occurring in 2015–16. Specifically, the estimate excludes $2 million in additional resentencing workload costs and $12 million in reclassification workload costs, for a total of $14 million. We note that the Governor’s budget proposes a $21 million General Fund augmentation to the courts to support these costs ($14 million), as well as reclassification workload that occurred in 2014–15 ($7 million). As we believe that all costs incurred within a given year should be accounted for, it is inappropriate to exclude the above $14 million costs from the estimate of 2015–16 court costs.

Likely Overestimates Court Costs Related to Reclassifications. The administration assumes that courts require the same level of resources to process resentencing and reclassification cases. However, preliminary data collected by the judicial branch suggests that it costs less to process reclassifications. Accordingly, it is likely that the $14 million request for funding for these hearings is overstated.

On Net, Administration’s Estimate Likely Overstates Court Costs. After adjusting for the various concerns raised above, we estimate that the actual level of costs incurred by the courts in 2015–16 due to Proposition 47 is lower than the administration’s estimate of $26.9 million. Specifically, we estimate that actual costs could be lower by at least $10 million. In order to obtain a more accurate understanding of Proposition 47 impacts on trial court workload, we recommend that the Legislature direct the judicial branch to recalculate estimated costs by using actual data on the time and resources needed for reclassification and resentencing workload by April 1. We also recommend the Legislature direct the judicial branch to report the level of costs incurred due to Proposition 47 separately for 2014–15 and 2015–16.

Recommend Withhold Action on 2016–17 Request for Additional Funding. Given our concerns with how the administration and judicial branch calculated Proposition 47 costs, we recommend the Legislature withhold action on the Governor’s proposal to provide the judicial branch with $21 million in additional General Fund support in 2016–17 pending the receipt of revised cost information.

Estimated SNSF Deposit in 2015–16

Based on its estimates of the savings and costs resulting from the implementation of Proposition 47, the administration currently estimates that $29.3 million from the General Fund would be deposited into the SNSF on July 31, 2016 for expenditure in 2016–17. As discussed above, we find that the administration likely underestimates the savings and overestimates the costs resulting from the measure. Specifically, we estimate that the SNSF deposit in 2016–17 could be around $100 million higher than the administration’s estimate. While the administration’s approach would benefit the General Fund from a lower deposit to the SNSF, it does so at the expense of the programs that receive funding from the SNSF. As previously indicated, the overall intent of the proposition was to have no net impact on state expenditures. Regardless of the administration’s ultimate decision on the transfer amount, the Legislature could chose to provide additional funding to the programs funded by the SNSF.

Back to the Top

How to Pay for SNSF Deposit in 2016–17?

As discussed earlier, we recommend that the Legislature reduce the budgets of programs that have realized savings under the measure in order to ensure that it does not result in any net effect on state expenditures. However, as we discuss below, the administration does not propose adjustments to DSH or the judicial branch to reflect savings to those departments, resulting in reduced budget transparency and increased General Fund expenditures.

Administration’s Proposal

Adjustments to CDCR’s Budget to Reflect Savings. The existing population–based budget process for CDCR results in adjustments to the department’s budget for estimated changes in inmate and parolee caseload. Given that the administration’s estimate of the effects of Proposition 47 on the inmate and parolee populations appears to be largely on target, the savings and costs to CDCR resulting from the measure are generally already reflected in the department’s appropriation for 2015–16.

No Adjustments to DSH and Courts Budgets to Reflect Savings. In contrast, the budgeting process for DSH and the courts is not based on workload so a separate action is required to revert savings realized by these departments. However, the administration does not propose any reductions to DSH to reflect the estimated savings. Similarly, it does not propose an adjustment for the courts to reflect estimated savings, despite the fact that the 2015–16 budget already provided an augmentation to the courts for costs incurred to implement Proposition 47, and proposes an additional augmentations in 2016–17.

LAO Comments

Proposal Reduces Legislative Oversight. The administration’s approach to allow both the judicial branch to keep the $1.7 million it is estimated to have saved and DSH to keep the $9 million it is estimated to have saved would reduce legislative oversight. This is because it would allow the judicial branch and DSH to redirect their savings to other programs and services without legislative review or approval.

Recommend Reducing Courts and DSH Budgets. Because we believe the overall intent of Proposition 47 is to have no net impact on state expenditures, we recommend that the Legislature reduce the budgets for the courts and DSH to account for the savings resulting from this measure. This reduction would result in additional resources becoming available to offset the funds that will be transferred from the General Fund into the SNSF. To the extent that the judicial branch or DSH require additional funding for Proposition 47 or other workload, separate funding requests for that workload should be submitted to the Legislature for consideration as part of the budget process.

Back to the Top

How Are Funds Deposited in the SNSF Spent?

As discussed earlier, funds in the SNSF are continuously appropriated and are required to be annually allocated to support (1) mental health and substance use services, (2) truancy and dropout prevention, and (3) victim services.

Proposed Allocation of Funds

BSCC to Award Grants Based on Recent Legislation. Chapter 438 of 2015 (AB 1056, Atkins) provides direction to BSCC regarding the allocation of grant funds from the SNSF for recidivism reduction programs. Specifically, the legislation directs BSCC to allocate the funds through a competitive grant process to recipients with proposals that meet all of the following requirements: (1) are designed to serve people who have been arrested, charged with, or convicted of a criminal offense and have a history of mental health or substance use disorders; (2) offer mental health treatment, substance use disorder treatment, misdemeanor diversion programs, or some combination of the three; and (3) have a public agency as the lead applicant. The legislation also requires BSCC to form an executive steering committee that includes relevant stakeholders (such as representatives of state and local governments, community–based treatment providers, and formerly incarcerated individuals) to make recommendations regarding criteria for evaluating grant proposals. Under the legislation, BSCC could use up to 5 percent of the SNSF funding for administration of the grant program.

CDE to Award Grants Aimed at Reducing School Dropouts and Truancy. The administration does not propose a specific plan for how the SNSF funds would be awarded to schools. Instead, the administration indicates that it will work with the Legislature to develop a grant program consistent with the requirements of Proposition 47. The administration does propose that SNSF funding count towards the Proposition 98 funding guarantee. (Proposition 98 is the state’s constitutional school funding obligation that generally requires a certain portion of state General Fund revenues go to schools.)

VCGCB to Allocate TRC Grants. VCGCB currently awards around $2 million in grants annually to TRCs, which are programs that directly assist victims in coping with a traumatic event (such as by providing mental health care and substance use treatment). To date, six TRCs statewide have received such grants. VCGCB indicates that it plans to distribute Proposition 47 monies through their existing grant process.

LAO Comments

Allocation for BSCC Grants Consistent With Legislative Requirements. The administration proposes to allocate 65 percent of the SNSF deposit to BSCC to administer a grant program consistent with the requirements of Chapter 438. Given that this is consistent with both legislative intent and the measure, we find that this is a reasonable approach for allocating these funds.

Provide Flexibility to Schools in Using SNSF Funds, Focus on Outcomes. We recommend that the Legislature allocate SNSF funds to schools with the highest concentrations of at–risk students and then give those schools flexibility in deciding how best to address their dropout and truancy issues. To this end, instead of tracking expenditures of SNSF funds, we recommend the state rely on its newly adopted accountability system to monitor student outcomes and intervene in districts that fail to improve outcomes of at–risk students. This approach would be consistent with the state’s decision to fund schools based on their students, focusing less on spending requirements and more on student outcomes.

Provide More Guidance to VCGCB on TRC Grants. Given that the state only began funding TRCs in recent years and because of their limited number, we recommend that the Legislature provide additional guidance to VCGCB on the use of these funds. Specifically, we recommend the Legislature:

• Structure Grants to Ensure Effectiveness. We recommend that the Legislature structure the grants for TRCs to ensure that funds are spent in a manner that effectively and efficiently provides services to victims. Specifically, we believe the Legislature should consider requiring TRCs to use a “trauma–informed” approach—an approach to delivering services that takes into account the unique needs of individuals suffering a trauma (such as providing multiple services from one location in order to limit the number of times victims must retell the story of their victimization in order to receive assistance). Additionally, the Legislature could consider specifying the length of grants to ensure that new TRCs have a sufficient amount of time to get established before needing to apply for a renewal of their grant, or requiring VCGCB take such timing issues into consideration. Finally, the Legislature could prioritize which types of organizations will receive grant funds in the event that more grant applications are received than can be funded with available Proposition 47 monies. For example, establishing TRCs affiliated with trauma hospitals (as is the case with some state–funded TRCs) provides a point of access for the most severely injured crime victims.
• Ensure Receipt of Federal Reimbursement Funds. Under the federal Victims of Crime Act (VOCA) grant program, the state is eligible to receive a federal reimbursement of 60 cents for every state dollar spent on qualifying victim services. Examples of qualifying victim services include mental health counseling and medical expenses. Some of the services TRCs are likely to provide to crime victims are eligible for federal VOCA funds. If the state is able to get VOCA funds for its expenditures on TRCs, it could increase the amount of available funding. Accordingly, we recommend that the Legislature direct the VCGCB to ensure that the state receives all eligible federal VOCA funds for services provided through TRCs. For example, the Legislature could consider requiring VCGCB to collect information on eligible expenditures from grant recipients and include those amounts when applying for federal VOCA funds.
• Expand TRCs to Additional Regions of the State. Currently, six TRCs receive funding—three in the Los Angeles region, one in San Francisco, one in Stockton, and one in Fairfield. Given the potentially significant benefits of TRCs in the provision of services to victims, we recommend the Legislature expand access to TRCs by requiring VCGCB to prioritize the allocation of Proposition 47 grant funding to regions without TRCs.
• Evaluate Grant Recipients Based on Outcomes. In order to ensure that SNSF dollars are being used effectively, we recommend the Legislature require the evaluation of TRC grant recipients and the outcomes they achieve. The Legislature could specify certain basic criteria (such as the number of victims served, the types of services provided, and improvements in victims’ mental health) and require VCGCB to develop additional criteria that it deems necessary. The Legislature could also have VCGCB periodically report on the outcomes achieved and any changes made to the grant program as a result of the findings. VCGCB could use the information gathered to inform future funding decisions. This would help ensure that TRCs are delivering services to victims effectively.

Back to the Top

Conclusion

Proposition 47 delegates the calculation of savings for deposit into the SNSF to DOF. These calculations will be finalized after the Legislature adopts the 2016–17 budget. In this report, we provide a framework for how we believe these calculations should be made. We also find that the administration likely underestimates the savings and overestimates the costs resulting from the measure. Specifically, we find that the SNSF deposit in 2016–17 could be around $100 million higher than the administration’s estimate of $29.3 million. Regardless of DOF’s estimate, the Legislature could choose to appropriate additional funding to the programs funded by the SNSF if it finds that the administration’s savings estimates are too low. If it so chooses, the Legislature will want to make use of additional information on the costs and savings to the courts described in this report.