April 20, 2012

Pursuant to Elections Code Section 9005, we have reviewed the proposed constitutional and statutory initiative related to public employee retirement benefits (A.G. File No. 12-0008).

Background

Existing Public Employee Pensions. California governments currently offer comprehensive pension benefits to their employees. The benefits are funded from public employer and employee contributions, as well as investment earnings generated from those contributions. These pension benefits are provided through 85 state and local defined benefit pension plans. Some governments also contribute to retiree health and dental benefits for their former employees. The state government, for example, will contribute an estimated $1.7 billion to health and dental benefits of state and California State University (CSU) retirees in the 2011-12 fiscal year, according to the most recent actuarial estimate prepared by the State Controller’s Office (SCO).

Types of Retirement Plans. In general, California public employees are enrolled in defined benefit pension plans, which provide them with a specified benefit—generally based on their salary levels near the end of their career, their number of years of service, and the type of job they had while in public employment. Public employees typically are obligated to contribute a fixed amount—as a percentage of their pay each month during their working careers—to these plans.

Contributions from public employers and employees combined usually must equal at least the amount estimated by actuaries as the “normal cost” for plans each year. These normal costs are the amounts estimated to be necessary—combined with future investment returns—to pay for benefits earned by employees in that year. To the extent that the plans do not have enough money over time to pay for benefits, an unfunded liability results—due, for example, to lower-than-expected investment returns or decisions to give retroactive benefit increases that apply to prior years of service. In general, public employers bear all of the responsibility to pay for such unfunded liabilities. As of 2009-10, the most recent year for which data are available from SCO, public employers paid a total of about $15.5 billion to public pension systems to cover benefit costs, including several billion dollars to pay for unfunded liability costs.

Many California governments also provide their employees with options to contribute funds to defined contribution retirement plans, which are common in the private sector. Defined contribution plans do not promise a defined benefit like those described above. Instead, these plans are able to provide retirees with income generated from prior contributions plus available investment returns. Employers have no obligation to provide additional money to these defined contribution accounts to offset lower-than-expected investment returns.

In addition to defined benefit and defined contribution plans, about one-half of California public employees also are eligible to receive Social Security benefits. Teachers and many public safety workers, however, generally are not eligible for such Social Security benefits.

Contract Clause. Most public employees in California enroll in public pension programs, and they accrue certain rights to pension benefits on the day they are hired. Contracts related to pensions sometimes are included in collective bargaining agreements or in statutes, but in some cases, they may be implicit (unwritten) commitments based on a public employer’s past practices. Both the U.S. and California Constitutions contain a clause—known as the Contract Clause—that prohibit the state or its voters from impairing contractual obligations. Interpreting these Contract Clauses, California courts have ruled for many decades that vested pension benefits for current and past public employees can be reduced only in rare cases—generally, when public employers provide a benefit that is comparable and offsets the pension contract that is being impaired or when employers previously have reserved the right to modify pension arrangements.

Proposal

This measure changes certain retirement benefit provisions applicable to all public employers in the state.

Hybrid Plans Required for Future Public Employees

“Hybrid” Retirement Plans. This proposal amends the State Constitution to require each of the state’s public retirement systems to provide one or more hybrid retirement plans for employees of each public employer who are enrolled in a defined benefit pension plan. The hybrid plans are required to consist of a defined benefit component, a defined contribution or alternative plan component, and, if applicable, benefits provided under the Social Security program. The measure states that the hybrid plans are to “reduce employer risk and cost.”

Key Plan Details Would Be Set by a State Official, Subject to Limitations in the Measure. This measure requires the Director of Finance, an official in state government appointed by the Governor with the advice and consent of the State Senate, to establish “initial criteria and requirements” for hybrid pension plans on or before January 1, 2013. These initial criteria and requirements would have to adhere to several key guidelines specified in the measure, including:

The state’s Administrative Procedures Act (APA) provides for a formal rulemaking process for state departments, including opportunities for the public to provide comments on proposed rules. This measure provides that the Director of Finance’s activities undertaken with regard to the design of the hybrid plans are exempt from the APA. The measure also provides that the state generally would not have to reimburse local governments for any mandated activities, programs, or levels of service related to the measure.

Only Hybrid or Less Costly Alternative Plans Allowed for Future Public Employees. The measure requires each public retirement system to make available one or more hybrid plans to public employers and their employees beginning on July 1, 2013. Public employers would be required to offer only a hybrid plan to public employees first hired on or after that date, unless the public employer develops an alternative pension plan determined and certified by the plan’s actuary and by the system’s board “to have less risk and lower costs to the employer than any available hybrid plans” designed by the Director of Finance. Any resident or corporation paying taxes within the jurisdiction of the relevant public employer would have the right to file a lawsuit to prevent implementation or continued implementation of an alternative plan if, for example, the taxpayer could successfully challenge the determination that the alternative plan is less risky and less costly to the employer.

Hybrid Plans May Be Made Available to Existing and Prior Public Employees. The measure states that “to the extent possible while preserving the beneficial federal tax treatment of contributions,” the hybrid or alternative plans described above shall be made available to existing and prior public employees who are now members of public pension plans. These members, however, would not be required to join the hybrid plans.

To date, the federal Internal Revenue Service (IRS) has not approved continuation of beneficial federal tax treatment for some public pension plans that have sought to allow existing employees to switch to new hybrid plans. Proposed congressional legislation to allow such changes has not yet been adopted. Accordingly, it is uncertain whether or to what extent this provision would have any effect.

Limits on Future Employees’ Pensions

The measure limits retirement benefits of public employees first hired on and after January 1, 2013 (described as “future employees” in this section), as described below.

Defined Benefit Calculations to Be Based on Highest Average 36-Month Regular Pay. Currently, many, but not all, public employees in California receive pension benefits based on their highest single year of compensation. This measure instead requires defined pension benefits of future employees to be calculated based on their highest average regular payrate during at least a consecutive 36-month period of service. The measure further specifies that, for future employees, final compensation shall not include bonuses, “unplanned overtime,” or payments for unused sick leave or vacation.

Minimum Service Retirement Ages Increased. Currently, most public employees are able to apply for retirement based on their years of service beginning at age 50—typically with a lower benefit than they might be able to receive by retiring later. This measure would allow future public safety employees to apply for service retirement no earlier than age 52 after five years of service and all other future employees to apply for service retirement no earlier than age 57 after five years of service. If minimum age requirements of the federal Social Security Act are subsequently increased, these age requirements would be increased by an equal number of years for any new public employee hired after the operative date of the federal law change.

Limits on All Public Employees’ Pensions

The measure also contains certain new limits on retirement benefits of all public employees—both future employees and those who are already members of California’s public retirement systems. These limits, as described below, are to apply only to the fullest extent permissible under the U.S. Constitution. (This means that if it is determined that the U.S. Constitution’s Contract Clause prevents implementation of any of the limits described in this section of the letter, they would have no force or effect.)

Moreover, if a labor agreement between public employers and employees that is in effect on November 7, 2012 would otherwise prohibit any of these limits, the limit would not apply to the public employer and employees subject to that agreement until the expiration date of the agreement in question. In some cases, this particular provision could delay application of one or more of these limits to certain public employers and employees for a few years after this measure takes effect.

Retroactive Benefit Increases Prohibited in the Future. This measure prevents future retroactive pension benefit increases—that is, benefit increases adopted in the future that would be applied to an employee’s prior years of service.

Contributions From Both Employers and Employees Required. This measure requires both public employers and employees to contribute payments to fund a defined benefit pension’s normal cost. Specifically, public employees would be required to contribute at least 50 percent of the actuarially determined normal costs each year. Public employers, some of whom currently pay all or a part of employees’ required contributions on their behalf, would be prohibited from doing so. (The statutory components of this measure specify that this change would be implemented for current public employees only if allowed under both the U.S. Constitution and the State Constitution.)

“Airtime” Purchases Prohibited. Using funds to purchase service credit (referred to as airtime purchases) generally would be prohibited in the future.

Limits Ability of Retired Annuitants to Work for Public Entities. Currently, public employers hire retired annuitants—past full-time public employees already receiving pension benefits. Often, public employers pay the retired annuitant a part-time salary—generally with few, if any, benefits—while they continue to draw pension benefits and, in some cases, retiree health benefits. Some public retirement systems limit how much a retired annuitant can work and still draw pension benefits. This measure applies such limits to all public employers in the future. Specifically, retired annuitants’ service could not exceed a total of 960 hours or 120 full-time days in a consecutive 12-month period in that public retirement system. Moreover, retired employees serving on public boards or commissions would not be able to earn any retirement benefits for that service unless he or she “reinstates from retirement” (that is, suspends their receipts of pension benefits during that period).

Felonies in Official Duties Would Result in Benefit Forfeiture. The measure requires public employees convicted of state or federal felonies for conduct related to their official duties to forfeit their retirement benefits.

State Retiree Health Benefits

Reduction in State Contribution to Future Employees’ Retiree Benefits. For state and CSU employees hired after the effective date of this measure, the measure limits future retiree health contributions by the state. Currently, the state subsidizes its retirees’ health premiums in an amount up to 100 percent of the health premiums currently attributable to state employees in their health plans. Under this measure, for future state and CSU employees, the state’s maximum contribution to their retiree health benefits would be equal to the last three-year average of the premiums the state paid for their benefit when they were an active employee. For most current state employees, the state pays around 80 percent of health premiums. Accordingly, the maximum state retiree health contribution for future employees would be reduced. In addition, future state and CSU employees would be required to work longer to receive the maximum state contribution. Under this measure, 50 percent of the maximum state retiree health contribution would be payable to future employees who retired after 15 years of service. This payment would grow somewhat for each year of service after 15 years until reaching 100 percent of the maximum state retiree health contribution after 25 or more years of service. Other limits also would apply to state retiree health contributions for future employees.

Public Retirement System Boards

Changes to Composition of Retirement System Boards. The State Constitution provides that retirement system boards have broad authority to administer their pension systems and oversee their actuarial analyses. The board of the largest such public retirement system in California, the California Public Employees’ Retirement System (CalPERS), would be changed by this measure. Several existing members of the CalPERS board (an appointee of the State Personnel Board, an official of a life insurer, and one public representative) would be removed from the board. In their place there would be the following new members: the Director of Finance, a gubernatorial appointee with expertise in health insurance who does not have a direct or immediate familial financial interest in a public pension or retirement system, a gubernatorial appointee who is an elected official of a public agency that contracts with CalPERS for pension benefits, and two other gubernatorial appointees to represent the public who have financial expertise and who do not have a financial interest in a public pension or retirement system.

Fiscal Effects

Very Difficult to Determine Fiscal Effects for a Number of Reasons. The fiscal effects of this measure are difficult to determine for many reasons, including the following:

Pension System Analyses of Long-Term Fiscal Effects

Staff of the two largest public retirement systems in the state, CalPERS and the California State Teachers’ Retirement System (CalSTRS), recently provided to a legislative conference committee an analysis of a proposal that matches some, but not all, aspects of this proposed initiative. Analyses by pension system actuaries are significant because these systems are the primary source of detailed data about their members and system financial characteristics. Moreover, under the State Constitution, pension system boards have very broad authority to set the assumptions related to their actuarial analyses. These analyses by CalPERS and CalSTRS essentially focused on the long-term potential savings from the plan—that is, the savings (principally in normal costs) in public employer pension contributions once all public employees are enrolled in this measure’s hybrid plans and covered by other provisions of this measure a few decades from now.

CalPERS Analysis of State and School Employer Fiscal Effects. The proposed initiative states the hybrid pension plan shall reduce employer and taxpayer risk and cost. On February 14, 2012, CalPERS staff forwarded to a legislative conference committee the analysis referenced above based on certain assumptions, including parameters provided to the system by the committee’s staff. This analysis assumed, among other things, that the CalPERS defined benefit pension system earns a 7.75 percent future average annual return, that the hybrid plans’ defined contribution plans earn an average 6.75 percent per year during employees’ careers, and that future employees use their defined contribution balances at retirement to purchase an annuity from a private insurance company at a 4.5 percent interest rate. The CalPERS analysis found that the expected state government savings would generally be “not significant” and that, for several groups of state employees, cost increases would “largely offset cost savings in other plans.” Specifically, given the various assumptions incorporated into their analysis, CalPERS found that the state’s correctional officer pension costs would increase by 2.1 percent of payroll, firefighter pension costs would decrease by 0.7 percent of payroll, Highway Patrol officer pension costs would increase by 0.5 percent of payroll, and state miscellaneous employee (generally, non-safety employees) pension costs would decline by 0.6 percent of payroll. It appears that these results, combined, would represent little or no net savings for the state related to state and CSU employees.

The CalPERS analysis indicated that school district pension costs for their classified (non-instructional) employees eventually would decline by 2 percent of payroll, which currently would be about $225 million per year.

While the CalPERS analysis makes an assumption about future investment returns, as described above, it also notes that hybrid plans would tend to reduce substantially the risk of future employer contribution rate volatility. This is because a hybrid plan would switch some of the risk of investment losses currently borne by public employers to pay more if investment returns are weaker than expected to public employees themselves. For example, CalPERS found that the hybrid plans could decrease the potential risk of future unfunded liabilities by 31 percent to 45 percent for state plans. At the same time, the CalPERS analysis noted that lowering such investment risks for employers in a hybrid plan does not necessarily mean lowering the overall cost of providing a given retirement income level for future retirees. In effect, the CalPERS analysis assumes a lower overall investment return and/or higher administrative costs for the defined contribution plans.

Local Government Fiscal Effects. With regard to local governments enrolled in CalPERS’ pension programs (including many cities and special districts), CalPERS indicated that cost savings under the proposal were not “easily quantified,” given the differing benefit formulas in place for these local governments and the fact that some have already adopted lower levels of benefits for future employees. Despite the difficulty in preparing such an analysis, CalPERS indicated it expected savings for both local miscellaneous and local safety employees to be greater than the state government’s savings for similar employee groups. The CalPERS analysis stated that this opinion resulted from the fact that local employees in that pension system generally were subject “to the same or better retirement benefits” than state employees and that “they contribute on average less” to their pensions than state workers.

In addition to the benefits provided through CalPERS, county retirement systems and several cities’ retirement systems also provide pension benefits. The provisions of this measure that would generally base future employees’ pension benefits to be calculated on the basis of regular payrates—excluding other forms of cash compensation—would overturn past court decisions applicable to the state’s 20 county retirement systems that now require these systems to consider some non-salary compensation items in such calculations, such as certain lump-sum payments for sick leave and vacation leave. For this reason and others, we expect that these other local systems would experience greater savings (as a percent of payroll) than the state government as a result of this measure. Currently, cities, counties, and special districts contribute over $4 billion per year to all public pension systems. Accordingly, local government cost savings of hundreds of millions of dollars (current dollars) per year seem possible a few decades from now.

CalSTRS Analysis for Teachers and School Administrators. CalSTRS recently reviewed the fiscal effects of a similar proposal. The system’s actuaries found that public entities—including schools, community colleges, and perhaps the state (which contributes to CalSTRS, along with school employers)—would, in the long run, experience reduced costs of over $800 million per year (current dollars). Administrative cost increases, while not quantified in the CalSTRS analysis, also likely “would be significant,” the system stated.

Assumptions Are Key to These Analyses. As the CalPERS and other such analyses note, the assumptions used in developing these estimates have a material effect on the outcome. Many such assumptions will prove to be incorrect over time, as is always the case with pension actuarial analyses. It is important to note that the CalPERS analysis described above incorporated several assumptions that may minimize state savings, compared to those that actually could be achieved if this proposal is adopted. It appears, for example, that some groups of future state employees are assumed to contribute less to their own pensions in the CalPERS analysis than those groups already do. In addition, while purchases of annuities by future hybrid plan participants could provide them with more certainty about retirement income, this assumption also could increase the estimated public employer and employee costs for a given retirement income package above what it would be without this assumption. Such purchases of annuities are not required or referenced in this proposal. The CalPERS analysis appears to assume that contributions to the hybrid plans’ defined contribution element are split evenly between future employees and public employers, while the measure would allow plan designs that permit less than 50 percent of these defined contribution payments to be made by the employer. The CalPERS analysis also did not discuss in detail the manner in which it incorporates elements of this proposal that generally limit future employees’ benefits to being calculated solely on regular pay, which may reduce public employer costs to some extent.

Conversely, other assumptions in the CalPERS analysis could understate costs of the proposal. The CalPERS analysis, for example, assumes a 7.75 percent defined benefit program investment return (which subsequently has been lowered by CalPERS for its pension plans to 7.5 percent) and a 6.75 percent defined contribution investment return. Actual investment returns could be lower than this (increasing such costs) or higher than this (reducing employer and employee costs). Moreover, the CalPERS analysis states that it includes no additional costs resulting from a likely increase in the incidence of disability retirement that could result from adoption of this proposal.

In prior initiative analyses, we have observed that the creation of hybrid pension plans for future employees could result in declines in income to defined benefit pension trusts and result in the need for changes in the trusts’ asset allocations. Depending on how the hybrid plans are designed and implemented, they could lead to reductions in future investment returns and resulting increases in public employer costs—particularly in the short term and medium term (the next few decades). The CalPERS analysis described above notes the possibility of such increased costs in some instances, but that analysis does not reflect any such increased costs in its quantitative estimates. Even minor changes in investment assumptions could lead to substantially increased public employer costs—perhaps totaling hundreds of millions or a few billion dollars per year—in current dollars. Though such changes in investment strategy could increase short-term and medium-term budgetary costs for governments, this could reduce the risk of future unfunded liabilities, which could result in substantial public employer savings in the long run.

The pension systems’ analyses also do not reflect the potential costs if public employers choose to increase other compensation items (such as salaries or other benefits) to offset reduced or changed pension levels for current and future public employees. Some, but not all, public employers likely would choose to do this in order to remain competitive in the labor market.

Short-Term and Medium-Term Fiscal Effects

Savings Likely Less Than They Will Be Eventually. As noted above, the full long-term fiscal effects of the measure described above would not materialize until all public employees are enrolled in the hybrid plans and covered by other provisions in this measure a few decades from now. Some such savings could begin immediately—for example, for current public employees for whom higher pension contributions are implemented pursuant to this measure. In the next few decades (the short term and the medium term), however, potential cost savings from the hybrid plans, the changes in employee pension contributions, and the other provisions of this measure would be smaller than they will be in the long term. In total, during the next few decades, potential retirement benefit cost savings for public employers could total in the hundreds of millions of dollars or more per year.

Additional Costs Possible. In addition to the short-term and medium-term savings noted above, additional public employer costs also are possible beginning in the near term for some, but not all, public employers. For example, additional, offsetting compensation increases for some public employees could occur. In addition, potential alterations by some retirement systems to their asset allocations due to this measure’s changes could result in lower actuarially assumed investment return rates in the short term and the medium term. Other additional costs could result, depending on how this measure is implemented and administered. It is unknown whether all of these potential added costs will be more or less than the short-term and medium-term savings described above for any given public employer.

Conclusion

In the next few decades, state and local government retirement benefit costs could decline by hundreds of millions of dollars or more per year (in current dollars) as a result of this measure. These short-term and medium-term savings, however, could be partially or entirely offset by a variety of factors discussed above, with the net savings or costs potentially varying from one public employer to another depending on how this measure is interpreted and administered, among other factors.

Over the long run (a few decades from now), once this measure’s hybrid plan and other benefit limits apply to all public employees, state and local government retirement benefit cost savings of a few billion dollars per year (in current dollars) seem possible, depending on how the measure is administered by public employers, courts, and other entities. These savings include hundreds of millions of dollars (in current dollars) of potential state retiree health and dental benefit savings and potential reductions in governments’ payments for future unfunded pension liabilities. Other cost savings are possible but impossible to quantify, including reductions in hypothetical future cost increases for retroactively applied benefits. Potential increases in other compensation paid to public employees to make up for the reduced pension benefits included in this measure could offset these various categories of savings to some extent.

Summary of Fiscal Effects

This measure would result in the following major fiscal effects for state and local governments:


Return to Propositions
Return to Legislative Analyst's Office Home Page