|Budget Issue:||Amendments needed for pension reporting bill that was passed as part of the 2010 budget package|
|Finding or Recommendation:||Amend Chapter 733, Statutes of 2010 (SB 867, Hollingsworth), to make pension reporting requirements of CalPERS and the State Treasurer more useful and workable. Require reporting based on more reasonable alternate investment return rates for the pension system. Focus reporting requirements more on state plans, instead of CalPERS' hundreds of local plans. Require an official other than the Treasurer to provide the independent analysis to the Legislature required under the bill.|
SB 867 of 2010 Was Well-Intentioned Pension Transparency Bill. As part of the 2010-11 budget package, the Legislature passed Chapter 733, Statutes of 2010 (SB 867, Hollingsworth), an urgency measure advocated by Governor Schwarzenegger that instituted new requirements for (1) the California Public Employees' Retirement System (CalPERS) to provide more information to elected policy makers concerning the system's actuarial assumptions for future investment returns, payments of unfunded pension liabilities, and estimates of future public employer pension costs in the event the investment return assumptions are not met and (2) the State Treasurer to address a publicly noticed floor session of the Legislature and give his or her perspectives each time that CalPERS changes public employer contribution rates.
The key goal of SB 867 was to promote heightened awareness among elected state policy makers, the CalPERS board, and the public of the possible future budgetary impacts from state and local pension liabilities. Moreover, there has been much discussion recently concerning the possibility that future investment return assumptions of public retirement systems, including CalPERS, are far too optimistic. If such assumptions are, in fact, too optimistic, this means that current taxpayers are not paying enough into the pension fund now, thereby deferring costs to cover current and past employee benefits to future generations. The measure aims to encourage scrutiny of CalPERS’ future investment return assumptions. We believe that all of these are worthy goals.
Bill Contained Serious Drafting Problems. During floor debate on SB 867, various problems with the bill were acknowledged by some legislators. Subsequently, staff of CalPERS also have noted various unworkable and/or unintended consequences of the measure, as it was drafted last year. For example, the new law requires reporting by CalPERS and a public address by the Treasurer to the Legislature each time that the system sets employer contribution rates for each of its state, school, and local agency plans (which are accounted for separately by CalPERS). With over 2,000 such plans, this requirement could be interpreted to greatly expand CalPERS' actuarial workload and require the Treasurer to appear numerous times before the Legislature each year. We doubt that CalPERS' existing staff resources are sufficient to handle this increased workload.
Moreover, detailed reporting on each local plan would be a poor use of time and resources by the Legislature and the Treasurer, inasmuch as the state government does not have direct financial responsibility for local plans in CalPERS. As an alternative, we suggest that the Legislature amend the measure, as detailed below, to require one annual report by CalPERS summarizing various state and local pension issues and one address to the Legislature (or legislative committees) every two years by the Treasurer or an alternate independent official concerning these matters.
Bill Requires CalPERS to Adopt Alternate Reporting Method. As part of its efforts to encourage scrutiny of CalPERS’ investment return assumptions, SB 867 requires CalPERS to calculate pension liabilities in its reports to the Legislature and others using “a discount rate equal to the rate of the 10-year United States Treasury (UST) Note as of 30 days before the date of the report.” This means that instead of calculating liabilities using CalPERS’ annual assumed investment return (currently 7.75 percent), this report would require liability reporting assuming a much lower discount rate. As of January 21, 2010, the 10-year UST yield is 3.4 percent. Using such a lower discount rate would result in CalPERS calculating a much higher amount of liabilities and, therefore, future state and local costs, compared to standard public pension reporting methods. (A group of Stanford University students recently calculated CalPERS and other public pension liabilities using a similar method as envisioned in SB 867. This Stanford report indicated that CalPERS’ state and local unfunded pension liabilities totaled about $240 billion as of 2008--substantially more than the amount then reported by CalPERS.)
Rationale for This Reporting Method. Currently, there is significant debate in the pension policy, accounting, and actuarial arenas concerning the appropriate method of discounting of pension system liabilities. Rules for public sector accounting for pension plans generally are established by Governmental Accounting Standards Board (GASB) Statements 25 and 27. Under GASB 25, pension discount rates are based on an estimated long-term yield for the pension plan, with consideration given to the nature and mix of current and planned investments. Currently, CalPERS estimates that this long-term yield is about 7.75 percent, on average, per year. By contrast, some—such as the authors of the Stanford study discussed above—believe that guaranteed public sector pension liabilities should be discounted at a “riskless rate” of investment return, such as the 10-year UST Note rate cited in SB 867. The riskless rate, some say, is more appropriate since current law provides almost no way—outside of a local entity's filing for bankruptcy—that public entities can avoid paying the full employer share of their accrued defined benefit pension liabilities.
LAO View: Information May Be More Alarmist Than Useful. While we view the accounting discussions of the riskless return rate as an interesting hypothetical debate, we conclude that it is not reasonable to assume, in effect, that CalPERS will earn an average annual investment return of only around 3 percent or 4 percent over the long term. (The average annual return over the last 20 years, for instance, has approached 8 percent.) Accordingly, the reporting required by SB 867 seems more alarmist than useful, in our view, for policy makers. In other words, reporting based on the riskless rate of investment return probably represents a “worse-than-worst-case” scenario for future funding health of CalPERS.
Conflict of Interest. The State Treasurer is a member of the CalPERS board. Under SB 867, however, he or she also is required to report annually to a public session of the Legislature on “the reasonableness of the board’s selection of the investment return assumption” and other matters. This creates a potentially inescapable conflict of interest. This has nothing to do with the qualities or opinions of the current Treasurer or any future Treasurer. Rather, it has to do with his or her role on the CalPERS board. If the Treasurer votes for a pension valuation as a member of the CalPERS board, with all of his or her fiduciary duties related to that board, he or she may have no option but to hail the merits of the CalPERS action in the required annual addresses to the Legislature. In our view, the Treasurer cannot credibly vote for a valuation and rate plan in the CalPERS board room and later provide a completely independent critique of that plan to the Legislature. Under the Constitution, the Treasurer, as a member of the CalPERS board, must ensure that his or her duty to system participants “shall take precedence over any other duty,” including state budgetary considerations. The Treasurer cannot be a truly independent source of opinion for the Legislature, as envisioned by SB 867.
We agree that there is value in the reporting requirements envisioned by SB 867. The measure, however, contains serious drafting problems. We recommend that the law be amended in the following ways: