Two actuarial associations did not publish a controversial paper by their joint task force, reflecting a split in the profession over whether public pension debt should be measured with risk-free bonds or the earnings forecast for stock-laden investment funds.
Using safe but low-yield bonds to offset or “discount” future pension obligations would cause pension debt to soar, creating pressure to raise the annual rates paid by state and local governments that are already at an all-time high for many.
Critics have been contending for a decade that overly optimistic pension fund earnings forecasts conceal massive debt and the need to take even more money from government budgets or find a way to cut pension costs.
The leading California critics, now mainly at Stanford University, are not professional actuaries. They have backgrounds in finance, like David Crane, and in academic economics, like Joe Nation and Joshua Rauh.
The paper of the joint Pension Finance Task Force paper of the American Academy of Actuaries and the Society of Actuaries, which did not make it through the usual peer review process, was based on the principles of financial economics.
“One of the assertions of the paper is that public pension plans are purported to be default-free obligations so they would be valued using default-free interest rates,” an anonymous former task force member told Pensions & Investments.
Although not as well publicized as criticism from outside the profession, a group of actuaries has been urging the adoption of a risk-free discount rate for about a decade, said Paul Angelo of Segal Company actuaries in San Francisco.
Angelo, chairman of the California Actuarial Advisory Panel, does not favor the use of a risk-free discount rate. He agrees with not publishing the task force paper, saying it lacked the “science” to support the change and relied only on assertions.
For the first time, Angelo said, the actuaries urging a risk-free discount rate went beyond simply reporting debt and seemed to be advocating its use to set the annual payments to the pension fund made by government employers.
The California Public Employees Retirement System and the California State Teachers Retirement System currently assume their pension fund investments, expected to pay two-thirds of future pensions, will average 7.5 percent a year in future decades.
The systems use the 7.5 percent long-term earnings forecast to reduce or “discount” the cost of future pensions, as if it were money in the bank. Thirty-year U.S. Treasury bonds, regarded as risk free, were yielding 2.23 percent last week.
When a much lower rate is used to discount future pension obligations the pension debt or “unfunded liability,” the shortfall in the projected money available to pay future pensions, balloons to a much larger amount.
An example is shown on the “California Pension Tracker” website directed by Joe Nation at the Stanford Institute for Policy Research.
The debt of California public pension systems in fiscal 2013 using a 7.5 percent discount rate is $281.5 billion. Using a lower discount rate of 3.723 percent (the CalPERS rate in 2013 for terminating plans) the pension debt more than triples to $946.4 billion.
California Pension Tracker
The giant California Public Employees Retirement System, covering half of all non-federal government employees in California, is deep in debt even when using its 7.5 percent discount rate.
CalPERS has not recovered from a $100 billion investment loss during the financial crisis. Its investment fund was $260 billion in 2007, dropped to about $160 billion in March 2009 and was $306 billion last week.
In 2007, CalPERS had 101 percent of the projected assets needed to pay future pensions. Now after weak earnings during the last two fiscal years, its funding level is lower than the outdated 75 percent reported in a previous Calpensions post.
A CalPERS spokesman said the funding level for the last fiscal year ending June 30 is an estimated 68 percent. Nearly a decade later, that’s not much higher than the CalPERS funding level of 61 percent in 2009 at the bottom of the financial crisis.
The reassurance that CalPERS long-term earnings average more than 7.5 percent has eroded, dropping to 7.03 percent for the last 20 years. And the outlook is dim: The 10-year earnings forecast from Wilshire and other CalPERS consultants is 6.64 percent.
Girard Miller is probably not rethinking a line from his widely circulated Governing magazine column in 2012 debunking a dozen “half-truths and myths” used by both sides in the public pension debate:
“Pension funds are not going to invest their entire portfolio in 3 percent Treasury bonds right now — or ever — so the risk-free model is not even descriptive of reality and has little normative value.”
The current Economist magazine says (“No love, actuary” Aug. 13-19 issue) it has seen a draft of the joint pension task force report and thinks the two actuary associations should allow the paper to be published.
“American public-sector deficits are more than $1 trillion, even on the most generous of assumptions,” said the magazine. “This is an issue in which debate should not be stifled.”
Some of the debate was aired when the Governmental Accounting Standards Board spent several years developing new rules that took effect in 2013 and 2014. Pension systems can use their earnings forecasts to discount future pension obligations.
But if the projected assets fall short of covering the pension obligations, the system must “crossover” and use a risk-free rate to discount the remainder of the pension obligation. CalPERS did not have to crossover.
The new accounting rules require pension systems to use the “blended” rate to report pension debt. But they can continue to use the previous method based only on their earnings forecast to set the annual rates paid by employers.
The California Actuarial Advisory Panel agreed with the blended rate and suggested a few tweaks in a letter to GASB on Sept. 17, 2010, from the chairman at the time, Alan Milligan, CalPERS chief actuary, who is retiring this year.
An independent “Blue Ribbon Panel” commissioned by the Society of Actuaries issued a report in 2014 that, among other things, endorsed the use of some version of a risk-free discount rate.
“The Panel believes that the rate of return assumption should be based primarily on the current risk-free rate plus explicit risk premia or on other similar forward-looking techniques,” said the panel report.
Angelo said the influential Actuarial Standards Board, which was essentially neutral in Actuarial Standards of Practice issued in 2013, may revisit the risk-free discount rate issue, possibly within a year or so.
Reporter Ed Mendel covered the Capitol in Sacramento for nearly three decades, most recently for the San Diego Union-Tribune. More stories are at Calpensions.com. Posted 15 Aug 16
Calpensions 
August 15, 2016 at 1:02 pm
Pension obligations are certainly not risk free. Someone bears the risk that investments will not perform as hoped or that costs will be higher than planned. Why not share that risk between future tax payers and pensioners? Why must tax payers bear it all? However, given that public employee unions seem to have taken over, especially in California and other “progressive” states, the chances of risk sharing ever being adopted seem nil. Thus, accounting for future costs/earnings should be done based on the most conservative actuarial assumptions, including a discount rate that is as free of risk as possible. The alternative is to place huge risks on future generations of tax payers. How can anyone justify that?
August 15, 2016 at 1:18 pm
A rate based on historical averages would actually be as low as the risk-free rate — if adjusted for expected inflation and inflated asset values.
If the actuaries were honest, in the past every time asset values bubbled up (2000, 2007, now) the expected rate of future return would have been cut down. And the future rate of return would have been increased following asset price crashes.
The idea that pensions were “fully funded” in 2000, so politicians could cut pension funding and hand out retroactive pension increases, was a lie. That was a temporary bubble. So is this.
https://larrylittlefield.wordpress.com/2013/11/29/pensions-the-nature-of-the-lie/
August 15, 2016 at 2:32 pm
This is a very balanced piece about investment and discount rates,i.,e, the annual amount of money that must be set aside, and assuming a certain rate of return will provide the funds to pay the retirements w/o creating deficits.
The hard fact is that if employees are promised unachievable retirements in the first place(like Sonoma Co. with its 90% of final salary at some age for all), it doesn’t make any difference at that level of benefits, whether a 3.5% or a 7.5% investment rate is used, the benefit is unaffordable either way. With the 3.5% rate, the annual contributions would bust all budgets and at 7.5% the deficits will become so costly that the pension systems will fail.
This investment rate issue is distracting from the problem, namely pension levels are too high at any rate level. Crybaby reformers ignore true reform because it will take guts in the form of political will.
August 15, 2016 at 3:08 pm
Paul Angelo admitted in a meeting of the Contra Costa County Employees Retirement Association that I attended that his firm recommended rates based on hiring/firing ability of the client (in this case the retiredment association. In a March 4, 2013 Editorial in the Contra Costa Times (Dan Borenstein had to have written it since he was there.) The Editorial was titled “Lower Pension investment rate still not enough.
Here is the quote from the editorial:
“The pension system’s actuary, Paul Angelo, who works under contract, didn’t help matters. He recommended lowering the rate to 7.25, but repeatedly said the 7.50 percent would be acceptable—even though his calculations showed the rate should be lowered to 7.09 percent.
As he explained his methodology, it became clear that the rate should be even lower. But if his firm recommended even the 7.09 percent “that would be our last day here because that is a big drop.” The supposedly independent actuary was making his own political calculation.
So, he explained, he rounded his recommendation up to 7.25 percent, leaving some board members scratching their heads. As Trustee Debora Allen correctly pointed out, if Angelo wanted to round to an even quarter point, the recommendation should have been 7 percent.
Allen voted for the 7.25 percent rate, but wanted it lower. As she said, “We should start living in the real world and facing the reality we have before us.”
August 15, 2016 at 6:21 pm
In setting the new procedures, the GABS caved-in* to the wishes of the WRONG stakeholders on the current structure …. the Public Sector workers/Unions, the Gov’t Legislature bought with Union campaign contributions and election support Union, and the myriad of “consultants” that earn high fees for selling their sole.
Their fear ?
If these Plans were PROPERLY valued, the enormity of the true cost would have Taxpayers DEMAND immediate and material reductions in the pension accrual rate (and increases in the unreduced retirement age) for the future service of all CURRENT workers.
None of these stakeholders wanted to tough that with a 10 foot pole.
The Taxpayers are eminently justified in reneging on all of these unnecessary, unjust, unfair, and clearly unaffordable pension/benefit “promises”.
————————————————————
* GABS apparently spend a great deal of time finding the change with the LEAST possible impact on these pensions. How absurd a choice to only require a crossover from the 7.5% to a risk-free rate only if and when plan assets hit zero WHEN THAT CALCULATION is done assuming that the plan will indeed earn the 7.5% rate. How can this be justified? In the very least, the risk-free rate should be used from day 1 (not starting from the crossover duration) for liabilities that exceed current assets.
And assuredly the PROPER (but far more costly) choice would be to use the SAME discount rate that the US Gov’t REQUIRES of Private Sector Plans in their valuations …. a long-term high-grade Corporate bond rate (of about 4%).
August 15, 2016 at 6:26 pm
@MG: Reminder – Pensioners are taxpayers! Stop trying to divide and conquer.
August 15, 2016 at 7:06 pm
Quoting John M. Moore above ….. “This investment rate issue is distracting from the problem, namely pension levels are too high at any rate level.”
John hits the nail on the head, and this is exactly what I have been saying for a decade in commenting on Public Sector pensions ….
The ROOT CAUSE of the problem is grossly excessive public Sector pension (AND benefit) “promises”. When properly factoring in not just the MUCH greater Public Sector DB pension per-year-of-service “formula-factors”, the MUCH younger full/unreduced retirement ages, and the existence of COLA-increases (all but unheard of in Private Sector Plans), Public Sector pensions are ROUTINELY 2 to 4 times (4 to 6 times for safety workers) greater in “value at retirement” than those granted Private Sector workers who retire at the SAME age, with the SAME pay, and with the SAME years of service, And that’s just the Plan’s “Normal Cost”.
AND ……… if you accumulate all of the worker’s own pension contributions (WITH investment earnings thereon), the sum accumulated at retirement would be sufficient to buy NO MORE than 10% to 20% of their incredibly generous pensions. Taxpayer contributions (and the investment earnings thereon …… earnings that in the absence of these absurdly generous pension promises would have stayed in THEIR pockets, perhaps to help fund their far smaller retirements) are supposedly on-the-hook for the 80% to 90% balance.
So what do we have ……. absurdly generous pensions with the Taxpayers responsible for 80-90% of total plan costs. How is this even remotely justifiable?
——————————————————
Those who oppose Public Sector pension reform … the workers, the Unions, the consultants, and our BOUGHT-OFF Elected Officials ….. like to say that the CAUSE of the pension mess we are in today is the lack of full funding, obviously ignoring the 1 to 1 relationship between the Plan’s generosity and the resultant calculation of the required funding contribution. Of course that’s false. The lack of full funding is not the CAUSE of the problem, but the CONSEQUENCE of the true ROOT CAUSE …. grossly excessive Pension “generosity”.
August 15, 2016 at 8:46 pm
Another misplaced emphasis is about the size of retirement benefits and not the unachievable cost of the benefits. Marin’s County Counsel now receives about $160,000 a year in retirement benefits(plus about $200,000 in salary) but the cost to his previous employers(Santa Clara and Sonoma county) for his annuities are huge because of deficits and POB for his era of employment.That $160,000 a year may cost those two employers that much because of unfunded deficits that compound at 7.5% a year and POB at about 5% a year.
The trend is for annual contributions to equal a pay as you go system as deficits build. Seeling was right on, only it is worse that he predicted. Scary, but probable in this “me me” era.
August 15, 2016 at 10:49 pm
John Moore, the “Size” and the “Cost” of pension benefits are directly related and move in tandem, so addressing the “size” clearly also speaks to the “cost”.
However, you do have a point. While my accurately stating (above) ………… “Public Sector pensions are ROUTINELY 2 to 4 times (4 to 6 times for safety workers) greater in “value at retirement” than those granted Private Sector workers who retire at the SAME age, with the SAME pay, and with the SAME years of service” ……….. speaks to the grossly excessive “size” of these pensions, I ALSO addressed the “cost” indirectly by pointing out that it’s the Taxpayers that are on-the-hook for all but the 10% to 20% of Total Plan costs typically funded by the worker’s own contributions (again, see my above comments).
However, perhaps more IS need to educate the Taxpayers on the “cost” side (as you suggest). Well, here’s another way to look at the extraordinary “cost” of Public Sector pensions …….. while most Private Sector workers today get little more in employer-sponsored retirement security than a 3%-of-pay employer “match” into a 401K Plan, the TRUE %-of-pay TOTAL COST to fully fund the TYPICAL full-career Public Sector pension (using the SAME assumptions that the US Gov’t REQUIRES of Private Sector Corporations in their pension valuations) ranges from 25%-45% of pay for non-Safety workers and from 40% to 60% of pay for Safety workers. If we use the mid-point of these ranges (35%-of-pay for non-Safety workers and 50% of pay for Safety workers) and subtract the workers own contributions (assuming 5% of pay for non-safety workers and 10% of pay for Safety workers) we are left with TAXPAYER contribution requirements of 30%-of-pay for non-Safety workers and 40% of pay for Safety workers.
BOTH of these %s are more than 10 TIMES what Private Sector employers are willing to contribute towards their workers retirement needs….. that 3% 401k “match”.
It would take an extraordinary demonstration of far far lower Public Sector “cash pay” to justify such HUGE pension contribution requirements………. demonstrations that clearly do NOT exist.
And, Taxpayers must ALSO factor in the HUGE cost of Public Sector retiree healthcare “promises”……. employer-sponsored retiree healthcare benefits being an EXTREMELY rare benefit offered by Private Sector Corporations today. Several studies have suggested a level annual %-of-pay cost of 10% to 12% to fund typical Public Sector retiree healthcare “promises”.
August 16, 2016 at 12:19 am
Oh, See Saw, you’re being silly. All pensioners are taxpayers, but not all taxpayers are pensioners. In fact, non-pensioner taxpayers are by far the majority. Also, we’re not talking about todays employees and today’s taxpayers. We’re talking about tomorrow’s pensioners whose risks are being foisted on future taxpayers.
August 16, 2016 at 3:44 am
Cry me a river. All taxpayers receive services which are provided by future pensioners. Those services come with a fee.
August 16, 2016 at 10:47 am
The problem is that the real cost of those “services” are not currently being revealed. By understating the real cost of pensions and pushing the costs of today far into the future means bad decisions are being made. Cities could spend differently if they knew the actual cost of a firefighter when accurate pension and retiree healthcare costs are included. Also, the people receiving the services today should be paying for them and not passing them on to the children of tomorrow.
August 17, 2016 at 3:05 pm
Recently, the Cook County [Illinois] Pension Plan restated their unfunded liability from about $6 billion to about $15 billion supposedly due to the new GASB pronouncements. As an Appointed Trustee of a far-western suburb of Chicago I have become aware of an issue related to the Illinois Pension Code or IPC that I think has some relevance. The issue involves the added investment utility to a police/fire pension plan of the conservative utility of the sale [or the synonymous term “write”] of listed, exchange traded options or LETOs, by said plans in conservative transactions known as covered calls. We advocate only the conservative uses of LETOs, including this suggestion. This issue may extend to other municipal plans.
The IPC is bifurcated, for municipalities with populations greater than 500,000, i.e., Chicago and for all other entities, i.e., “Downstate”. Section 113.1 through 113.4(a) of the Illinois Pension Code or IPC governs Downstate Police/Fire Pension funds. Section 113 applies to Chicago funds, where at 40 ILCS 5/1-113 [Sec. 1-113] it partly stipulates at item [10] the…
“Trading, purchase or sale of listed options on underlying securities owned by the Board.”
The above stipulation is unavailable in the IPC Downstate provisions. As you know, the conservative use of LETOs may provide added portfolio income to Plans otherwise potentially provided by the General Funds of their respective communities via a required annual levy. The State of Illinois, in population size, has established an imprudent investment standard for downstate police pension plans.
August 31, 2016 at 6:08 pm
The majority of public pensions are assuming the 7%+- rate of assumed rate of return. Illinois in its TRS plan recently reduced is assumption from 7.5% to 7%. Sounds reasonable until you look at their funds historical financial statements and see that they have only earned barely 6% over the last 10 years and that period was not even equivalent to the low to zero rate environment of today. Actuarial firms get paid by the funds to perform their educated guesses and will not anger the hand that feeds them so will do little to rectify the situation. The federal government will not mandate change since they are in the same pickle although they exempt themselves from most of the pension rules anyway.
In Illinois, another 10 years or less, they will show us what happens when excessive assumptions and gimmicks regarding “smoothing” reduce the cash flow to the point of not having funds for benefits.
California is currently better off than Illinois at least.
September 1, 2016 at 1:43 pm
As an Appointed Trustee of a suburban police pension board, I quickly came to the conclusion that the combination of current monetary policy and the Illinois Pension Code’s requirement for a 50/50 fixed income – equity mix would not produce the 7% annual return required to meet the plans obligations. Simple math would suggest that if one were to get 3% from FI over time [a generous assumption?] then one would also need to return 11% from equities annually to make that bogey. The result – an increasing annual tax levy from the populace as required by state law, which is not yet an issue at least in the suburbs. Our suburban plan, one of the better run ones I would suspect, has also failed to achieve the 7% return over the last decade. Close but not quite there.
September 10, 2016 at 1:29 am
Here is the latest working paper referred to in the article: http://www.pensionfinance.org/papers/PubPrin.pdf
We authors gave this paper to the SOA on 9/5. We told the SOA that they were free to publish it this week to meet their promise to their members made 8/26. On 9/7 the SOA published a 10-month old version of our paper.