STUMP » Articles » On Full Public Pension Funding: A Follow-Up

On Full Public Pension Funding: A Follow-Up

by meep

22 September 2017, 12:20

I got several emails over my recent post on the never-fully-funders advocates with respect to public pensions.

I am going to address a few below.

WHAT IS FULL FUNDING?

I got the following question via email:

What exactly IS “100% funding”? Does it mean that the funding is there to pay the benefits WITHOUT considering future earnings in the fund? Does it include assumptions about mortality, and earnings rates? Does it factor in COL increases, and survivor pensions?

Short answers: Wait a moment, no, yes, and yes.

Long answer (which answers the first):

There’s two parts to this:

- the contributions being made (a cash flow statement item)
- the funded ratio (the balance sheet item)

Let me take the second first: 100% funded ratio means that the asset measurement equals the liability measurement. In most of the official funded ratios you see, the asset measurement may take into account smoothing or book value — that is, not really a fair value or market value of the assets at a point in time.

The liability value is a value of the benefits currently accrued, and implicitly contains what future investment earnings will be (and sometimes future contributions). It includes all sorts of assumptions: when people will retire, how fast they’ll die off in retirement, if they’ve got a survivor’s benefit, if they’ll never vest, if they’ll die before retirement, etc.

So here’s the actuarial black magic: sometimes, COLAs weren’t included in the valuation, because they weren’t guaranteed. If I remember correctly, this is what happened with Detroit and the “13th check” — that wasn’t in the valuation, because it wasn’t guaranteed. If it’s written into the benefits that there’s a guaranteed 2% per year COLA, then that absolutely is included in the liability valuation.

Different valuation approaches also consider future salary increase scales, but in such a case, they also consider future contribution rates.

So now I come back to the first part. 100% funding contributions means covering the normal cost (additional benefit accrued this year) and amortized cost of unfunded liability (making up for prior year shortfalls). These are done separately. Determining both is involved in the kind of balance sheet valuation mentioned above. But if you decide to do a level % of payroll contribution assumption, the assumption that payroll is going to increase by a certain % per year has an outsize effect.

The assumption with the largest effect is the discount rate/assumed rate of return, but when you’re amortizing over 30 years an assuming that the contribution will increase 2% per year, so assuming that the final payment would be 80% higher than the initial one. There’s a reason that most 30-year mortgages are level-dollar, not increasing by 2% per year.

As it is, from my recent blog post shows that covering 100% of these “required” contributions doesn’t lead to an increasing funded ratio:
Public Pensions: Why Do 100% Required Contribution Payers Have Decreasing Fundedness?

I went and checked some specific details (because I have long-running threads on public finance and public pensions at the Actuarial Outpost — I do not know for certain, but the Detroit city pensions did have a 13th check problem for sure, which was a part of the bankruptcy workout.

Our dispute was whether they were part of the actuarial valuation.

Comment by pension actuary Dan Moore:

My opinion that the 13th check was not included in the liabilities is based on the fact that they were treated as a discretionary benefit, and the fact that they are not even mentioned in the plan provisions section of the most recent General Retirement System actuarial valuation report. Also, this report that the 2 pension funds paid 13th checks through 2011 and still maintain ‘savings plans’ for active employees -i.e., cash balance arrangements (apparently informal and supplemental to the main pension plan?).

In general, “bonuses” in pension payments like discretionary COLAs and 13th checks are not included, as they’re not guaranteed elements of the pensions.

But the Detroit pensions kept paying those 13th checks each year, even when the pension fund did not do well.

So, the pensioners came to regard them as “guaranteed”. Turned out they weren’t – these were part of what was whacked in the bankruptcy deal.

By the way, San Diego has a 13th check problem, too.

WHAT ABOUT CHANGING VALUATION RATES?

Someone else asked me about the California plans that got their payments whacked — see, Calpers has been using a 7.5% valuation rate for its “going concern” valuation… but the moment you want to leave, they drop it down to a “risk-free” rate… I think about 3%. (Could be higher, could be lower… the exact number right now is not what’s in contention):

From my email:

I just read your most recent one rebutting the paper from Berkley. You mention the orphaned entities and their huge cuts. Wouldn’t the retirees have taken huge cuts even if they had been 100% funded?

My response:

Yes, they would have gotten sizable cuts, even if they were 100% funded under the Calpers “going concern” valuation. The difference, of course, is between the 7.5% valuation rate used for “going concern” plans, and the ~3% rate used for plans in run-off. I forget the specific details for the closed Calpers plans, as to what their funded ratios were before they closed up.

There is a really blunt way to try to adjust a liability value given different interest rates, which is what a lot of the “adjusted funded ratios” you see coming out of Moody’s and research papers. If we assume a liability duration of about 10 years (this may be high), then the adjustment would be: (1.075/1.03)^10 = 1.53 — So about a 150% funded ratio with 7.5% discount rate would be a 100% funded ratio for a 3% discount rate.

That’s somewhat exaggerated, because I’m not sure what the duration is/should be.

But let’s assume you started out with 70% funded — i.e., Assets = 70% * liabilities measured at 7.5%

Then if it were re-calculated at 3%, liabilities at 3% = 153% * liabilities measured at 7.5%

So Assets = 70%/153% * liab measured at 3% = 45% * liab measured at 3%

That actually sounds about right. A cut of 55% of benefit levels at valuing using 3% vs. a cut of 30% of benefit levels at valuing using 7.5%.

This is a really rough calculation, and I was planning on doing a followup post, so this would be a good thing for me to address.

Consider it addressed.

If you want to see me comparing numbers when this whacking was being considered… and, by the way, two more Calpers plans are likely to get similarly whacked,… here are prior posts:

Much Ado About Discount Rates: Welcome to the Pension California

California Watch: Calpers Valuation, Exits, and Governance:

As a recap, the issue is that California uses vastly different valuations depending on whether you’re staying in the plan or if you’re exiting.

My high-level take:

The inherent value of what was promised shouldn’t depend on:
-who did the promising or
-how they plan on fulfilling the promise

The promised amounts are what they’re worth.

That said, there’s the embedded option that the promise won’t be fulfilled. That option’s value does depend on who is promising — some sponsors are much more likely to renege than others.

If these valuations were broken into the two parts, I would be just fine with that approach to valuation. You would have the inherent promise value, using risk-free rates, and then the difference between that and what was held on the books would be the value of the default option.

To bring Detroit back into it, at one point the emergency manager Kevyn Orr had an actuarial firm do a rough pension valuation using much lower valuation rates than had been used, partly as a tactic to get the pensions to play ball. When they agreed on pension cuts as part of the bankruptcy deal, they did not use that harsh valuation… though it would have been a better idea of what those pension guarantees are actually worth.

Because Detroit still has a pension problem.

OTHER PEOPLE’S REACTIONS

Thank goodness some sensible people are involved in San Diego pensions.

An oped by someone who has been on pension boards:

Commentary | Why unfunded public pension plans amount to theft

The opinion piece “Why full funding of public pensions wastes money” (Sept. 15) was very disturbing. To try to convince taxpayers there is no reason to fully fund pension plans is a disservice to public employees and retirees.

Today, taxpayers are paying part of the annual pension costs for employees who provided their services to the community years ago. Taxpayers of those previous years did not pay the total employment costs of the firefighters or the librarians for their services back then. Some of those costs were pushed out to later years’ taxpayers — up to 20 years in many cases. Our children will be paying part of their future taxes to pay for pensions earned for services performed many years ago.

That is what happens when there is an unfunded pension liability. This constitutes an intergenerational theft. Delay the street repairs today, we need to first finish paying for the pension the street repair man earned 10 years ago!

…..

The authors recognize that corporations need to fully fund their pension plans because the corporation might go bankrupt and thus be unable to pay pension benefits. But in contrast, they argue that governments never “go away.” The argument is governments will always be there and will have sufficient taxing power (and the political will to use such) to tax future years’ taxpayers whatever is needed to pay the then-current retirees’ pension checks. This is simply not true!

Look at the millions of Greeks who had their pensions cut severely due to an inability of the Greek government to pay for those promised pensions. Greece did not “go away,” but the pensions sure did!

Public entities may not “go away,” but they can and do go bankrupt.

How about Loyalton, California, where retires saw pensions cut 60 percent, or East San Gabriel Valley where pensions were cut 63 percent?

How about Detroit, the largest-ever municipal bankruptcy? Retirement checks were reduced and the annual cost of living increase was eliminated.

…..
Municipal bankruptcy is relatively new; the priority of pensions in bankruptcy is still unsettled law. But the inescapable fact is there is only one guarantee the retiree will receive his pension check. And that is to have a fully funded pension plan.

Arguing that “fully funding a public pension plan is not needed” and is actually a “waste” of money is a disservice to public retiree constituents.

Vortmann has served on the boards of the pension agencies of both San Diego County and the city of San Diego. He’s also been a member of the mayor’s Blue Ribbon Finance Committee and the city’s Pension Reform Commission.

I completely agree with Vortmann.

Another reader sent me his own annotated comments to the article here: A reader’s op-ed rebuttal.

I will pull out a few of his comments:

INVESTING vs GAMBLING

“Op-Ed” illustrates what these individuals are doing is “gambling” because when betting on sports (or really any other pure gambling activity), there are no loss-mitigation strategies. This is a key difference between (risk reducing) investing and (risky) gambling.

Reader’s reaction to January 2017 piece on spiking pension costs for San Diego:

basically the question to be asked is why did TPTB
deliberately ignore items w/ in the portfolio like
the 13th pension payment which has been SOP for over
three decades???

Ah, that 13th pension check keeps coming up….. funny how these things get ignored.

(and if you’re not familiar with internet lingo — TPTB == The Powers That Be)

MY BACKSTORY

Public pension funding is a topic I visit fairly frequently. Important posts on this issue:

That last one is from 2014, where I’m hacking at a particular myth.

While I originally got into the public pensions debate about a decade ago, and I explain my motive here:

I have many motives. One motive is finding out what’s really going on — when I get interested in a topic, I start digging into it, trying to figure out how we got where we are. Digging into the various actions going on helps me learn more about the dynamics. The public pensions problem involves multiple interacting dimensions, and I’m still learning about these.

One motive is to make people quit saying stupid things. I am, at heart, a teacher, and I can’t abide flat untruths to be spread willy-nilly. Others go after the you only use 10% of your brain myth – I go after 80% funded is good enough myth.

One huge motive, though, is I don’t want public pensions to fail.

I have lots of public school teachers in my extended family, and I would rather pension promises made be fulfilled. Thing is, because everyone assumes “government doesn’t go out of business,” serious problems with pension funding and design are not addressed or addressed in ways that make the pensions even more likely to fail.

I am not happy about the impending disaster. No, not all public pensions are going under. But many are not in as good a state as official numbers show. Many people will not get their full promised pensions. They won’t get nothing, but they will get less than they had planned on, just like with Detroit retirees and employees.

And the issue is that all these public pensions disasters are years in the making — these disasters do not come from nowhere, without warning.

The problem is that once the disaster comes, you have people like retirees getting hit. If they had been warned years ahead of time, they could have been prepared.

And I am really pissed at the never-fully-funders, who blithely assume that all will be well.

It won’t.

And your type of fantastical thinking is making it even more likely that things will fail.

STUMP » Articles » On Full Public Pension Funding: A Follow-Up » 22 September 2017, 12:20

On Full Public Pension Funding: A Follow-Up