After all the stürm und drang of the Legislature, the final piece of next year’s puzzle has fallen into place. The PERS Board today adopted its new assumed rate for the 2018-2019 calendar years. The Board spent very little time debating between the extremes of 7.1% recommended by the OIC and its consultant, and the slightly more optimistic forecast of other consultants used by Milliman at 7.2%. In the end, the Board went to 7.2%, largely because the actuaries gave them the latitude to adopt any rate between 7.0% and 7.25% as a responsible choice. A few wanted to split the difference at 7.15%, but there was no opposition when Board Member Pat West (the member rep on the Board) moved to adopt the 7.2% rate. It was quickly seconded by the employer rep on the Board, Lawrence Furnstahl of OHSU. The rest of the Board quickly approved the motion and, in a blaze of light, the meeting was over.

After the meeting, I checked with Matt Larrabee, the principal actuary for Milliman, who confirmed for me that the setback would be 4 months for a typical retiree. This means that if you delay retirement past December 1, 2017, it will take you 4 additional months of working to recover the benefit you would have received if you retired on December 1. While the most directly affected members are those who remain eligible to retire under Money Match (less than 13% of all non-retired members), it will have an impact on beneficiary options for Full Formula retirees as well. The changes to mortality had virtually no impact on the rates, as changes in one element were offset by other changes. Overall, the totality of the economic assumptions other than the assumed rate itself, had a near zero impact on liabilities for the system. The impact to employers on the uncollared rates will be approximately 1.9% of payroll, less than it could have been.

There was no opposition by any stakeholder to the change, at least not at today’s meeting. In fact, the Board didn’t even offer the possibility of public testimony, and the Board Chair John Thomas repeatedly interrupted almost any speaker at the podium to sermonize. While he’s done a good job as Board Chairman, and clearly knows financial analysis, I find him personally tiresome as though he is lecturing small children. Thankfully, I’ve chosen not to attend Board meetings because of his overall mien. I went today solely because the decision was important to a lot of people, and because I’m still adjusting back to Portland time from my time spent in Iceland.

All in all, the title pretty much covers my feelings. It could have been a lot worse.

This will be my last post for awhile, in large part because we are moving into “ordinary time”, where nothing of the moment will take place. The next time for something significant to occur will be the working report from Governor Brown’s group studying how to lower the UAL by $5 billion. That will, in turn, lead to some legislative momentum that **might** occur during the short Legislative session next February. However, any substantial changes to PERS will probably not come before the 2019 Legislature. If nothing positive happens between now and then, we might expect to see some attempts to significantly alter PERS.

## 7 comments:

Hello Marc,

In your article you wrote: "... the setback would be 4 months for a typical retiree. This means that if you delay retirement past December 1, 2017, it will take you 4 additional months of working to recover the benefit you would have received if you retired on December 1."

Just to clarify, it might help readers to make this statement a little more specific, such as the following revision: ".. if you delay retirement past December 1, 2017, it will take you 4 additional months of working to recover the MONTHLY benefit AMOUNT you would have received if you retired on December 1." (I simply added the two words in capital letters.)

The reason I put it this way is because the way I view the situation, the prospective retiree also would be losing out on the four months of payments (December 2017 thru March 2018), since my assumption is that he would receive four fewer monthly retirement checks since his life assumption (assumed date of death) would be unchanged under either chosen retirement date.

Therefore the net present value of all monthly checks would be even higher if he retires by December 1, 2017, relative to a retirement date of April 1, 2018. So, to obtain the same net present value of all future monthly retirement benefit checks, he would need to work more than just four additional months past December 1, 2017.

Now, admittedly there are many other considerations to one's retirement decision such as salary earned each month, medical and other benefits earned each month, etc. However, if you are simply determining how long it would take to work past December 1st to make up the drop in the value of one's PERS tier-one payments (present value of all future monthly payments) then the answer is somewhat longer than four months. The period would be impacted by the person's age at retirement, number of years under Tier 1, etc. So, the number of months to wait would vary somewhat for each individual. I haven't done the calculations, but I'm guessing that one would have to wait perhaps five or six months to regain the total value of all Tier 1 retirement payments.

Regards,

David

Marc,

What about the actuarial (life expectancy) tables?

Did they get changed yet?

Did the numbers change much?

Thank you,

David

@David Cooke. You are, of course, correct; however, the Actuarial Equivalency Factors do not scale exactly linearly so it is hard to project out NPV for a December 1, 2017 vs an April 1, 2018 retirement. The longer you live, the higher the probability that you will live a bit longer, so actuarial tables don't exactly follow a precisely linear relationship. That's one of the reasons I've avoided too much specificity or clarity at this point. Until I see the revised tables, I can't say with certainty how any single individual will be affected. The four month setback is just a number given by the actuaries as a ballpark estimate of how much longer an individual would have to work to remain at the same place relative to the December 1, 2017 retirement date preserving the current tables and assumed interest rate. I'm not sure I completely agree with your statement "...since my assumption is that he would receive four fewer monthly payments of retirement checks since his life assumption (assumed date of death) would be unchanged under either chosen retirement date". The assumption could indeed change, even in as short a period as four months. Living longer is a reason that people live longer.

The actuarial equivalency tables will NOT be available for review until much before the end of October, maybe slightly later. The four month setback is a rough estimate of the recovery time for a 30 basis point reduction in the assumed rate. The actuarial tables are a combination of expected longevity at every month of every age beginning at earliest retirement age, assumed rate, and assumptions about inflation over the period of time over which an individual is expected to draw benefits. It is arcane math, not for the faint of heart, and, although I minored in math, I still don't fully understand the way all the different pieces combine to produce new AEFs. But, until they come out, there is no way we can draw any more conclusions than I've drawn in my blog post. Whether or not NPV will change as a result of 12/1/17 retirement vs a 4/1/18 retirement remains to be seen.

Mark--

No idea if you are following this blog during the "off-season." Do you know when a new estimate of the shortfall will come from PERS? I did not see it in the briefing book. A lot of factors influence this shortfall, but one is investment returns. I believe it was around the beginning of 2016 when the board released its $20-22 billion estimate (of course the politicians used the higher number for effect.) But since 1/1/2016, the S&P 500 is up 21%. That's only one factor, but surely investment performance has helped the shortfall. When will the estimates reflect investment returns?

And of course a bit THANK YOU for all of your informative accounts and wry narrative. This is very much appreciated!!!

@befuddled. The next formal estimate of the shortfall will come with the 2017/ system valuation, done after final earnings crediting next April. Assuming PERS hits the assumed rate target for 2017 (a not-unreasonable guess at this point in the year), the UAL will probably increase about $2 billion from the rate cut. If PERS manages to do measureably better than the assumed rate in 2017, almost all the xcess would go to paying down the UAL, and lowering projected employer rates for the next round. But even if PERS were to earn a net 9%, the extra 150 basis points would probably only reduce the UAL by about half what the reduction in lng term rates would increase it. Better than nothing, but one of the reasons that the actuaries argue that it would be difficult to grow the fund out of the current UAL.

Post a Comment