Saturday, July 24, 2021
House of Pain
OK. I lied. THIS is my last post. Yesterday the PERS Board voted to lower the assumed interest rate from 7.2% to 6.9%. This falls between the maximum recommended by Milliman (7.0%) and their lowest bound recommendation (6.8%). This means that anyone still in Tier 1 (active or inactive) who expects to retire under Money Match will see benefits lowered from retirements during 2021 versus retirements taking place in 2022. Similarly, anyone retiring under Full Formula (from any tier) WITH A BENEFICIARY will see reductions in benefits beginning with retirements on or after January 1, 2022. We don't know how much the reductions will be. As a general guide, I've been telling people who ask that a 25 basis point reduction in the rate works out to be about a 3 month setback in benefits before crossing over into a higher benefit. The current reduction is 30 basis points, but other changes may make the setback slightly longer. We won't know anything with certainty until PERS releases the updated Actuarial Equivalency Factors sometime in late November or early December. The AEF tables combine information about mortality, assumed interest rates, inflation, salary growth, and a few other economic factors. We only know the assumed interest rate change at this point. I have no idea how the various factors come into play, because actuarial math is one of those places where I never received any formal education. I get math, but I don't have any idea how Milliman produces those tables.
Now you can return to your normal channel and ignore this page in the future, unless you are researching PERS history.
Subscribe to:
Post Comments (Atom)
13 comments:
I knew you couldn’t stay retired. Thanks for all you’ve done.
It was low hanging fruit. I couldn’t let it go.
Well , I spoke with Pers today about the lowering of the assumed rate. This only applies to Tier I according to them. I'm Tier II and she stated it will not affect me. I then asked about the age/experience tables. She state they would , although to check back in September.
So maybe I wont get hit to bad, for I'm retiring in July 1st 2022.
Oy. The amount of misinformation PERS itself contributes. You were asking one question and she wasvasking acquestion you didn’t ask. The assumed interest rate applies to different entities in different ways, but in retirement applies in precisely the same. In the case of TIER 1 members, the assumed rate is what members earn on contributions made prior to September 2003. For Tier 2, you get market rates for money contributed in the same period. So, in that sense, the person at PERS was 100% correct. Bit you weren’t asking that question. You were asking how the assumed rate reduction would affect your benefit at retirement. That depends entirely on the actuarial equivalency factors that convert your account balance and the employer match (or the bulked up account balance used to get to a full formula benefit) into a stream of monthly payments. The actuarial equivalency factors take into account all the variables I’ve been talking about - assumed interest rate, mortality factors, salary growth, inflation, your and your beneficiary’s age and optional form of payout. When I talk about the impact of rate changes, I refer specifically to their effect on those actuarial factors in terms of a “setback”. If your benefit were to be calculated under the current (soon to be outdated) tables, the new tables will reflect reduction that would make the later retirement reach parity with the dec 1, 2021 benefit on or about May 1, 2022. So, by retiring on July 1, 2022, your benefit should be slightly higher than it would have been had you retired on Dec 1, 2021 (assuming you had a choice).
When you ask PERS, you have to be very precise in the question you are asking. Otherwise you will get a technically correct, but irrelevant, answer. Unfortunately, the answer you seek isn’t available to PERS staff right now.
Hopefully, like Sherlock Holmes at the Reichenbach Falls, you'll miraculously return by public demand.
Hello all - I have so enjoyed this BLOG. I am a Tier 1 member and am closing in on finishing my 38th year in the PERS system. Still too young to benefit from Social Security and such but have plans on trying to be in a position to retire in November of 2022. Didn't like the 6.9 reduction and I worry a bit about going from the 3 highest years FAS calculation to 5-years but it's the best I can do under the circumstances.
Very best regards,
Dave
PERS had a great year for earnings. How can they ethically rationalize cutting the assumed rate? PERS does not work to benefit its members, it seems.
Retiring? A spreadsheet showing differences for each month going forward was helpful in choosing just when to retire. Not all months are created equal ... for example, retiring in June resulted in only a $10 per month increase in my benefits, whereas previous months were about a $40 per month increase in permanent benefit each month. Also, be sure to retire on July 1 or before for the COLA, which is an increase that would take several months to work to achieve that benefit. In my experience, a cut in the assumed rate takes longer than the period PERS stated it would take to reach parity with an earlier retirement date. Spreadsheets were very helpful showing these differences. Important to calculate your benefits with the current benefit calculator and print them out, and do the same when PERS gets a calculator for the new assumed rate, and put these in the spreadsheet.
LK
PERS is required to maximize benefits to members, while not unduly burdening the employers. The parity problem is well-known and depends on age, beneficiary age, option.
Let me elaborate more fully now that I'm not engaged in four conversations at one time. PERS' earnings crediting are constrained by a law passed in 2003. PERS cannot credit more than the assumed rate to Tier 1 members unless (a) there is no UAL and (b) they've earned more than the assumed rate in 3 consecutive years. The assumed rate is not tied to current earnings at all. It is a forecast based on economic data provided by a whole host of independent data brokers with expertise in the overall market. The PERS Board has been excessively conservative in cutting the assumed rate in the face of strong evidence that long-term earnings are on the decline. The cut this time was hardly surprising except that it wasn't lower. In order to earn higher than the assumed interest rate, the Oregon Investment Council, which is independent of PERS, has to take some significant risks. Sometimes the risks pay off; other times they don't. The OIC has been agitating for years that the assumed rate is too high, and has been engaged in a long term derisking strategy that is guaranteed to fail to hit the assumed rate target. That is one of the reasons that the Board has been forced to lower the assumed rate. They ignored all the recommendations in 2019 to lower the rate effective 2020. They held to the 7.2% rate that all but forced the more significant decrease starting in 2022.
As for the reductions to benefits for not-yet-retired, the setback figures are always based on Milliman's "average PERS retiree". Since people aren't "average", then actual results of the setback might be significantly longer, or slightly shorter. There are a lot of variables that go into the calculated of the Actuarial Equivalency Factors, and very few people actually fit the profile of the "typical" or "average PERS retiree". As such, the numbers are given out only with the notion that they are a rough "back of the napkin" estimate of how much longer you might have to work in order to achieve parity with benefits achieved under a different, more favorable, set of assumptions. Your age, your Tier, your "best" calculation, whether you have a beneficiary or not and how much age difference there is between the two of you, and what option you choose as the payout method all influence the time to reach parity.
As far as the ethics of cutting the assumed rate, the answer lies in the law, not the ethics. It resides in the fact that the assumed rate is fixed at the rate when you retire. Your retirement is expected to last about 20 years in most cases, but longer in many. How can they ethically forecast out 20 years without taking into consideration what the tea leaves are projecting for the future. It literally has nothing to do with how they did just before you retire, or in the previous year or two. They have to look at trends going back 20 years and forecasting out 20 years. As they always say when you invest money, "past performance is no guarantee of future performance". Ethics don't matter; the law does.
I had an "interesting" conversation with a PERS rep on the phone. I will retire effective December 1 or July 1 of next year. I asked when PERS would be able to give me an estimate of the benefit under the new assumed rate. She said after the first of the year! Kind of tough to compare a number you know to a number you can't know. Sort of a Kafka meets Heller moment, I suppose. Rules of thumb can help provide the months of further work to make up the difference in benefits, but I worry a little about the actuarial tables and how changing those might influence the benefit. Looks like November 30 is my last day...
None of this is surprising. PERS is a prisoner of the Board and the actuary. They cannot provide information they don’t have.
Yes, of course Mark. However, I see no reason why PERS can't give estimates once that information is finalized. Part of the point of making the decision on assumed rates in July is to provide the requisite time to provide information. We now know at least the assumed rate that will be in effect for retirements starting 1/1/22 and thereafter. Once finally confirmed in October, the formal estimate for 2021 and 2022 should reflect the whatever assumed rate would be in effect at the time of retirement. My point is that PERS will have the information necessary to give two estimates (each incorporating a different assumed rate), one for a 12/1/21 retirement and another for a 7/1/22 retirement. I don't know why they can't do that. Am I missing something?
PERS does not have the information now, and the Actuaries haven’t yet prepared their recommendations on other crucial details (like mortality factors) to calculate Actuarial Equivalency Factors until the October Board meeting. Only at that point will PERS staff be able to do the tedious task of reprogramming the benefit calculation and Benefit estimators into their system and rigorously test it. PERS a rigorous suite of test data that they use to validate changes to benefit calculations. Mistakes are costly to both PERS and to the retiree/prospective retiree.
Unfortunately, lots of things have changed over the years including the timing of the decision on the assumed rate, which used to be finalized in April. The experience study, which is the basis for all the assumptions that go into making the Actuarial Equivalency Factors is also an integral part of determining employer contribution rates. And the Experience Study gets more complex every time the legislature makes any type of changes to the retirement system. SB 1049, enacted in the 2019 session, is not yet fully implemented. The legislature makes the changes without any consideration of the complexities of turning a statute into a fulling realized law with all the necessary administrative rules required to implement the law as written and as intended. If you want to complain, take it up with your Legislator.
Post a Comment