Without getting into a discussion about "actuarial recovery", I do want to talk about the "factors" that PERS plans on using to compute the monthly reduction of benefits due to this "actuarial recovery". For me, these tables are more than an academic interest. I've been preparing a "calculator" that will help retirees figure more closely what their net benefits will be after such a reduction scheme. The calculator works in two stages - first it figures out what the "revised" benefit would be as of a date certain (now 9/1/07) and how that benefit is smaller than or greater than the current "fixed" benefit. This stage also determines the extent to which the member has been allegedly overpaid and computes the invoice amount due PERS. The second stage is to determine how the invoiced amount will be repaid over one's actuarial lifetime according to the retirement Option (i.e. 0, 1, 2, 2A, 3, 3A, 4) selected at retirement. Here is where the TABLES come in handy. Without them, there is no way to determine with any precision, how much any specific PERS window retiree will be affected by the actuarial reduction method.
At my request, PERS has provided me with the Actuarial Recovery Factors for the Strunk/Eugene implementation. I now have the tables for Option 1, Option 2, and Option 3 retirements, but do not yet have the tables for Option 0, Option 2A, Option 3A, or Option 4. In perusing these tables, I was struck by the expected mortality implied by the recovery "factors". In one instance, for example, an Option 1 benefit for a 15 year old (not sure how that works, but the factor is there) translates into a life expectancy of 147 additional years - to age 162!!!!! In another case - my own - my wife's and my joint life expectancy work out to be that PERS thinks that it will be paying us a revenue stream until the last of us reaches about 103 years old. As I considered this, I began to have all sorts of good and evil thoughts. The good thoughts were that for "window retirees" the repayment schedule is considerably longer than any of us had imagined and that, for most "window retirees", the worry about overpaying beyond actuarial life expectancy is probably a moot point. But then my mind began to wander to not-yet-retired PERS members (Tier 1 & Tier 2). If this was Mercer's first salvo into the mortality factor arena, could this be a harbinger of more evil deeds to come as the actuarial tables are revised every two years. Could this portend a lowering of benefits for future retirees by using much more "generous" life expectancy tables? To be fair, I wrote of my concerns to Paul Cleary and to David Crosley, Executive Director and Communications Director, respectively, of PERS. David wrote back to assure me that the "actuarial reduction factors" were nothing more than the 2003 (Milliman) mortality tables adjusted (by Mercer) to reflect the 2% cola over the retiree's life expectancy. This results in a longer repayment period. He also disabused me of any suggestion that Mercer uses any different methodology for mortality factors than did Milliman, and while the 2007 mortality tables may change, the change will be based on experience data as they have in the past.
So, at this point I'm satisfied with the explanations offered. There appears to be nothing sinister going on and current (non-retired) PERS members don't have anything NEW to worry about. I truly had visions of the ghost of Tom Joad when I first encountered these factors. Now all I'm stuck with is trying to figure out whether Audrey Raines is just another example of poor personnel screening by the US Government or whether she turned bad after she was hired. (hint: if you don't watch '24', you won't have a clue what I'm talking about here. If you don't, what's wrong with you?).