I was reading a good article by Wade Pfau at Forbes.com today and it made me think about a couple of things that have been rumbling about my mind on Sequence of Returns Risk and the various spending rules.
1. I love Wade's stuff and I think he is a smart guy with an unusually good bead on the retirement problems of today's world. He certainly has better answers than I do. But when reading articles like this I have to admit that am always a little surprised (not by him as such, but by everyone who has written on this topic over the last 10 years) that either: a) the current state of retirement finance in 2016 is still so fuzzy (which I'm thinking it is and will remain because that is the way the future works), or b) the essential nature of the uncertain-ness of the retirement formula(s) has been so inelegantly communicated to the world for so long.
2. Sequence of returns is a serious problem and the math absolutely needs to be well understood by anyone that has retired, doesn't still work a bit, can't or won't adjust spending, happens to have the return assumptions of someone living in 2016, has high-ish expected longevity, and has an endowment that might or might not work over the planning horizon. On the other hand, I gotta say that spending shock risk makes sequence risk look like a piker. For example, go to TheRetirementCafe.com and search for posts related to "Chaos."
3. One of the other things I've been thinking about on sequence risk lately is that it is not really necessarily a phenomenon of "early retirement" but I have zero quantitative proof of this and in the end it probably doesn't matter...probably an "angels on the head of a pin" kind of thing. To start, here is a sliver of Pfau on this: "If a retiree’s portfolio drops in value during the early retirement period, portfolio withdrawals will dig a further hole." Here is my idea (or theory or hypothesis): there is no such thing, up to a point, as "the early retirement period." My reasoning is as follows: a.) each year that one survives to a new age, statistically and demographically speaking one's longevity gets a new estimate that is just a little bit further out -- the goal post has shifted, and b.) in each new year or day or hour or instant that one is still alive, one, in effect, starts a new retirement all over again in that very instant. Retirement, then, is always in a continuous process of 'becoming.' The beginning is a process and the end recedes. For those two reasons I think there is no such thing as a discrete "early retirement period" after which one is "safe." What I do think, however, is that there can in fact be a critical mass of sufficiently bad financial-environmental-phenomena (low returns, high inflation, spending shocks, high volatility, etc.) over enough time proceeding from whatever "the instant" is -- totally independent of when retirement "technically" started -- that a retirement can clearly be put at risk of running out of money. That means the risk is more or less continuous rather than some kind of step function that somehow stops dead at 5 or 10 years from the retirement date. Of course I also think that the risk does abate the closer one gets to the end game. Sequence risk at 110 is probably a silly concept. For an early retiree, on the other hand, it'll hang in there for a while whether or not you think you are either in or out of an "early retirement period."
No comments:
Post a Comment