Adding Monevator to my blog list plus the false precision of spend rates

*(I think abnormal returns was directing you here???--> A hidden variable if the link was about errors in spending or wealth compounding over time)

I was reading an enjoyably reasonable post at the Monevator (How to improve your sustainable withdrawal rate) yesterday.  One of the things I like about what I read is that it looks like he/she has skin in the game when it comes to the idea of decumulation. That makes the content, right or wrong, worth 10 times a lot of the academic dreck that I read.  The exploration feels essential.  Few have "it." Dirk Cotton does. Darrow Kirkpatrick does. Francois Gadenne is not a blogger, but he does. Ken Steiner does. Most academics with W2 income and pensions don't really feel it as far as I can tell.  Most quant-finance bloggers are hyper focused on portfolios, single period math and haven't even thought about consumption or longevity yet. Newfound Research is an exception. Moshe Milevsky is an academic but he's also a genius retirement-quant and knows this stuff to a depth I can't even aspire to.  I want to lease part of his brain for a year or two.  Monevator?  I will add it to my blog list.


One of the comments to that article, correctly in my opinion, calls out what he/she calls the "false precision" of spend rates.  Given the uncertainty of changing portfolio values, two decimal precision seems absurd. And it is.  Add longevity uncertainty, an uncertainty that changes every year, plus estimation uncertainty in returns, inflation and spending -- and there is no way to know.  4% as a blind rule? Bah!  Monevator talks himself into a 4.7% rate but then retreats to 4. But that was a well-reasoned choice and his to make.  A lot of articles I see point to Wade Pfau recommending sub 3% rates in the current environment. But they forget that he also said a safe withdrawal rate is unknown and unknowable. That's about right I think. They also forget age and longevity.

I was thinking about it while reading and I don't recall that I have ever promulgated a hard spend recommendation on RH. Yes, I came up with an age based spend rule as a kind of tongue-in-cheek thing but even that I fudged with a factor that makes it whatever you want within a sliding age scale.  On spend rates, this is about all I know for sure these days:

1. The first and second derivatives of lifetime risk with respect to spend rates are both positive.  You can take a stand anywhere you want on spend rates as long as you have internalized what this means. This principle also has an inverse that Gordon Irlam once pointed out.  There is a cost to safety: to move towards 100% safety is not a linear path in what it requires of you.

2. Errors compound. Fine, spend 7% at age 50 but that money is gone forever and next year when you make the same mistake again it will now be on top of a wounded portfolio, a pair of choices that magnify the second error.  And then there's year three.  At that point you are getting closer to betting on death and capital markets (title of a paper by Olivia Mitchell and Raimond Maurer).

3. Spend rates depend on one's age which is another way of saying they are horizon dependent which is another way of saying they depend on your opinion of longevity in general as well as your specific expectation for yourself, something that changes every year.  Using "30 years" might be fine for you but doesn't cut it for me and it certainly won't at 80. As it didn't at 50. As for spend rates with respect to age: 4% will probably be right somewhere in the age progression but not at 45 and not at 85. Though for some it might be...but probably not for me.  Milevsky made the strong case in some paper I can't put my finger on that longevity is the factor in retirement when it comes to spend rates.

4. No tool or software is going to give you a correct answer.  They all have errors and/or biases and/or oversimplifications.  This is why I triangulate and monitor over time. Things change. Nothing is certain.  No one knows. Opinions flow like water. The stakes in any given year are relatively small...until they aren't.  Watching is worth the effort even if your friends and relatives think you are perseverating a wee bit.

5. In the absence of annuitization and in the presence of superannuation risk, being a little tight early in retirement and a little looser later probably makes sense.

6. It is likely but not certain that partial annuitization by some means will enhance spend rates and lifetime consumption utility. For legatees worried about getting "ripped off" by the pool if the annuitant dies "early," well then they need to be willing to take the other side and support them at age 105 or 110 when the well is dry. Also, as I recall, multiple papers have made a point, about which I now agree, that  waiting and/or using deferred annuities and/or feathering in to a pool in tranches of lifetime income that displace fixed income allocations can sometimes make a lot of sense.

There are probably other things to say about spending but that's all I had for now.