Gilgongo wrote:Not sure I understand that. If SoR risk is about eroding the value of your assets too much early on by what is essentially the reverse of pound cost averaging, then what does it matter what happens next?
That’s a good point, and my trite answer didn’t do the subject justice.
I jumped straight to thinking: ‘you won’t solve the problem that way, but I know how to, so you’re wrong’. But you wished only to minimise the problem. So, I think your first question can be correct: draw down on the ‘least affected’. But will it always be correct?
Think your approach through:
Before the market disaster that elicits the SoR risk you are 50/50 stocks and bonds. Crash! Now you’re stocks are down 10% and bonds down 3% (or even up 3%) over the last 3 months. So you cash in bonds, a la your method. Then, bonds outperform stocks for the next 10 years such that you’d have been better off having more bonds and less stocks. The following 10 years might vindicate your choice - who knows. You can dream up any scenario to (in)validate one’s ideas - as you can see.
I think your approach is trying to identify ‘you sell whatever asset or asset class has the least upward potential in the future’; you want to keep the stuff which will go up the most. Which stuff will? No one knows how anything will perform in the future, long term or short term. Some of us think we do: study charts, look at history, do fundamental analysis, whatever. Some will turn out to assess it right, others won’t; we don’t know who is in which group until the end.
So I’ll stick by my earlier answer that your 3 month approach won’t reliably tell you which asset will perform best in the future. No doubt, in some situations it would have.
Deciding which asset to cash in during a SoR crisis is fundamentally the same as deciding what to invest your next spare cash in during stable times. You might get a better or worse return than choosing differently. No one can know ahead of time. That’s the reason people choose the ‘fixed % withdrawal/belt tightening’ or ‘go back to work’ or variable % withdrawal approaches; because they put you in control.
You can view ANY withdrawal at any time as a SoR risk issue: the markets are chugging along happily, you’re spending an asset that would otherwise have risen more beneficially at the next market upturn. So you lose out in the long term. The SoR risk we all picture occurs when markets plunge, but the same process is eroding your future wealth is you spend your assets, however the market is moving. So the SoR problem we’re trying to find an answer to is simply part of a continuum of spending/returns relationship. Those thoughts help me view solutions, which are intended to avoid the ‘no money left well before dying’ outcome, like ‘tighten your belt’ or ‘go back to work’ or ‘save more before you retire’ because they are solutions on a their own continuum which can be executed in a measure to suit the times. In contrast, the approach you suggested will be the best on only some occasions, I think.
This is one of the holy grails of draw down investing. There are complex books on it like Otar's Unveiling the retirement myth, McClung's Living off your money, and Kites' Strategies for managing sequence of returns risk in retirement.
Looking forward to hearing how you think your 3 month rule might work better than I’m picturing. Sites like portfoliovisualiser might allow back testing to see how it works.