Here's a really really obvious point that escaped me until a couple of days ago, although other posters have touched on it.
4% of what? How do you put a suitable value on your retirement portfolio at the start of retirement?
The usual way is to take the market value of the portfolio at the start of retirement, the number printed on your brokerage statement. But that's wrong.
Or let me put it this way: the risk of exhausting the portfolio has two components:
1) The effect of fluctuations in portfolio value during the withdrawal period--which, problematical as it is, is somewhat spread out in time and partially smoothed and averaged. A bull market later in retirement may not fully balance the effect of a bear market early in retirement, but it certainly helps.
2) The effect of the instantaneous snapshot of your portfolio value that you use to gauge the size of your withdrawals forever after.
Well, the effect of #2 is awful, and it's huge, and it's completely concentrated at one instant in time, a single data point, no averaging.
Think of it this way. Suppose your retirement portfolio were worth $1 million, but on the day you retired you received an erroneous brokerage statement with a computer glitch that said your portfolio was worth $1.5 million, and on the strength of that you decided to withdraw $60,000 the first year, then COLAed, believing that was following the 4% rule. A month later you receive a corrected statement, but you ignore it and continue withdrawing what you think is 4%, but which we know is really 6%. Obviously disaster will result.
And it's not fair to attribute this disaster to the vagaries of the market during retirement. The disaster should be attributed to using an erroneous starting value. It's not fair to blame the "4% rule" when you're really withdrawing 6%.
Well, I think most of us will conceded that the stock market as a whole can be very significantly mispriced, even if you can only see it in hindsight. I say the market was overpriced in mid-2007, S&P 1600, and underpriced in early 2009, S&P 800. Accept that for the sake of argument.
I think you'll agree that someone following the 4% rule who retired in mid-2007 is likely in trouble, and someone following the 4% rule who retired in early 2009 is likely in good shape... and that the difference in outcomes depends almost entirely on the single choice of the day on which the initial 4% was calculated.
The person retiring in mid-2007 and using the brokerage statement is in a very similar situation to the person with the erroneous statement. The number he's using to gauge his withdrawals is wrong in both cases--it's too high. And in both cases, the essence of the problem is not so much what the market did in the future, it's what the brokerage statement said in the present.
If you follow the 4% rule, a bull market immediately before retirement is just as bad as a bear market just after retirement, because a bull market immediately before retirement tricks you into withdrawing too much!
I don't see how all the Monte Carlo simulations of thirty years of retirement can help you deal with the intense concentrated risk of the accidental moment when you value your portfolio.
Now of course I'm being a little overwrought here. But it seems odd to me that I haven't seen any suggestions that the 4% should not be calculated on the brokerage-statement portfolilo value at retirement, but on some weighted average of the portfolio's past history preceding retirement... a history over a considerable period of time, at least twenty years.
It is an operationally meaningless statement, but what you really want is not 4% of what Mr. Market thinks your portfolio is worth, but 4% of what your portfolio is truly worth.
This resonates with something else I've been groping toward: the risk of a high stock allocation is not merely the risk that it will be down when you need to draw on it, or the risk that it may not yield as much over 20 years as you planned. There is also the risk of being unable to judge your progress toward your goal, and thinking you're in good shape when you're really just riding a temporary and evanescent peak.
Last edited by nisiprius
on Mon Jan 25, 2010 5:51 pm, edited 2 times in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.