Request for ideas

Peteyperson’s Idea about Withdrawals
and
Should We Rebalance Portfolios?

The more that I investigate safe withdrawal rates, the more that I question the value of rebalancing portfolios. Repeatedly, I have found that rebalancing removes a considerable upside potential while providing only a very little in the way of downside protection. I consistently find that we do best to let our stock allocation grow except when stock valuations are highest. Even then, rebalancing only approximates the benefits of switching (that is, varying) allocations. It does so in an awkward, inefficient manner.

Behind the theory of rebalancing is the notion that we cannot discern what the market is going to do. This assumption is very good for the short-term, but it is lousy for the intermediate-term and the long-term. Ben Graham was right. Prices matter. A measure such as P/E10 allows us to predict future performance for the intermediate-term. Or we can apply the Gordon equation directly or John Bogle’s variant, both of which are based soundly on the dividend discount model, to predict overall performance in the intermediate-term. We come up with the same result. Even though there is an element of uncertainty, there are sound reasons to apply this information about the intermediate-term.

Because we can estimate intermediate-term performance much better than making a coin toss, we should be able to do much better than mindlessly rebalancing.

Peteyperson, a prominent contributor to several boards, favors drawing money away from the fixed income side of a portfolio when stocks are doing poorly. His rationale is flawless. We know that selling stocks when their prices are low is what kills retirement portfolios. We are best off selling stocks only when we can get good prices. We should plan on leaving our stocks untouched over an extended period of time until we can get favorable prices.

Every time that I have looked into something like this, I have used two portfolios, one with stocks and fixed income together and the other with fixed income alone. I have always found that you end up doing best with a single portfolio. I have never found that having a second portfolio works out well.

But the problem is not with peteyperson’s idea. It is with our calculators and with what I have looked at. I have never made withdrawals in the manner that peteyperson suggests. I have always come as close as I can, but I have never actually checked out his approach.

We need to make a major calculator upgrade to test out peteyperson’s ideas. [We will probably end up being me.] I need to hear some recommended algorithms. Right now, I am thinking of changing withdrawal percentages as a function of P/E10, a new form of switching. I am hoping to avoid having to change a couple lines in every sequence. I think that I will have to do just that.

Then again, I never thought that I would ever use calculators. And I never thought that I would ever set up my own web site.

Please send me your ideas as to what would be some good algorithms for implementing peteyperson’s idea.

Have fun.

John Walter Russell