Asset allocation period for rebalancing advice is poort

However, something about the how-to information on asset allocation in both books disturbed me even more than one writer wanting to see patterns in short-term random results.

The advice about rebalancing.

Both of them say that after a period of time you should rebalance your portfolio, selling the assets which have most risen in price and buying more of the assets which have risen the least (or fallen!).

There is no clear cut consensus on how often you should do this, nor even any guidelines. The books mention periods of one year, quarterly, monthly and even weekly -- but leave it up to the reader.

This seems to me an incredible weakness, for several reasons:

1. Using a short period means you have little opportunity to take advantage of long term trends. For example, if you'd started an asset allocation program in 1995 would it have made much sense to sell off all your stocks in 1996? That bull market ran through March 2000.

Also, bear markets can go on for a long time. Some asset allocation programs contain gold. If you'd started an asset allocation program in 1981, your portfolio would by now consist almost entirely of gold . . . how happy would you be about that?

2. If your accounts are not tax-sheltered, you're realizing taxable capital gains, so part of your portfolio's gains are going to the government.

Even tax-sheltered accounts will have increased transaction costs. Brokerages like to rebalance client accounts every quarter. I wonder why they do it so often?

Seems to me that a poor asset allocation program is little better than an HYIP con game.



asset allocation rebalancing flaws
asset allocation rebalancing flaws

Asset allocation assets exhibit random covariance

Recently I read several books on asset allocation. This is not a full review of them, but I did notice several disturbing points.

One of the books wrote a lot about the variance between difference types of assets, such as stock and bonds. This variance is a critical point for asset allocation, because the theory behind it, Modern Portfolio Theory, comes from Harry Markowitz's work showing that overall portfolio risk is reduced by holding assets that don't go up and down in tandem.

Anyway, the book went on at length about how tricky this is, because the variance of stocks and bonds changes over time, and came up with different variances over 3 year periods in the past.

I had one of those flashes -- and it goes like this.

The stock market moves in a random walk.

The bond market moves in a random walk.

Therefore, whether stocks and bond prices go in different directions through various 3 years periods is . . . ta! da! --

Random.

I mean, at any given moment either market go only be going up, down or sideways. Therefore, sometimes they'll go in the same direction, sometimes they won't.

Yet this book wants you to adjust your allocation periodically on the basis of these random moves.

Look, the whole idea is that you're reducing risk because these two markets both are random, but can go in different directions. But it's obvious that sometimes they'll go in the same direction.

If you want uniformity, put your money in a passbook savings account in a bank.


asset allocation illusion
asset allocation illusion