Walking The Dogs

A little year-end wrap up: The Dow Jones Industrial Average closed 1998 at 9181, up 16% on the year. The S&P 500 index closed the year up 26% while the NASDAQ composite index recorded a new high today, closing the year up 39%. Despite its ups and downs, a fine year to be an investor.

Today is the day followers of the Dogs of the Dow theory are supposed to adjust their portfolio for the year. This strategy, much touted by The Motley Fool is relatively simple. Let me quote the Dogs of the Dow web site: "After the stock market closes on the last day of the year, of the 30 stocks which make up the Dow Jones Industrial Average, select the ten stocks which have the highest dividend yield. Then simply get in touch with your broker and invest an equal dollar amount in each of these ten highest yielding stocks. Then hold these ten "Dogs of the Dow" for one year. Repeat these steps each and every year."

This theory runs smack into another stock market theory, the Random Walk hypothesis (also known at the Efficient Market Theory - EMH). This says that the fluctuations of stocks will resemble a random walk, i.e. the daily changes will look like random numbers. The reason for this is that at any given moment, a stock price reflects all that is known about the stock and the world. Changes to what is known (economic shocks, bad weather, new inventions) occur randomly, and the stock prices will reflect this. Unless you have inside information (not known to the market), all stock market strategies are doomed to fail, i.e. to work no better than chance.

Another way to show this is to consider what would happen if a (publicly known) strategy for beating the market were to exist. Naturally, people would start to follow this strategy to make money. As more and more people pile on, every time a "buy" signal occurs, the stock will jump upward. Every time a "sell" signal occurs, the stock will jump down. Those trying to buy will find they pay a higher price than they thought, and when selling a lower price is obtained. To get around this, people will start to buy early, when a buy signal looks about to happen, driving up the price, and vice versa on the sell side. Sometimes they will be wrong and have to bail out when the expected signal doesn't happen. These actions will erase the price differentials that make the system profitable.

This is happening to the Dogs of the Dow system now. I refer you to a fine article in the December 28, 1998 issue of Barron's, entitled "Bound for the Pound?" by Andrew Bary, for the details. (By the way, I subscribe to the online edition of the Wall Street Journal, which includes Barron's. Much cheaper than a print subscription, and you get the information right away, no waiting for the paper boy.) The Dogs of the Dow theory has gotten so much press that it is self-defeating. Not only articles, but also whole books, even a mutual fund, are marketed.

If you like the Dogs theory, my advice is to cook up your own variant. First, make up your own version of the Dow Jones average. Pick 30 blue chip stocks that are not in the Dow. What's a blue chip? To me, it's a large cap stock that's been around a long time (50 years at least), has a long record of positive earnings and has a long record of paying regular dividends. Then apply the Dog theory to your index. Don't re-allocate on a yearly basis - that would saddle you with short term capital gains taxes - choose a period longer than a year, maybe even longer than 60 months to take advantage of the coming lower rates for ultra-long-term capital gains. Don't pick the first of the month either, pick some other day.

Advocates of these simple theories often say things like "it's been proven to yield such and such a percentage over the last so many years." Well, I say that's nonsense. They didn't have the theory at the beginning of that time period, for one thing. And the record of that period does not reflect masses of people trying to follow the same theory. It's like saying "Microsoft has gone up so many percent over the last 10 years, so buy it now for a similar gain in the next 10 years." Good luck with that kind of thinking.