Quarterly Newsletter – July 2005

As you know, our method combines two basic ideas – diversification and rebalancing. Diversification means owning many different asset classes – with the expectation that some will do better than others. Thus the return of the total portfolio should be smoother (less risky) than the return of its separate components. Rebalancing means readjusting the components of the portfolio back to their desired percents – which again reduces risk as it prevents any one asset class from becoming too large a portion of the whole.

In our last letter we presented the results of our investment method over the past ten years. We observed that diversification and rebalancing was successful. Risk had been much reduced while return was not sacrificed. In this letter we will extend the comparison. Using data going back to 1955, we compared the results of a diversified and rebalanced portfolio to the S&P 500 (see page three for details of the data) using rolling ten year holding periods. We began the study in March of 1955 because this is the first time for which monthly return data is available for non-US markets. This provided a much broader result than using only the last ten years.

Rolling ten year holding periods look at the investment results from March 1955 through February 1965, then April 1955 through March 1965, and so on. There are 479 such ten year holding periods between March 1955 and December 2004.

The accompanying chart presents the results. It is divided into four sections, each with two columns. Each column displays data for one of the two investment methods under study. The first three sections display, respectively, the maximum, average and minimum annualized returns for each method for the 479 ten year periods studied. The last shows the standard deviation of the ten year rolling returns. Standard deviation measures the degree to which the data points making up a series differ from the average of the series. It is a measure of dispersion, volatility or risk. In the case of investment returns, all things being equal, a lower standard deviation is better than a higher one.

The results show that in all three cases, i.e. maximum annualized, average annualized and minimum annualized returns, the undiversified method had the lower returns. This is striking because investing using the S&P 500 as an index is pretty much the “gold star” method – a method that very few actively managed mutual funds manage to exceed. And yet, diversified investing has done better. (The returns of the diversified model were reduced by 1.3 % per year to cover advisor and fund management fees, and also adjusted for estimated transactions costs. The S&P 500 results were not adjusted for fees at all. Please see the disclosure pages for a description of the model used and its limitations.)

Now look again at the two risk measures – minimum annualized returns and the standard deviation. Not only did the diversified and rebalanced portfolio deliver higher maximum and average returns, it did so with lower long-term volatility and a better minimum annualized return. This is a remarkable result. Better returns with less risk.

The primary reason in our minds for diversification and rebalancing is risk reduction, not increased return. Although our method has outperformed the S&P 500 for most of the ten year periods in the study, it would not have outperformed some other asset classes, such as the riskier U.S. small capitalization value stocks. But owning a portfolio tilted toward riskier assets is dangerous. The small stocks fell more than 70% in 1931. The large growth stocks fell more than 60% during the market collapse that started in 2000.

We compare our method to the S&P 500 because that is a benchmark that many investors use to compare performance, and they do so because owning only the S&P 500 is a level of risk they might imagine taking. In the time period beginning in 1955, diversification and rebalancing do well when compared to that benchmark.

In this study, the diversified portfolio is rebalanced back to the desired tolerances each month. This means each ten year period studied begins with the same portfolio structure. It also, however, means that any benefits that may accrue to our general method of less frequent rebalancing are not shown. This is a study of diversification, maintained through time by rebalancing. It is not a study of the benefits, or harms, of less frequent rebalancing.

We believe this study shows that the results of the past ten years were not a fluke. Diversification pays. Rebalancing pays. They pay in a smoother ride, and in peace of mind. Of course, the past is not a guarantee. Perhaps ten years will not be a long enough holding period in the next market bubble. We just don’t know. And that is the point. The future is unknowable. Given the risks, we believe our method is the best way to invest.

With this letter we are enclosing our Form ADV Part II and our Privacy Policy statement. The securities laws require us to send the Privacy Policy once a year, and to offer to send the ADV. We are sending both. Please call with any questions.

Jim & Dean