Our investment strategy is based on the philosophy that the best way to “beat the market” is to be safely protected when it goes down, not dangerously exposed when it goes up.

This means always thinking about the unlikely and unpleasant things that might happen rather than the most likely or most hoped for pleasant outcome. A diversified portfolio that takes account of all possibilities is like a household that is always paying a fire insurance premium. Most years the premium reduces income; but when misfortune strikes the insurance provides protection. Misfortune strikes the stock market far more often than fire strikes houses.

The insurance premium can take the form of investing in cash or even bonds in an “all equity” portfolio, or more cash and bonds than normal in a balanced portfolio. It can also mean reducing investments in expensive and fashionable sectors of the markets and increasing their opposites, or simply reducing the average risk of individual holdings.

We focus on well-managed companies with sustainable business practices. Such companies should have high returns on their capital and highly motivated and innovative workers. Often these companies are well positioned to grow rapidly and share the benefits of their growth with investors. Indeed, prospects for a growing final market or a growing share of the market are important investment criteria for us. We prefer not to commit ourselves to owning a “growth stock” portfolio, as the term is understood in the investment business. However, the characteristics just described lead us in the direction of owning many companies that are likely to grow faster than their industries or than the economy.

Before purchasing a stock, we calculate an expected return over the next four to ten years. We try to limit new purchases to stocks with expected returns that are at least 5% a year greater than a comparable maturity U.S. Agency bond. We sell a stock when the projected return is less than or equal to the yield on a U.S. Agency.

We calculate risk-efficient portfolios across sectors, asset categories and countries based on subjective scenario forecasts. After many decades of doing this with complex quadratic programming systems designed to calculate the efficient frontier, we now use more intuitive techniques which are less restrictive, able to incorporate more complex ideas, and faster and easier to employ.