As Benjamin Graham's book entitled "The Intelligent Investor" mentions, there are four rules for buying common stocks for the defensive investor. Before we list those rules, we should explain a few definitions. The Defensive Investor is someone who has little to no knowledge of investing but wishes to make a decent/average return. The defensive investor checks his portfolio at most once per year. He also does not do any research of companies or care about how the market is acting. At most he should expect a decent return, not a great one though. Defensive investors are typically people with full time jobs that aren't in finance (or perhaps are) and wish to simply put as much time and effort into common stocks as they do with a tax free savings account.

1. There should be adequate though not excessive diversification. This might mean a minimum of ten different issues and a maximum of about thirty.

A minimum of ten different issues is imperative to allow for the necessary risk to be somewhat minimal in the portfolio. The more, the better -- usually. However, when the number of different common stocks reaches around 30, the investor is putting too much time into his portfolio. 30 is not a strict number, but just a decent estimate.


2. Each company selected should be large, prominent, and conservatively financed. 

Large companies tend to be less volatile and have more steady dividends. Moreover, we assume the large company is an industry leader. A company is conservative if its common stock at book value is at least half of the total capitalization, including all bank debt. Also, to be "prominent", a company must rank at least among the first quarter or first third in size within its industry group.


3. Each company should have a long record of continuous dividends.

To be specific, a company should have a continuous dividend payment for at least 10 years. This ensures that the company is at least somewhat stable in its earnings and dividends, which is always a good sign.


4. The investor should impose some limit on the price he will pay for an issue in relation to its average earnings over, say, the past 7 years. We suggest that this limit be set at 25 times such average earnings, and not more than 20 times those of the last 12 month period. This would ban virtually all growth stocks.

Simply put, the P/E should be no more than 25 of average earnings (say over a 5 or 10 year period). The P/E should also not be more than 20 over a 1 year period. These keep your estimates conservative and make sure you don't overvalue companies.