Graham’s Simplest Approach to Selecting Stocks

For over 70 years the works of Benjamin Graham have served as the bible for value investors. Successful money managers such as Warren Buffett and John Neff swear by the simple message put forth by Graham of looking for values with a significant margin of safety.

Reviewing the philosophy of successful investors such as Benjamin Graham can often prove enlightening. Graham’s philosophy continues to flourish primarily through two books—Graham and David Dodd’s "Security Analysis" and Graham’s "The Intelligent Investor." "Security Analysis" was first released as a college textbook; it examines the fundamental investment process by covering analysis of the economy, industries, financial statements, and bonds and stocks. McGraw-Hill owns the rights to this book and still publishes a number of textbook editions including the original 1934 version.

The first edition of "The Intelligent Investor" by Graham was released in 1949 and geared toward the individual investor. Grahams last revision was released in 1973. Subsequent revisions have left the original chapters written by Graham in the early 1970s largely unchanged, with additional commentary presented separately. "The Intelligent Investor" presents Graham’s basic philosophy on holding a mix of bonds and stocks and selecting stocks for both the defensive investor and the enterprising investor. HarperCollins Publishers offers several editions of "The Intelligent Investor," ranging from the original 1949 edition with a foreword from John Bogle to a revised 2003 edition with commentary by Jason Zweig.

Graham’s approach focused on the concept of an intrinsic or central value that is justified by a firm’s assets, earnings, dividends, financial strength and stability, definite company prospects, and quality of management. By focusing on this intrinsic value, Graham felt that investors could avoid being misled by the misjudgments often made by the market during periods of deep pessimism or euphoria. This contrarian view dictates that stocks will appear most attractive when they are relatively unpopular with the market. The selection process takes great conviction and discipline because the momentum of the stock market will seemingly be against the investor, and there may be no clear indication as to when the market will come around to agree with you. However, in Graham’s opinion the possibility of extraordinary gains only exists when the investor disagrees with the market.

In "The Intelligent Investor," Graham laid out specific quantitative rules to follow when selecting stocks for the conservative investor. These rules shaped the Graham defensive and enterprising screens featured on AAII.com and built into Stock Investor Pro, AAII’s fundamental stock research and screening program.

Recently, an AAII member informed us that Heilbrunn Center for Graham & Dodd Investing at Columbia University maintains an on-line archive of articles by and about Benjamin Graham (http://www7.gsb.columbia.edu/valueinvesting/research/schlossarchives). These articles highlight the fact that Benjamin Graham continued to study and test different stock selection approaches until his death in 1976.

An interview with Graham published in the September 20, 1976, edition of Medical Economics lays out a simplified value approach to building a stock portfolio that any investor should be able to easily follow.

Determining the appropriate price to pay for a stock is the critical issue for a value investor. Graham believed that at any point in time there are a large number of securities that are priced too high as well as a large number of stocks that are priced too low. Graham felt that if you put together a large enough diversified portfolio of low-priced stocks, then you could skip getting involved in the fundamental analysis of specific companies or industries.

The key is using a rule for purchasing stocks that indicates a priori that they are undervalued. Secondly, when focusing on a single rule to select stocks, you must purchase enough stocks to make the approach effective. Finally, a sell discipline must be established.

Price-Earnings Ratio

The price-earnings ratio, or earnings multiple, is one of the most popular measures of company value. It is computed by dividing the current stock price by earnings per share for the most recent 12 months. It is followed so closely because it relates the market’s expectation of future company performance, embedded in the price component of the equation, to the company’s actual recent earnings performance. The greater the expectation, the higher the multiple of current earnings investors are willing to pay for the promise of future earnings.

If the market has low earnings growth expectations for a firm, or views earnings as suspect, it will not be willing to pay as much per share as it would for a firm with high and more certain earnings growth expectations.

However, investors often pay too much for companies that appear to have the best prospects at the moment and react too negatively to companies considered to have the weakest prospects. This mistake tends to be a self-correcting process that value investors can use to their advantage.

Graham’s simplest investing approach relies on selecting stocks trading with a low price-earnings ratio. How low? Graham suggests one way to determine the maximum acceptable price-earnings ratio is to look at what high-quality bonds are yielding. If bond yields are high, you would want to select stocks selling cheaply—screening for relatively low price-earnings ratios. If bond yields drop, then you could pay more for a stock and therefore screen for stocks with a higher price-earnings ratio. Graham’s rule of thumb is to select only those stocks whose price-earnings ratios are less than inverse of double the AAA bond rate. You would double the bond yield and divide the result into 100.

At the time of interview, the average current yield of AAA bonds was around 7%. Doubling the rate gives you 14, and 14 goes into 100 roughly seven times (100 ÷ 14). The most Graham would pay for a stock in 1976 was seven times earnings. If a stock’s price-earnings ratio was higher than seven, it would not be included.

If the AAA bond yield drops to 6%, then the maximum acceptable price-earnings ratio is eight—double six is 12; 100 divided by 12 is around eight.

High-grade corporate bonds are currently yielding around 5.25%. This translates into a maximum acceptable earnings multiple of around 9.5 [100 ÷ (5.25 × 2)].

However, Graham set some absolute limits. He felt that one should never buy a stock with a price-earnings ratio above 10 no matter how low bond yields go. Conversely, a price-earnings ratio of seven would always be acceptable no matter how high bond yields may go. When calculating the price-earnings ratio, Graham believed it was important to relate the price to the historical trailing earnings, not projected earnings. While Graham thought items such as projected earnings were significant in theory, he felt it was more practical to use the observed historical earnings figure.

Financial Position

While Graham believed that low-price-earnings ratio stocks would help establish a pretty good portfolio, portfolio performance could be improved if investors also selected companies with a satisfactory financial position. Graham used many tests of financial strength, but favored a simple rule that a company should own at least twice what it owes. There are a couple of easy ways to test for this situation. You could compare the ratio of stockholders equity to total assets and require a ratio of a least 50%. Alternatively, you could look at the ratio of total liabilities to total assets and require a ratio of 50% or less.

A strong financial position provides maneuvering room as a company expands or if it experiences trouble. Graham used this simple rule because it considers all forms of liabilities.

Minimum Portfolio Size

Graham felt that when using this simple selection method investors had to hold a larger number of stocks. Lack of in-depth company analysis can only be overcome through a diversified portfolio. A portfolio of 30 stocks was considered the ideal minimum. If your capital was limited, Graham suggesting purchasing odd lots (less than 100 shares of stock) when building your portfolio.

Sell Discipline

Graham suggested that investors set a profit objective when buying stocks, and he thought that a 50% profit goal should provide good results. Under this rule, a stock is sold once it is up 50%.

Furthermore, a time limit must be set for a stocks maximum holding period in advance. Graham’s research indicated that a holding period of two to three years worked out best. If you establish a maximum two-year holding period, you would sell a stock after two years if it did not meet its profit objective.

Following the System

While Graham felt that this approach could produce consistent results over the long run, there was the possibility of poor short-term performance. Graham suggested that investors looking to follow this approach should be committed to a minimum five-year horizon. This means being prepared, both financially and psychologically, to handle a short-term loss. Graham pointed out how the strategy would have done poorly in the bear market of 1973 to 1974, but then would recoup its losses in 1975 to 1976. Overall, Graham felt that this strategy was capable of producing a long-term total return (capital gains plus dividends, less commissions) of 15% per year. Like all systems, it is important to trust it and let it run its course so that the statistical probabilities would operate in the investors favor.

Applying the Screen

We thought that it would be interesting to test how this simplest of Graham strategies would have performed in our mixed market environment of the last several years.

Stock Investor Pro, AAII’s fundamental stock screening and analysis program, was used to construct a hypothetical 30-stock portfolio at the start of 1998. Grahams sell and buy rules were then used to manage the portfolio over time.

Some slight modifications and additions to the screening filters were made in constructing the hypothetical portfolio. Starting with the December 31, 1997, edition of Stock Investor Pro, the first filter excludes non-exchange listed stocks. Over-the-counter stocks are typically smaller companies with more limited liquidity that may require additional analysis. The December 31, 1997, edition of Stock Investor Pro contained a universe of 8,261 companies; excluding over-the-counter stocks left us with 7,786 stocks.

Next, stocks classified as part of the Miscellaneous Financial Services Industry were filtered out. This industry classification is normally applied to closed-end mutual funds by Reuters—Stock Investor Pro’s primary data source—and the Graham strategy focuses on common stocks. This filter further reduced the list of passing companies to 7,663 stocks.

The next conditioning filter excluded foreign-listed stocks trading as ADRs. Many of these companies release their primary financial statements in a currency other than dollars and may have different reporting requirements than domestically listed companies. While values might exist within this group of companies, they typically require additional analysis to ensure that their financials and ratios are comparable to domestic stocks. Excluding these foreign ADRs reduced the list of passing companies by 145, leaving 7,518 stocks. Liquidity, or the ability to buy and sell a stock quickly and without substantially moving the price is an important consideration for investors. There are many possible screens for liquidity. The screen already required that a stock be listed on an exchange, but we also required that a stocks price be at least five dollars. This minimum price is used as a test for determining if a stock can be purchased on margin and was applied to the Graham screen to ensure minimum marketability and to avoid penny stocks. This filter reduced the list of passing companies from 7,518 to 5,763 stocks.

The two primary Graham criteria were then applied to this universe of stocks. The prevailing long-term yield for high-quality bonds at the end of 1997 was around 9.5%. This results in a maximum allowable price-earnings ratio of around five [100 ÷ ( 9.5 × 2 )]. However, Graham indicated that a stock with a price-earnings ratio below seven should always be considered, so seven was used for this screen. Only 215 companies out of the total of 8,261 companies tracked by Stock Investor Pro at that time had a price-earnings ratio of seven or lower. Adding this filter to the other criteria dropped the number of passing companies from 5,763 to 89.

Graham’s financial strength requirement was then applied to the screen. Graham does not want total liabilities for a company to be more than half the level of total assets. Stock Investor Pro reports on the ratio of total liabilities to total assets, so a criterion specifying that this figure be no more than 50% was added to the Graham screen. Around 3,500 out of the 8,261 companies in the Stock Investor Pro universe met this requirement. Adding this filter to the Graham screen left the 30 companies that were used to construct the initial hypothetical portfolio.

Testing the Strategy

To test the strategy, roughly equal dollar amounts were invested in the 30 stocks passing the Graham screen. The closing monthly price was used to determine the number of shares to purchase, and the date of purchase and cost was noted to determine portfolio sells.

Figure 1. Performance of the Graham Screen CLICK ON IMAGE TO SEE FULL SIZE.

Normally when we test a strategy, the screen is run every month and only those stocks passing the screen continue to be held, while new stocks that pass the screen are added. For this test, however, we held each stock in the hypothetical portfolio until it was up at least 50% on a split-adjusted basis during a month-end portfolio status check. If the stock was up at least 50% over the original month-end purchase price, it was sold and the proceeds invested in a company that passed the screen that month. Stocks were also sold if they failed to reach the 50% price appreciation objective after two years. We also sold stocks if they were delisted, did not meet exchange filing requirements or were acquired. We did not include any commission costs in our calculation, take any time slippage (time between analysis and actual purchase) into account, consider the bid-ask spread when buy or selling stocks, or account for dividend income. Stock positions were not rebalanced, so some individual positions became larger than average, and the reinvestment of losing stocks would lead to smaller-than-average positions.

The results of the screen are presented in Figure 1 and Table 1. As the chart and table show, the strategy got off to a terrible start—losing 27.1% in 1998. The market was focused on growth stocks, and the stocks initially selected by the strategy were out of favor and falling further out of favor. The year 1998 is the kind of year that would test your conviction in a strategy. As subsequent years highlight, the strategy righted itself and started to clearly outperform the S&P 500 index starting in 2000.

Portfolio Characteristics

The companies that make up the current hypothetical portfolio are presented in Table 2. The companies in the table are sorted by the date they were added to the portfolio, with the oldest holdings listed first. The portfolio consists of a diverse collection of firms.

As revealed by the portfolio characteristics in Table 3, the portfolio as a whole has a very low median price-earnings ratio, 7.2 compared to the 20.1 median value for all exchange-listed companies. All of the stocks added to the Graham portfolio possessed low price-earnings ratios when they were added to the portfolio, but price appreciation and/or earnings declines have pushed the ratio up for a number of securities. Janus Capital has the highest price-earnings ratio of 29.4, due to declining historical earnings per share since purchase. A couple of stocks currently have negative earnings, so a price-earnings ratio cannot be calculated.

The median market capitalization (share price times number of shares outstanding) for the stocks in the hypothetical portfolio indicates that the portfolio would currently be characterized as a small-cap portfolio.

Over 70 companies currently pass the simple Graham screen. The 30 stocks with the lowest price-earnings ratios are presented in Table 4. The table is dominated by companies in the iron and steel industry and shipping industry. In building portfolios, Graham emphasized diversification, so you would need to pass-up some of these companies.

Conclusion

The simplest of Benjamin Graham’s approaches to selecting bargain stocks illustrates that it is possible to beat the market over the long run with a very simple, but well-thought-out strategy. Graham’s approach reinforces the importance of establishing a sell discipline to take reasonable profits, and to let go of stocks if they do not pan out as expected. It is only possible to follow such a simple strategy when building a relative large portfolio of stocks. If you carefully analyze companies before adding them to your portfolio, it is possible to hold a smaller portfolio, but with a simple screen, Graham felt it was important to hold a larger number of securities. Most importantly, before following any strategy, you must be prepared emotionally and financially to deal with the downturns that will inevitably occur.