Security analysis involves analyzing the business, its operating results and financial position, its strengths and weaknesses, opportunities and threats, future estimates with certain assumptions, comparisons with similar companies or the same company historically, and a judgment of the security’s attractiveness.
Graham immediately cautions against focusing on the future estimates of growth and sales, as these are the least reliable figures we have and are most vulnerable to miscalculation. These estimates are also a large factor in the high P/E assigned to so-called growth stocks, which, as we learned in previous chapters, tend to fall the hardest.
The interest coverage ratio is the chief factor in assessing the safety, or quality, of a bond. For preferred shares (which are subordinate to bonds), you must look at both the interest + deferred dividend coverage ratio to determine the safety.
Graham suggests applying a threshold you feel comfortable with (generally 4 - 7 times for EBIT) and deriving the average results for 7 years, while also considering the minimum coverage shown.
The analyst should also consider:
- Business Size - Larger indicates the company has the ability to ride through poor economic times
- Stock/Equity Ratio - The more common stock a company has used for financing vs. debt, the more likely the debt will be repaid (since the common stockholders only get paid after all the debt is repaid)
- Property value - The greater the realizable value of the company’s assets vs. the amount of debt, the more likely the bonds will be repaid in full.
To analyze a common stock, you must come up with a valuation of the share and compare to the current price. In valuing the share, you must estimate the average future earnings and discount them back. To estimate average future earnings, you look at average past data for revenues and expenses and project them into the future with sound assumptions as to the change from the past. These assumptions are based on expected growth in the general economy and the industry and company in particular.
Not all earnings are alike. Two companies may have the same earnings, but wildly different valuations because of a number of factors affecting the discount rate:
1. General Long-term Prospects - How does this company grow? Beware companies that grow by acquisition only or have excessive cash from financing activities (e.g. WorldCom), or companies that rely excessively on a single customer. What are the company’s competitive advantages and are they sustainable?
2. Quality of Management - this is a really nebulous factor because there are no objective, quantitative tests of managerial competence. Also, it is already considered in the past performance and future estimates and long-term prospects. Graham says it should not factor into the discount rate again!
3. Financial Strength and Capital Structure - Companies with lots of surplus cash and no debt are preferred if all else is equal (e.g. share purchase price). Look at Net Income + non-cash expenses like Amortization & Depreciation.
4. Dividend Record - A long, uninterrupted record of dividend payments indicates high quality.
5. Current Dividend Rate - What is the ratio of dividend payout to reinvestment? Compare to what the company is using its reinvestment money for - if it is investing in low-return projects, then perhaps this is a sign of management ego and empire building rather than sound investment policy. A good, current article on this topic by my friends over at Barel-Karsan covers the recent Best Buy dividend revocation decision in favour of a low-ROI investment.
Discount Rate for Growth Stocks
Graham suggests a straightforward formula for valuing growth stocks:
Value = Current (Normal) Earnings x (8.5 + twice the expected annual growth rate over the next 7 - 10 years)
This can be rearranged to find the implicit annual growth rate in the market’s pricing.
This formula will lead to a valuation on the conservative side, which introduces a margin of safety.
Graham’s Suggested Practice for Analyzing Common Stocks
First, find the past-performance value - how much the stock would be worth, absolutely or as a % of the market - if we assume that its relative past performance will continue unchanged (using the previous actual average 7-year growth).
Second, consider how much this should be modified due to future expectations.
This process is elaborated on in future chapters with concrete examples.