5-Year Cash Flow

Ben Graham spends a good portion of the common-stock section of Security Analysis outlining proper analysis of the income statement. He warns of companies charging costs directly to equity accounts, completely bypassing the income statement and thereby inflating reported earnings; of companies taking huge reserves in one year with the intent of using the reserves to inflate earnings in subsequent years; of deferred charges that should really be expensed; and so on.

Since the small investor today must filter through thousands of stocks with the tools he or she has on hand, often with limited time since for most this is a part-time endeavour, I decided that in order to circumvent any tomfoolery going on in the income statement, I would use cashflow as my earnings proxy.

I start my cashflow with the total Cashflow from Operations as reported on the cashflow statement. From this I deduct any capital expenditures. Now because I want to compare the ability of companies to generate cash irrespective of their capital structure, I next add back the net interest expense. By net interest I mean the interest expense less any interest earned from cash balances. I also want to ignore differences in cashflow caused by different tax rates, so I remove any cashflows relating to income tax. This would be income taxes charged on the income statement adjusted for changes in deferred taxes and/or taxes payable/receivable. While it is important to recognize and consider permanent differences in tax rates and the value of tax assets, for the initial filter I feel it's better to remove these effects and consider them afterward.

Once I've made these adjustments (or come as close as I can since I don't have the resources to pull this information exactly; instead I use MSN's on-line financial data, which I can access for free. It often pays to be stingy in the investment game, particularly if you're a small investor) I have what I call CashFlow Before Interest and Taxes, or CFBIT. I look at the average value over the past 5 years, since that's how far back MSN data goes. As mentioned previously, Graham was adamant that several years' earnings be considered, not just the most recent year. He writes "The [earnings power] record must cover a number of years, first because a continued or repeated performance is always more impressive than a single occurrence, and secondly because the average of a fairly long period will tend to absorb and equalize the distorting influences of the business cycle." Good advice that's rarely followed by small investors.

So once I have my average CFBIT, I compare it to the firm's Enterprise Value (EV) to derive an Earnings Multiple. The EV is calculated as the total value of the outstanding shares at the current market price plus the book value of debt plus the value of any outstanding preferred shares, less any cash and short term investments. The idea of EV is to estimate what it would cost to purchase the entire firm so that all the CFBIT returns to you. Some analysts use the market value of debt in the EV calculation but since we're looking at this from the position of an equity investor, it makes more sense to me to use the book value. After all, the bondholders and other creditors would have first claims on the company and if you really did plan to buy the entire firm, you'd likely have to pay out the book value. Obviously there are exceptions but we want to keep this simple.

So with the Earnings Multiple we are halfway to coming up with a "cheapness rating". Next we turn our attention to the balance sheet.