Reports on a company’s earnings, whether reported in terms of accounting earnings or cash flow, are by their nature historical. Investing, on the other hand, looks to the future. What economic profit is a company likely to create? What internal profit growth can it sustain, and what dividends and other returns will those profits afford to the owners of the business? And what do all these components imply in today’s market for the value of the company’s shares? A company’s history is often the best guide to its future, and deciphering past results and building them into a forecast of cash flows, and then into a prospective valuation, is the beginning of the investment process.

Flaws in Traditional Valuation Measures

Over time investors have devised many measures for turning up attractive stocks, by deriving the potential value of a share and comparing that estimate to its current price. One classic is a comparison of a stock’s price to its current or projected earnings per share—the price-earnings (P/E) ratio. Another is comparing the current price to the book value, which is the portion of a company’s net worth, according to the financial statements, that is attributable to each share. Both the ratios of price to earnings and price to book value per share can be easily calculated, and are widely used as shorthand measures of a stock’s price versus its true value.

But the P/E ratio and price-to-book value per share fall short for a number of reasons. One is that they try to compress information on a company’s past, present, and future into one statistic, at one point in time, and are too simple as a result.

Additionally, P/E or price-to-book ratios are not absolute measures: faster-growing companies typically show higher ratios than laggards, and in a favorable market ratios for all firms will generally be higher. In effect, each sector and market regime requires a different measuring stick, and price-earnings and price-to-book value ratios thus call for a lot of interpretation.

These criticisms are not new: in 1972, Jack Treynor, an investment scholar, former editor of the Financial Analysts Journal, and one of the great thinkers in the financial world, criticized both simple measures of valuation that took reported earnings without adjustment as gospel, and the investors who rely on them: “Needless to say, if the analyst insists on being provided with a single number so simply related to market value, then he is delegating away to whoever provides that number most of the real task of security analysis.”¹

But in our view, the underlying reason for rejecting such simple valuation measures is that they typically are derived from accounting measures of profits and assets, as reported and without adjustments. Even though corporations employ brigades of accountants to prepare financial statements, and turn them over to regiments of auditors who evaluate their fairness, reported results in many cases are simply not representative either of companies’ current financial health or their prospects.

Net book value, for example, measures the assets of the corporation net of debt, but it does so from the accounting perspective of historical cost—the amounts originally recorded for assets and liabilities. For assets in particular, net book value seldom reflects current value, and can thus present a misleading estimate—either overstated or understated, depending on the case—of a company’s value per share.

The P/E ratio is severely limited as well, even apart from the distortions created by the accounting rules that go into the calculation of the earnings figure. The P/E ratio compares earnings per share for a given period—typically the latest 12 months, or a forecast of the coming year—to a share’s price. This calculation yields a number that an investor has to subject to a complex comparison. Long ago—in the early decades of the last century—stock valuation was fairly simple, as investors tended to apply a standard price-earnings ratio of 10 times to all stocks. Then in the late 1920s, investors became more inventive with valuations, and higher and higher multiples were applied to stocks thought to have superior prospects for profit growth (in the market of the time, these included public utilities and chain retail stores).²

In the aftermath of the U.S. stock market crash in 1929, John Burr Williams, a Harvard-trained investment manager, developed a method for valuing stocks that was similar to the valuation of bonds—taking into account the stream of payments, that is, dividends, an investor receives into the future. This method has both greater practical and theoretical appeal than the P/E ratio. For one thing, it looks further ahead in time than the one-year forecast typical for P/E valuation, trying to take account of a changing future. For another, it addresses the valuation question more directly: the investor doesn’t receive earnings per share every year—just that portion that is paid out in dividends.³ Such dividend discount models are elegant solutions, but are highly detailed, and rather than cover them here, we present a review in Appendix B.

In summary, both P/E ratios and price-book ratios are badly flawed and long outmoded, so that many professional managers have replaced them with alternative measures. In the case of Epoch, those measures are based on free cash flow. We define free cash flow for a given period as the company’s cash available for distribution to shareholders after the investment in all planned capital expenditures and the payment of all cash taxes.⁴

Why is free cash flow so important? We’ll be discussing many aspects of companies’ cash flow, but to reiterate our earlier point, a company’s free cash flow is the origin of the value of a company. In the normal case, companies pay dividends from free cash flow, so understanding how much free cash flow a company might be able to consistently generate, and its ability to pay future dividends, is essential.

Accounting versus Finance: A Case Study

For an illustration of the shortcomings of accounting principles, and the potential peril of measuring a company’s performance solely on accounting earnings, we turn to a classic case study: “Feathered Feast,” appearing in the November-December 1993 edition of the Financial Analysts Journal, and authored by Jack Treynor. The case study is reproduced in Appendix C.

“Feathered Feast” tells of a fictional pension portfolio manager, Shephard Saunders, who is about to face his investment board to account for the fund’s unsuccessful investment in Feathered Feast, Inc., a chain of fast food restaurants, also fictional, where the signature dish is the Featherburger.⁵ Saunders typically preferred to invest in solid companies possessing tangible assets, but he had fallen prey to a broker’s story that Feathered Feast shares, already trading at a lofty 40 times earnings, were headed higher on the back of the company’s steady earnings growth (10 percent per year) and generous dividend payout (85 percent or so of annual net income). All was on track in December 1991, but in the following year profits rapidly turned to losses, due to cutthroat competition and excess capacity in the fast food business. By the end of 1992 the company was selling its trademark restaurants— 30-foot high chickens sculpted from stainless steel—for scrap.

The central question in the case study is what defines the value of those restaurants. The company recognized their cost in earnings through annual depreciation, spread over 12 years to match the term of lease contracts—an estimate thought to be fairly conservative, as the physical life span of the steel structures was 40 or 50 years.

By the end of 1991 Feathered Feast had invested over $700 million in its restaurants, and net of depreciation, they were on the books for about $600 million. For 1991, they earned the company a net income of $85 million, but as noted, by the end of 1992 the restaurants showed large losses.

Through this example, Treynor successfully contrasts accounting measures with financial valuations, particularly for fixed assets with long lives. In accounting terms, the restaurants seemed to be a valuable asset, with a remaining cost of hundreds of millions yet to be depreciated. However, their ability to generate positive net cash flow had disappeared in the competitive battle. Shephard Saunders, the case study’s portfolio manager, consults an accounting oracle during his post mortem investigation of the failed investment, who observes: “A brick wall is nothing but mud on edge if its capacity to render economic service has disappeared; the molecules are still there, and the wall may be as solid as ever, but the value has gone.”⁶

In a company’s financial statements, plant and equipment are stated in terms of historical cost, net of depreciation. “Net book value” is how that measure is known, but that is a misnomer: as any well-trained accountant would explain, net book value is not—and not meant to be—a representation of an asset’s market value. Instead it is the result of a systematic allocation against earnings of an asset’s purchase price over time. But what meaning does it have? Only by coincidence would it equal the asset’s economic value. Thus accounting measures are usually a poor foundation to use in estimating future value of an enterprise. Even so, out of habit or insufficient knowledge, many investors rely on net book value and other accounting measures as primary metrics of what an asset or company is worth.

Instead, as “Feathered Feast” demonstrates, the economic value of an asset, or an entire company, is determined by the future cash flow an asset will be able to generate. Valuing companies on their free cash flow offers a superior method, incorporating the entire value chain—revenues, margins, working capital requirements, and capital expenditures—into one consistent analysis.⁷ Thus, it is able to answer crucial questions on a company’s “story”: how and why have revenues grown over time? What has been the cost, in materials and capital expenditures, of producing those revenues? Has management’s past capital spending provided the company with profitable growth opportunities? Generally accepted accounting principles (GAAP) measures of earnings, in contrast, are highly condensed, and provide too little of the information needed for competent forecasts.⁸

The distinction between the opinion of accounting profit and the fact of cash creates opportunities for investors attentive to the difference, and is the basis for the investment philosophy of Epoch Investment Partners.