The relatively placid financial markets of a few weeks ago may have been the calm before the storm. During the time I was writing this piece, the S&P 500 dipped into negative territory for the first time this year, although it immediately bounced back somewhat. October has of course traditionally been a rough month for the stock market, and 2014 is turning out to be no exception.

Many market mavens are disturbed by the apparent collapse of oil prices. The important oil benchmarks, such as West Texas Intermediate and Brent crude, have been plunging through key support levels. Not that long ago, experts were worried about stronger economic growth bringing back inflation; now it seems like deflation may be a better bet.

In an uncertain market, it’s crucial to know the precise value of the company that you are investing in. The days of riding the general trend upward are over, at least for a while.

Valuing a stock — and buying below the estimated value — is the key to successful investing.  In How to Read a Corporate Balance Sheet, I discussed how shareholder’s equity (i.e., book value) on the balance sheet can be used as rock-bottom liquidation value for a company. Deep value investors like Benjamin Graham liked to buy stocks below book value, but such opportunities are extremely rare these days except in the high-risk areas of deeply-troubled, illiquid U.S. microcap stocks and Chinese stocks with questionable accounting. 

In his early days, Warren Buffett followed the “cigar butt” deep-value quantitative approach of his Columbia Business School professor and mentor Graham, but Buffett’s investment approach began to evolve closer to growth with the 1958 publication of Philip Fisher’s book Common Stocks and Uncommon Profits. Unlike stodgy Graham who was born in the “old world” of London, England and worked most of his adult life in New York City, Fisher was a free-wheeling Californian from San Francisco who took an adventurous and optimistic view of stock investing, not focused on current assets but future growth.

Unless a company plans to liquidate, which is rare, measuring a stock’s value based on book value is arguably irrelevant. A company that plans on continuing in business as a going concern will never sell its productive assets, so the proper way to value a company is on its current cash-generating ability and potential to grow that ability in the future. Although the future is unknowable, Fisher analyzed qualitative “scuttlebutt” (e.g., management expertise and integrity, along with the company’s competitive position) to make educated guesses. Consequently, whereas Graham focused on the “here and now” balance sheet, Fisher focused on the “forward-looking” income statement, which measured changes and trends in the balance sheet.

In honor of Philip Fisher and growth investing, let’s take a look at the income statement as a source of stock valuation for companies that are ongoing concerns and have no plans to liquidate. In theory, the best way to value a stock is to estimate all of its future free cash flows on an annual basis and discount  them at an appropriate annual interest rate to reach a net present value. Most DCF models calculate free cash flows for the next 10 years (based on a constant or slowly-declining growth rate) and then add a large terminal value based on a multiple of the 10th-year free cash flow to simulate in one final number the net present value of all future cash flows in perpetuity from year 11 to infinity. However, as I wrote in Value Investing and Value Traps: Separating Winners from Losers:

Performing a full-fledged discounted cash flow (DCF) analysis is not only time consuming, but requires an endless number of input assumptions that are likely to turn out to be wrong.

Many legendary value investors feel the same way about DCF. For example, Jean-Marie Eveillard said in a 2008 interview:

We never use discounted cash flows. Buffett does not consider discounted cash flow either, because the way things work, after 10 years you have a residual value which is often about half the net present value. So not only do you pretend to know what is going to happen over the next 10 years but even beyond. So we never do discounted cash flow, which I think is garbage. It’s as bad as the efficient market hypothesis.

Not only is future cash-flow growth uncertain, but so is the appropriate interest rate used to discount that future growth. In his classic investment book Margin of Safety, value investor Seth Klarman argued that calculating a  stock’s value requires “predicting the future, yet the future is not reliably predictable.” Consequently, one should be humble and conservative in one’s predictions and then discount those predictions by a substantial margin of safety in case the prediction is overly optimistic.

How large a margin of safety depends on the stock; for small-cap stocks with a limited financial history, a 30-40% discount makes sense whereas a discount of only 15-20% would be reasonable for a large-cap blue-chip stock with decades of financials. Considering the macro-economic backdrop is also important, especially today when interest rates are near record lows and corporate profit margins are near record highs. Stock valuations get crushed when interest rates rise and/or earnings fall. If your financial adviser claims that he doesn’t need a margin of safety, he is engaging in counterproductive “future babble” and I suggest that you find another adviser! As financial blogger Barry Ritholtz recently wrote:

Investing is about making probabilistic decisions with limited information about an unknowable future. The variables are well known, as are the possible outcomes. Anyone who claims to know the future, who says they can tell you what the economy will do, what earnings will be and, therefore, where the stock market is going is lying to you. Understanding the variables and valuation should help you make better investing decisions.

A much-simpler valuation method than DCF is to skip over the estimate of 10 years of free cash flows and just use a multiple of today’s free cash flow (or earnings or book value) to calculate a stock’s value. In essence, a multiple-based valuation just calculates a terminal value from the get-go, where one takes a “snapshot” value from the current year’s income statement and assigns a multiple to it to get the stock price.

For example, if a company’s earnings per share are $1.00 and the multiple of earnings you choose is 10, then the stock value would be $10 ($10 * $1.00). This begs the question how you determine what the proper multiple should  be. One possibility is to look to the past for guidance about the future. One could look at the average multiple of earnings the company or the industry has sold for in the past, but a company’s future could look very different from its past and a particular company’s business prospects could be very different from the industry average.  

Another possibility is to use the multiplier formula for the terminal value in a DCF analysis:  1/(cost of equity capital – growth rate). But, again, borrowing from a DCF analysis requires us to estimate cost of capital and a terminal growth rate, which is guesswork. Still, at least the formula illustrates the two factors that go into choosing an earnings multiple. Cost of equity is the rate of return demanded by investors to compensate them for business risk. The average cost of equity is around 10% (page 3) and the average long-term annual growth rate in earnings per share is around 3.8%. 

It makes sense that the higher the cost of generating equity returns, the lower the value of that equity (i.e., lower P/E ratio), and the higher the rate at which equity can grow earnings, the higher the P/E ratio. So, if you subtract average EPS growth of 3.8% from the average 10% cost of equity, the result is 6.2% and the reciprocal (1/0.062) is 16, which just so happens to be the long-term average P/E ratio of the stock market.

Here’s a crucial fact to remember: Using a p/e ratio for stock valuation only works for companies with reliable earnings.

In reality, Lynch’s valuation method is too simplistic because it assumes all companies with equal growth rates have equal business risk and that is not the case. One company currently growing earnings at 30% may face a high likelihood of an earnings deceleration in the near future,whereas another company growing at 30% may easily be able to maintain a high growth rate for the foreseeable future. Both companies would be valued the same even though one company’s earnings growth was much more sustainable and that wouldn’t make sense. 

Such is the flaw of using a snapshot multiple. I would only consider using a P/E ratio on stable stocks with a prolonged operating history and a modicum of earnings-growth reliability. For value investor Joel Greenblatt, reliable earnings are critical to his stock-valuation methodology:

I care very much about long term earnings power, not necessarily so much about the volatility of that earnings power but about my certainty of “normal” earnings power over time. My goal is to buy a company at a low multiple to normal earnings power several years out and that the company earns good returns on capital at that level of normal earnings. I usually just look at a simple multiple to normalized earnings. 

The stocks in my Roadrunner portfolio have all been carefully evaluated based on underlying value, not the whims of today’s volatile market.