The PEG ratio: this valuation shortcut doesn’t work

The price/earnings (p/e) ratio is a widely used way of valuing stocks, largely because it's simple. A company with slow-growing or very volatile profits should trade on a low p/e ratio, whereas one that can keep growing at a much faster rate merits a higher valuation. However, deciding exactly what p/e ratio is justified by a given rate of growth is not so straightforward.

To do this, many investors use the price/earnings to growth (PEG) ratio, which is the company's p/e ratio divided by its earnings growth rate. A fairly valued stock is assumed to have a PEG of one; hence a lower ratio implies a bargain. But does the PEG ratio really work? Let's take a look.

A crude shortcut

The PEG is easy to calculate, although there are a variety of different approaches: some investors use estimated growth for the next year or for the next five years, while others use the historical rate of growth. Many use a p/e ratio based on last year's earnings, but some use forecast earnings for the current year. For this article, we're going to use a p/e based on last year's earnings and a forecast annual growth rate over the next five years, which seems the closest to a standard formula.

Subscribe to MoneyWeek

Subscribe to MoneyWeek today and get your first six magazine issues absolutely FREE

Get 6 issues free

Sign up to Money Morning

Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter

Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter

Sign up

There's no doubt that PEG is simplistic. While a p/e ratio has solid foundations (see below), a PEG ratio is just a rule of thumb. And it has obvious flaws, as we can see by comparing it to a more rigorous method. To do this, let us assume that a company is set to grow earnings at a given rate, g, over the next five years. After that, the company starts paying out 50% of its earnings, and grows dividends at a rate of 5% per year. Finally, let's also say we require an annual return of 10%.

The red line in the chart shows the value we'd put on the stock if we took a PEG of one to be fair value. In this case, the p/e ratio we'd be willing to pay would simply be equal to g. For example, a company expected to grow at 25% per year for the next five years gets a p/e of 25. Meanwhile, the blue line shows the p/e we'd demand if we use the dividend discount model (see below) to value the stock. This line is very sensitive to the assumptions we use: changing them will move it up or down.

But that doesn't matter for our purposes. What's important is that the two linesare different shapes (a straight line andan exponential curve) and can give the same answer at only two points. That means the PEG ratio ruleis not a useful approximation for an accurate valuation.

It will tend to put a lower value on low-growth and very high-growth stocks and a higher value on stocks with growth rates in the middle. But even where it gives similar results, that's an accident rather than a valid estimate.

Not backed up by the facts

So PEG is not theoretically sound. However, sometimes flawed rules of thumb still work better than sophisticated models. So the real test is whether picking stocks with low PEG ratios has been a reliable way to beat the market. But a quick back test using Bloomberg data over the past ten years is not encouraging: it suggests that investing in the 20% of S&P 500 and FTSE 350 stocks with the lowest PEG ratios would have returned almost exactly the same as their parent indices. After allowing for costs, investors would have done worse.

There are few academic studies on the PEG ratio and those that exist are patchy. However, the most detailed ones also seem to indicate that PEG ratios add no extra value to the basic p/e ratio*. So despite the popularity of PEG ratios, both theoretical and empirical evidence suggests that they have little value as a valuation measure. Investors should not be tempted to use them as a substitute for thorough analysis.

Stick to the p/e ratio instead

One popular model for valuing a stock is the dividend discount model (DDM), also known as the Gordon growth model. This says that the price (P) of a stock is equal to the value of next year's dividend (D), divided by the difference between the rate of return the investor wants (r) and the long-term growth rate of the dividends (g). Expressed as a formula, that's P=D(r-g).

The DDM can be adapted to cope with more complicated situations, such as stocks that don't pay dividends initially or grow faster for a period of time before settling down to a long-term trend. But it is highly sensitive to changes in any of the assumptions and will give very different results for seemingly small changes.

Since it's difficult to forecast growth ratesover even short periods, it's wise to be sceptical about precise valuations based on long-term estimates. Nonetheless, the DDM is still a powerful tool not least as a way of checking whether a valuation looks sensible.

If we divide our stock price (P) by an estimate of the current year's earnings (E), we get the forecast (or forward) p/e ratio. Using our formula, we can see that (P/E)=(D/E)(r-g). The expression D/E is just dividends divided by earnings (the dividend payout ratio). So we can see that the fair or justified p/e ratio for a stock is linked to the DDM. Hence we can explore what assumptions may implicitly be priced into a stock's p/e ratio, helping us think about whether they seem at all plausible.

*Adjusting P/E ratios by Growth and Risk, Javier Estrada (2004).