The shorter the pay back period of the investment is for a person buying the share, the better. Thus the lower the PE of a share the better investment it is. The other side of the coin is that the higher the PE, the greater investor confidence in the share and its underlying future earnings.
This is simple, clean...and not always true.
Allow me to elaborate:
A negative PE is simply a function of the company reporting a loss. With plenty of profitable businesses on the JSE, a negative PE is generally something to avoid. This is perfectly obvious to most investors, but I include it here for completeness.
Share trading close to tangible NAV ("liquidation cases")
When companies are beginning to fold and / or future earnings include a high level of uncertainty, investors begin to value the shares on the basis of what they would get if the company liquidated. The tangible net asset value per share (tangible NAV) strips out all intangible assets and should be close to the liquidation value of each share, thus in these cases the shares ignore earnings and trade close to tangible NAV.
Perhaps the company had a massive one-off profit, but investors remain sceptical concerning its future profits. As the PE uses the historical earnings per share to calculate itself, this share would probably trade close to tangible NAV...yet at a misleadingly low PE. In this case, the PE is indicative of the risk of future earnings versus liquidation.
Also, sometimes the company's earnings are on the decline and investors see major problems looming for the company. Once again they would probably value it around tangible NAV, but in this case the PE would probably be extremely high (as a function of the earnings being extremely low). Once again, this is not a sign of confidence in the share or its future underlying earnings, but in fact, pessimism being supported by a base of liquidation assets.
Consider the external auditor's report where comments on the "going-concern" might be made in attempting to indentify situations like this.
(Short-term) Earnings Irrelevance ("dividend plays")
This is a subtle one and often mistaken, but some shares are valued by investors as what they call "dividend plays". In other words, the market is valuing the business based on the dividends it declares and not so much based on its earnings.
In the long-term earnings drive dividends and not the other way (you have to earn the money before you can pay it out), but in the short-term established dividend policies, shareholder expectations and internal politics that can influence dividends. Also, one-off "special" dividends can change share prices and unduly influence the market PE.
Imagine the case of a company who has healthy cash reserves and an established dividend policy, but profits have had a short-term knock due to a one-off costs etc. This company would probably declare a dividend in line with its historical dividends (as it wouldn't want to disappoint shareholders and it has the resources to do so) and its share would trade (relatively) in line with its historical dividend yield, despite lower earnings creating a higher PE.
Making use of the Dividend Model is perhaps a method of understanding market valuations of share like this.
The major flaw in evaluating shares based on their PEs is that the ratio is calculated based on historical earnings. Future earnings may differ significantly!
As the global recession hit the markets share prices all over the world slumped. Despite these drastic stock market movements, it took a while for corporate earnings to audited and then reported. Due to this corporate earnings reporting lag many stocks looked like they were trading at ridiculously low PEs just after the crash...albeit, based on pre-recession profits.
Forward PEs are used as a means of over coming this limitation.
Relative Overvaluations ("using the expensive to value expensively")
The PE of a sector or related company is often used to value a share. While this is an accepted valuation approach, it sometimes doesn't take into account that the original benchmark is in fact over-valued itself (or even, sometimes undervalued).
During the ‘tech bubble' if you had used the Tech Index's PE to value an IT firm it would have returned a ludicrous value, as the entire sector was overvalued.
Making use of the DCF Model is a one of the ways of avoiding this mistake when performing a valuation.
While the above list is far from complete and each share and its PE should be analysed on a case-by-case basis, these are a number of more common pitfalls investors face when interpreting market determined PE ratios.
So long as the pitfalls are born in mind, though, the PE ratio remains a powerful market-determined gauge of a share price and, in conjunction with other ratios, valuation techniques, and fundamental research can help guide an investor down the right path.