Designating low-PE stocks as value is a flawed concept. On average and over many years, low-PE stocks have generated much lower EPS growth, were more likely to cut their dividend, earned lower returns on capital and had higher financial leverage. There’s more trash than treasure in the bargain bin!

Asserting that a company trading on a lower PE multiple is better value than one trading on a higher multiple assumes away any reason why two businesses should be valued on different multiples. In fact, it assumes that every business should trade on the same earnings multiple. This is clearly flawed. The appropriate measure of value is the price relative to worth.

A value manager selecting stocks based on low-PE has, perhaps unwittingly, chosen companies that are lower quality, lower growth and higher risk. The analysis that follows is based on the largest 2,000 publicly traded companies globally by market cap in 2014 and excluded loss-makers. Let's see what you got for your money over the last five years, based on starting PE quintiles at the end of 2014, starting with growth. The value quintile delivered the poorest EPS growth outcome over the last five years. 

Let's now look at quality, as measured by return on invested capital (ROIC) for the non-financial companies. Growth and quality and intricately linked - the higher the ROIC the more free cash a business has to reinvest and drive future EPS growth. We can see that the value quintile earned the lowest ROIC of the five groups. 

Let's turn to earnings risk. It's one thing to pay a low multiple of earnings, but if profits decline then you are perhaps paying a much higher multiple than you think based on prospective earnings. 

It should be no surprise that the value group was over-represented among the dividend decliners.

Turning to financial risk, we can see that the leverage ratio of non-financial companies was highest in the value cohorts.

Is it possible that we have chosen a five-year period that simply disadvantages low-PE businesses ? No. In fact, if use PEs at the end of 2007 (pre-GFC) and look at performance over the subsequent seven years, the differences are even starker.

The central flaw in the notion of value stocks or value investing is that it expresses value in terms of a PE multiple relative to an average. Whether a company is good value or not is, rather, a function of its share price relative to its worth. Companies differ in their growth prospects and their risk characteristics, and these differences should be reflected in their PE ratios. Choosing stocks based on a PE discount relative to a market average biases one's portfolio to lower growth, lower quality and higher risk businesses. 

Simon James

I think a lot of the other contributors on Livewire need to read your article twice, print it out and put it on the wall. The constant "is [X] a bubble" refrains, absent notions of reinvestment that artificially reduces the Earnings in PE, is very tiresome. Most don't even use a PEG ratio or EV/SALES comparison to try some form of apples with apples. Young companies shouldn't be measured like Old ones and vice versa.

Sam Green

I mean if you judge value purely by P/E then definitely. But what about PEG value? What about share price performance over that time? We know that low P/E companies are the 'dogs' and as such you'd expect lowest ROC, weakest growth, etc, that is why they are on low P/Es.

Stephen Arnold

Sam, thanks for your comment. Re PEG, earnings growth is not something that the market forecasts with any degree of reliability, whether over one year or longer, so the PEG ratio doesn't, in my view, improve the relevance of the PE ratio. In terms of share price performance, over the last five years performance has been directly correlated with starting PE, with value/ low-PE doing the worst. Measured over other time periods you will get different results.

David Lau

The article does present value managers as being a little simple minded. I'm sure good value managers also look at quality, earnings and other types of risk. At the end the article says "Whether a company is good value or not is, rather, a function of its share price relative to its worth". I'm sure most value managers would certainly agree with this, as would growth managers. The difference is that value managers are willing to look at companies that perhaps are structurally challenged on the basis that there's a right price for every profitable business. But they still think in terms of value relative to worth and won't just buy something because it's low P/E. In a few years it may be that value investors are seen as the geniuses and growth investors the dunces.