Brendan Byrnes: Another one of the things you look for is stocks of the highest quality so that can mean many different things. What does that mean as far as the book goes and as far as your research?
Wesley Gray: Yeah so one of the ways that we'll -- well first off let me got to the very beginning here because a lot of people are very familiar with like the Magic Formula, right? An eloquent concept, you've focused partly on price, partly on value, combine the two and you want to try to find the best -- you want to maximize that value price quotient by those firms, right? So you may get Coca-Cola (KO +0.99%) at 20 times earnings. That might be a great deal, where some junkie manufacturing firm at three times earnings is a terrible deal.
Now, before we even get the quality what we found is that framework is actually flawed empirically. Intuitively it makes sense but it just doesn't work empirically. The reason being is before you even think about quality, you've got to ensure that you're in the bargain bin.
Right? Because what we've found is that when you just look at price and quality simultaneously sometimes it forces you to pay a high price for a firm, say a 20 PE. And even though it may look cheap because the firm is of such high quality, it turns out that Warren Buffett is an anomaly here, just doing that on average is a bad bet.
So before we even look at quality it's so important that you first just get to the cheapest stuff. And we use just top deciles EBIT total enterprise price value. Now, once you've got there to make sure that you don't just focus on quality at the out-set, you're in the cheap stuff, now we think about quality.
And the way that we do quality is just how they talk about security analysis and how Ben Grahams been telling it for a hundred years. And what we look for is elements of the franchise or economic mote. So some of the things that we look for there is we just look at returns on capital, return on assets. But unlike many researchers they just look at the past year. The problem with that is these ratios are highly mean revert, right? So if I look at an oil and gas firm that made 100% last year that doesn't mean they're quality because they just may be in a highly volatile commodity business and next year may be negative 50%.
So what we want to look for is we look at the eight-year geometric mean return on capital and return on asset because what that's going to do, that's going to help you identify a good, steady path of actual generating value.
So, for example, like a Coca-Cola, if you look at their eight-year return on capital it's like 15%, the geometric mean. So that's something that we look at, long-term returns on capital.
Some other things that we look at is like long-term ability to generate free cash flow so we literally over a whole business cycle, which we just consider eight years, we add up all of the free cash flow you generated and just divide that by your total assets and rank that against the whole universe.
And the reason we do that is a lot of times firms say, well, you know like Jeff Bezos on Amazon (AMZN +0.20%) for example, he always says we're doing this capex investment not because we care about current earnings, because we think this is a high MPV project.
And that's fine, managers can say this, and we should be able to identify over an eight-year cycle we're going to see some of your capex investments actually come to fruition and so that's what that measures meant to capture; firms that actually invested in capex and were actually able to generate free cash flow.
And then the final aspect we look at is margins. Margins are another great way to kind of identify quality at least historically so if you're a firm like Apple (AAPL +1.05%) that can actually grow margins systematically over time that's clearly a potential indication of a mote.
And the second one is a firm like say Procter & Gamble (PG +1.11%). If you look at Procter & Gamble's margins over time it's 50, 50, 50, 50, 50 and so you think if microeconomics is true and competitions going to go after them, if they didn't have a mote it would start to come down and be more volatile but it's not, it's literally 50, 50, 50, 50 forever. So this would be another indication of like a mote.
So these are just subsamples, some of the things that we look at to try to quantitatively identify a firm that probably had, in the past at least, some sort of economic mote or franchise, whatever you'd like to call it.
