Someone emailed me this question:
The biggest short-cut is finding those situations where a high-quality business is under-reporting its current earnings for some reason. This is probably the area where I have had the most success. If I could only invest in one technique, it would probably be this one. Find great businesses that are temporarily reporting worse earnings than you expect them to have five years down the road. An example of a good one that I never bought was Greggs in the U.K. Greggs is a fast food company. It started as a bakery, but has moved to something closer to a sandwich shop. There is no perfect analogy in the U.S., because Greggs sells some things like savory pastries ("pasties") which are part of British cuisine but have no role in American cuisine. If you imagine a cross between Starbucks and Subway, that's not too far off what a Greggs is.
There are other differences, though. For example, Greggs historically hasn't had much seating. That's another U.K. versus U.S. thing. American fast food places have quite a lot of seating even if they mostly do take-away business. There are also differences in terms of the time of day when Greggs sells things. It doesn't do much breakfast business compared to what American investors would expect. This is because the away from home breakfast market in the U.K. is a lot less developed than it is in the U.S. People in the U.K. grab fast food for lunch much more so than for breakfast.
But, overall, we aren't talking about something unique here. It's not impossible for an American investor to understand what Greggs is and what it does. Finally, I should say there are a ton of Greggs locations in the U.K. It's the least recognizable name to American investors among the big fast food companies in the U.K., but that's because it's a U.K. based concept. By location, Greggs is actually big even compared to Starbucks ( SBUX ) and McDonald's ( MCD ). We're talking cheap, fast food. The concept is well known in the U.K. The operating margin is very predictable. This is something I look at a lot. In the case of Greggs, I had about 25 years of data regarding sales, operating profit, same store sales and the number of locations. I knew how much sales they were doing per location over each of the last 25 years. I knew what the operating margin of the business was and what the same store sales trends were. Greggs stock price had sometimes been at the kind of levels you'd expect for an established fast food concept. Generally, an established fast food business has a very high P/E ratio. But after the 2008 financial crisis, Gregg's shares declined.
About five years ago, the stock got really cheap. I read some articles about why this was. The articles said it was because of the kind of food Greggs served. It was out of touch with a societal shift in the U.K. - as there has been in the U.S. - toward greater concern with fresh, healthy food. I looked at the numbers, and I didn't buy that story. There might be a societal shift going on, and it might be bad for Greggs in the future. But, that's definitely not what caused the same store sales declines during the recession and immediately after. Those were clearly caused by a decline in foot traffic near Greggs locations. Fewer people were walking into Greggs locations. In my view, the reason fewer people were walking into Greggs locations was that fewer people were walking near Greggs locations.
Greggs had a lot of stores on what the British call the "high street." In theory, the "high street" is comparable to a "main street" in the U.S. In reality, it's different. The concept of a main street in the U.S. might've been important circa 1980 or earlier. But, a lot of big shopping malls and then "power centers" (places where you stick a Walmart, a Home Depot, a PetSmart, a Best Buy, etc.) have been built across the U.S. Also, population density is totally different in the U.S. and the U.K. So, Greggs locations were in places you wouldn't stick a fast food concept in the U.S.
That's the explanation my newsletter co-writer Quan and I came to. It's also what the company's management was saying. They said they were going to re-locate stores toward places between where people live and work. Put more of them near train stations and where commuters go. This makes sense. Because it's where you put fast food concepts in the U.S. You don't generally put fast food locations in shopping areas. I mean, I've seen a few. There's a Chipotle not far from me that is in a shopping area, but that's rare. There's also a nearby Starbucks. Both of them do a lot of business, and, yes, they get some foot traffic from the shoppers. But, what they really get is a lot of business from people driving on the main road that runs along the edge of this shopping zone.
This seemed like a surmountable problem to me. Greggs could always adjust its menu slightly toward healthy food if it wanted to. Most fast food concepts change their menus a lot over the years. People don't notice. But the McDonald's menu you grew up with is a lot less like the menu you see today than your memory probably tells you. That's because they never make the change overnight. They slowly test things out. Eventually you've got a place selling a lot of salads and yet you don't think of them as having made this shift. I figured Greggs could do the same. The whole reason for wanting to pick Greggs for the newsletter was simply this: the operating margin was below average. EBIT divided by sales was the lowest in more than 25 years. Historically, EBIT margin had been very stable at Greggs. There were a couple years of same store sales declines. The EBIT margin plunged to historically unseen levels. I figured that if foot traffic either improved near the company's locations or the company moved to locations where foot traffic was better. it would return to the old operating margin levels. That meant you got a double whammy.
The company's P/E was below average. The P/E ratio should have, in my view, been above average, because the "E" was cyclically depressed. This issue comes up a lot. With something like a fast food concept, you shouldn't value it on EV/EBIT. You should value it based on EV/Sales. If you can't predict the operating margin, you shouldn't buy the stock. But if you can predict the operating margin, as you almost always can with an established fast food concept, you should buy the stock based on the average historical operating margin times the current sales level. That's how I valued Greggs. I just took the 25-year average operating margin and then I applied it to the current sales level. The company looked like it was trading at about a 50% discount to what an established fast food concept normally trades for.
Greggs was a great business facing a temporary problem. The problem was similar to a cyclical one in that it was depressing earnings a lot more than sales. Another example, one that did go in the newsletter, is Hunter Douglas (HDUGF). Hunter Douglas is also a "great" business in the sense of its competitive position. The return on equity is less great. It takes more capital to run an integrated manufacturer of shades and curtains than it does to run a fast food chain. But, I thought Hunter Douglas should, in a normal economy, be able to earn about a 15% after-tax return on equity while growing in line with nominal GDP. A company that earns about a 15% return on equity and grows about 5% or 6% a year is a good business if it can be expected to earn that 15% with almost perfect regularity. I thought Hunter Douglas was such a business. The company had poor earnings but no losses in the period from 2008 to 2014. But that was the worst period probably since the Great Depression for housing related stocks. Here, again, the stock looked like it had a "somewhat" low P/E ratio. But, in reality, I thought it had a very low P/E ratio. We have the same situation as with Greggs. The company is out of favor. In favor, it should probably have a P/E in the 15 to 25 range. Instead it ends up with a P/E in the 10 to 15 range. Okay, so maybe a little cheap. But not a deep value stock right? Can a stock with a P/E in the neighborhood of 13 be a value stock?
It can if the 13 multiple is being put on a year that isn't the best in terms of earnings. That's what happened to Greggs and Hunter Douglas. These companies had lower than normal P/E ratios at the same time they had lower than normal margins. Think about that. It means their price-to-sales ratios were absolutely plunging. Sales are more stable than earnings. If you think you understand the competitive position of the company going forward, you should always focus on price-to-sales instead of price-to-earnings. I'm saying you should use price-to-sales when comparing one point in a stock's history to another point in that company's history. I'm not saying you should compare Luxottica and Village Supermarket ( VLGEA ) based on price-to-sales. That would be crazy. They are in different industries. But, don't use the most recently reported "E" as your input for the P/E ratio. That makes no sense. It will betray you at exactly the time you need an accurate reading.
A stock with an abnormally high or low operating margin will have a misleading P/E ratio. Greggs and Hunter Douglas were good stock picks because the market put some of the lowest P/E ratios in those stocks' history on them at exactly the same time those companies were recording some of their lowest operating margins. So, you had high quality stocks with historically low price-to-sales ratios. That's the best kind of investment you can make. If there was just one screen I could run, it would be to have a list of the most predictable companies with the strongest competitive positions that were trading at the lowest price-to-sales ratios in their history.
Price-to-sales doesn't work for everything. Quan and I also picked Frost ( CFR ) and Progressive ( PGR ) for the newsletter. We didn't use price-to-sales. In the case of Progressive, we used price-to-premiums, which is similar to price-to-sales. In the case of Frost, we used price-to-deposits or price-to-earning assets (at Frost, these two figures are almost identical). A bank earns its profits on its deposits. Deposits are the stable source of the company's long-term earning power. If I was analyzing a copper producer, I'd value that producer based on price-to-tons of proven reserves. If I was analyzing a timber company, I'd value that stock based on price-to-acres of timberland.
In fact, Quan and I looked at U.S. Lime (USLM), and that's exactly what we did. We didn't value the company based on reported earnings at all. Instead we looked at the company's proven reserves and then we made our best guess about how quickly the company could turn these reserves into output and we did a DCF type approach. The important part to that analysis is that we didn't use the most recent earnings. We didn't really care what the price of lime was this year. Nor did we really care how many tons of lime the company was selling this year. Instead, we started with the company's existing reserves. Then we came up with an estimate for the future (real) price of lime. We discounted future production to the present using those estimates. The process of converting present day reserves into future earnings is more complicated than using a price-to-sales ratio. But, it's really just another application of the same principle. We focused on the constant aspect of the company's business that would drive its normal earning power. Progressive's normal earning power is fueled by its premiums, Frost's normal earning power is fueled by its deposits, and U.S. Lime's normal earning power is fueled by its limestone reserves.
Focus on normal earning power. It doesn't matter what Progressive's combined ratio is this year. It doesn't matter what the interest rates paid on Frost's loans are this year. And, it doesn't matter what the price per ton of lime is this. What matters is the normal level. So, don't think in terms of P/E. Look for great businesses where the P/E is lying to you.
Disclosure: Long CFR.
This article first appeared on GuruFocus .