Someone who reads my blog emailed me this question:
“I wonder how do you evaluate a company? Based on P/E? DCF? EBIT? Others? When I invest in a net-net or a company with low book value it’s easy, but with others I’m not sure that I’m doing it right.”
The formula I use is always distributable free cash flow plus growth. If a company has a stock price of $50 a share and free cash flow of $3 a share – then it has a free cash flow yield of 6%. I do not want to invest in a company where I think the return will be less than 10% a year. In this example, I would need to believe the company could grow free cash flow per share by 4% a year. In theory, I would be fine with any combination that adds up to 10% a year. So, a 5% free cash flow yield and 5% growth rate is fine. But so is a 7% free cash flow yield and a 3% growth rate. The reverse is also fine. A 7% growth rate and a 3% free cash flow yield would be fine too.
The formula seems simple. It is in theory, but applying it is trickier. For example, I would probably have more confidence in a stock I was buying on the assumption of a 7% free cash flow yield and a 3% growth rate than a stock I was buying on the assumption of a 3% free cash flow yield and a 7% growth rate. It is probably easier to count on a 3% growth rate than a 7% growth rate, but it all depends on what company you are looking at. Companies with high levels of customer retention can often count on growth rates at least as high as inflation. So if you ever have the good fortune of finding a company with a 90% or higher customer retention rate and a 7% free cash flow yield – that is probably a stock you can comfortably buy.
The actual figure I look at is always adjusted for debt and normalized in terms of the business cycle. Plenty of companies can have a free cash flow yield of 5% to 10% if by “free cash flow yield” you mean the leveraged yield on the market price – not the enterprise value. It is especially common for a stock to have a high free cash flow yield when it is cyclical, but this is misleading. It is important to focus on the normal level of free cash flow. I will look at something like the 10 or 20-year average free cash flow margin (free cash flow divided by sales) and take the median figure in that series. Then, instead of using the current free cash flow figure for this year – I will take this year’s sales as if the level of sales is normal but replace this year’s free cash flow with a normalized figure. The normalized figure I use is: Current Year Sales * Median Free Cash Flow Margin = “Normalized” Free Cash Flow. This is a starting point. It is the number I use before starting to really research the stock. But I am not going to focus on a stock unless the current level of sales times the long-term average free cash flow margin looks good.
Sales are more stable than free cash flow. Free cash flow is not a stable number. It is an important number but it varies at least as much as earnings – sometimes a lot more – at most companies. I also look at the long-term growth rate. It is rare for me to consider investing in a company where the past record is not sufficient to justify the investment. I will often see a good past record and yet decide not to invest in the company because I am unsure that they can achieve the same free cash flow margin in the future. Likewise, a company may have a long history of strong growth – but I am not sure it will repeat that growth in the future. So what I am talking about here is just the starting point.
The starting point for me is that I must believe the stock’s free cash flow yield plus its growth rate may add up to 10%. If I do not think that is possible, I do not investigate the stock any further. Now, where this gets complicated is that some companies that do not look cheap could actually qualify as good investments under this formula. One example is Costco (COST, Financial). I do not think I would invest in Costco because it is hard for me to have confidence in any sort of offline retailer that sells more than just food. I could have confidence in a food store. When Quan and I were writing the newsletter, we picked three retailers: Tandy (TLF), Village (VLGEA, Financial) and PetSmart (no longer a public company).
Tandy has a lot of buying power in leather working accessories. It can get lower prices for anything it has contract manufactured on its behalf and for other things it buys than much bigger retailers. Costco, Wal-Mart, Michael’s, etc. do not have better buying power than Tandy within Tandy’s niche. So, Tandy is a special case. It dominates its niche.
Village is a supermarket. It has stores that average about 60,000 square feet. Village is focused on the New Jersey market, but the size of its stores is not that different from Kroger (KR) stores. Supermarkets are the business that is potentially the most insulated from online competition. A supermarket offers a wide selection of items you can buy in just one short trip. Unlike a Costco or Wal-Mart, a supermarket is very close to its customers. Very few Americans drive more than four miles to get to the supermarket they use most. It is difficult for an online retailer to compete with a supermarket as the main shopping destination for something like a family in a suburban area. Online retailers can certainly compete with supermarkets for customers like single people in cities. It would be easy to win that kind of business. So, a company like Village that runs big stores in the suburbs is potentially a durable business. I would consider it.
We were going to pick PetSmart. However, the company went private before we could release an issue on it. Quan and I did a lot of work trying to figure out if it is possible to sell dogfood online cheaper than you can sell it at PetSmart. We decided it was impossible. Some companies – like Amazon – will sell dogfood, but they must be doing this for some marketing reason (like pleasing Amazon Prime customers) rather than because they can offer low prices and earn profits while selling the same dogfood as PetSmart. So, I have enough comfort with retailers like Tandy, Village and PetSmart to consider buying them. However, I do not have enough comfort with a retailer like Costco. I cannot imagine ever buying this stock, but other people could. Charlie Munger (Trades, Portfolio) certainly could. So, we will look at Costco and discuss whether it scores well on the formula I use to decide how cheap a stock is.
GuruFocus does a calculation for a “Yacktman Forward Rate of Return”. The “Yacktman” in that formula’s name is Donald Yacktman (Trades, Portfolio). He runs a mutual fund. GuruFocus says that Costco has a slightly higher than 10% Yacktman Forward Rate of Return. This uses a normalized free cash flow number instead of the current number. One can sometimes approximate – very quickly – what a company’s distributable free cash flow might be by looking at how much the company spent last year on dividends and share buybacks and then divide that sum by the share price. But, again, that is a leveraged number. I want to look at normal free cash flow (as I define it) and then divide it by the company’s enterprise value instead of the company’s market cap.
Basically, my approach is Yacktman’s approach. You can look at what GuruFocus has to say about Yacktman’s Forward Rate of Return approach. That approximates the theoretical approach I would use. I do not think price-earnings (P/E) is important. EV/EBIT is a useful figure, but I think it is most useful when comparing one company to possible peers and to itself in the past. It is helpful to know how expensive this stock is relative to other stocks in the same industry group. It is also helpful to know how expensive this stock is relative to what it traded at in the past. But is EV/EBIT more useful than sales to enterprise value?
I do not think so. If you understand a business, I think the best metric to look at is enterprise value to sales. When I talk about Frost (CFR, Financial) – I do not talk about the stock’s price-earnings ratio. I talk about the stock’s price-to-deposits. Deposits are a more stable number than earnings. Earnings in any one year are not important, long-term average earnings are what matters. And knowing what sales are today – or deposits for a bank – is probably going to be more useful than looking at earnings. I do not think it is a good idea to focus on actual reported earnings. For very stable businesses, it may not matter. I think it is always better to look at 10, 20 or 30 years of the company’s free cash flow margin and then look at this year’s sales. Even if you want to use EV/EBIT – I would not use this year’s EBIT figure. Instead, I would look at EBIT/Sales over the last 10, 20 or 30 years. Then I would look at this year’s sales and multiply it by the EBIT margin. I think that is a much better approach than using any one-year figure for EBIT, earnings or free cash flow.
A lot of investors – especially value investors – rely on earnings figures. They may prefer free cash flow, but these items are way down the income statement or the cash flow statement. Companies can earn very different amounts on the same level of sales, assets, etc. from year to year. You want to ignore these fluctuations. So, you always need to come up with a normalized number.
You also always have to do a free cash flow calculation for yourself. A lot of people skip this step. When I am talking about free cash flow – I do not want to use a number obtained from a website. The free cash flow numbers seen on websites are not a substitute for doing the calculation yourself. That is the problem with the approach outlined here. The Yacktman type approach is the one that makes theoretical sense. But both “normal” free cash flow and the expected growth rate are estimates on your part, you have to calculate them yourself. So, you will never be as sure of yourself while doing these things. You will never feel as certain that you have done the calculation “right.”
The appeal of something like the net-net approach is that it is easy to measure. The problem with the net-net approach is that it is not measuring the things that matter most. The things that matter most are free cash flow yield and growth in the business, but those things are hard to measure. So you will always feel more alone in your calculation than you would when doing a Ben Graham type approach.
With a Ben Graham approach, you have confidence that you did the calculation right. However, with each individual stock – you are going to have less confidence that the calculation matters. A single net-net pick is not reliable. A group of 20 net-nets will perform well, a group of 10 net-nets will probably perform well, but a single net-net is unpredictable. It could give you a 200% return or it could give you a 100% loss. If you are right about a Yacktman type investment – you can buy and hold it forever.
If you feel comfortable with Costco, you can buy it now. According to GuruFocus’s Yacktman Forward Return Rate, COST is now priced to return 10% a year. So, you can buy it today and – if that figure is predicting the future correctly – you can keep making 10% a year in Costco indefinitely.
Start a free 7-day trial of Premium Membership to GuruFocus.
