I’m going to assume here that you are a value “investor,” not a value “trader.” You are investing over the kind of timespan that Warren Buffett (Trades, Portfolio) invests over (buy and hold forever) or maybe just the kind of timespan Ben Graham invested over (buy and hold for a few years).
If you look at the Graham-Newman letters to partners – they are really mostly just a list of positions – you’ll see the turnover rate is quite low. It may be 20% one year. It might be 40%. But it’s never 50%, 75%, 110% – like many mutual funds and many brokerage accounts are today.
I’m going to use five years as my standard for an investment. You can use three years. It doesn’t matter much to what I’m about to talk about. Either way, when you’re buying a stock, what the business looks like today doesn’t matter.
Let me explain that. The stock market is always forward looking. Let’s say Starbucks (SBUX) just reported earnings and talked about its rewards program and so on. The market doesn’t really care. Mr. Market doesn’t get himself worked up about what Starbucks just reported last week. Mr. Market cares about what Starbucks is going to earn next quarter and next year. But that’s the way a trader thinks. You’re an investor. Why does that matter?
Let’s say you are buying a stock today and planning to sell it in three years. You’re buying in 2016, and you’re going to sell in 2019. Now, I’d say this might just be a long trade. Or maybe it’s a short investment. Either way, what matters here? Is it what the stock is going to report in earnings in 2019?
Probably not. Especially not if we are talking about a big, liquid stock. There will be analysts covering this stock. The company may be giving guidance. In November 2019 – when you’re thinking about selling this stock you bought in 2016 – the market is going to be focused on 2020 and maybe even 2021. Mr. Market won’t care at all about 2019 anymore. Even if you are buying a stock today and only planning to hold it for three years and then sell – you should be thinking about what that stock is going to report in earnings four or five years from now. The present isn’t what sets the price at which you sell. The expected near future is what sets the price.
Some value investors – like Graham and Richard Pzena (Trades, Portfolio) – think in terms of “normal” earnings. If we were looking at a stock like Carnival (CCL, Financial) when oil is priced at $40 a barrel and yet we believe $60 a barrel oil is normal, we’d have to adjust that stock’s earning down to reflect the 50% higher fuel expense in a normal year. If we’re thinking about a lender to subprime car buyers, like America’s Car-Mart (CRMT, Financial), we might think that underwriting results will be unusually bad right now because competition in this space is unusually high.
Credit is super loose in that space. A few years from now, more subprime auto loans might be going bad. More lenders might be pulling out of this space. But America’s Car-Mart will still be there. And when credit is tighter, the company might be able to demand tougher terms. It may have loans with shorter lengths in months. That would lead to lower losses. We might normalize for something like oil prices or losses in industries where underwriting is cyclical (insurance, lending, etc.).
I’ve talked about normalizing bank earnings before. The two biggest banks in Texas are Frost (CFR, Financial) and Prosperity (PB, Financial). Frost is interest rate sensitive. Low rates hurt Frost. High rates help. Prosperity is interest rate neutral. That bank reports similar earnings regardless of where the Fed Funds Rate is. It might appear that Frost and Prosperity are similarly priced. In fact, they may be in terms of price-earnings (P/E) and dividend yield and yet really be differently priced in normal times. That’s because Frost would earn a lot more if the Fed Funds Rate was 3% right now.
If you think a Fed Funds Rate of between 3% and 4% is normal – as it has been over the last century or so – then you’d think Frost is much cheaper than Prosperity. This kind of normalization is easy for investors to understand. It’s especially easy for value investors, I think. It fits the Graham mold. But now I’m going to talk about something that might seem a little more speculative. We aren’t going to talk about adjusting results to reflect the historical average level of oil prices, loan losses or interest rates. Instead we are going to talk about what would happen if a company was run on autopilot.
I’m going to use a pair of businesses for this comparison. Both are publicly traded, and both have excellent business models. If you haven’t researched either of these companies yet, I encourage you to read their annual reports right now. These are two stocks you want on your watch list. They are interesting businesses. They are the kinds of businesses a long-term investor might want to buy and hold. That doesn’t mean they are priced right for a value investor now. I’m not going to discuss their prices. What I’m going to discuss is their inevitable futures.
The stocks are Cheesecake Factory (CAKE, Financial) and Howden Joinery (LSE:HWDN), a U.K. company. Cheesecake and Howden are both kind of category killers. Cheesecake is a U.S. restaurant chain. It’s not very big by number of sites, but each site is huge. Cheesecake has some of the highest square footage per store and some of the highest sales per square foot of any full-service restaurant you’ll ever see. As a result, Cheesecake has high sales per site. That’s fine. But what interests us in this discussion of a stock’s “cruise control” future is how quickly Cheesecake generates those sales.
The opening of a new Cheesecake location is an event. The company doesn’t need to spend a lot on advertising because its grand opening is the advertising. It has buzz. Landlords want Cheesecake to move to their malls. They ask to be the next mall to get a Cheesecake location. It brings traffic. Cheesecake can fill its restaurant from the first day. The location starts making money right away.
What’s the difference between what a Cheesecake will earn in the first month of its first year and the first month of its third year? Answer: not much. Cheesecake doesn’t tend to get more traffic in later months than it does in its opening month. In fact, when a Cheesecake Factory location first opens, it may actually be more full at first than it will be later. It opens the way a blockbuster movie opens. It has a lot of opening weekend buzz. And then it tapers off a bit from there and normalizes at a very high but slightly lower level. OK. What does that mean for growth?
A Cheesecake Factory location doesn’t generally have any traffic growth. It doesn’t seat more people this year than it did last year. The same-store sales growth of the location is going to be equal to menu price inflation. Let’s say Cheesecake raises prices by 2% this year. Let’s say it raises prices by 2% next year. And then it raises prices by 2% the year after that. Well then what will this location be making in 2019 relative to what it made in 2016 when it first opened? It’ll be making about 106% (a little more) of what it made when it first opened.
This isn’t something we even have to factor into our analysis much at all. The age of Cheesecake’s locations doesn’t matter. From the moment they open, they really only grow sales at the rate of menu price inflation. When we are researching Cheesecake’s growth prospects, we don’t really need to inquire much about whether the average location has been open two years, four years or eight years. In all cases, if menu price inflation is 2%, then same-store sales growth is going to be 2%.
Now, let’s talk about Howden Joinery. A new Howden Joinery location doesn’t open the way a blockbuster opens at your local multiplex. It doesn’t open like Cheesecake. Howden doesn’t serve the public at all. It only serves builders. It sells what builders in the U.K. need to install kitchens in homes. It has granite countertops and its own private label appliances and so on. It provides credit to these trade account clients. Each location is run rather independently. It has its own manager. It takes time to build up repeat business with these kitchen installers in the local area. There is no buzz.
A Howden Joinery location is like the rollout of a little indie film. It opens with a couple of screenings (customers), then a week later it’s in a few more theaters (has a few more customers), and it keeps growing the customer list. But it’s slow. There’s nothing high profile about the grand opening of a Howden Joinery location. This is important to the per site economics. A Cheesecake Factory is already firing at 100% of its “normal” sales capacity in the first month. Want to guess how long it takes a Howden Joinery location to hit 100% of its potential? Seven years.
A new Howden Joinery location usually loses money in its first year after opening. Then it reaches the break-even point sometime near the end of year 2. The location then starts contributing to the company’s earnings in years 3, 4, 5, 6 and 7. Now, let’s think about what this means for the corporate result. Let’s be somewhat conservative. We’ll assume a new Howden Joinery location loses money in year 1 and breaks even over all of year 2. That’s not the way the company explains it. It says a location starts breaking even by the end of year 2. And the company also says that the location keeps growing for seven years.
We’ll assume it hits max (real) sales potential on day one of year 7. If those are the facts – here’s how we can think about drag and acceleration from new locations. All year 1 Howden locations cause drag on the corporate performances. In any given year, Howden’s earnings are worse off to the extent it opens centers. Let’s say Howden opens 50 centers this year. If it opened 25 centers instead, it would report higher earnings for the entire year. The less Howden grows its store count, the better its present-day earnings picture will be.
Year 2 Howden centers are neutral as far as earnings. I think they also probably cause net drag. But we’re being conservative here. Now, what happens in years 3, 4, 5 and 6? We know Howden centers in Year 1 lose money. And they don’t make money in Year 2. They do make money in years 3, 4, 5 and 6. They also make money in year 7 and beyond. However, the centers are “mature” by year 7. They are profitable but immature in years 3, 4, 5 and 6.
What does this mean? It means Howden’s “cruise control” rate of earnings growth will be highest when it has the greatest proportion of centers in the 3-, 4-, 5- and 6-year-old age group. Actually, that’s not quite what it means. Because the rate at which very new centers switch from losing money to making money might provide a big corporate boost. And then there’s the issue of how many centers are being opened. Let’s find a simpler way to put it.
How can Howden report the highest possible rate of earnings growth right now?
It can stop opening new centers. If Howden has, say, 600 centers, and it opened 60 of those in the last two years – the worst thing for the company to do as far as this year’s EPS is concerned would be to open another 60 stores. The best thing would be to stop opening stores all together. The young centers have a high natural rate of same-store sales growth. The old stores don’t. Those 60 young centers might have same-store sales growth in the 5% to 10% range while old centers have growth in maybe the 0% to 5% range. New stores lose money. Stopping all new depot openings and letting your money-losing new depots turn into moneymakers will drive the best EPS growth.
Of course, Howden should never actually do this. GEICO loses money in the first year or two after signing up business. That’s how the business model of companies like GEICO and Progressive (PGR, Financial) work. They should never stop trying to sign up new customers. But we – as investors – need to remember that GEICO and Progressive and Howden – especially when they are growing quickly – are underreporting their future earnings in a way Cheesecake isn’t.
If Cheesecake stopped opening new locations, it would only grow EPS a little faster than menu price inflation. If Howden stopped opening new locations, it would grow EPS a little faster than same-store sales. At first – when a lot of its centers were still just newly opened – this EPS growth could be something like 10% a year for a couple of years without the company seeming to do anything.
In this sense, Cheesecake’s current P/E ratio is an accurate representation of its price. Howden’s current P/E ratio is not an accurate representation of its price. It’s not speculative to assume that all a company’s existing locations will “mature” along the lines its past locations did. If Cheesecake stopped opening new sites, its growth would limp along at the rate of inflation over the next five years. If Howden stopped opening new centers for the next five years, its EPS would grow at rates you’d normally expect from a growth stock.
That’s why you shouldn’t – as a long-term investor – look at what Cheesecake is earning today and what Howden is earning today. If you are buying these stocks now and holding them for three or four or five years – what matters isn’t what these companies will look like in 2016. It’s what they will look like in 2021. In other words, you need to adjust Howden’s centers for “maturity” the same way you need to adjust Progressive’s business to make sure you’re looking only at “seasoned” business.
New business is a drag at first for these companies. It only becomes profitable in later years. If you don’t make this adjustment, you’ll think that Howden is normally priced when it’s actually cheap. It doesn’t matter what Howden earns today. It matters what Howden would be earning in 2021 if it were in a “steady state” at that point. Don’t worry what a stock is earning when you buy it. Only worry about what a stock will earn when you sell it.
Disclosure: Long Frost.
Start a free seven-day trial of Premium Membership to GuruFocus.
