Someone who reads my blog emailed me this question:
“I had a question regarding return on capital employed. So I have read various views on what should and should not be included. I believe that Buffett believes you should look at return on tangible capital employed. I struggle with the exclusion of intangibles because some of them are very relevant. For example, if you have a company that is a serial acquirer surely goodwill should be included as a means by which management’s capital allocation is evaluated? What is the counterargument to my question? Also, what do you think should and should not be included?”
I agree with Buffett. It’s important to exclude intangibles. Capital allocation needs to be evaluated. But it needs to be evaluated separately from the business. The day-to-day managers of the business – as it exists today – have no control over the level of intangibles. In fact, intangibles are mostly just the amount that a company has “overpaid” for something in the past. The company could have really overpaid in an economic sense.
For example, if Newscorp buys Myspace it will pay above book value. This will be listed as goodwill. If Myspace declines in popularity in the future, that goodwill is then written off. But, the accounting goodwill added to the books when Newscorp bought Myspace was immediately irrelevant to both the managers of Myspace specifically and Newscorp generally. It’s unfair to judge the profitability of the business based on the price one person agreed to pay for it once.
You should evaluate capital allocation. I spend a lot of time doing that, but I separate capital allocation from the economics of the business. For example, I’ve written about ad agencies. Omnicom (OMC, Financial), Interpublic (IPG, Financial), Publicis (XPAR:PUB, Financial), WPP (LSE:WPP, Financial) and Dentsu (TSE:4324, Financial) have similar economic measures of return on capital employed. They have very different figures in terms of return on their retained earnings.
Capital allocation has been best at Omnicom (it mostly buys back stock), second best at WPP (it mostly acquires good enough companies at good enough prices), then it gets worse at the other companies. As an example, Interpublic made a series of mistakes in the late 1990s to maybe very early 2000s. We shouldn’t penalize Interpublic as it exists today for the level of goodwill it built up through overpaying for stuff back then. In many of these cases, the company writes off the goodwill. These aren’t the best examples, but they are the examples that come to my mind first.
What we want to know about is the incremental return on investment. If Interpublic was to grow 5% next year, how much of its earnings would it need to retain? The answer is not much. If we included goodwill, we might think that a company needs to retain more earnings to support organic growth than it really does. Now, yes, if Interpublic keeps making the same kind of acquisitions it did in the past, then the return on those acquisitions will be poor.
But it’s easy to test for capital allocation apart from return on capital. Here’s the simplest way to do it. Go to Google Finance. Type in “OMC,” then check the “IPG” box over the chart to add Interpublic stock to the chart. Then set the stock chart to “ALL.” Google Finance will give you its (not always entirely accurate) estimate of stock returns in Omnicom and Interpublic from 1978 to 2016. That’s 38 years of returns.
I can give the compound figures for each stock based on the Google Finance figures. According to Google Finance, Interpublic stock has returned 10% per year over the last 38 years. Meanwhile, Omnicom stock has returned 13% per year over the same 38 years. That doesn’t sound like much of a difference. Just three percentage points. But over four decades those three percentage points translate into about a tripling of your money in Omnicom versus Interpublic. You can make these stock return comparisons over shorter periods of time. I use a test like this all the time to see if the high returns on unleveraged net tangible assets that I see in my Excel history for the company have translated into actual solid returns for the stock.
In the short run, the stock price performance is not relevant to a value investor, but a company that is successfully earning returns on equity of 20% or more per year decade after decade really can’t post bad stock market returns over 30 to 40 years. It is hard for a stock to post even 25-year returns that are different from the underlying business. I can’t really think of any examples where this has happened. Right now, about the longest I can come up with is that some darlings of the tech boom in the late 1990s have had not-so-great stock performances over 15 to 18 years or so even when the underlying business did fine. It is almost impossible to find stock market returns of 25 years or more that diverge sharply from the performance of the underlying business.
I’ll give you another example. Right now, I’m reading the annual reports for Southwest Airlines (LUV, Financial). On its website, Southwest has annual reports (in PDF form) going back to the 1970s. It is a good stock for me to analyze. Obviously, I got interested in the stock when I heard that Berkshire Hathaway (BRK.A, Financial)(BRK.B) had bought shares in the four biggest domestic airlines. Of those four, Southwest has the best long-term performance. It has the best financial strength. There are a lot of reasons why – if I was interested in an airline – I’d be interested in Southwest.
I entered data from those annual reports into a big, giant Excel sheet running from the 1970s through today. The purpose of this Excel sheet is to measure things like sales, assets, equity, return on sales (margin) and return on equity. An especially important part of the calculation is looking at the variation in these figures. How much does Southwest’s operating margin fluctuate? How much does Southwest’s return on equity fluctuate? These are important things to know. If I ever did buy Southwest, it would be on the basis of EV/Sales or EV/Tangible Equity – not on the basis of this year’s most recently reported earnings. In the Excel sheet, Southwest has a long and unusually profitable history. Not just for an airline. It has a long history of high profitability for any kind of business. But is that true? Are we missing something?
Well, what we can do is go to Google Finance and look for the long-term stock performance. We can’t really compare Southwest to peers. There aren’t other long-lasting public companies in the airline business. They tend to enter bankruptcy a lot. They don’t stay public the way giant ad companies do. We can compare Southwest to the Standard & Poor's 500 though. Google Finance gives the compound rate of growth in the index as 8.7% per year over the last 38 years. So the S&P 500 did 8.7%. It gives the compound annual return for Southwest Airlines stock as 11.9% per year over the last 38 years. Southwest appears in my Excel sheet to be a more profitable business than most in the stock market.
Then when we look at the long-term record of Southwest Airlines stock versus the S&P 500, we again see Southwest creating more value. In fact, Southwest’s long-term stock market returns fall somewhere between Interpublic and Omnicom. The airline hasn’t done as well as Omnicom. But it has done better than Interpublic. That’s impressive.
It probably has to do with capital allocation. Southwest probably allocated more capital for its successful domestic airline business and didn’t use much of its free cash flow to do anything else. Meanwhile, Interpublic didn’t just buy back its own stock. The company couldn’t reallocate capital to growing its business organically. That’s not possible in the ad agency business because – unlike airlines – ad agencies don’t require any capital to grow. An ad agency can’t reinvest in its own business. Actually, it can, but it has to buy back its stock to do so.
That’s what Omnicom has chosen to do. Omnicom has allocated far more free cash flow to buying back its own shares than other ad agencies have. In fact, that’s the reason Omnicom has outperformed some other ad agencies' stocks over the years. It didn’t buy back as much of its own stock as Omnicom did. And – for an ad agency – buying back your own stock is almost always the best possible use of your capital.
So, yes, I do use net tangible assets. I never include goodwill, intangibles or any idle cash and cash equivalents. I only include what is needed to run the business this year.
When I wrote the newsletter, I included a whole section on capital allocation. I don’t mean to de-emphasize capital allocation by ignoring goodwill. Instead, I want to separate capital allocation decisions from business decisions. I’ll use another example from the past –Â Fair Isaac (FICO, Financial). I bought this stock just after the financial crisis. I had been aware of the stock for a long time. I like the core credit scoring business.
But the company had done an acquisition of which I wasn’t especially fond. That’s not because it bought something bad. It’s because Fair Isaac used stock to do the deal. Now, Fair Isaac was at that time a monopoly business. It had near infinite returns on capital. And it could grow at least a little. It’s a very bad idea for a company like Moody’s (MCO), Dun & Bradstreet (DNB), IMS Health (IMS), etc., to ever sell off a part of the business. The economics of those businesses are so superior to the economics of any businesses they’d acquire that it’s difficult to get any value from issuing your shares in a swap for some other business. Companies like that should simply never use stock in acquisitions and never issue shares. I don’t even like the idea of compensating employees using shares. This sounds extreme.
But if you start looking at the math from a long-term buy-and-hold-forever shareholder of one of these companies, you’ll see it’s really not that extreme a position. Omnicom was able to do probably 3% per year better than some of its peers simply because it devoted most of its capital to buying back stock while some of its competitors made acquisitions. Some of those acquisitions were smart. But you’d have to be very, very smart to make up for the loss of any shares of a business that can raise prices faster than its expenses and doesn’t have to increase assets at all to do it. If you have a business that can raise real prices and not lose customers, you shouldn’t issue stock in that business for any reason.
Here’s my point about Fair Isaac. The stock was cheap on a normalized free cash flow basis coming out of the financial crisis. But that’s not the reason I bought the stock – or, at least, it is not the only reason I bought the stock. I bought Fair Isaac because I believed free cash flow would be high relative to its market cap, and I believed the company would use that free cash flow to reduce its share count. I felt that if, say, I knew I’d be getting at least a 10% return in the stock, I also knew the company would be getting at least a 10% return on its purchases of its own stock.
In early 2009, it was possible to find lots of companies at 10% free cash flow yields. But I didn’t think most of them would take that free cash flow and use it to just buy more and more of the same company at that same 10% yield. For most stocks, I thought the 10% free cash flow yield was a ceiling of sorts. They’d probably allocate the rest at rates below that level. At Fair Isaac, I felt management was going to reinvest the free cash flow exactly the way I would if the decision was left to me.
That’s capital allocation, and it’s critical. It’s also why I’ve mentioned Bank of Hawaii (BOH) before. I believe that company will use more of its earnings to buy back its own stock than most other banks do. For me, that doesn’t make Bank of Hawaii a better or worse business. It just makes the stock a better investment. I separate the concept of goodwill form the concept of capital allocation. There are good businesses that don’t make good capital allocation decisions. I don’t want to confuse the issues because one day that good business might have a different management team that makes different capital allocation decisions.
Disclosure: No positions.
Start a free seven-day trial of Premium Membership to GuruFocus.
