During a speech at the University of Florida in 1998, Warren Buffett (Trades, Portfolio) was asked for his thoughts on the stock market. His answer has always stuck with me:
“I have no idea where the market is going to go. I prefer it going down. But my preferences have nothing to do with it. The market knows nothing about my feelings. That is one of the first things you have to learn with a stock. You buy 100 shares of General Motors (GM) - now suddenly you have this feeling about GM. It goes down, you may be mad at it. You may say, "Well, if it just goes up for what I paid for it, my life will be wonderful again." Or if it goes up, you may say how smart you were and how you and GM have this love affair. You have all these feelings. The stock doesn't know you own it. The stock just sits there; it doesn't care what you paid or the fact that you own it. Any feeling I have about the market is not reciprocated.”
The point, which is particularly relevant for active investors who (by definition) presume some level of market inefficiency, is that your involvement doesn’t suddenly make the world rational. Looking to the short-term price signals from Mr. Market for answers is a fool’s errand.
This problem is exacerbated in an uncertain world. Even a mispriced bet or a great opportunity that should work in your favor on average can go against you. That's something you have to learn to live with if you're going to be an investor (particularly a concentrated investor). You can't stick your head in the sand and remain confident when the facts change. This is something you learn quickly when you play Texas Hold’em: Starting with a great hand doesn’t matter if the cards don’t fall your way (and it can be very expensive if you refuse to accept that fact). All you can do is assess new information as it becomes available and use it accordingly. If things don't go as expected, the best answer may be to throw in the towel and play the next hand.
But this can be easier said than done. In a recent appearance on the Capital Allocators podcast, professional poker player Annie Duke discussed research from Dan Kahan at Yale that shows people work incredibly hard to find ways to discredit any evidence that is counter to their world view. Importantly, his research shows that (contrary to what we may like to believe) being smart doesn’t help. The idea that you are intelligent enough to recognize these biases and remove them from your thought process does not mesh with reality. Mrs. Duke nailed the key takeaway: "The smarter you are, the better you are at slicing and dicing data to support your prior beliefs."
Clearly this is a real problem that we must try to get their arms around. So, what can we do about it? How can we stop volatility from pushing us away from a solid investment, while simultaneously ensuring we appropriately account for new evidence or data points that may invalidate our thesis?
I think there are two key ways this can be dealt with.
The first thing I try to do to is put my thesis in writing (this one of the many reasons I enjoy writing for GuruFocus). That’s important because our brains have a way of convincing us that the information we accumulate over time has been part of our hypothesis all along (I’ve showed some real-world examples of this in articles that reviewed my old investment journal entries).
Putting your thoughts in writing prevents you from editing your prior conclusions to mesh with the current reality. It’s a useful tool for avoiding the very real risk of “thesis creep".
The second thing I try to do is an extension of the first tool: I try to roughly quantify my arguments and assumptions. I think 21st Century Fox (FOX)(FOXA) offers a good example. When I bought the stock, I explicitly called out domestic revenue growth in the Cable Network Programming business. My research led me to believe that Fox could continue pushing for rate increases that more than offset subscriber declines. In terms of the financials, this was core to the thesis.
When FOXA shares were flat quarter after quarter in late 2016 and throughout 2017 despite (what I viewed as) solid financial results, this was a key data point that led me to conclude that the thesis was still intact and that I needed to stick to my guns and keep being patient (it’s still too early to call this a win, but I think recent developments support this conclusion as well).
Now let me present another example that has been less encouraging. I made a small investment in Chipotle (CMG) a few months back under the belief that the business had not structurally changed and that the historic financials were a useful indicator for thinking about the future.
For what it’s worth, I think Bill Ackman (Trades, Portfolio) may have been operating under a similar thesis when he made a large investment in the company. Here’s what he wrote about Chipotle in his third quarter 2016 shareholder letter:
“While traffic and sales have begun to recover, average unit volumes are still 19% below peak levels… While Chipotle’s reputation has been bruised, we think that with the passage of time and improved marketing, technology and governance initiatives, the business will not only recover but become much stronger. Chipotle’s sales recovery will be neither smooth nor predictable over the next few quarters; yet, we believe that all of the key drivers of Chipotle’s powerful economic moat and long-term success remain intact.”
He included one addition comment that I think is worth highlighting:
“We conservatively have assumed that profit margins will be at a discount to peak levels reflecting the cost of new food safety procedures as well as increased investments, offset over time by thoughtful management of overhead costs and increased operating leverage.”
Considering it has been more than a year since he wrote those words, let’s see how the results stack up. We'll start with revenues. This is from Chipotle’s 10-K for fiscal 2017:
“In 2016 we experienced lower total company sales than the preceding year for the first time in our history as a public company, and our average restaurant volumes declined from $2.532 million as of September 30, 2015 to $1.940 million as of December 31, 2017.”
As those numbers suggest, Chipotle’s average unit volumes (AUVs) are still down about 20% from the peak. The sales recovery Ackman is waiting for has not materialized, at least so far.
Then there’s profit margins. From 2010 to 2015, Chipotle reported average operating margins of 16.4%. By comparison, the company’s operating margins were 6.1% in 2017. Clearly, there’s a lot of work left to be done for Chipotle to get anywhere near peak levels.
When I invested in the company, I thought mid-teens EBIT margins were still sustainable if AUVs recovered (that’s a big if). I bet Mr. Ackman would've agreed (otherwise it's a lot more difficult to make the numbers work at $410 per share). Over time, I’ve become less convinced that this is an achievable target -- or at least not as a base case for an investment thesis.
First, there’s the ongoing cost of the food safety procedures that Chipotle implemented after the crisis. Here’s what CFO Jack Hartung said on this topic at the ICR conference in early 2016:
“Our margins are going to suffer. We think that on an ongoing basis, once all is said and done, that the ongoing costs probably won't be less than 100 basis points added to our margins and that would be mostly in food. There might be some in labor as well, and we don't think it will be more than 200 basis points added. And so, if you take our historic margins, if we get our sales back, we can get our historic margins back minus this 100 basis points to 200 basis points."
When Ackman talks about a discount to peak levels, I bet that’s what he’s (mostly) referring to.
But I don’t think that’s the end of it. I’m starting to think Chipotle may face some ongoing costs as a mature business that it has been able to live without in the past. For example, on the fourth quarter conference call, management indicated that they expect marketing and promotional expense to be “at or slightly above 3% of sales” in 2018 (despite this investment, guidance suggests store traffic will be roughly flat). By comparison, Chipotle spent $57 million, or less than 1.5% of revenues, on advertising and marketing costs in 2014. I think there’s a legitimate case to be made that Chipotle will look more like a traditional restaurant chain in the future. This may require a larger commitment than it's made in the past on marketing spend to drive traffic.
Without going into other line items (like labor costs), I think there’s sufficient evidence in the recent results to make me question whether mid-teens EBIT margins are attainable. I’m not sure that’s a deal breaker for the thesis, but it clearly suggests there have been some notable (and likely permanent) changes in the business over the past few years. I think it’s a good example of how quantifying your thesis can help you spot developments that impact intrinsic value.
That's a good stopping point for now. I look forward to your thoughts!
Disclosure: Long FOX, FOXA and CMG.
