Importance of a Multiple

Icarus is flying too close to the sun

I'm going to start my story with an example. Dominos Pizza. On the last day of 2009, Dominos was selling at five times its earnings. At peak it traded at just over 40x earnings. At this same time its earnings had multiplied tenfold. Over twelve years, there has still been a very high return. But the fact you had more than an eightfold increase on the earnings multiple speaks to the real magnitude of the return. 

I am going to make peace with a crowd here really quickly. I think that long-term investors have a lot to shake on. Ben Graham works well when you focus on periods in the around three year category; that number comes from Joel Greenblatt, who found that 36 months is the longest for a thesis to play out. Much longer and your IRR really suffers. But for those who buy and hold forever Charlie Munger types growth and value really start converging. Over one-year periods, the big thing is the multiple plays. Companies can experience radical multiple swings but the swings typically occur once, and typically within that three year period. The three year test goes to show can the business improve on margin and can it grow faster than market expectations. The five year period is where you see the beginning of the shakeout from any transformations or initiatives and over a ten year time horizon almost the entire return will be driven by revenue growth. That comes from research from Morgan Stanley that breaks down the return drivers of a stock over a given time horizon. I am typically in the one to three year camp and I will tell you why.

Because most companies don’t stay great investments over a period of ten years. Maybe five, rarely ever ten. Microsoft, Apple, Google, Nvidia, Tesla, Amazon. Great, the mag7. If your portfolio is lucky enough that you started your career in 2012 and your investments include the Mag7, Dominos, Monster, Fico, Adobe, NVR, Autozone or Oreilly, Copart, and Thermo Fisher, I will shut up. Because you had a diversified portfolio of these businesses and everything worked for you, the revenues went up and the multiples expanded. That's something exciting. Looking at Thermo Fisher on a pure earnings basis, you still would have had a 7x multiple on the stock. However, due to the multiple expansions, the return is a little under 13x. I am certainly on the lookout for these businesses. If I can get a PE of 25 trading at a PE of 12 or 15 but that can grow earnings at north of 12% per year over an extended period of time that would save me a lot of headaches. The reality is it is simply not that easy. Disney is a great example. The company has gone through multiple booms and busts. Buffett sold a stake in the 60s and did so again in the 90s. He talks about regretting that decision. The reality is he shouldn't. For all intents and purposes he bought and sold at pretty good times on both occasions. Disney did really well in the 60s but was in an uncertain place when Walt died. The business went on a huge decline for most of the 70s and 80s in terms of financials. There was some good times in the 90s with some iconic animated movies like The Lion King, The Little Mermaid, Tarzan, Aladdin, etc. but for most of the 2000s the house of mouse was declining again. It was really the transformations of the business through the acquisitions of ESPN(that would not start to print for years), Pixar, and especially Marvel that Disney would become the goliath we know today. But we saw how goliaths can fall when studios run out of ideas, get involved in culture wars, and streaming wars how companies can go into decline. This is not trading. These are ideas that take place over periods of at least 18-36 months. Charlie Munger’s beloved Costco is one of the patron saints of capitalism but there is so much that company has done right so consistently. That level of consistency is so incredibly hard to find. 

I should also mention that great companies that were once great investments can become bad investments. Buffett made basically a mid teens multiple of his initial investments in a decade compounding I think at around 40% per year. But since 98 coke has had capital appreciation of about the nominal rate of GDP growth plus the dividend. Now by this time Berkshire was running out of things to buy already so selling out then would have caused problems But in 98 coke dropped and didn't come back to its previous prices until 2012. 14 years of dead or negative money after that multiple is going to hinder your long-term returns. A lot of this is true with the Nifty Fifty companies. Not the Indian stock exchange. This was a list of companies that had powered the US stock exchange in the 60s to the early 70s. These were companies like American Express, Coke, Eli Lilly, General Electric, and 3M. But you also had plenty of companies that wound up getting acquired, like Gillette (By Procter and Gamble), Lubrizol (Bought by Berkshire), and Ponds (bought by Unilever), as well as mergers (Burroughs merged with Sperry to form Unisys and Bristol Myers later merged with Squibb) but some companies, are either nonexistent or obsolete like JCPenney, Sears, Xerox, Polaroid, Simplicity Pattern, S.S. Krege(Kmart) and plenty of others that are now tiny assets in larger companies. Held for too long the young and hungry entrepreneurs are going to take your lunch money.

There are plenty of businesses that people thought could never be disrupted, Walmart, GE, Exxon. But this also applies to tech, IBM, Intel, and Cisco. These businesses once held patents for incredibly dominant ideas. 

I am, in particular, pointing out Nvidia. With Nvidia, you have a business trading at over 30x sales. If the revenue does in fact increase by over 60% year over year the margins stay stable, and the stock does not move at all then the business will trade at just over 20x sales with a 40x earnings multiple. At that point, you might be able to argue that if you believe Nvidia can sustain 40% yearly earnings growth, which would have to come from stable margins at north of 40%, a net margin level that fewer than ten large caps in the world have. At this point, you can expect between 7-10% yearly returns. But God forbid that the AI trend does not become 5% of the whole US economy over the next decade you can Expect Nvidia's ten year return to be no greater than inflation. We can see that this happened with Cisco in the 2000s. The underlying business grew well, but it was in fact subject to competition, which killed growth and margin, and Cisco is still well off from its 2000s highs. Nvidia is a quality business with a great culture that is sure to grow fundamentally. But the valuation is so disconnected from these fundamentals that I am very comfortable staying away.

 This is why I prefer the Seth Klarman method of value investing. I look for a company with an upcoming value unlock. Typically the way that manifests is an event that causes the market to shift their perspective on the business. Perhaps the margins or revenue were better than expected, or maybe the market has their sector or theme “in play” regardless. The drivers of the return are primarily a multiple expansion. A P/E of 4 or 5 goes to 8 or 10, maybe 12 or 15, and you make a double or a triple. Now I am saying this as someone with very little actual capital. Greenblatt shut down his fund at a billion dollars (adjusted for today's money) Buffett shut down the partnership to pursue Berkshire full-time when the partnership had 100 million in today’s money but event driven is basically a 3-36 month strategy. Because we are nimble, look at the Buffett Partnership letters. He was not holding companies for five six years, he was holding them for two or three until the market re-rated them. He also invested when the expectations were low for the company. Nobody is surprised when the A student gets an A+ and they may even be incredibly disappointed if that student gets a B-. But if the D- student even gets a C- that might be a cause for celebration. I'm looking for convincing situations where C and D students might be worth Bs.