Amazon and Damodaran prove valuation is hard… And not always accurate
– David Wanis, May 2019
Investing is hard. But investing is hard in a different way than mathematics or physics is hard. In science, difficulty can be overcome by those of superior intellect, who know the formulas and how to apply them to the science at hand. Investing requires a different type of intelligence, and previous Amazon valuation attempts shows just how difficult it can be.
Although Amazon (AMZN) listed in 1997 and its financial performance has always attracted controversy, AMZN caught the attention of the market in a big way in 2014 given the divergence between the reported financials and the performance of the stock. What happened over the next five years should serve as a reminder to any investor who believes knowing a valuation formula means knowing the investment outcome.
The punchline to the story is the disclosure that sometime in the first quarter of 2019 – probably at prices between US$1,400 and US$1,800 per share, Berkshire Hathaway purchased shares in AMZN for the first time. Not only that, at the AGM in early May, Warren Buffet disclosed that although he wasn’t directly responsible for the purchase, “The people making the decision on Amazon are absolutely as much value investors as I was when I was looking around for all these things, selling below working capital years ago.”
But let’s go back and look at why in 2014, AMZN caused such debate.
Here was a stock that defied many of the previous models of how shareholder wealth is created and measured. The bull case was that the company chose to reinvest its entire profit margin (from businesses that were very profitable) into new growth opportunities of which there were many. At some point in the future, the company would release the economics to cash flow – remembering that since day one, CEO Jeff Bezos has measured value as free cash flow per share – but until that point there was too much new opportunity to pursue.
The bear case was that AMZN had convinced the market of this narrative which allowed them to grow their business with no margin discipline and compete unfairly against traditional companies whose shareholders demand a reasonable profit. This margin, the bear case argued, would never materialise because the lack of margin was not a reinvestment decision, rather it was a result of their low prices (and low pricing power). This model of reinvestment in new growth ventures wasn’t even unique, rather it was a revival of the conglomerate model dismantled by corporate raiders in the 1980s.
In late 2014 revenue was up (more than 20% y/y on an annualised base of over US$80bn) but operating income (EBIT) was deteriorating – recording a loss of US$374m in 3Q 2014 which was worse than both the prior quarter and prior year performance. Despite this, the stock had outperformed the market considerably over the prior five years (2009 to 2014) and the market cap was now north of US$150bn. This for a company that had been listed almost 20 years and had yet to sustain consistent profits.
Aswath Damodaran is widely considered the doyen of valuation. Almost all professional investors are likely to have learnt some of their valuation knowledge from his work. There is almost no higher authority on valuation than Damodaran. Fortunately for practitioners, he is very active in working through real world examples, and he tackled AMZN in 2014.
In a post titled “If you build it (revenues), they (profits) will come”: Amazon’s Field of Dreams! published October 2014, Damodaran ran through the history of AMZN and sought to address and value this idea of AMZN holding back profitability until some future point. The analysis is very detailed, but the conclusion arrived at was if AMZN could continue the current rate of revenue growth and sustain an operating margin similar to the weighted average of the retail and media sectors, then AMZN stock would be worth US$175/share or US$81 billion. This at a time when the actual stock price was US$415/share. In case there was any uncertainty over his investment view, Damodaran concluded;
“Mr Bezos has delivered on half of his field of dreams vision by building up the revenues for Amazon, but the other (and more difficult) half of the vision requires that the “profits” arrive. Much as I would like to believe in miracles, it will take far more work to make Amazon profitable than it will to make Shoeless Joe Jackson show up in a cornfield in Iowa!”.
Few value investors saw AMZN as undervalued in 2014.
Less than five years later, the world’s best-known value investor discloses to have purchased shares in AMZN at between 8x and 10x the Damodaran valuation. A whole generation of investors are now trying to reconcile the cognitive dissonance of what just happened. The point here is not ‘Buffett is right’ (time will tell), but rather ‘valuation is hard’. Even when done by not just experts, but literally the guy who wrote the book, the real-world margin for error is far more than 10% but can be 10x!
Markets change. Investors today are very comfortable with the idea that Warren Buffett evolved from the Benjamin Graham investment model (buy at below the value of hard assets) to the Charlie Munger model (buy companies that have a moat and can sustain and reinvest at high returns on capital). Investors are perhaps less open to the idea that Jeff Bezos may be using an investment model that is an evolution of Mungers, adapted to a different world.
We estimate fair value for our portfolio holdings and potential investments, however valuations are just one of many models we use to try and understand the risk and return opportunity set we face. We consider valuations with a level of precision that history, rather than our personal biases and desires, suggests they have.