Amazon is the cheapest of the American megacaps on trailing earnings and the lowest-scoring of them on Investily's report card. It scores 62 out of 100. Those two facts are not a contradiction, and understanding why is the most useful thing a beginner can take from this company.
The score: 62/100
A 62 is not a bad company. It is a company where the pillars disagree. Amazon's growth is respectable, its margin has roughly doubled in five years, and its balance sheet is fine. What drags the score down is what it earns on the enormous and rapidly growing pile of capital tied up in the business.
Growth: steady, and better on profit than sales
Revenue grew 14.8% over the trailing year against 11.2% over three years and 10.5% over five — so growth is accelerating, modestly. Operating income grew 21.9% over the trailing year, comfortably ahead of revenue, which is the pattern you want. Net margin has climbed from a five-year average of 6.6% to 10.5%.
Read only that paragraph and Amazon looks like a business getting meaningfully better. The income statement is not where the problem lives.
Returns: where the score is lost
Return on capital employed is 15.4% over the trailing year. That is a respectable number in isolation and a poor one for a company trading as a technology compounder. But the figure that should stop you is the cash-flow version: 2.7%.
That gap — 15.4% on an accounting basis against 2.7% in cash — is the whole story. Amazon reports profit on capital that is not, for now, arriving as cash. The reason is visible one line further down:
- Book value growth, trailing year: 63.8%
- Revenue growth, trailing year: 14.8%
- Five-year book value growth: 35.3% a year
The capital base is compounding more than four times faster than sales. Returns are a ratio, and when the denominator grows like that, the ratio falls even while profits rise. That is not fraud or mismanagement — it is what building datacenters and logistics capacity looks like in the numbers. But it is also why the score is 62.
Valuation: cheap on the wrong measure
Amazon trades near 21 times trailing earnings, the lowest multiple among the American megacaps and roughly a third of what it commanded a year ago. On the surface that reads as a bargain.
The forward numbers argue otherwise. Consensus has revenue growing only 6.6% next year and earnings per share falling 6.5%. Because earnings are expected to decline, the forward multiple is higher than the trailing one — around 23x. A stock is only cheap against earnings that are going up. Amazon's are not, on current estimates.
The bottom line
Amazon is the clearest live example of why a low P/E is not a discount. The multiple is low because the market has looked at the capital being consumed and the earnings expected next year and priced accordingly. The score sees the same thing from the other direction.
The bull case is straightforward and might well be right: this is a company deliberately trading present cash for future capacity, exactly as it has before. If the capital earns its return, the 2.7% cash return on capital rises, the score rises with it, and today's multiple looks obvious in hindsight. The bear case is that the capital keeps compounding at 60% and the returns never catch up. For the framework behind that judgment, see return on capital employed.