Stef

Thoughts on $AMZN $AMZN reported but I won’t bore you with another recap of the quarterly results. A hundred other people have already done that, and repeating the same numbers. Instead, I rather focus on a few things that caught my attention. $AMZN p added $13 billion in quarterly AWS revenue, that’s equivalent to $50 billion in annual revenue. To put that into perspective, AWS is adding the equivalent of an enormous technology company to its revenue base every year, while operating margins approaching 39%. That’s an extraordinary accomplishment for a business already operating at this scale. Management expects AWS revenue growth to exceed 40% in both 2027 and 2028, which is a remarkable outlook for a business of this magnitude. We’re not talking about a startup doubling from $100 million to $200 million in revenue, but one of the world’s largest cloud businesses compounding at rates normally associated with much smaller companies. The absolute dollars involved are staggering, and I think it’s worth exploring what those numbers might eventually mean for $AMZN’s earning power. Let’s do some destination analysis, because this is where things become fascinating. Suppose AWS eventually reaches $180 billion in annual revenue and subsequently grows 40% for two consecutive years. Revenue would reach approximately $252 billion after the first year and $353 billion after the second. At a hypothetical 35% operating margin, AWS alone would generate $124 billion in annual operating income. For perspective, $AMZN’s entire business currently generates approximately $104 billion in operating income. In other words, under those assumptions, AWS alone could eventually produce more operating profit than all of $AMZN generates today. That’s a remarkable possibility, although the $180 billion starting point and 35% margin are assumptions. Nevertheless, it gives you some appreciation for the enormous earnings potential. But here’s where my enthusiasm becomes considerably more measured, because revenue growth and shareholder value creation are two entirely different things. $AMZN generated $168 billion in operating cash flow over the past twelve months, an extraordinary amount of money by almost any standard. Unfortunately, it also spent approximately $211 billion on capital expenditures, resulting in negative $42 billion of free cash flow. During the most recent quarter alone, capital spending approached $76 billion, while management expects approximately $220 billion in cash capital expenditures next year. Now, I have absolutely no problem with a company sacrificing current free cash flow to create substantially greater future earning power (ie $MELI). In fact, some of my most successful investments have involved businesses that deliberately prioritized long term opportunities over short term profitability. But calling something an investment doesn’t automatically make it a good investment, and spending enormous amounts of money doesn’t establish that those expenditures will generate attractive returns. What matters is the relationship between the capital invested and the incremental economic value created. Consider the economics of AI infrastructure, because I think this is one of the most important questions facing $AMZN today. A data center building might remain economically productive for decades, but the computing equipment inside it could become obsolete much faster. If $AMZN spends $100 billion expanding capacity, how much of that spending creates durable earning power, and how much will eventually need to be repeated simply to maintain its competitive position? The difference between growth CAPEX and recurring maintenance could have enormous implications for future free cash flow. On a side note, it’s quite fashionable at the moment to ignore this truism. 1/👇