I've been reading and listening to a lot of interesting commentary on Amazon's financial situation from tech writers and a lot of it focuses on the question of whether Amazon can or ever will be able to "flip the switch" and go from being a high-growth high-revenue company to being a profitable one.
That's a fascinating question, but it's worth introducing the idea that you could imagine a successful company that never posts GAAP profits. Now you can't have a company that spends $100 to make widgets that it sells for $99, but that's not the sense in which Amazon is unprofitable. Amazon is unprofitable because it spends ~100% of its cashflow on capital expenditures oriented toward doing future business. The important thing here is that from a shareholder's viewpoint, a dollar of capex isn't like lighting a dollar bill on fire. The shareholders own the commpany's capital equipment. If for some reason the Amazon board decided to liquidate the company tomorrow, it's not as if the shareholders would be left emptyhanded. You'd auction off all of Amazon's office buildings and fulfillment centers and warehouse robots and AWS servers and patents and whatever else they have. By contrast, if you liquidated Apple you'd not only have the proceeds from that kind of auction but an enormous pile of US dollars.
So good for the shareholders.
But the point is that though stockpiling cash is a way of increasing the value of your enterprise so is stockpiling capital goods. Imagine 20 guys who pool their money together to buy a building and open a restaurant in it. Then instead of having the restaurant be "profitable" it spends all its surplus on buying a second building and opening and second restaurant. And so it goes year after year for twenty years. At the end of that time, they own 50 buildings and 50 restaurants. Then they decide to close the restaurants and sell the buildings (and the kitchen equipment the restaurants contain) to other investors and divvy up the proceeds. Depending on how the local real estate market evolved over that time, they might earn themselves some healthy capital gains for their trouble.
Now as far as I know, this kind of scenario doesn't play out in the real world. And certainly it doesn't play out with giant publicly traded companies. But I do think it goes to show that the idea that a rapidly-growing unprofitable firm might be a sound investment doesn't genuinely depend on the theory that the firm will be able to "throw the switch" at some point. Your share gives you a slice of the firm, and as long as the firm is growing in the sense of "amassing more firm-owned stuff" then the value of your investment is growing.