Legend has it that in 1929, businessman Joseph Kennedy, the father of the future president, realized it was time to get out of the stock market when the shoeshine boy started offering him trading tips.

I had my own such moment a couple of years back when I started hearing people say they were selling their cars because "it's cheaper to take Uber everywhere!"

It wasn't that I doubted them, mind you. I just started to wonder about the math.

Uber and Lyft are functionally taxicabs — better dispatched and more convenient but, still, taxicabs, pretty much. There's a reason that, before the Uber/Lyft revolution, almost no one said, "I'm going to sell my car and take taxis everywhere!" Unless you are a hermit or live in a dense urban core, a month of taking cabs costs more than a month of Corolla ownership.

Boosters of the ride-share revolution like to point out that most of the nation's cars spend most of their time parked; there ought to be money in liberating all that unused capital. True enough — except that someone has to drive the car, including the time spent circling as they wait for rides.

In 2014, journalist Timothy B. Lee spent a week driving for Lyft. He drove for 50 hours but spent only 14 of those hours actually ferrying passengers. All that circling wears out the car and burns both gas and the driver's valuable time.

So how can Uber and Lyft, both of which are planning initial public offerings this year, be price-competitive with car ownership outside of places such as Manhattan?

Answer: heavy subsidies, from both the companies and the drivers themselves.

Uber and Lyft have long used investor money to subsidize operations. Lyft's IPO documents, filed last week, indicate that in 2018 the company booked $8.1 billion in rides, collected $2.2 billion in revenue — and lost more than $900 million after expenses. Uber is also losing money, although perhaps not quite as much.

This despite the fact that many drivers seem to be underpricing their services. Whenever a driver arrives to pick you up in a massive truck or a luxury automobile, you're either looking at a driver who took up driving as a form of charity work or one who doesn't understand that ride-sharing income should be calculated after deducting gas and vehicle depreciation. Not every driver makes quite such a blatant error, but there's considerable evidence that earnings are low after accounting for expenses, and drivers don't necessarily realize that.

Thus, the ride-sharing market offers a real-life illustration of the old economist's joke: "We're losing money on every unit, but we'll make it up in volume!" Unfortunately for us riders, there's only so much cheap investment money, and only so many inexperienced drivers, out there. Once Uber and Lyft have burned through those, they're going to have to charge us what the rides are actually worth. Customers will be in for a rude shock.

Nor are theirs the only customers due to relearn that there's no such thing as a free lunch. After 13 years, the music-streaming service Spotify finally (barely) turned a profit last quarter on its nearly 100 million subscribers, only to forecast substantial losses for the coming year. Netflix is burning through borrowed cash as it races to build out its content library. In the journalism business, a host of digital start-ups are running out of investment funds without ever having run into a viable business model.

That's not to say that all the subsidized businesses are headed for the same kind of trouble that has beset BuzzFeed or the Huffington Post. There's obviously a market for ride sharing, for streaming and, yes, even for digital journalism. And in some of those businesses, notably streaming, the massive economies of scale really might deliver Facebook-style windfalls to early investors.

But Lyft and Uber are a different story. They're not selling a song or a movie that can be endlessly replicated for little incremental cost; they're selling a physical service that's pretty expensive to deliver. At some point, we're going to have to pay for it.

Heavy users of ride sharing should start getting used to the idea that the cost will soon go up, and plan their lives accordingly. And investors should prepare for demand to drop when customers and drivers discover the true price of the service.

In other words, as a modern-day Joseph Kennedy might say: Don't buy frothy stocks unless you're willing to lose a bundle. And don't sell your car unless you're ready to walk.