There were few surprises in the filings of ride-sharing firm Lyft Inc. as it prepared to go public. Anybody paying attention already knew the company didn't make money. The net loss last year approached $1 billion on revenue of $2.2 billion.

Lyft didn't have much of a balance sheet, either, besides some investment in scooters and the cash remaining from its private capital raises. Of course, there was no big number on the asset side representing cars.

Lyft and its rival Uber Technologies Inc. have shown you can be in the car ride business and not have cars.

In a way it's really quite an achievement, Uber and Lyft managing to recruit lots of people to drive around their customers while paying them as little as possible to do it. And here's the real brilliance of their approach: They found drivers to do the driving in their own cars.

Calling the whole thing ride-sharing to make it part of the promising age of the "sharing economy" turned out to be just another way for Uber and Lyft to effectively sell a business idea that's nothing more than using somebody else's capital asset.

Somehow this seems to mark a new milestone in the evolution of a business model called asset light.

As an idea, it's not new. Businesses have always tried to use as few assets as they could get away with to meet customer needs. It just wasn't always this easy to do.

Franchising is a good example of a traditional asset-light approach, where much of the economic value comes from the brand supplied by the franchise company. It's up to the franchisees to pull together the money to buy or build an Applebee's restaurant or Super 8 Motel.

Obviously the brand dies if individual franchisees can't make money. But one motel owner in an unlucky location tipping over won't mean much to the franchise company.

It's much the same formula for big manufacturers, too. They may not have franchisees, but they do have suppliers willing to invest in the equipment they don't want to own.

The manufacturer could easily afford a team of machinists on a fully equipped shop floor, but it's hard to keep all the machines working all the time. An outside machine shop can always line up more customers to keep its plant and workers busy.

Big companies lose some control by outsourcing, but if parts can be made correctly and arrive on time, the system works.

Big manufacturers have long found ways to push even more of the capital requirement onto smaller suppliers. One way is by requiring them to build parts without getting paid for them, even though the parts may sit inside the big customer's assembly plant awaiting use. Then when the supplier does get to send the bill, it may take three or four months for the check to come.

This may not feel fair, but there's no point for an entrepreneur to go to war over terms with a Fortune 500 company. And if cheaper suppliers can be found overseas, that, too, becomes the local entrepreneur's problem.

An asset-light approach doesn't work in every business, of course. But aspiring beer barons can easily get into the business without buying any fermentation tanks, and semiconductor engineers don't need to build a $10 billion fab to get their chips made.

Taxicab companies maybe had drivers who owned their own cabs, said Evan Rawley of the Carlson School of Management at the University of Minnesota, but what was different when Uber launched in 2011 and Lyft a little later is that these companies had nothing to do with the cars. And the two upstarts grabbed customers quickly by making it so painless to use a smartphone to get and then pay for a car ride that it was easier than even taking your own car.

Neither company will make money until they raise prices. That is, Rawley said, until they get better at "colluding," a jarring thing to hear from a business school professor.

No one's going to set prices over cigars in a backroom of the California Club. But, as he put it, the companies will eventually have to become satisfied with their respective market shares and then raise prices in concert.

That means the people with the pressing economic problem of being in a commodity business with zero influence over prices are the drivers, he said. The driver compensation that will keep attracting drivers, he added, "is always going to be very, very low. Because anybody can do it."

And, of course, even small cars aren't cheap. A new Honda Civic has a five-year cost of ownership that exceeds $30,000, and that assumes only driving 15,000 miles per year. One good yardstick for the total cost of driving a car is the mileage rate put out by the Internal Revenue Service, now 58 cents per mile.

This was a useful number to remember when digging into the economics of driving from "The Rideshare Guy," provider of a robust blog site just for drivers. A contributor was writing about a short-lived strike in California last week due to another rate cut by Uber.

In a typical week in early March, 90 rides generated "pay" of $16.65 an hour assuming a full week. But then there's the cost of the car. Maybe a car can be operated cheaper than the IRS rate, but by applying that cost the pay slips to about $10 an hour.

And that's before the rate cut by Uber. The minimum wage in California is $12 an hour for firms with 25 workers, so no wonder there was a strike.

Another attention-grabbing headline appeared in February atop a chat thread on UberPeople, another site for Uber and Lyft drivers. "Many Thousands of Rideshare drivers defaulting on car loans," it said, pointing to recent news from the Federal Reserve Bank of New York that car loan delinquencies have shot past the peak after the Great Recession nearly 10 years ago, to more than 7 million Americans at least 90 days behind.

Lyft came public last week valued at more than $24 billion, while the people providing the assets to make this business work, thousands of them anyway, maybe can't afford the payments on a Honda Civic.