Caution: Lyft out of order

Unlock the Editor’s Digest for free

Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.

Firstly, lol.

Shares of Lyft soared more than 60 per cent on Tuesday before falling back sharply, after an error in the ride-hailing company’s quarterly earnings release exaggerated the outlook for margin growth in 2024 by 10 times. The company reported that it would improve adjusted earnings margins by 500 basis points, or 5 percentage points, in 2024 compared with the previous year.

Lyft stock surged in the minutes after the announcement to as much as $19.70, the highest price since August 2022. The shares then retreated to a gain of about 15 per cent after chief financial officer Erin Brewer said on a call with investors and analysts that the increase would in fact be 50bp, or 0.5 percentage points. “This is actually a correction from the press release,” Brewer said.

Secondly, what? Here’s the 2024 guidance in full from the corrected release, with the relevant line highlighted.

In what world could 500bp be right? Analysts (and presumably investors) had been expecting Lyft to guide for an adjusted ebitda margin of 2 per cent for 2024, up from 1 per cent in 2023. Six per cent is quite the jump from 1 per cent. Note also that Lyft guided for the current-quarter adjusted margin to be “approximately 1.4 per cent to 1.5 per cent”, so something very dramatic would have to happen after March.

Ebitda margin is a simple case of dividing one number by another, so it’s possible to hit the typo target by reducing the divisor, which is gross bookings. Forecasts going into the print were for 2024 annual gross bookings of $15.6bn or thereabouts. If this number were below $5bn Lyft would have a 6 per cent margin, ceteris paribus, though ceteris paribus really isn’t how P&L statements work. And if Lyft had expected gross bookings to fall by 60 per cent-plus this year, it probably wouldn’t have said in the line directly above its typo that gross bookings would rise.

That leaves adjusted ebitda, which has to change from the pre-print expectation of $320mn to around $1bn. That’s approximately fourfold year-on year growth, an upgrade to consensus forecasts of about 200 per cent, and implies annual free cashflow going from negative to half a billion dollars.

In what ways might Lyft — America’s second-biggest minicab dispatch company, which since its foundation in 2012 has never broken even — become dramatically more profitable overnight?

Maybe it could hike the average take rate, which is the cut it takes from each driver’s fare in addition to the service fee. Six per cent margins are possible at a total take rate of 40 per cent, from 32 per cent currently. But since Lyft has only just settled a lawsuit over the charges it levied on New York drivers, it would be a very brave strategy.

What about expenses? Lyft’s insurance cost per trip is about $2.43, or 50 cents per mile. If Lyft stopped insuring its drivers entirely the adjusted ebitda margin would go up to 6 per cent, though again there may be some unwelcome externalities.

Look, all this is ridiculous. That’s the point. The back-of-the-envelope numbers above were worked out in 10 minutes by a person who’d never previously looked at Lyft, cared about Lyft, hailed a Lyft, or been within 4,400 miles of an available Lyft. A person with any understanding of Lyft’s financials should have identified instantly that the 500bp guidance was implausible.

So who was trading at the wrong prices on what was an obvious typo?

Below (per Bloomberg) is the post-close price and volume. It shows a slightly fishy 1mn shares traded in the seconds before the Lyft release dropped, followed by another 2mn exchanged in the minutes before the conference call, after which there’s a near-3mn dump after the correction was issued.

For a small-cap that trades just 10mn in the average session, that’s decent volume. And frankly, we just can’t rationalise it.

But maybe you can. Do you reckon this latest breakdown in EMH was some combination of:

1. Bots that can do arithmetic but not logic?

2. Idiots with the same problem?

3. Anyone who wanted in on the inevitable class-action lawsuit?

4. Something else entirely that you’ll explain in the comment box?