2 ms·
That comes down to the "Nature of the Firm" dynamic[1]; there's a tradeoff between market efficiency and transaction costs, and every organization faces it. Yo
by SilasX 10y ago
That comes down to the "Nature of the Firm" dynamic[1]; there's a tradeoff between market efficiency and transaction costs, and every organization faces it.
You can e.g. ask the same question about "why not auction every shift?", "why not auction every internal work product transfer?" (extreme example: why not auction every fetch-coffee task?)
The answer is that having to re-auction every subcomponent introduces transaction costs that may eclipse the efficiency gains of selling at the optimal, market price.
In the case of Uber's drivers, that "transaction cost" includes the additional lag in taking bids before you get matched, and higher unpredictability of fares.
Of course, it actually gets kinda complicated here: Sidecar did more what you're describing, and (IIUIC) eliminated that lag by letter drivers submit a kind of formula that calculated and submitted their bid based on those factors (e.g distance to pickup), so it's not necessarily some kind of huge dealbreaker to do auctions; drivers don't have to sit down and scratch their heads about how much they'd need to do every request.[2]
Sidecar didn't succeed though -- they went out of business on this model (and keep in mind they had a much more solid defense of workers' contractor status for exactly this reason). But this might not have been the cause of their failure.
I'm guessing it comes down to there being a big premium on being able to (for lack of a better term) "commoditize" the ride hailing services so that eliminating such bidding was a net positive.
[1] https://en.wikipedia.org/wiki/The_Nature_of_the_Firm https://en.wikipedia.org/wiki/The_Nature_of_the_Firm
[2] though note that in Sidecar's case, driver's got a list of bids and consciously chose the one they liked based on their implicit criteria rather than being auto-matched