The Ponzi threshold and the Armstrong principle

Interesting convergence of ideas, as this reminds me a lot of the (Lance) Armstrong Principle: If you push people to promise more than they can deliver, they’re motivated to cheat.

Here’s Palko’s example (see also here):

First, hype and next-big-thingism push Uber’s value far beyond any defensible level, then, as reality sets in and investors realize that the original business model, though sound, can never possibly justify the money that’s been put into the company, Uber’s management responds with a series of more and more improbable proposals in order to keep the buzz going.

What can be hard to understand with Uber, as with Armstrong, is how things went on like this for so long. It’s practically conventional wisdom now that Uber is trapped, and it was no secret within the cycling world that Lance was doping. But in both cases lots of influential people stayed on the train for a really long time. Part of it must be the sunk cost fallacy or its intellectual/reputational equivalent, part must be the stock pumping idea (if you have money invested in Uber or reputation invested in Lance, then you’d rather keep the bubble afloat for as long as possible), and part of it must be the calculation that it will all work out (in Uber’s case, that might be public subsidies such as juicy citywide Uber contracts for public transit; for Armstrong perhaps the hope was that his anti-doping adversaries would eventually give up).

In any case, whatever one thinks about these particular cases, it’s interesting to see the connection between the Ponzi threshold and the Armstrong principle.