>https://irle.berkeley.edu/files/2020/07/Parrott-Reich-Seattl... suggests that net driver pay is less than half of gross pay, which doesn't account for the portion that Lyft takes. Taking that at face value, maybe it gets 40% cheaper? Driving is expensive!
I'm not sure what expenses goes into calculating "net driver pay", but based on a quick skim there are some questionable items in there. "Exhibit 28 Total Seattle TNC driver expenses" lists stuff like "health insurance costs" and "independent contractor taxes", which obviously wouldn't be needed for robotaxis. If we drill down into "vehicle operating costs", there are also some questionable items, like $1560/year expense for "cellphone". I agree that robotaxis being 90% cheaper is unlikely, but 66%-75% cheaper (ie. a quarter to a third of the price) seems to be within the realm of possibility.
I'm not sure what expenses goes into calculating "net driver pay", but based on a quick skim there are some questionable items in there. "Exhibit 28 Total Seattle TNC driver expenses" lists stuff like "health insurance costs" and "independent contractor taxes", which obviously wouldn't be needed for robotaxis. If we drill down into "vehicle operating costs", there are also some questionable items, like $1560/year expense for "cellphone". I agree that robotaxis being 90% cheaper is unlikely, but 66%-75% cheaper (ie. a quarter to a third of the price) seems to be within the realm of possibility.