4 ms·

Two problems mentioned in the rebuttal: the study counts every Airbnb unit as a rental taken permanently off the market, and the dataset counts every blocked-of

Two problems mentioned in the rebuttal: the study counts every Airbnb unit as a rental taken permanently off the market, and the dataset counts every blocked-off calendar date as a night the unit was booked. Both of those are obvious over-counting. A study that doesn't even try to account for those is pretty obviously political. Airbnb should have posted estimates of how much this over-counting could distort the data, because it sounds like a lot. $616M is one of those numbers that sounds high but in context probably isn't, so it wouldn't surprise me if those two distortions wiped out the number, or at least cut it to below headline worthy.

loceng 9y ago
On the surface you can't assume either way if a listing on Airbnb means the unit is permanently off the market or not, likewise with counting or not counting every blocked-off calendar date: if it's not a period of 4-8-12 months blocked off, then it's likely not being rented to a long-term renter; the lister perhaps even perhaps booked those block-off calendar dates through a different service, or did the transaction not through/off of Airbnb. Whether it's $100MM or $616M - it's a sign that there are external costs that that city and society must shoulder because of the higher prices people can charge with shorter-term rentals, primarily to non-locals who may or may not be staying or moving to the city.
- neonate 9y ago
  Of course you can't assume either way, which is exactly what they did, which means they over-counted, inflating the numbers in the direction of their conclusion.