3 ms·

The authors spend 15 pages or so describing the calculations, its limitations, and alternative calculations. Focusing on one parameter to throw out the whole p

The authors spend 15 pages or so describing the calculations, its limitations, and alternative calculations. Focusing on one parameter to throw out the whole paper would be junk science. http://ftp.iza.org/dp13670.pdf http://ftp.iza.org/dp13670.pdf

nradov 6y ago
The authors spend 0 pages describing the basis for the $11,000 cost of asymptomatic cases. In fact they just blindly accepted the estimate from Kniesner and Sullivan (2020), who picked a number without doing any calculation of actual costs. So I stand by my assessment that this is junk science.
hn_throwaway_99 6y ago
I didn't see where the $11,000 estimate for asymptomatic cases came from (can anyone point to exactly where that's from?) However, if it is the case that the authors did use that estimate for a per-asymptomatic case cost, then I think it's perfectly valid to throw out the rest of the paper as nonsense. I mean, the only reason anyone is talking about this paper at all is because the top-line number is so outrageously high. If a fundamental calculation going into that number is that patently ridiculous, I don't care if you have 15 pages or 1500 pages, adding more pages of numbers can't get around the absurdity of the underlying primary assumption.
- hn_throwaway_99 6y ago
  Replying to my own comment, I found the paper where they got the 11k from, it's page 7 on this link: http://ftp.iza.org/dp13632.pdf http://ftp.iza.org/dp13632.pdf. This analysis is complete, utter, and total bullshit. They took a "value per statistical life" number of $11 million, then pulled a semi random 0.3% out of their ass to account for an $11k cost per asymptomatic case. At no point did they actually attempt to account anywhere for the true cost of treating (or, in this case not treating) an asymptomatic case. These numbers are complete garbage.

nradov 6y ago

The authors spend 0 pages describing the basis for the $11,000 cost of asymptomatic cases. In fact they just blindly accepted the estimate from Kniesner and Sullivan (2020), who picked a number without doing any calculation of actual costs. So I stand by my assessment that this is junk science.

hn_throwaway_99 6y ago

I didn't see where the $11,000 estimate for asymptomatic cases came from (can anyone point to exactly where that's from?) However, if it is the case that the authors did use that estimate for a per-asymptomatic case cost, then I think it's perfectly valid to throw out the rest of the paper as nonsense. I mean, the only reason anyone is talking about this paper at all is because the top-line number is so outrageously high. If a fundamental calculation going into that number is that patently ridiculous, I don't care if you have 15 pages or 1500 pages, adding more pages of numbers can't get around the absurdity of the underlying primary assumption.

hn_throwaway_99 6y ago
Replying to my own comment, I found the paper where they got the 11k from, it's page 7 on this link: http://ftp.iza.org/dp13632.pdf http://ftp.iza.org/dp13632.pdf. This analysis is complete, utter, and total bullshit. They took a "value per statistical life" number of $11 million, then pulled a semi random 0.3% out of their ass to account for an $11k cost per asymptomatic case. At no point did they actually attempt to account anywhere for the true cost of treating (or, in this case not treating) an asymptomatic case. These numbers are complete garbage.