lottery to incentivize testing? - Printable Version
+- The CardBoard (https://thecardboard.org/board)
+-- Forum: Emergency (https://thecardboard.org/board/forum-11.html)
+--- Forum: Covid-19 (https://thecardboard.org/board/forum-12.html)
+--- Thread: lottery to incentivize testing? (/thread-19953.html)
+- The CardBoard (https://thecardboard.org/board)
+-- Forum: Emergency (https://thecardboard.org/board/forum-11.html)
+--- Forum: Covid-19 (https://thecardboard.org/board/forum-12.html)
+--- Thread: lottery to incentivize testing? (/thread-19953.html)
lottery to incentivize testing? - g1313 - 06-26-2020
This is a bit old but I thought it was an interesting idea, from https://freakonomics.com/podcast/covid-19-reopen/
Quote:LEVITT: Well, I think the answer is: you’ve got to make it worth people’s while to take this test. It’s what economists call “internalizing the externality.” So we’re going to need a lot of apparently healthy people, people without symptoms, to take this test. So I think we should pay them, and a sensible way to do that might be in the form of a really big lottery. So you might even call it, like, Pandemillions, or something like that. So you could imagine we could put something like $500 million, $1 billion a week into this lottery. And in order to get a lottery ticket, you’d have to go and get tested for Covid. And the social benefit would so swamp the costs of doing this. A billion dollars a week or something, it’s peanuts compared to even the existing CARES Act and almost vanishingly small compared to the costs overall of this disease.
DUBNER: What’s the difference if you test positive or negative, though? Do you get more chances at the lottery if you test positive because we want to incentivize people then to stay home for an additional two weeks or whatnot?LEVITT: So I think if you test positive, it maybe gets simpler because you’re talking about a smaller group of people. I would simply just pay people to stay at home. I would pay a big enough number that even if you don’t feel sick, you’d want to stay home. So if something like, I don’t know, $2,000 per week, and you get paid that as long as you’re testing positive. I would pay handsomely for people to stay at home. I really think if the incentive plans that I’m pushing get put into place, our problem will not be getting people to stay home or to take the test. Our problem will be that people are going to cheat like crazy to try to get certain results and get into the lottery and whatnot. I’d much rather have the problem of people too eager to get tested and faking Covid than the problem we have, which is a pandemic in which people are out and about doing things and we don’t know how to stop it.
RE: lottery to incentivize testing? - JustAnotherFan - 06-26-2020
(06-26-2020, 12:49 PM)g1313 Wrote: This is a bit old but I thought it was an interesting idea, from https://freakonomics.com/podcast/covid-19-reopen/
Quote:LEVITT: Well, I think the answer is: you’ve got to make it worth people’s while to take this test. It’s what economists call “internalizing the externality.” So we’re going to need a lot of apparently healthy people, people without symptoms, to take this test. So I think we should pay them, and a sensible way to do that might be in the form of a really big lottery. So you might even call it, like, Pandemillions, or something like that. So you could imagine we could put something like $500 million, $1 billion a week into this lottery. And in order to get a lottery ticket, you’d have to go and get tested for Covid. And the social benefit would so swamp the costs of doing this. A billion dollars a week or something, it’s peanuts compared to even the existing CARES Act and almost vanishingly small compared to the costs overall of this disease.
DUBNER: What’s the difference if you test positive or negative, though? Do you get more chances at the lottery if you test positive because we want to incentivize people then to stay home for an additional two weeks or whatnot?LEVITT: So I think if you test positive, it maybe gets simpler because you’re talking about a smaller group of people. I would simply just pay people to stay at home. I would pay a big enough number that even if you don’t feel sick, you’d want to stay home. So if something like, I don’t know, $2,000 per week, and you get paid that as long as you’re testing positive. I would pay handsomely for people to stay at home. I really think if the incentive plans that I’m pushing get put into place, our problem will not be getting people to stay home or to take the test. Our problem will be that people are going to cheat like crazy to try to get certain results and get into the lottery and whatnot. I’d much rather have the problem of people too eager to get tested and faking Covid than the problem we have, which is a pandemic in which people are out and about doing things and we don’t know how to stop it.
This is risky. Some prior research has shown that incentives can sometimes backfire, such as when it comes to doing things for the good of others.
https://hbr.org/2009/03/when-economic-incentives-backfire
RE: lottery to incentivize testing? - g1313 - 06-26-2020
(06-26-2020, 01:49 PM)JustAnotherFan Wrote: This is risky. Some prior research has shown that incentives can sometimes backfire, such as when it comes to doing things for the good of others.
https://hbr.org/2009/03/when-economic-incentives-backfire
True, and this phenomenon was explored in the Freakonomics book which was co-written by the same economist who made the above lottery proposal. However a few reasons it may not apply:
1) Apparently the seminal study on blood donations and incentives by Titmuss in 1970 was not actually an empirical study, it was theorized based on survey results. Many of the follow up studies also suffer from this - they ask someone how willing they are SAY that they will donate blood with or without an incentive, but they don't actually see whether someone actually donates blood.
2) In addition, the Titmuss study is not directly applicable to COVID testing. He actually did not argue that paying for blood donations would DECREASE the blood supply, he only argued that it would not INCREASE relative to the cost of paying, and also the quality of the blood supply would decrease (because people who are donate blood just for money could be more likely to use IV drugs, for example). For COVID testing, any test is in theory a good test.
3) I found some other follow up studies that suggested that it's specifically direct cash payment that makes people feel weird about blood donations, but if you abstract the financial incentive in some way (as with a lottery, or a voucher) then that effect diminishes. (https://freakonomics.com/2013/06/05/is-paying-for-blood-a-good-idea-after-all/ and https://www.sciencedirect.com/science/article/abs/pii/S0167487010000735)
RE: lottery to incentivize testing? - JustAnotherFan - 06-26-2020
(06-26-2020, 02:33 PM)g1313 Wrote:(06-26-2020, 01:49 PM)JustAnotherFan Wrote: This is risky. Some prior research has shown that incentives can sometimes backfire, such as when it comes to doing things for the good of others.
https://hbr.org/2009/03/when-economic-incentives-backfire
True, and this phenomenon was explored in the Freakonomics book which was co-written by the same economist who made the above lottery proposal. However a few reasons it may not apply:
1) Apparently the seminal study on blood donations and incentives by Titmuss in 1970 was not actually an empirical study, it was theorized based on survey results. Many of the follow up studies also suffer from this - they ask someone how willing they are SAY that they will donate blood with or without an incentive, but they don't actually see whether someone actually donates blood.
2) In addition, the Titmuss study is not directly applicable to COVID testing. He actually did not argue that paying for blood donations would DECREASE the blood supply, he only argued that it would not INCREASE relative to the cost of paying, and also the quality of the blood supply would decrease (because people who are donate blood just for money could be more likely to use IV drugs, for example). For COVID testing, any test is in theory a good test.
3) I found some other follow up studies that suggested that it's specifically direct cash payment that makes people feel weird about blood donations, but if you abstract the financial incentive in some way (as with a lottery, or a voucher) then that effect diminishes. (https://freakonomics.com/2013/06/05/is-paying-for-blood-a-good-idea-after-all/ and https://www.sciencedirect.com/science/article/abs/pii/S0167487010000735)
+1 great response
RE: lottery to incentivize testing? - burger - 06-26-2020
There's a bigger problem with paying people with the virus: giving people a financial incentive to get the virus could result in more cases. If I'm young and healthy and broke, and I see that I can get $2000 a week to get the equivalent of a mild cold, what's to stop me from getting infected on purpose? Sure, I'd have to get tested and be isolated, but I could be spreading the virus before I get a positive test result. This seems really short-sighted.
As for testing, the number of tests being taken is going up all over. In Arizona, they're overrunning their ability to do the tests in labs now now. But there is something to be said for an intervention that massively increases the number of tests. I can't find it now but I've seen epidemiologists discussing ubiquitous testing (like everyone in a city being tested twice a week) as a way to rapidly get the virus under control. Find the cases, isolate them, voila. If some sort of financial reward for the testing alone--not just the positive results--got us to this goal, I'd be fine with that.