3 ms·

Here's the summary: → a new Harvard study found giving low-income Americans cash isn’t that helpful. Researchers gave 5,000 US participants $2000, $500, or no

Here's the summary: → a new Harvard study found giving low-income Americans cash isn’t that helpful. Researchers gave 5,000 US participants $2000, $500, or nothing. “Within 4 weeks, we see no significant difference in bank balances between either cash & control groups,” wrote one of the study authors. And: “We find NO positive effect of cash—if anything, in some prespecified analyses, we even see negative effects.” Apparently the cash infusions added stress to people’s lives, which makes sense (likely family and friends came knocking).

jaclaz 4y ago
... and after having spent more than 50 pages in describing the experiment and its results and analyzing any number of (wild) guesses, the researchers actually talked to 15 (fifteen) people of the 5,000 (five thousand) involved in the experiment and half of them told the researchers how the money was not enough: >Indeed, after the study concluded, we conducted qualitative interviews with 15 participants, where—among other things—we asked them how much money they would have needed to receive for it to have made a difference in their lives at that time. About half of participants indicated that they would have needed to receive more than $2,000, with those responses falling between $4,000 and $25,000. Although this is by no means a representative sample or ideal methodology, these responses may at least be indicative of the general order of magnitude needed. How surprising is it that someone who is living at the poverty level or under it when given a "small" amount of money (no matter if 500 or 2,000) after one month has spent it?
- jonah 4y ago
  It really seems like an insultingly small amount of money. It's almost like they're trying to disprove something like UBI by association while not actually doing something that's useful. If they'd given those randomized amounts monthly for a year or 5 years, I would bet the results would be in line with similar studies which show marked benefits.
  - naillo 4y ago
    Yeah were they expecting them to get $500 and within 4 weeks start a company and become cash flow positive? Being poor is often a consequence of education and with a few years of guaranteed income to make them feel safe to take that risk, they probably would have been able to tackle that. But it's not surprising at all that nothing happens in just a month.
  - em500 4y ago
    Gook luck trying to fund a sufficiently powered RCT to give people $2/month for a year, let alone 5. $2k for 5 years is $120k per participant, so about $120M for a 1000 sample size study.
    - jaclaz 4y ago
      Sure, but imagine making a medical experiment where you give once 1 mg of the drug (that should be dosed at 1 mg/day as per results of previous experiments for at least 2 weeks) and then writing an article in which surprisingly it comes out that it didn't work well.
jka 4y ago
Worth mentioning that the experimental payments during this research were one-time-only. (a few descriptions[1][2][3] of Universal Basic Income seem to define it as including recurring/periodic payments) [1] - https://en.wikipedia.org/wiki/Universal_basic_income https://en.wikipedia.org/wiki/Universal_basic_income [2] - https://basicincome.stanford.edu/uploads/Umbrella%20Review%20BI_final.pdf#page=3 https://basicincome.stanford.edu/uploads/Umbrella%20Review%2... [3] - https://www.investopedia.com/terms/b/basic-income.asp https://www.investopedia.com/terms/b/basic-income.asp
- erklik 4y ago
  Nearly all UBI is a recurring payment. Otherwise it's not really capable of living on.
- jaclaz 4y ago
  Yes, it is usually intended as a monthly payment, BUT the actual UBI (in theory) doesn't (and shouldn't) pass through a "means test", i.e. (still in theory) it is "Universal" and as such all citizens should get it regardless their income or estate, as the Wikipedia article highlights: >This article is about the system of unconditional income provided to every citizen. For the means-based model of social welfare, see Guaranteed minimum income. https://en.wikipedia.org/wiki/Guaranteed_minimum_income https://en.wikipedia.org/wiki/Guaranteed_minimum_income also sometimes called "basic income" (without the "Universal") which is more pertinent and actually the one on which some (mostly failed) experiments have been done or are being made.
- retrac 4y ago
  Indeed. One of the main criticisms of both Canadian major experiments with UBI (Manitoba in the 70s, and Ontario in the last few years) was that while they paid regularly, both programs were relatively short-lived -- a few years. This doesn't appropriately model the incentives and behaviours people would likely engage in if they knew with reliability the program would still be paying them in ten or more years. It wouldn't capture either negatives like people leaving the workforce permanently, or positives like going back for a four year degree they couldn't otherwise afford. Participants likely treated it as a one-time windfall (if over a couple years).
stevenicr 4y ago
This is the funniest thing I've seen all week. "found giving low-income Americans cash isn’t that helpful ... see no significant difference in bank balances" Funniest because it proves the complete ignorance of those without poor problems. Of course there is no more money in the bank. There are dozens of things more important than money in the bank.. the brakes that should of been replaced 8 months ago. The tires that are unsafe. Dental health. Many more. 'We find NO positive effect of cash—if anything, in some prespecified analyses, we even see negative effects.” Apparently the cash infusions added stress' Of course - it's stressful trying to choose - do you spend the 500 on your kid's teeth or do you get your teeth worked on? Do you spend 500 at the vet or buy healthier food for the family this week. Do you replace your brakes or the tires? Do you fix the brakes or go to the dentists? Do you feed your kids better or stop the parasites sickening the pets. Do you X or Y? Should you Y or Z? Yes those things are stressful, and certainly more stressful than 'I can't afford any of these things and that's the way it is so stop stressing it' - I'd like to say "Duh!" - but I don't think the high brow harvard study would really get it. Although I disagree with 'found giving low-income Americans cash isn’t that helpful' - I mean it's stressful - but at least it is helpful in fixing on or two of the life changing things that need fixing - so it is helpful. It's just when you have a dozen life altering debts accruing life changing interest already (deferred maintenance leading towards bigger problems) - a couple grand does not fix all the things that need fixing in the American life. Maybe there is some other place in the world where you can get all of your healthcare, vehicle maintenance and other necessities all taken care of for $2,000 (?) - maybe those folks could put some money in the bank.
phkahler 4y ago
If you gave a Harvard professor two grand, you would not perceive any change in their bank account a month later either.

jaclaz 4y ago

... and after having spent more than 50 pages in describing the experiment and its results and analyzing any number of (wild) guesses, the researchers actually talked to 15 (fifteen) people of the 5,000 (five thousand) involved in the experiment and half of them told the researchers how the money was not enough: >Indeed, after the study concluded, we conducted qualitative interviews with 15 participants, where—among other things—we asked them how much money they would have needed to receive for it to have made a difference in their lives at that time. About half of participants indicated that they would have needed to receive more than $2,000, with those responses falling between $4,000 and $25,000. Although this is by no means a representative sample or ideal methodology, these responses may at least be indicative of the general order of magnitude needed. How surprising is it that someone who is living at the poverty level or under it when given a "small" amount of money (no matter if 500 or 2,000) after one month has spent it?

jonah 4y ago
It really seems like an insultingly small amount of money. It's almost like they're trying to disprove something like UBI by association while not actually doing something that's useful. If they'd given those randomized amounts monthly for a year or 5 years, I would bet the results would be in line with similar studies which show marked benefits.
- naillo 4y ago
  Yeah were they expecting them to get $500 and within 4 weeks start a company and become cash flow positive? Being poor is often a consequence of education and with a few years of guaranteed income to make them feel safe to take that risk, they probably would have been able to tackle that. But it's not surprising at all that nothing happens in just a month.
- em500 4y ago
  Gook luck trying to fund a sufficiently powered RCT to give people $2/month for a year, let alone 5. $2k for 5 years is $120k per participant, so about $120M for a 1000 sample size study.
  - jaclaz 4y ago
    Sure, but imagine making a medical experiment where you give once 1 mg of the drug (that should be dosed at 1 mg/day as per results of previous experiments for at least 2 weeks) and then writing an article in which surprisingly it comes out that it didn't work well.

jka 4y ago

Worth mentioning that the experimental payments during this research were one-time-only. (a few descriptions[1][2][3] of Universal Basic Income seem to define it as including recurring/periodic payments) [1] - https://en.wikipedia.org/wiki/Universal_basic_income https://en.wikipedia.org/wiki/Universal_basic_income [2] - https://basicincome.stanford.edu/uploads/Umbrella%20Review%20BI_final.pdf#page=3 https://basicincome.stanford.edu/uploads/Umbrella%20Review%2... [3] - https://www.investopedia.com/terms/b/basic-income.asp https://www.investopedia.com/terms/b/basic-income.asp

erklik 4y ago
Nearly all UBI is a recurring payment. Otherwise it's not really capable of living on.
jaclaz 4y ago
Yes, it is usually intended as a monthly payment, BUT the actual UBI (in theory) doesn't (and shouldn't) pass through a "means test", i.e. (still in theory) it is "Universal" and as such all citizens should get it regardless their income or estate, as the Wikipedia article highlights: >This article is about the system of unconditional income provided to every citizen. For the means-based model of social welfare, see Guaranteed minimum income. https://en.wikipedia.org/wiki/Guaranteed_minimum_income https://en.wikipedia.org/wiki/Guaranteed_minimum_income also sometimes called "basic income" (without the "Universal") which is more pertinent and actually the one on which some (mostly failed) experiments have been done or are being made.
retrac 4y ago
Indeed. One of the main criticisms of both Canadian major experiments with UBI (Manitoba in the 70s, and Ontario in the last few years) was that while they paid regularly, both programs were relatively short-lived -- a few years. This doesn't appropriately model the incentives and behaviours people would likely engage in if they knew with reliability the program would still be paying them in ten or more years. It wouldn't capture either negatives like people leaving the workforce permanently, or positives like going back for a four year degree they couldn't otherwise afford. Participants likely treated it as a one-time windfall (if over a couple years).

stevenicr 4y ago

This is the funniest thing I've seen all week. "found giving low-income Americans cash isn’t that helpful ... see no significant difference in bank balances" Funniest because it proves the complete ignorance of those without poor problems. Of course there is no more money in the bank. There are dozens of things more important than money in the bank.. the brakes that should of been replaced 8 months ago. The tires that are unsafe. Dental health. Many more. 'We find NO positive effect of cash—if anything, in some prespecified analyses, we even see negative effects.” Apparently the cash infusions added stress' Of course - it's stressful trying to choose - do you spend the 500 on your kid's teeth or do you get your teeth worked on? Do you spend 500 at the vet or buy healthier food for the family this week. Do you replace your brakes or the tires? Do you fix the brakes or go to the dentists? Do you feed your kids better or stop the parasites sickening the pets. Do you X or Y? Should you Y or Z? Yes those things are stressful, and certainly more stressful than 'I can't afford any of these things and that's the way it is so stop stressing it' - I'd like to say "Duh!" - but I don't think the high brow harvard study would really get it. Although I disagree with 'found giving low-income Americans cash isn’t that helpful' - I mean it's stressful - but at least it is helpful in fixing on or two of the life changing things that need fixing - so it is helpful. It's just when you have a dozen life altering debts accruing life changing interest already (deferred maintenance leading towards bigger problems) - a couple grand does not fix all the things that need fixing in the American life. Maybe there is some other place in the world where you can get all of your healthcare, vehicle maintenance and other necessities all taken care of for $2,000 (?) - maybe those folks could put some money in the bank.

phkahler 4y ago

If you gave a Harvard professor two grand, you would not perceive any change in their bank account a month later either.