4 ms·

> In addition, the UBI creates bad incentives, and requires enormous taxes if funded at an “acceptable” level. > A UBI experiment, in contrast, might be a good

> In addition, the UBI creates bad incentives, and requires enormous taxes if funded at an “acceptable” level. > A UBI experiment, in contrast, might be a good way to convince people of the folly of the UBI. For a modest cost, you could persuasively demonstrate the strong disincentive effects, reducing support for this massive waste of resources. Of course, this assumes that fans of the UBI actually care about evidence! > ... > Score: 10/15. GPT-4 fails to explain that a UBI is bad by EA standards because it does the opposite of targeting. “Might not have the same impact” is a gross understatement. It also misses the real point of a UBI experiment: To convince believers that this obviously misguided philanthropic strategy is misguided. The author seems VERY against UBI. I don't think it is really fair to ChatGPT to mark it down because it does not also strongly oppose UBI. I also don't know what "bad incentives" UBI creates. UBI literally creates incentives for the poor to work because it means working does not deprive them of governmental benefits.

sterlind 3y ago
ChatGPT lost points because of the alignment problem - ChatGPT's explanation didn't align with the author's political ideology! But yes, personally I know several people on disability who would love to try working part-time, but making more than the (extremely low) cutoff puts you at risk of losing disability income and even your health insurance entirely. And a lot of people are in the boat of being sporadically able to work, or able to work a little, but not enough to survive. SS(D)I also forces people to divorce their spouses because spousal income counts 100% against your own income. It's an extremely cruel system.
- SoftTalker 3y ago
  It's a cruel system because of the history of massive fraud.
  - toomuchtodo 3y ago
    Can you provide a citation as to how prevalent fraud was and what percentage of total benefits the fraud incurred as a cost?
    - yowzadave 3y ago
      Also, the cost of fraud must be balanced against the cost of enforcement, both in terms of the cost to run the enforcement apparatus, and the cost of not providing benefits to people who are owed them because of the onerous requirements to get those benefits!
  - jychang 3y ago
    There’s massive tax fraud at the high end too, does that mean you agree we should be cruel to anyone earning over 6 digits just in case they are committing fraud?
246235462 3y ago
UBI /is/ a government benefit. If they were able to subside on their government benefits before then all that UBI does is give them even more generalized benefits thus removing even the minimal incentives they already had to work.
annoyingnoob 3y ago
If I know the government is now giving my tenants $500/month, the rent is going up $300/month. As an example of bad incentives.
- grayhatter 3y ago
  While I think the action is hostile and clearly unethical... I'm sorry you're getting down voted, for your hypothetical example... seems HN doesn't like different opinions. This is a good answer.
- chimprich 3y ago
  If you think everyone is getting a net $300 then you've misunderstood the proposal of UBI. Taxes increase and government benefits decrease to compensate.
- maxerickson 3y ago
  In a scenario where a UBI made sense, it isn't clear that you would have that leverage. Like if there is so much productivity that there isn't work to go around, building additional housing is a matter of choice, not a matter of cost. Same with pretty much every day to day need. Of course there wouldn't be unlimited yachts or whatever.
- gamegoblin 3y ago
  In a free market, prices are set by supply and demand. If you decided to raise rent $300/mo and the landlord next door chose to not raise them, your tenants will move. Obviously, if you live in a city without a free market in housing resulting in an artificially reduced supply (e.g. San Francisco), your strategy will work. Or, if you and all the landlords in the area collude (forming a cartel), you can also make this work, assuming you can avoid legal action.
  - setr 3y ago
    If every landlord knows every tenant suddenly has more money, they should all be raising their rent. And they will compete in doing so, and the renters will compete for space/price, and because they all have more money they can/will bid higher. You don’t need collusion for prices to increase with UBI, you just need supply and demand to do its thing
fijiaarone 3y ago
[flagged]
- grayhatter 3y ago
  > Being pro UBI is very strong evidence of lack of intelligence, even if your reason for supporting UBI is purely self-serving. Do you have an actual argument to go along with this insult? > It’s like killing old ladies for spare change in front of their sons biker gang and police level of amoral stupidity. I can't even pares this analogy; it's nonsensical to me.
- wizofaus 3y ago
  Even Elon thinks UBI is almost inevitable once AI/Automation makes most current human work unnecessary. I'm inclined to agree though I doubt it'll be a smooth path to get there (and may not even happen in my lifetime).
bloaf 3y ago
> A UBI experiment, in contrast, might be a good way to convince people of the folly of the UBI. For a modest cost, you could persuasively demonstrate the strong disincentive effects, reducing support for this massive waste of resources. This response is insidious, but not because of the professor's position on UBI. If we already had overwhelming evidence that a new UBI study would "persuasively demonstrate the strong disincentive effects", then that evidence would by itself "persuasively demonstrate the strong disincentive effects." Furthermore, if such evidence existed, the proponents of UBI would already be ignoring it. Therefore, the first strike against the suggested response is that it lacks basic logic: either the professor lacks enough evidence to knows the outcome of the study before it is done, or the professor is not justified in asserting that more such evidence will be persuasive. The second strike is this: If the professor expected students to be familiar with the evidence against UBI, then we should expect the professor to be looking for the students to demonstrate that familiarity by referencing it in support of their conclusion (i.e. that a UBI study would have a certain outcome.) That the professor does not expect this indicates that he is not training students to make arguments based on evidence. The final strike is the evaluation of the UBI as a persuasive tool. Even if we assume that the professor is correct on both counts (i.e. that a UBI study will be negative and persuasive), that by itself isn't a justification for a "modest cost" in an Effective Altruism framework. I wouldn't spend a "modest amount" to convince some people that the earth is not flat, for example, because it would have no real impact on any of the metrics Effective Altruists use. Because the professor stops short of connecting the educational outcome to the EA metrics, he has presented a failure of a response as a suggested response. So to summarize: the response is internally incoherent, expects authoritative phrasing over evidence, and ultimately fails to directly connect the claims to the question.
jillesvangurp 3y ago
Exactly, subtracting points for failing to meet his bias. Arguably, gpt-4 gave the better, more neutral answer here. Suggesting an experiment might be enlightening either way.
Izkata 3y ago
> I also don't know what "bad incentives" UBI creates. UBI literally creates incentives for the poor to work because it means working does not deprive them of governmental benefits. The "B" in UBI means "enough to live off of". I'd expect more people to take advantage of it for the freedom and stop working than would start working.

sterlind 3y ago

ChatGPT lost points because of the alignment problem - ChatGPT's explanation didn't align with the author's political ideology! But yes, personally I know several people on disability who would love to try working part-time, but making more than the (extremely low) cutoff puts you at risk of losing disability income and even your health insurance entirely. And a lot of people are in the boat of being sporadically able to work, or able to work a little, but not enough to survive. SS(D)I also forces people to divorce their spouses because spousal income counts 100% against your own income. It's an extremely cruel system.

SoftTalker 3y ago
It's a cruel system because of the history of massive fraud.
- toomuchtodo 3y ago
  Can you provide a citation as to how prevalent fraud was and what percentage of total benefits the fraud incurred as a cost?
  - yowzadave 3y ago
    Also, the cost of fraud must be balanced against the cost of enforcement, both in terms of the cost to run the enforcement apparatus, and the cost of not providing benefits to people who are owed them because of the onerous requirements to get those benefits!
- jychang 3y ago
  There’s massive tax fraud at the high end too, does that mean you agree we should be cruel to anyone earning over 6 digits just in case they are committing fraud?

246235462 3y ago

UBI /is/ a government benefit. If they were able to subside on their government benefits before then all that UBI does is give them even more generalized benefits thus removing even the minimal incentives they already had to work.

annoyingnoob 3y ago

If I know the government is now giving my tenants $500/month, the rent is going up $300/month. As an example of bad incentives.

grayhatter 3y ago
While I think the action is hostile and clearly unethical... I'm sorry you're getting down voted, for your hypothetical example... seems HN doesn't like different opinions. This is a good answer.
chimprich 3y ago
If you think everyone is getting a net $300 then you've misunderstood the proposal of UBI. Taxes increase and government benefits decrease to compensate.
maxerickson 3y ago
In a scenario where a UBI made sense, it isn't clear that you would have that leverage. Like if there is so much productivity that there isn't work to go around, building additional housing is a matter of choice, not a matter of cost. Same with pretty much every day to day need. Of course there wouldn't be unlimited yachts or whatever.
gamegoblin 3y ago
In a free market, prices are set by supply and demand. If you decided to raise rent $300/mo and the landlord next door chose to not raise them, your tenants will move. Obviously, if you live in a city without a free market in housing resulting in an artificially reduced supply (e.g. San Francisco), your strategy will work. Or, if you and all the landlords in the area collude (forming a cartel), you can also make this work, assuming you can avoid legal action.
- setr 3y ago
  If every landlord knows every tenant suddenly has more money, they should all be raising their rent. And they will compete in doing so, and the renters will compete for space/price, and because they all have more money they can/will bid higher. You don’t need collusion for prices to increase with UBI, you just need supply and demand to do its thing

fijiaarone 3y ago

[flagged]

grayhatter 3y ago
> Being pro UBI is very strong evidence of lack of intelligence, even if your reason for supporting UBI is purely self-serving. Do you have an actual argument to go along with this insult? > It’s like killing old ladies for spare change in front of their sons biker gang and police level of amoral stupidity. I can't even pares this analogy; it's nonsensical to me.
wizofaus 3y ago
Even Elon thinks UBI is almost inevitable once AI/Automation makes most current human work unnecessary. I'm inclined to agree though I doubt it'll be a smooth path to get there (and may not even happen in my lifetime).

bloaf 3y ago

> A UBI experiment, in contrast, might be a good way to convince people of the folly of the UBI. For a modest cost, you could persuasively demonstrate the strong disincentive effects, reducing support for this massive waste of resources. This response is insidious, but not because of the professor's position on UBI. If we already had overwhelming evidence that a new UBI study would "persuasively demonstrate the strong disincentive effects", then that evidence would by itself "persuasively demonstrate the strong disincentive effects." Furthermore, if such evidence existed, the proponents of UBI would already be ignoring it. Therefore, the first strike against the suggested response is that it lacks basic logic: either the professor lacks enough evidence to knows the outcome of the study before it is done, or the professor is not justified in asserting that more such evidence will be persuasive. The second strike is this: If the professor expected students to be familiar with the evidence against UBI, then we should expect the professor to be looking for the students to demonstrate that familiarity by referencing it in support of their conclusion (i.e. that a UBI study would have a certain outcome.) That the professor does not expect this indicates that he is not training students to make arguments based on evidence. The final strike is the evaluation of the UBI as a persuasive tool. Even if we assume that the professor is correct on both counts (i.e. that a UBI study will be negative and persuasive), that by itself isn't a justification for a "modest cost" in an Effective Altruism framework. I wouldn't spend a "modest amount" to convince some people that the earth is not flat, for example, because it would have no real impact on any of the metrics Effective Altruists use. Because the professor stops short of connecting the educational outcome to the EA metrics, he has presented a failure of a response as a suggested response. So to summarize: the response is internally incoherent, expects authoritative phrasing over evidence, and ultimately fails to directly connect the claims to the question.

jillesvangurp 3y ago

Exactly, subtracting points for failing to meet his bias. Arguably, gpt-4 gave the better, more neutral answer here. Suggesting an experiment might be enlightening either way.

Izkata 3y ago

> I also don't know what "bad incentives" UBI creates. UBI literally creates incentives for the poor to work because it means working does not deprive them of governmental benefits. The "B" in UBI means "enough to live off of". I'd expect more people to take advantage of it for the freedom and stop working than would start working.