Rossi's Revenge
thecausalfallacy.com13 points by exolymph a day ago
13 points by exolymph a day ago
Well, that means the AGI fantasy/myth is dead too.
> "a large amount of money per month (between $333 and $1,000)"
A substansive amount, but does this really change the class of problems facing anyone in the U.S., aside from acute hunger? Does this remove any risk of debt hell or structural issues around the types of issues people around the poverty line face?
I am not really surprised by the results. Capable, poor people tend to use all the tools they have available from federal assistance to leaning on their community. Helicoptering money might invalidate them from certain aid. A chunk of it goes back to taxes. Certainly the money was appreciated and used, but no surprise it didn't "solve poverty".
I don't understand blindly throwing money at lots of people and asking them to do a survey and calling it science. Most people who win the lottery end up equally or more unhappy. Does that mean if you give everyone a million dollars, there would be no increase in leisure or happiness? Too little thought is given to wider societal changes in favor of individual impact, but were are very much the result of our environment.
People are not islands, and people with little money are likely more entwined with their communities. If you want to affect change you either raise the community or transplant an individual from a community. Studies like this completely ignore the social and emotion aspect of humans but then ONLY look at those factors (are you happier?) to judge effectiveness... with expected results.
Quoting one of the studies: "The Employment Effects of a Guaranteed Income: Experimental Evidence from Two U.S. States" by Vivalt, et al. https://evavivalt.com/wp-content/uploads/Vivalt-et-al.-ORUS-...
> Participants reduced their work hours as a result of the transfers by 1-2 hours/week and participants’ partners reduced their work hours by a comparable amount. Among other categories of time use, the greatest increase generated by the transfer was in time spent on leisure.
> We observe no significant effects on investments in human capital, though younger participants may pursue more formal education.
Sounds like positive effects, actually.
Effectively, everybody treated the money like a one-off windfall every month that they could unwind if it went away. Yeah, that's smart thinking when you live in poverty.
> In our setting, we observe very limited asset accumulation (on the order of $0 to $2000) and in-creases in debt (of around $1000 to $2000) in the treatment group relative to the control over the course of the study (Bartik et al., 2024). Participants appear to spend approximately the full amount of the transfers each month, on average.
So, basically these people are sufficiently strapped that $1000 per month is insufficient to do much more than handle an extra crisis or two. If you have a child, $1000 will get eaten up immediately--although it probably lets you spend a day or two with a sick kid (that would be 1-2 hours per week, no?) or possibly take them to a doctor when you previously might have gone to work.
In addition, these people know that there is a horizon in which the money will go away--and that horizon might be MUCH sooner if political will changes.
No one is going to rely on that $1000 to go to school for several years when it is going away. No one is going to change their housing situation when the money could go away. No one is going to rely on that $1000 to invest in anything when they have immediate needs.
And all this assumes that $1000 doesn't wipe out some other assistance that they are already getting. The US system is notorious for dropping aid if you suddenly get paid even slightly too much.
It would be more interesting to me to see what giving them a check for $36,000 would do rather than $1,000 per month.
Strawman detection activated.
The goal of giving money is not to make those who receive it more educated, better parents or making them politically active. I mean, why did I have to say that?
(The amount of money is also unlikely to enbale to make savings for the future.)
The goal is that their immediate poverty decreases. Why is the author insistent on the measurement of unrelated stats? Because he has an ax to grind.
So to be clear you would consider UBI a success if the only detectable change in metrics is simple arithmetic in the form of “we gave them money and then they had more money”?
I’m sorry to sound snarky but I’m struggling to read this comment any other way. You seem to claim any metric that isn’t a number representing dollars a person has (however ephemeral) is “unrelated” which seems completely bonkers to me.
Early advocates (e.g. Friedman) for UBI (and also NIT) often focused on broader distributional outcomes for society and the economy as a whole. One is the reduction of the "welfare cliff". Another is the increase in relative spending power of the lower class as a group, which could lead to growth in businesses that serve these cohorts. Neither of these effects is assessed at all by RCTs that look at a small population of individuals for a period of less than a decade, because they are effects that occur over a long time and among a necessarily large (millions) group of people.
There is a sort of unstated assumption among some social policy critics that goes, roughly, "we can test most policy effects that matter with well-designed trials". People believe this less because there is any evidence for it and more because modern experimental science is impressive and successful in many ways and so therefore must have the answer to any question. However, many policy questions remain "wicked" and outside of a reasonable experimental domain.