▲ 117 points
back
94 comments
"while there were large economic gains for recipients (targeted as the poorest), gains for non-recipients were so substantial that the gap between the poorest and least-poor within villages widened slightly." Now that is a novel result!
Publicising that might be a way of overcoming the resistance of the large number of people who are mainly interested in their own relative status, and so dislike giving assistance to the poorest members of society. If you can show people that yes this intervention will help the poorest but it will help you more then those concerned mostly with their own relative status might be more likely to back it.
If you can show people that yes this intervention will help the poorest but it will help you more then those concerned mostly with their own relative status might be more likely to back it.
Folks have been trying this. When the poorest folks have enough money to spend on things, everyone tends to benefit - it isn't a new theory. But mostly, teaching folks this has failed. It is either that, or you need to settle on "people simply dislike giving others help, even if they reap rewards."
I think part of the issue is that the rewards aren't nearly as visible or shockingly impactful as actually being given money to thrive.
One important caveat is that in the study, as far as I can tell, the money is coming from a different country (presumably the USA) than the target country (Kenya), so there is a baseline effect of increasing the net wealth of Kenya. If the money is collected from and distributed to the same community, the results may not necessarily be the same. I think it would be very interesting to see a study along these lines, but I’m not sure how feasible it would be.
Which is key - a system that had money injected into it showed the results expected of increasing the money supply.
The same may not happen on a national scale if you’re circulating inside it.
The general trend of history seems to suggest this isn't true.
Why do you think it is?
I mostly see propaganda and lies directed at stopping this from happening rather than a basic human response to it.
Like say the 'welfare queen' myth, or the trope that 'giving people money will just cause them problems or make them lazy', seems you have to really stretch to convince people that helping the poor is a bad idea and the standard human response is to help.
> Like say the 'welfare queen' myth, or the trope that 'giving people money will just cause them problems or make them lazy', seems you have to really stretch to convince people that helping the poor is a bad idea and the standard human response is to help.
In my experience, these attitudes are more common among those who live around poor people than among those who don't. It's my wife who taught me the difference between "people who are good at being poor and ones who aren't." Coming from a privileged background, I had no idea.
It's a sociocultural adaptation to scarcity. Historically, whether you were in a Bangladeshi village or rural Oregon, even the rich don't really have sufficient surplus to help more than a few people. And there were no rich foreigners or central government to swoop in. In that scenario, anticipating and expecting help is maladaptive. Everybody in the village needs to be incentivized to work as hard as possible for themselves before relying on anyone else.
It not that simple. The same theory say that government should spend more money during recessions since that send more money into the hands of citizens, thus increasing peoples ability to consume more and drive up the economy. However, by doing so the amount of money in the system increase and thus you get inflation, forcing the government to spend even more money.
This doesn't happen every time as is demonstrated by this study, but history has plenty of example where it has and hasn't worked. The open question is to figure out when it work and why it doesn't work in other situations.
The cause of the inflation matters in how the response plays out. The inflation we're seeing currently isn't being driven by the general population having too much money but the response by central banks has been to act like it is.
Personally I view the two major factors to be post pandemic shipping costs and the Ukraine. A meaningful response from government could have headed this off but instead we're seeing energy producers hit record profits while large parts of society are going to be unable turn their heating on over the winter and on top of that rising interest rates are going to see every other aspect of their budget become even more expensive as well.
Government can always respond by taking money back out of the system in the form of taxing wealthy people. You don't have to print money. You can just redistribute what's already there.
They might dislike rational helping because they see their own place in the economic game as irrationally hampered, and under conditions of perceived injustice, would rather get their own before extending access to others.
Some years ago I was talking with a friend of mine about how things were going badly in a particular country that is culturally of interest to us. I was saying something about how it would be so much better if they had better economic growth so that people there could be less poor, and she said something like "but then they would consume more!". She's not rich, but she's definitely not poor -- she's a divorced professional, she works at home, and she has a wonderful, large house in midtown Austin (in a great location) that she owns outright. Her house is probably worth a lot of money, but she and her ex-husband bought it decades ago. I probed a bit and she was not embarrassed by what she said. She consumes a fair bit, but she doesn't want others consuming more than they do, even though they consume a lot less than she does.
The problem is that the conservative perspective would largely rather go without these net benefits than see someone they think is undeserving, or worse "immoral" in some way, get any benefit.
That's why we have means testing at all: not to prevent people from needlessly getting benefits, but to prevent those who are undeserving of them from getting them.
A lot of things make more sense when you understand this imo, and stop trying to frame other people's views into your own moral framework (giving you the mistaken impression that they're hypocrites about it).
The conservative perspective is not, and has never been, that poor people deserve to stay poor. It is that I can and will give my money to poor people, and you're free to do likewise, but we're not going to point a gun at old Steve and make him do the same.
Now that gun-pointing redistribution programs are fully entrenched, some people are still trying to have some say - ANY say - in where their money goes.
"Trickling up" instead of "trickling down"!
Or trickling from the people who do nothing to the people who have to produce something nice for the lucky ones to buy. From the non-workers to the workers.
Also seems to suggest that minimizing wealth gaps may not be the best objective function
(by itself) relative to other possibilities. Curious.
Did anyone suggest it was? That seems like a very weak straw man of the very many sensible arguments from multiple angles, for why vast wealth inequality is a bad thing for society, including the members of the society with the extreme wealth but surrounded by poverty.
You can have a large wealth gap, while still having a high bottom standard of living. A great example of this is Sweden https://www.youtube.com/watch?v=2E0dWHCnic8
Supplying handouts does not affect the wealth gap - giving the needy some real capital like own housing, shops, etc would actually contribute to that goal. Handouts don't help enough because the money is spent on their necessities (which was the point), and obviously that means the money immediately flows to the rich ones that supply those items (housing, food, etc).
Reducing the wealth gap is crucial, I think you're severely misunderstanding.
Let's imagine this in the most extreme form in a thought experiment. An identical twin is stranded in the desert. You have the means (water, transport) to save one of them. Is it better to save one (which one?) or leave both to die? (I have not settled on a firm answer so far.)
It's interesting, this is the sort of thing that Angus Deaton suggested when he argued against the EA program in a Rationally Speaking podcast[1]. His premise is that cash transfers to the needy won't work in a country where an oppressive government can just take the cash for themselves. Perhaps this is part of the effect here? (Bribes, taxation, etc.) I'm a bit skeptical that this is a knock-out argument against the GiveDirectly program, but it's definitely something to keep in mind and explore further.
[1]: https://podcasts.apple.com/us/podcast/deaths-of-despair-effe...
But you can also say the opposite: handouts increase inequality and poverty.
Who is going to suffer because of the enormous covid handouts and subsequent inflation?
It was nice while it happened but a lot of rich people really benefited and a lot of poor people are going to go through the worst consequences of inflation and recession.
> Who is going to suffer because of the enormous covid handouts and subsequent inflation?
Don't know how it played where you live, but in my country, most covid handouts went to the rich, not to the poor.
1. Cash was given to companies they could then give to the people, instead of straight to the people.
2. Cheap/free loans were given to companies/rich people, not poor people, because poor people can't pay off loans so nobody gives them any.
The insane stock market profits were the result of this - rich people getting excess money. If it went to the poor people, they'd buy food and shelter, not stonks.
> But you can also say the opposite: handouts increase inequality and poverty.
They didn't say anything that would let you conclude either of those things in the article, one they said the exact opposite of. Not read the whole paper yet though.
Even if I also have a hunch that your attribution of inflation to handouts is true, I do not believe that it has been established.
What is clear however, that increased prices for labor, wheat and energy, combined with post-covid logistics issues, have very clearly increased consumer pricing.
handouts increase inequality and poverty.
Or perhaps another way to say it is this: cash doesn’t matter. It’s all about capital. If you want to reduce inequality, give the poor some capital. Unfortunately, I think they’ll likely sell their capital to buy food or pay rent.
We use handouts to prevent revolutions and political instability. You don't recognize the absolutely massive benefits that those payments have to everyone because they get to ignore the poor.
This is common sense.
When the government was handing out money indiscriminately to everyone right after Covid, we just put the money in the bank. That helped us. But it didn’t stimulate the economy. When they gave the same amount of money to some people we knew that lost their jobs, they spent it on necessities - helping the economy.
Giving money to people who will spend it obviously helps everyone. But there have been plenty of studies showing that people will hurt themselves financially even if it will help other people. Especially if it helps “them”.
I like givedirectly very much, especially because they try to do actual research on the outcomes. However I would love to see some research that is not done directly by them, but is more independent.
Givewell has pretty in depth reports on the charities they recommend. They’ve spent a very large amount of time trying to figure out the efficacy of giveDirectly, largely because it is effectively an infinite sink that all other philanthropic activities can be compared against. Interestingly givewell just recently demoted giveDirectly to instead recommend a charity that does cash handouts to parents who get their children vaccinated. I’ve got the report on my reading list but haven’t had a chance yet.
Note that in this case one effect these cash payments don't have is: coming out of the pockets of those who don't receive them. That's because these cash payments are from abroad. There's a very big difference between cash payments from influxes versus domestic cash payments paid for by loans, taxes, or inflation.
Trickle down economics?
In fact it's almost exactly the opposite. Rather than the supply-side economics which trickle-down focuses on, this sounds like a clear example of demand-side economics. By improving the income of some consumers in a community, they are able to buy more goods and services that they wouldn't have otherwise, thus improving the local economy altogether.
Does it improve that economy by at least as much as the cost of the transfers? It seems like a lot of economic benefits can be created by injections of money from outside the system under study.
That's the point of the fiscal multiplier figure: according to their estimates it improves the economy by a multiple of 2.6x the cost of the money transferred.
-
Seems like reverse "trickle down" in another respect too: the poorest recipients got better off but the gap between rich and poor still widened, presumably because the poor spent much of their increased budget buying things from the rich.
It wouldn't surprise me if that were the case.
Imagine someone who sewed things by hand, given (or even loaned) the money to purchase a sewing machine, making them many times more productive. Now the community can have things sewn faster and more easily with less effort, with tons of downstream effects (more sails, better clothing, stronger soft goods, etc) and it's easy to see how the net improvement to that community can far exceed the price of the machine.
Another way to ask that question is "does there exist tools whose price is less than the total amount of increased economic activity they enable", which when phrased that way, it seems clear that the answer is "yes".
I would think that the more consumers that can "vote with their wallet" on which products are the best, the more quickly the market will identify and reward good products. Identifying and rewarding the best products (i.e. those processes that create the most valuable outputs from the inputs) is one of the primary benefits of market-based economies. The positive feedback loop of reinvesting profits helps them lift above the general noise.
Yes now go ahead and pay out the benefits in a demurrage currency so that benefits payments can't be withheld from the economy which means the amount you need to inject will be much lower.
How is supply side economics, with tax reductions the main focus, not also driving demand?
The people who get their tax reductions tend to be rich and not spend their money.
It's probably worth applying some scepticism to this study:
1. It's an economics paper. Not a field with a reputation for accurate predictions beyond the basics.
2. It's commissioned by a charity that wanted academic evidence of the benefits of its own policies, so it's as conflicted as can be.
3. It's research by academics who are as a group notoriously left wing, so they're being asked to do research to support their own ideological preconceptions.
That doesn't mean the conclusions are wrong, but it does mean they shouldn't be accepted at face value, and you'd have to review it very thoroughly - ideally with external double checks of validity - to be able to conclude anything from it.
Doing even a very fast check of the paper immediately reveals a major problem: the headline claim of 0.1% price inflation excludes the price of labour, which inflated significantly. This pretty much explains the entire effect. Not surprisingly, if you start handing out cash at random in poor communities then the people who aren't lucky enough to get it start demanding some of that cash too. But you don't need research to know this. It's economics 101, that's what's happening right now in western economies (big cash transfers from governments -> inflation).
We may suspect the paper is being undermined by ideology and funder bias because its early statements on inflation try hard to hide this fact. At the start, it's saying:
"Importantly, we document ... minimal price inflation"
(abstract). Then it becomes:
"For inputs, we find positive point estimates [of inflation], but they are not always statistically significant. For outputs, we document statistically significant, but economically minimal, local price inflation."
(page 4). Unless read very cynically this sounds like good news. But then we notice how obfuscated this statement is. "Positive [input] point increases but they are not always statistically significant", i.e. some are and the magnitude of the increase isn't mentioned. Later we find:
"we do see significant increases in the factors that we directly measure, and particularly in the wage bill: enterprises in treated (control) villages increase spending on labor by USD PPP 82 (70), a sizable change relative to the control mean"
(page 20). So we've gone straight from "minimal price inflation" to "significant increases ... a sizeable change". I've read so many academic papers in the last few years and whenever you see such large slippage between claims in the abstract and claims buried deep in the paper, it's always a bad sign. It's a leading indicator that there's going to be other kinds of problem, probably involving bogus stats/models/assumptions.
Moving on. We might have hoped that people would use the money for investment that could permanently increase their standards of living after the cash transfers end, but that doesn't happen:
"Strikingly, we do not see strong evidence of a firm investment response"
But this isn't something that needs research to know, it's the story of decades of massive and sustained foreign aid to Africa. Aid money gets spent on increased consumption but not sustainable investment.
> So we've gone straight from "minimal price inflation" to "significant increases ... a sizeable change".
I'm not sure the fact that a different variable (total wage expenditure) from the one discussed in the abstract (price inflation) recorded a different outcome is a particularly pertinent critique of the paper.
Since the paper also records that the increased wage expenditure allowed wage earners to buy more goods because the price of goods inflated minimally, and the enterprises spending more on wages nevertheless experienced a [non-statistically significant] increase in profit, leaving all groups materially better off, this is GDP growth, not inflation.
A more pertinent observation is that this is pretty much what you'd expect from injecting additional money into a local economy operating well below capacity and buying most of its consumer goods from areas not receiving cash injections, so it doesn't necessarily work at national scale or if the jurisdiction self-funds its poverty alleviation funding.
Although I agree with your conclusions that the study doesn't tell us much, I don't think wages should be excluded from claims about price inflation, especially in non industrial economies heavily reliant on cheap labour. The charity advertises the headline 0.1% rate and the abstract makes no mention of the fact that actually some of the most important prices in that society increased significantly. If they'd admitted to this up front of would have seriously reduced the apparent success of their initiative.
But there is really a deeper problem here. Every time researchers do this they're training people to assume their claims are deceptive in some way. Large groups of people are just tuning out academic claims because of this sort of thing, they don't care. But this is bad for social cohesion because the people who take academic output on faith then conclude that they must be a superior breed of person: "reality based", "understands the science" etc. We already have this problem and it's getting worse. To wit: you can't afford the time to double check every claim presented as important or that will affect social policy that affects you, so you have to generalize, and increasingly that means assuming that if an academic makes a claim convenient for their prevailing ideology, it's probably a trick or misleading in some way. If caught they tend to blame journalists for "misrepresenting" their work, or they'll point to a footnote on page 67 where they redefine a standard term and use it to claim nobody should have ever assumed the obvious interpretation of what they were saying. It's just so tawdry. Then for people who stop listening, it gets used as a weapon to beat them around the head.
So I think academics have a moral obligation to be brutally honest in their claims and abstracts. This sort of word game where they arbitrarily exclude the prices that went up from their definition of price inflation, is ultimately self defeating.
> 1. It's an economics paper. Not a field with a reputation for accurate predictions beyond the basics.
If your first reason for dismissing an economics paper is because you don't like economics, that's a bad start. Especially since the rest of your comment reviewing this economics paper consists of you doing economics. You're preemptively dismissing everything you yourself have to say.
I mean, if economists suck, then surely the economic theories of people who don't think economics is real suck more.
I think skepticism towards individual psychology papers is warranted given the replication crisis in their field.
I think the same is true of economics — and is the sentiment that person was expressing.
Similar to this, a few weeks ago, centered on the US:
Worth pointing out the summary (2nd link) by the people is arguably pretty bad. They are basing being better off on reactionary feelings and money left in the bank. That makes no sense at all for people with little wealth. It seems like they purposefully picked things to make the study a failure.
Many of the children comments explain this well.