GiveDirectly Gives Poor People a Year's Income
huffingtonpost.com
huffingtonpost.com
I am curious what the effects are on those who don't receive the transfers. (Those who don't meet the requirements, or live in neighboring towns, for instance.) While I'm certainly not saying it outweighs the positive effects, there must be some negative unintended consequences. Someone uses the funds to start a business, which is enough to make an existing business in the field unprofitable for instance. I wonder how much research has gone into those types of side effects.
My understanding is also that direct cash infusions can have undesirable macro effects - basically Dutch Disease caused by aid money instead of natural resources. Again I'm curious to what extent those effects are understood.
If anyone has knowledge or links to resources along those lines, please share! I did find this somewhat more critical review[1] but it still just focuses on the recipients, rather than knock-on effects.
[1] http://www.ssireview.org/blog/entry/givedirectly_not_so_fast
> Do grants distort local markets? It seems possible to us that a large infusion of cash into an area could alter economic opportunities for both recipients and non-recipients. Such effects could be positive (for example by spurring investment and job creation or by increasing the availability of retail goods) or negative (for example, by leading primarily to local inflation). The limited evidence addressing this issue in the RCT of GiveDirectly's program in Rarieda[1] and the broader literature on cash transfers points to no distortion. There is an ongoing RCT[2] of GiveDirectly's program that is testing for macroeconomic effects.
Not a ton of concrete results as yet, especially for the larger macroeconomic effects, but at least from that one study the local effects appear to be minor. Certainly more research would be valuable.
[1] http://www.givewell.org/international/technical/programs/cas... [2] http://www.givewell.org/international/top-charities/give-dir...
It's also worth pointing out that the "minor effects" are minor because they're not statistically significant for the sample rather than because they're not necessarily there. Haushofer and Shapiro, for example, finds that whilst recipients spend $36/month more, comparably poor non-recipients in a village spend $7.5 per month less. The sample size and volatility means the latter part of result isn't statistically significant, but on the other hand it is more consistent with the effects on the poor that didn't receive handouts in these particular villages being negative, and indeed if only a fifth of the village received the subsidy it's even consistent with the benefits being entirely positional, even though the dispersal of the losses amongst a larger population makes it difficult to precisely gauge.
GD gives to the poorest, like raising the minimum wage. They don't make anyone richer than a neighbor.
The SSI article is a hit piece with dodgy math, written by a competitor. It claims that a $1 gift that findances $1 of consumption PLUS $.80 investment profit is somehow a negative return. Ignoring the fact that the baseline for poverty relief is $0 investment profit.
Obviously there's potential for abuse, but that's the case with regular aid, too, and it doesn't end up making a big difference. Also maybe not entirely fair, but it's a net gain, which is better than no gain at all.
Google can probably find better references, but here are a couple of links backing the idea:
http://www.economist.com/blogs/freeexchange/2014/08/costs-fo...
If there are any studies about how direct donations affect social dynamics within a community, I'd be very interested to read them.
https://news.ycombinator.com/item?id=9664983
jbniche points out: "Just think the impact that would have had if it had been divided up and given directly to each Haitian adult. The adult population of Haiti is roughly 6 million, so we're talking about almost $100 per adult Haitian. Roughly, it's about 1/8 of the per capita GDP of Haiti. Imagine what you could if you were a poor American and received 1/8 of the American per capita GDP after a disaster (~6000 USD). That's a new roof, or a replacement vehicle, etc. It's not life changing, but it would have been very significant, and massive in scale."
https://news.ycombinator.com/item?id=9666390
and the article supports his point, saying that "A 2013 study in Uganda found that people who received cash enjoyed a 49 percent earnings boost after two years, and a 41 percent increase after four years, compared to people who hadn't gotten a transfer. Another study in Sri Lanka found rates of return averaging 80 percent after five years. In Uganda, not only were the cash recipients better off, but their number of hours worked and labor productivity actually increased."
But for smaller amounts (like the U$ 100 proposed), "One program gave $200 to at-risk Liberian men who were either homeless or who made their income from dealing drugs or stealing. The lead researcher, Chris Blattman, summarized the findings in an op-ed in The New York Times:
"Almost no men wasted [the money]. In the months after they got the cash, most dressed, ate and lived better. Unlike the Ugandans, however, whose new businesses kept growing, the Liberian men were back where they started a year later. Two hundred dollars was not enough to turn them into businessmen. But it brought them a better life for a while, which is the fundamental goal of any welfare program. We also tested a counseling program to reduce crime and violence. It worked a little on its own, but had the largest impact when combined with cash.""
(I'm not defending the Red Cross, but I would unsurprised if "just give them money" is less effective in the case of disaster relief than in the case of development.)
Probably a good thing, because AFAICT, they're pretty much indefensible at this point.
If you just give poor people any money in Port au Prince after the earthquake, you simply atract predators that will beat and steal those people living in the streets easily, collecting all this money in the blink of an eye. Rich people in Haiti did not seem to care much for the bad luck of the poor people either.
I think this and other efforts are pretty clearly demonstrating that this doesn't necessarily happen.
EDIT: Not to say it can't happen, but I'm pretty tired of this quasi-religious assertion that giving money to poor people will just inflate that money into meaninglessness, particularly when it has repeatedly not happened.
(edit to avoid further confusion: The point of this post is simply to point out that local economies behave very differently after a disaster. In particular, if you need X of something in order for people to not die and you only have << X available (perhaps 0), then cash handouts aren't really helpful in the short-term.)
Show me cases of price gouging for vaccines during an outbreak, or tents during a disaster or whatever. (And not just isolated assholes attempting to profiteer, either — because there's always going to be "that guy" — but systemic inflation.)
Otherwise, all I'm hearing is a lot of "seems" and "shoulds", which sounds pretty belief-based to me, particularly when I keep seeing evidence to the contrary.
Or: http://newsroom.uber.com/nyc/2012/11/hurricane-sandy-pricing...
(Note that inflation was systemic, because Uber basically became the entire market. Yellow cabs, which couldn't inflate their prices, just stayed home.)
No matter how much economic denialism you want to engage in, you still can't magically make scarce things plentiful. Similarly, I can complain about how much my server monitoring sucks, but that doesn't mean my latency isn't high.
The other example is closer to what I'm looking for, but I'm not convinced it's entirely what I'm talking about, either. Yes, there was a (broadly) localized price increase, but it was in the context of a global price up-trend (which can be blamed on biofuels as much as it can poor harvests).
Your cited article points out that global corn prices, for example, were up 84%, and had doubled (or, up 100%) in the horn of Africa. (About which, btw: we're talking about one of the most lawless places on the planet. I very much doubt the usual rules of economics apply unmodified in such an environment.) Those aren't egregiously divergent price increases.
I'm not sure why you believe the "lawlessness" of a region changes the laws of arithmetic. Scarcity is scarcity, regardless of whether the UN refuses to recognize your government.
So it's important to have a clear idea of what is and isn't needed when trying to help people and that can depend very strongly on the circumstances they face.
WRT vaccines in particular, please read my post as charitably as possible within the surrounding context of the thread -- I can only invest so much time in internet commenting and sometime my grammar isn't perfectly unambiguous. Sorry :-). To help crystallize the point I was making, I edited my parent comment.
In general, as long as we're being pedantic, I suppose that "completely unavailable" -- meaning you can't locally buy a vaccine even for infinite money -- is a rather degenerate form of price inflation. And vaccine stores can be ruined by power outages, which are common during natural disasters.
Edit: Also, my parent comment states "price inflation for bare essentials [immediately following a disaster]", not "systemic inflation" in the sense of a country or region's entire economy. Cash handouts may very well be an effective mechanism for providing long-term redevelopment aid to an area effected by a natural disaster; in fact, that's essentially what happens with insurance payouts when middle class areas of the US are hit by disasters. But redevelopment is quite different from emergency response, which is what this thread is about.
I think the onus is on you to somehow demonstrate that the price of goods would not be affected by such a blindingly obvious combination of supply and demand changes.
I have no idea where you think this "evidence" is coming from that somehow in a disaster prices don't rise. If they don't, it would only because the entire market has been destroyed and "prices" don't even apply anymore.
During this past winter when ice dams were a massive issue in MA, national chain hardware stores were trucking in roof rakes and selling them for $150+ (normal prices of around $25-30).
Edited to add: and looking at the concerns of one of the aid recipients, something should probably also be done about radio programmes spreading conspiracy theories about devil worshipping aid workers.
I don't have any idea what and how can be done with that, given that the first world's situation is much, much worse in that regard. Most of what the media does is regurgitate bullshit about everything; we may not have "devil worshipers", but we do have "big pharma".
In general, soft power as an alternative to hard power is a serious question in national security strategy -- trhway's comment is not a troll, it's something security/IR researchers and policy makers take quite seriously.
http://www.thebestpageintheuniverse.net/c.cgi?u=penny_drop
(warning about the site: contains lots of offensive and obscene language that can mask the author being usually right about things)
The article directly talks about how the Red Cross did a very poor job of hiring locals, especially in positions responsible for making choices, and had a tendency to dismiss their value.
> If you could turn back the clock would you just write a check to all the parents of the kids who go to that school?
> “Definitely,” Meyers said.
http://thebillfold.com/2013/05/giving-money-directly-to-poor...
I predict a "fashion" trend in the poor architecture of the region to eschew better roofs even if you can afford one.
I also think that an editor of a newspaper in Kenya and Uganda should do a cartoon where a villager uses his GiveDirectly money to buy a better roof, and then the company takes the money back and accuses the villager of fraud.
As economists I know they know this, thanks to the Lucas Critique ;) But I didn't see what was being done about it.
Suppose GD will give 10 if you have not upgraded your roof. Suppose a roof costs 100 and benefits you 150. Suppose fertilizer costs 100 and benefits you 145.
In the absence of any potential GD grant then buying a new roof is both socially optimal and the thing the farmer is incentivized to do.
But with the potential for a GD grant then buying a new roof is still the socially optimal choice, but the farmer is no longer incentivized to do it. Thus, a farmer might forgo buying a new roof in the hopes of securing a cash reward, which is inefficient.
As for perverse incentives, GP is right that as the news spread, people may save the money they have instead of investing it in a roof in a hope of getting additional money from GiveDirectly in the near future. That's, unfortunately, human behaviour 101. So in time, they'll have to change the criteria they use for distributing grants, and the article hints that they're thinking about it already.
I'm not saying they should change the criteria to food security/malnutrition/etc. I'm saying that,
1) suppose the criteria was food-related; people wouldn't likely game the system by voluntarily not getting food because, well, eating is pretty high on people's list of desires.
(there are food-related metrics like "reported food consumption" and "malnutrition" that are used to assess the performance of GiveDirectly and charity/aid/welfare programs in general. They are well-defined, even if subject to manipulation. In any case, I understand that the roof criteria is more practical at the enrolment phase)
2) analogously, people aren't likely gaming the system by keeping non-metal roofs, as having good shelter is also very high on most people's list. In fact, one of the most common ways recipients spend the grant money is getting an iron roof.
Also, it's not like having a non-metal roof automatically entitles you to a grant, as this is isn't a nation-wide government program.
GiveDirectly is considering changing requirements not because they think they're including too many (relatively) high-income families with non-metal roofs, but because they might be excluding some people who live in metal-roof houses but are still poor.
For scale: that's the population of Switzerland.
It takes the two richest people to provide for 0.0022% of the people on the planet.
Too high of a cost for too low of a return. I believe those men think they can solve more than 0.0022% of a problem with their fortunes.
Those 8 million will no longer have to worry about feeding themselves, educating their children, paying their rent. They can create. A % will create businesses to take advantage of the money flow, supporting themselves and their communities, paying taxes, jobs, export.
Over a single generation you could move from 5% literacy to 95%, for 8 million families. That's a lot of kids and a lot of creators. Fewer babies next gen too if history is right.
Those 8 million will essentially have their greatest fears and worries removed and replaced with joy and freedom. Just imagine the possibilities.
well, for a while. then the money runs out.
8 people per year * 20 years = 160 Million people.
Also, the example was a family of 4 which would bump it to 640 Million.
PS: Those 7000 million people include plenty of people that are doing just fine.
Bono's brave enough to admit this.
Kenyan economist James Shikwati: "for God's sake, stop the aid:". http://m.spiegel.de/international/spiegel/a-363663.html#spRe...
A portion of the corn often goes directly into the
hands of unscrupulous politicians who then pass it
on to their own tribe to boost their next election
campaign. Another portion of the shipment ends up
on the black market where the corn is dumped at
extremely low prices. Local farmers may as well
put down their hoes right away; no one can compete
with the UN's World Food Program.
Let me give you an example. In africa there used to be a great many tailors. This was a profession that was viable and quite respected on the continent long after the western world had gone to pre-made clothes. In part this was because the country had not modernized. Then the aid came. The aid came in many forms, but one of the forms was millions upon millions of T-shirts and jeans. The entire profession of being a tailor has been eliminated (of course except for the very rich people.)
I'm not saying africans should be tailors, but pointing out that aid distorts the market, and distorts the economy and can undermine it if it's not well thought out.
This is what Bono was saying and this is one of the lessons that GiveDirectly seems to have learned.
If we go from aid to commerce the tailors won't come back, but billions upon billions of consumer items will flood into Africa. As the country gets richer there is going to be at least a 300 million women who aspire to owning a washing machine, a generator (or other source of power) and indoor pluming. Say what you want about capitalism and industrialization, but short of something like a first contact with the Culture, no other system on the planet can satisfy those needs. And they deserve to be satisfied.
It seems to mean that basically they get 9% of donations to cover various costs. Both non-negligible but also not too bad.
Still would like to know what the "overhead costs" are.
The grocery store can set the money on fire for all I care. The charity, not so much.
Randomised Controlled Trials aren't at all novel or unusual in international development, despite what the article says. (though some of the parties conducting them aren't particularly inclined to share the data, especially the microfinance guys...)
The survey referenced is here http://www.princeton.edu/~joha/publications/Haushofer_Shapir...
The margin for error in these studies is high. The study notes that their relative measure of "women's empowerment"[1] for women whose neighbours received money increased by a similar degree to the food security of people that actually did receive money. Plausible? The only viable causal mechanism the survey authors and I can think of that leads women who haven't received any money to be more "empowered" as a result of a few neighbours receiving it is the spreading of rumours that aid money is linked to people treat their women (something randomised study authors would really prefer didn't happen. Subjects trying to second-guess the surveyors' intentions is a big problem) The only other hypothesis that seems to fit is secular trends or differences in village behaviour contaminating the study results, despite the best efforts to control for them. Either way that result is unhelpful, not least because it puts into perspective the surprisingly small proportionate increase in food security
Taking the statistics at face value, they strongly suggest the recipients don't waste the money and don't have it stolen from them, but the non-cherry-picked stats aren't as spectacular as the article makes out
Ultimately, it shows that households receiving $287 in cash had $278 more in wealth later on. Their businesses are larger but generate no more profit than before. People receiving a total of $1085 are only $531 better off in assets and actually have slightly worse returns on the businesses than the people who received lower handouts[2]
The one thing this wealth does enable people to do is consume more, quite a lot more ($40 more for people previously spending presumably because they have less need for savings or debt repayments. Whether this pattern continues further into the future (and how that affects their income/assets) isn't studied. Effects on the other villagers' income and wealth appeared negative, but not to a statistically significant degree.
I'm reasonably convinced that at this end of the scale the cash transfers are better than the more fashionable approach of loaning the small amounts of money (with high overheads and commensurately high interest payment burdens on the recipients) but the study supports the view that its a band-aid rather than a solution.
[1]which, reading the full text, is actually a measure of reduced domestic abuse [2]whilst not statistically significant, the result that larger handouts resulted in a smaller increase in revenue is unhelpful
Consumption isn't so bad to be worthless. At this level , consumption includes stuff life increased food for better health , which isn't reported in ROI. And a slightly silly analogy: Britain's elevate handouts kept many people afloat but generated one $billion profit enterprise (Harry Potter)
For background, GiveWell is a meta-charity that evaluates charities by the amount of suffering they reduce per dollar. It was started by former hedge-fund managers and highly transparent, and about as rigorous as you can get in this field.
They publish all of their research, for both their "top charities" research and their more long-term Open Philanthropy Project division (which researches more abstract threats like labor migration, AIs becoming evil, land use regulation, scientific research funding, and more). There's a wealth of knowledge there for anyone interested in making the world a better place through altruism.
For background, GiveWell is the most prominent charity evaluator in the effective altruism space, and does much closer inspection of its recommended charities than anyone else. They're highly data-driven and fully transparent. They not only vet the charities expenditures, but the core mission itself to ensure it's a highly effective way at reducing suffering: http://www.givewell.org/international/top-charities/give-dir...
You can rest assured that GiveDirectly has been well-vetted.
Don't tell FOX News, as I'm sure they'll find a way to give this a negative spin.
This seems to be part of a larger turn towards more data and study driven aid than was the standard. Hopefully the larger nation state players can get around to using aid programs to reap longer term improvements vs the current short term and occasionally overall harmful effects some programs have been reported to create.
Everybody is spinning their small business or new charity as a `startup`. Maybe culturally they are aligned with startup principles but my idea of startup was a small group of people looking to shake things up (a market, a technology, etc) rapidly, grow fast and cash out.
A guy selling shaving cream on etsy is not a startup by this definition.
And according to PG, http://www.paulgraham.com/growth.html
My issue with charities being called startups is more in the for-profit vs charitable goals. Companies have goals that just don't jive with the goals of a good charity. If we take a definition based purely around 'designed to grow fast' yes it can fit. There's just more connotation to the word startup than the purely growth based definition Paul Graham puts forward.
addendum edit: There's nothing inherently wrong with the 'designed to grow fast' definition, it just doesn't sit quite right (with me) applied to charitable orgs because of the additional meaning I talked about above.
https://www.ycombinator.com/nonprofits/
> We’d been thinking about this for a while. We had a hypothesis that many newly founded nonprofits could benefit from the same techniques we use to help startups. We tested this with Watsi in the Winter 2013 batch and it worked wonderfully. They did all the same things as the other startups, including present at Demo Day (which is after all a room with lots of rich people in it).
> Since some people were confused when we funded Watsi, I’d better clarify that the money we’re putting into the nonprofits will be a charitable donation, rather than an investment in the narrow sense. We won’t have any financial interest in them.
Regardless, the main point is that charity or not, for-profit or non-profit, they're all still fundamentally businesses and should be run accordingly.
The challenges of a company built for exponential growth are very different than, say, a small business, a lifestyle business, or a charity.
Nor do you need to be a for-profit business to seek rapid growth. If we're going to measure based on "rapid growth", GiveDirectly has been a far more successful startup than most YC companies.
implies rapid growth.
What is an example of a repeatable and scalable business that is not rapid growth, relative to other businesses in the industry?
Backblaze is a great example of a company that could not have grown any more rapidly than it did. If they had, they would not have been able to source enough drives to continue and would have been faced with some difficult decisions.
Rapid growth is a great goal in some industries. But in others, a long, slow ramp up is better. That doesn't mean you can't have startups in those industries.
That's why the local pizza join that just opened isn't a startup. They might be new, but they have no plans for innovating on their business model to reach larger scale.
[1] http://www.givewell.org/international/top-charities/give-dir...
[2] http://www.givewell.org/international/top-charities/give-dir...
These people sometimes make bad decisions, but by and large they're more responsible than most people I know.
This seems to be just one of those misconceptions like drug abuse, which AFAIR studies actually found more common in well-off people than in poor ones.
> Longer-term research into anti-poverty interventions is rare, but it exists for cash transfers. A 2013 study in Uganda found that people who received cash enjoyed a 49 percent earnings boost after two years, and a 41 percent increase after four years, compared to people who hadn't gotten a transfer. Another study in Sri Lanka found rates of return averaging 80 percent after five years. In Uganda, not only were the cash recipients better off, but their number of hours worked and labor productivity actually increased.
The opposite conclusion in a false dichotomy.
There are multiple ways to give money. The way the US does it in some cases, where earning 5K more in a given year can cost 10K in benefits, does creates a perverse incentive. The way this charity does it, you get 1K and that's it; it does not create perverse incentives.
Consider the following example. In one example, a person receives X a week to help, but the aid rolls back when they get a job, M dollars less of aid for each dollar earned. So a job comes along where they can earn Y dollars for 40 hours of work. We could call M the means testing coefficient or something fancy like that.
Say that M is 1, and Y < X. In this case, they will now spend 40 hours of their week for no immediate benefit. Having a job may lead to possible longer term benefits (say beginning a career where eventually Y > X), but the cost of 40 hours of their life a week for such little possible benefit would make not working the better choice.
If X > Y, but only by a small amount, and M is 1, then they will see some small benefit, but spread over 40 hours it may still not make the job worth it.
And if M > 1 and Y < X, it would actually be harmful to get a job.
On the other hand, say that M was .25. In this case, taking the job would almost always be beneficial, despite the relationship between X and Y.
As for this charity in question, M is 0 (but the aid lasts only for a year). It is the least likely to cause any such harm to getting a job (except for when M is negative, which is to say the more you earn, the more aid you get, which oddly enough happens in some cases of corporate welfare).
The question was my way of suggesting that the comment I was referencing was probably unsubstantiated - while still giving him a chance to come up with data if he happened to have any.
The idea that under privileged people can simply work and save their way out of poverty is naive and dangerous.
I believe the reasoning for why that should be goes back to the cycle of poverty (http://en.wikipedia.org/wiki/Cycle_of_poverty). It's harder to make more money if you don't already have more money. By giving a one-time lump sum, you give people the opportunity to dig themselves out of that hole. Eating better, dressing better, buying things that last longer, these all have big impacts on how much one can save and earn. Even getting some basic education helps enormous amounts. In the end, it might even lead to people being more willing to work harder and actively save more money.
Take out the means testing, and you remove the incentive not to work. They can still work, and, if anything, it lets people get out of the psychological hole where they have to make decisions that aren't utility-optimizing in the long term just in order to survive.
It can only be a one-time donation if no one ever does this again, so that's basically saying, "it's important that this whole model of charity never becomes so dominant as to give people multiple kicks at different cans of the same type."
The problem with long-term poverty is rarely lack of money as such. It is lack of access to capital (which is a different issue) and most importantly lack of the rule of law (which makes access to both money and capital difficult.)
There are charities that promote the rule of law (Amnesty, World Justice Project) that are more likely to change things in the long run than charities that respond to immediate needs (athough I'm all in favour of people supporting both.)
I think that allows families to learn to better manage their money and not blow it on a single large purchase, and also prevents disincentive to stop working entirely.
As the article mentioned, some folks would prefer to receive the money in a few payments so that they're less likely to have family members come begging, but some prefer it all at once so they can allocate it to a big project.