Two birds with one stack
As my long term readers know, there are two world-changing causes I am passionate about: donating to the global poor and increasing sample sizes in RCT studies. In my wildest dreams I dared to imagine lifting people out of poverty while simultaneously decreasing error bars. That opportunity has arrived: GiveDirectly is planning a decade-long trial of guaranteed basic income in Kenya. I hope to inspire you to donate money to this groundbreaking project, and I
pledge to match at least the first $1000 donated by Putanumonit readers donated $2,182, and my readers more than $10,000.
I’m going to argue that:
- People living in Kenyan slums can buy much more direct happiness per dollar than you or I.
- Research shows that you and I will be happier ourselves for donating the money to really poor people.
- Basic Income Guarantee (BIG) could be the best way to promote equality, prosperity and freedom in rich countries like the US.
- Or, BIG could be a terrible idea, but not for all reasons brought up by people who oppose BIG politically. In any case, we need a large-scale high quality trial of long-term universal income to find out what the consequences of BIG are.
- GiveDirectly is great at both getting the money to people who need it and at producing high-quality research.
- Either poverty alleviation or basic income research would be enough reason by themselves to donate. An opportunity to do both at the same time is reason to donate now, donate a lot, and donate publicly.
Since I started writing this post both Slate, Vox and FiveThirtyEight published excellent summaries of the potential benefits of BIG and the GiveDirectly experiment with 90% of the links and arguments I was going to provide. I encourage you to read those. Instead of repeating the same things, I’m going to give a briefer, personal perspective and describe what convinced me to commit $1,000 to the cause.
Hopefully, this will nudge a few readers to donate. These readers will probably be similar to me in trying to incorporate an altruistic heart with a libertarian brain to implement consequentialist ethics and bring the greatest good to the greatest number. If you disagree entirely with my values and don’t want to donate money to East Africa, enjoy this picture of an East African hedgehog instead and come back next week.
1. The Global Poor
The absolute worst paying college major in the US is education, with an average starting salary of $38,887, that comes out to $34,000 after tax. This means that practically every single college graduate in the US with a job is in the richest 3% of world population. This category includes me and my friends. In spite of our affluence, sometimes things suck in our lives:
- The Wi-Fi is slow.
- There’s no good Sushi in the neighborhood.
- The soccer game is cancelled because of rain.
- I run out of toilet paper.
- Happy hour ends before I leave work.
- Charged $150 copay by the health insurance.
The poorest 3% of the world earn $100 a year. Here are some things that suck in their life:
- There’s no electricity in the village.
- The daily meal is either ground maize paste or nothing.
- When it rains everyone gets wet because the roof is made of grass.
- Flying toilets.
- The water is only 90% water.
- They can’t afford any medicine whatsoever, so they get typhoid fever, malaria, cholera, dengue fever or HIV and die.
If I was living on half a buck a day I would need a stiff drink, but a bunch of research papers show that poor people receiving cash transfers don’t spend any more on alcohol or other “temptation goods”. Instead, they mostly spend it on better food, save up for medical emergencies and buy productive assets like metal sheets (used to rebuild homes and replace roofs) and goats (probably used for racing).
The best returns are achieved when cash transfers are given to the young and unemployed. These are people who are earning much less that their potential their entire lives because they have no cash to jump-start their careers and no access to credit given that they have no income. Young people receiving $400 each in Uganda got an average return of 40% in a year on the money by buying training and tools for vocations like carpentry, tailoring and hairstyling.
Finally, there are lot of organizations that donate things like livestock or training directly, but they add huge overhead costs and remove personal choice and dignity. There is no real evidence that people in poor countries make bad choices about spending direct transfers.
You know what? I spent $14 last week on flowers that my girlfriend turned out to be very allergic to and $32 on a Swiss knife that was promptly confiscated by the TSA when I forgot it in my travel backpack. Whatever the world’s poorest people choose to spend money on, they probably get more out of every dollar.
2. ‘Tis Better to Give
A while ago I wrote about the dumbest way to spend money, and my most enthusiastic commenter challenged me to write about efficient ways to spend money in pursuit of happiness. Being a conscientious researcher (not really), I turned to the leading literature in field. The literature provides a lot of evidence that giving to others makes people feel better than almost any other spending when the donation is voluntary and when it is impactful. $1,000 lifts an entire household out of abject poverty into tolerable poverty for at least a year, so I think that the “impactful” part is easily covered.
Honestly, I trust the research but I have never donated such a sum to charity before so I have no idea how it would make me feel. I promise to report how it feels like when I get there.
3. Give Me Freedom or Give Me Death (and Taxes)
We’ll leave Kenya aside for the moment and land back in the USA for the next couple of sections. The American governments spends about $3,000 a year per person (see footnote 2) on welfare, and maybe $2,000 more a year on a variety of tax breaks that could become unnecessary with a Basic Income Guarantee. Putting everyone in the US above the poverty line would cost at least $6,000 per person, so BIG will almost definitely raise tax rates for employed middle-class Americans like me. I also described myself as a libertarian just a few paragraphs ago. How on Earth could a libertarian support the government redistributing more of my income in taxes?
First of all, let me explain what kind of libertarian I am not. I am not a Randian-objectivist “taxes are theft” libertarian because “taxes are theft” is a central example of the non-central fallacy, AKA the worst argument in the world. I don’t think that governments are inherently evil as a law of nature. I do observe governmental suckitude as an empirical fact.
Below is a very simplified version of how I see the US political system operating with respect to its claims of helping poor people. Since this is going to be as angry a political rant as you’ll see on a blog sworn to avoid getting angry over politics (and also a one-sided and hedgehogian point of view), here are two hedgehogs for you to enjoy before/instead of reading this section.
In the US, there is a desire by liberals to give things to poor people based on need. It stands in opposition to the conservative desire for poor people to get things based on justice and merit.
The first thing that happens is that like every political issue, the question of helping the poor quickly degenerates into a vortex of tribal warfare. Instead of talking about specific plans to help the poor, one of our presidential candidates prefers to talk about waging war against the greedy and immoral bankers. Instead of describing specific plans to support American business, another presidential candidate seems to care mostly about fomenting hate for immigrants based on race and religion. As long as helping the poor is a political issue, efforts towards “helping the people I like” turn into “attacking the people I hate” faster than you can say toxoplasma.
The result of this political standoff is that Americans live neither in an efficient welfare state (like what Americans think Norway is) or in an efficient unrestrained market economy (like what Americans think Singapore is). Instead, the US isn’t a fun place to be on welfare, while simultaneously, in the name of helping the poor, the government destroys an insane amount of wealth by limiting free trade and distorting to the point of perversion every single human endeavor: the housing market, the labor market, the food supply, the financial industry, the education system, manufacturing, healthcare, weed and taxation itself. This destruction of wealth is supported by an ever expanding, ever expensive bureaucracy.
No one supports adding BIG to the current mess of welfare, subsidies and entitlements. Libertarian advocates imagine a possible amendment to the constitution, perhaps something like this:
Henceforth, federal, state, and local governments shall make no law nor establish any program that provides benefits to some citizens but not to others. All programs currently providing such benefits are to be terminated. The funds formerly allocated to them are to be used instead to provide every citizen with an equal and unconditional cash grant from the age of majority until death.
I feel the harm done to be by the US government in confusion, fear, frustration and constriction of my liberty, not just in dollars out of my pocket. If every upper middle class American made 20% less in after-tax income, our standard of living would at worst dip to the level of Canada or Belgium and at best wouldn’t change at all because the positional goods we spend our money on like houses in good neighborhoods, Ivy League degrees, Chanel bags and organic kale will just become 20% cheaper.
In contrast, removing even a few of the current destructive policies will make me much happier and reduce the government’s ability to harass me. This is why most existing governments are strongly opposed to BIG.
4. Jobs and Losses
One of the main arguments against BIG is that a guaranteed income will disincentivize people from working. Before we note that current welfare schemes themselves disincentivize employment to a huge degree and look at the evidence regarding BIG, I want to analyze a very technical argument that people often fail to grasp: work sucks.
I am utterly baffled by the obsession with employment as an end in and of itself. Why does the institution whose main job is printing money measure itself by the number of people who are unemployed? Why did the sentiment of an extreme Calvinist sect make us replace a utopia of leisure with endless workweeks of bullshit jobs? I’m writing this section at 3:06 pm on a Tuesday, guess where I’m sitting right now. If work creates prosperity, here’s a wild idea – let’s f%&#ng measure prosperity directly. Or wealth, or wellbeing, or progress, or innovation, and not the f@*%ing unemployment rate. And if work is so indispensable to people’s self esteem and happiness, why the f$!@ are we worried that giving them a meager check will prevent them from working?
To be honest, I don’t even mind my own job that much. The opportunity to pursue an interesting and rewarding career is the result of effort (studying math), planning (a degree in math) and a lot of luck (my parents are really good at math and gave me math genes and math textbooks). It’s a privilege. My company sells software to big corporations who use it for tasks that are normally done by college educated people with slightly above average IQ. With time, there are going to be fewer and fewer jobs for these people and for everyone less skilled than them. The reason that I’m not worried about robots taking my job isn’t that computers can’t be smart enough to write good software, it’s that once they are smart enough, humanity has about 5 minutes to worry about things like employment.
If the only workplace I was qualified for was McDonald’s, I would prefer to sit at home collecting a BIG check and playing Metal Gear Solid V while McDonald’s figures out at its leisure how much my labor as a burger-flipper is worth to them in a free market. The only thing worse than flipping burgers at a McDonald’s is being replaced by a robot after the government unilaterally imposes burger-flipper wage limits on McDonald’s while claiming to help you.
By the way, evidence shows that receiving money decreases working hours either a little or not at all. I just think this entire argument is perverse.
I don’t even intend to argue that minimum wage laws are harmful, I think it’s likely but far from certain. In any case I’m a humble blogger and not a labor economist. My argument is that every single existing government policy has many intended and unintended consequences, both known and unknown. Claiming that a proposed policy could have unpredictable consequences in a general sense is no argument at all against.
Thinking about predictable and harmful consequences is a good argument. Here are some actual reasons to worry that BIG could turn out to be a terrible idea:
- Some people (a disabled cancer patient with 10 kids) need more help than an amount we can afford to give everyone. Neglecting them is inhumane, and sticking additions and conditions on BIG will open the door to the same degeneration into political rope-tugging we have right now.
- A national BIG will turn citizen’s attitudes strongly against immigration, harming potential immigrants and the global poor.
- BIG will not actually increase the standard of living of people in poverty because it will be offset by inflation and an increase in predatory rent extraction.
- People will run out of goats to buy and actually start spending the money on drugs.
Are these problems real? Are they solvable? I lean towards “yes”, but that guess is based on pretty limited evidence. You know what would be really awesome? If someone did an RCT of long-term universal basic income with a big enough sample to measure all these effects.
5. I Trust GiveDirectly
There is no reason to suppose that an organization that is good at transferring cash to poor people very well will also do research well, but that seems to be true of GiveDirectly.
I’m not an actual expert on assessing NGO effectiveness, but the actual experts on assessing NGO effectiveness have GiveDirectly as one of their top four global charities. GiveDirectly is excellent at targeting people who need cash and getting the cash to them with very little waste on overhead and transaction costs. GiveDirectly currently operates at a rate of about $16 million transferred per year so allocating $30 million over the decade of the BIG pilot is well within their capabilities.
I spent some time browsing GiveDirectly’s research, it seems pretty solid. Even the study that measures cortisol is legit (the difference? a sample of n=1,372, vs. Cuddy’s n=39). The researcher they are currently proposing to lead the RCT is Abhijit Banerjee who has published both domain specific research on poverty interventions and methodological papers on experiment design. Truly a man after my own heart.
Neither GiveDirectly nor Banerjee have long histories of advocating for basic income and both have strong reputations to protect independent of whether BIG is a success of a failure. Everyone involved probably hopes to find that BIG is the best thing since giving poor people sliced bread, but I am fairly confident that the research will be conducted objectively. Most importantly, GiveDirectly has pledged to preregister the measurement and analysis methodology of their research. Preregistration is the single best way to convince me that someone is planning to do rigorous and honest research.
GiveWell has the track record in both cash transfers and research, now they just need our money. For every $1,000 we donate, our money will pull a person out of the worst poverty for at least a decade and increase by 1 the sample size of an experiment that could change the world for rich and poor alike.
Let’s Do It
Here are the possible outcomes of the GiveDirectly basic income pilot:
Worst case – The research is flawed/inconclusive, the recipients spend the money on food and get a temporary reprieve from hunger.
Better case – The research shows that BIG is no better than conditional transfers and normal welfare, the recipients invest the money at high returns (goats) and some of them escape crushing poverty forever.
Best case – BIG is proven to improve outcomes (health, security, income, education, aspirations, productivity) across the board, inspires people in rich countries to stop making it complicated and solve poverty and inequality by simply giving everyone money.
Here’s what we’re going to do: you’re going to click here and donate to the project. Then you email the receipt to putanumonit at gmail dot com and tell me if this post played a role in encouraging you to give. If it did, I will do two things:
- Match at least the first $1,000 of donations. Maybe more, I’ll see how it goes and how I feel about it, I could end up donating significantly more if I’m inspired by you. Edit: I was sufficiently inspired to donate $2,182 this year, thanks!
- Give you glowing acknowledgement in this very space. I encourage everyone to give loudly and proudly. In a world full of shameful virtue signaling, real action that improves lives should be publicized, boasted and accorded as much status as my humble blog can give.
To the awesome people who sent me donation confirmations so far and the ones who are about to: please let me know if I can acknowledge you by name and disclose the amount you have donated. I of course encourage everyone to do both.
24 thoughts on “More Power, Less Poverty”
The logical conclusion of EA style utilitarianism seems to be drastic redistribution (in light of the decreasing marginal return of wealth). This is true now, and probably would be true at any point in history. Do you think that drastic redistribution would have been correct in the 1500s? Would the scientific revolution have happened? What about in the 1800s? Would modern physics and computing exist?
The concentration of wealth has many downsides, I acknowledge that (much of wealth is used to perpetuate itself), but the concentration of wealth in the past centuries has proven to be an extremely productive way of advancing technology and prosperity.
This is a bit tangential to the point of this post, which I don’t really disagree with – BIC might be a much more effective way of doing welfare and since its politically impossible now in any developed country, trying it out in poor countries is a decent way of figuring out whether it works. I bring this up because have seen many EA proponents and people who share a similar rationalist utilitarian framework propose policies such as open borders, which seems equivalent to drastic redistribution. All of this without any consideration for the long term.
@Alex, I guess it depends on what sort of 1500s redistribution you ended up with. It’s generally accepted that the biggest advantage that lords had over serfs was:
– childhood nutrition –
because malnutrition in childhood causes permanent brain damage.
-education – so you can build on top of others discoveries and old discoveries are not lost.
– free time – if you are a genius peasant with an idea for a better plow, but you need to spend 16 hours of every day on your crops, you probably won’t build that time saving plow.
A county in the 1500s where all citizens get enough food, education, and gee time to reach their full potential would have a big advantage over it’s neighbours. A county that dissolved it’s nobility and then everyone became starved, overworked, and uneducated would be a disaster. I don’t think many countries in the 1500s even had the resources to adequately feed all their citizens (we can infer from historical height records that serfs were malnourished through most of history), much less to educate them all. But in the modern day, basic health and education for children is more of a distribution problem than a lack of resources problem. And the need for radical redistribution tapers off once people are out of poverty.
Another good article. Re one objection: ‘Some people (a disabled cancer patient with 10 kids) need more help than an amount we can afford to give everyone’ – some proposals deal with this:
(a) If some services are government-provided free at the point of use, such as healthcare (e.g. the UK’s existing National Health Service), then there is no need for exceptional extra payments to these people. You just provide cancer treatment for free; only genuine cancer patients will get it. Similarly, governments could provide elements of childcare (e.g. creches) for free; only genuine children will use them.
(b) Many universal income proposals include smaller payments for children (paid to their parents) to cover basic child rearing costs.
Right. These proposals make a lot of sense on a first-order analysis, but they do create second order problems that can be fatal to BIG.
First of all, providing anything that isn’t paid for by the people using it like crèche (day care) will cause it to be overconsumed. Maybe my younger sister will babysit my kids for a mere $5 an hour, but if I can use free day care (that costs taxpayers $30 an hour) I’ll use that instead. Now you’re wasting a lot of taxpayer money that is already stretched thin by BIG.
But the main problem is that any welfare that isn’t applied equally to everybody will become a source of political contention. 30 year olds want free day care, but 60 year olds want free cancer treatments. If there’s only money for one but not the other, they will fight it out in the political arena. The more exceptions you carve out, the more people have reason to jump into the fray, and you end up with the same welfare system we have now of distributing resources in a terribly inefficient way based on political expediency.
I think that BIG in rich countries can only work if the exceptions are few, simple, and immutable (maybe every welfare change will require a 75% majority to pass). Something like partial payments for children (as a fixed % of adult BIG) can work. Something like deciding which diseases get treated for free and which don’t is much more problematic.
Thanks (only just saw your reply, a couple of years later!)
For what it’s worth, the UK’s National Health Service (NHS) does do a basic cost-benefit analysis for treatments, with an upper limit for the amount it will spend per year of life saved. With the result that they will provide at least basic treatment for almost any condition (but avoids prohibitively expensive treatments). So it’s not grossly inefficient, though there is some political interference in it, and of course trade-offs between spending on the NHS vs other things.