On Tuesday we started a deep dive on Universal Basic Income by looking at what the concept actually entails.
We explored the different problems that people believe UBI can solve and the different visions that groups of supporters have. We explored Universal Basic Income as a replacement for the safety net, as a cure for poverty, and as a protection against job loss.
But all of this is based on one major premise: Universal Basic Income works.
The question we need to tackle next, then, is whether this is true. Would Universal Basic Income work? Is giving people cash the best approach to ending poverty?
Adopting a UBI is a huge undertaking, both economically and politically. Before we jump in, we should make sure there is evidence to support it.
That’s what we’re exploring today.
The research on Universal Basic Income specifically is controversial and spotty at the moment. Most experiments have been small, not universal, or canceled early. Sometimes all three.
The United States conducted a few experiments in the 1960s and 70s with a negative income tax. These experiments were only for a limited period of time (three years for most of the trials and five for one of them) and only featured a relatively small number of participants (ranging from 1,357 households to 4,800 households). Some of these (including the five year 4,800 household study) never published their results.
An experiment was launched in Ontario in 2017 that looked a lot more like Universal Basic Income. However, it was canceled this year after a conservative government was elected. A 2017 pilot in Finland was also ended prematurely this year.
GiveDirectly, a non-profit focused on ending extreme poverty, launched a 10-year Universal Basic Income pilot in Kenya in 2016 and has teamed up with research groups to help design the experiment and analyze the results as they come in. There does not appear to be any risk of this project ending early, but it could be a while before we have useful results.
Y Combinator, the tech start-up incubator, is also running their own UBI experiment. After conducting a short-term and small-scale test in Oakland, the company has decided to begin a larger scale $60 million experiment next year. Michigan’s Survey Research Center has signed on to help design and conduct the study to make sure it meets scientific standards. This is promising, but it will be a long time before we see any useful results from this study as well.
There is a program that has similarities to UBI going on in Alaska. Essentially, Alaska has taken the oil royalties that the government receives and used them to create a publicly owned mutual fund. The money comes in, is invested in stocks, bonds, real estate, and other investments, and then dividends are paid out to every Alaskan. Instead of the dividends being a set dollar value, as they would be with a UBI, the distribution is based on the 5-year moving average of returns. When the investments do better, the payout is higher. When the investments do worse, the checks shrink.
Despite the lack of studies on UBI, however, there is plenty of research on giving people money.
Much of this research comes from one of my favorite charities: GiveDirectly.
GiveDirectly is a non-profit that takes the money you donate and gives it to people in extremely poor communities in Kenya and Uganda. As noted above, they are now running a trial of Universal Basic Income under the same model, although most of the research discussed here is based on more sporadic transfers of cash.
The most obvious finding is that giving people money is extremely efficient as a mode of helping people.
GiveDirectly’s overhead costs are significantly lower than other charities. This is for the same reasons that Libertarians prefer Universal Basic Income to the current system of aid to the poor.
Giving people money is just a much simpler endeavor than most other forms of aid. This means that far less money gets wasted on overhead costs and far more goes directly to helping people.
Giving People Money Works
More importantly, though, the research shows that giving people money works.
Research finds that households that receive cash distributions show significant positive impacts on the health of the children in that household. Studies in South Africa have found height and weight increases. A study in Malawi found reductions in HIV and psychological distress in children. A study of households in Uruguay found reductions in low birth weight.
Studies across the board have found decreased child labor and increased school attendance.
It has also been proven that giving people money increases assets, earnings, food security, and mental health while decreasing domestic violence.
Give People Control
Giving people money works because it lets the recipient make their own decisions. Poor people know what they need better than we do.
Annie Lowrey wrote a book about Universal Basic Income and went to a village where GiveDirectly is running its UBI experiment. She wrote about seeing piles of unworn Toms shoes in people’s huts. They didn’t have enough food and their roofs were leaking, but they had lots of shoes.
William MacAskill’s Doing Good Better, a book about effective altruism, talks about all sorts of projects that looked good on paper, got tons of support from the West, and then didn’t actually help poor people that much. The problem with these projects is that people were trying to address issues based on theory and analysis from a distance.
If you want to help people you should give them what they say they need, not what you think they need.
What Do People Need?
So what do people actually need?
It turns out that needs vary from person to person and family to family. Go figure.
On the whole, though, GiveDirectly finds that cash transfers lead to increased spending on food, medical and education expenses, assets, and home improvements.
The assets in this case include things like livestock or fishing nets that can then be used to secure food or business in the future. The most common home improvement was adding an iron roof to a family’s hut so that water doesn’t leak in when it rains.
Letting People Speak
I generally prefer statistics to anecdotes when first trying to understand things. When understanding how people spend their cash distributions, though, I think anecdotes are helpful. The statistics make it a bit hard to grasp how this money can really change lives.
One of the first stories that I heard after donating to GiveDirectly was of a woman who used the money to have a well dug close to home. This allowed her to supply her family with water without having to leave her children for hours every day to go back and forth to the next village.
Here are some quotes from GiveDirectly recipients on how they spent their money and how it changed their daily life:
“I have been able to add capital to my business and increase the stock. This has increased my sales and improved my profits.” Irene, 23.
“I can have 3 meals in a day.” Dorcus, 87.
“I spent the entire transfer received from GiveDirectly to purchase a fishing net and a floater.” Erick, 40.
“I spent the money received from GiveDirectly to buy clean water, food, soap, and used most of the amount to pay school fees.” Fredrick, 70.
“I want to pay school fees for my child who is to join secondary school in 2018.” David, 40.
“I want to repair the house that my children spend their nights in, since it is leaking and has dilapidated walls.” Jane, 52.
Obviously some of these don’t apply to poor Americans. A poor family in Washington, D.C. is not going to use money from a Universal Basic Income to dig a well.
The principle holds, though.
A person knows better than the government whether their particular situation would be best served by spending on a new place to live or school activities for their kids or transportation.
Maybe they want to invest in a certification program to get a higher paying job. Maybe they want to buy tools to start their own home repair business. Maybe they just want to work less to spend more time with their kids.
Cash allows people the flexibility to pursue whatever is in the best interest of themselves and their families.
Drugs and Alcohol
One very common concern with cash transfers is that people will spend the money on drugs and alcohol.
This is the same thinking behind the limits that the government places on SNAP. Basically: we don’t trust poor people to make good decisions.
Part of me wants to reject this concern out of hand due to its sheer condescension. This is the same type of thinking that leads people to believe that poor people deserve to be poor. That poverty is a choice. It is wrong and harmful to society.
Regardless, though, this concern is also provably misplaced.
An analysis of the various studies on cash transfers shows that there is no evidence of any impact on alcohol, tobacco, drug use, or crime generally. In fact, one study in 2010 specifically gave cash to addicts and criminals living in Liberian slums. Even that money was spent on “subsistence and legitimate enterprise.”
As already established, people know what they need. We need to trust them to make their own decisions.
Cash Distribution vs. UBI
Admittedly, most of this research is from cash distributions generally rather than Universal Basic Income. Additionally, most of the studies are done in the poorest parts of the world. These mean they are not perfect analogues for a Universal Basic Income program in the United States.
The main difference between the cash distributions that are most often studied and the UBI proposals are mainly in the fact that the UBI proposals would provide consistent, regular payments. This would allow much more long-term planning than the more unpredictable cash distributions. This would seem to amplify the benefits that people receive from the money.
The only negative change I can think of on this front is that maybe some people would take UBI payments for granted more than they do random cash distributions and would therefore spend it less mindfully. I don’t know of any research that would suggest this is true and I don’t actually believe it is likely, but if I am stretching to find a reason that the research would not be applicable, this is all that comes to mind.
How Far Does a Dollar Go?
The second part is a bit trickier.
Based on the data that we have from poorer countries, I would guess that the findings all apply to cash transfers in the United States, but that larger dollar values are required to provide the same level of life-improvement.
Living on $5,000 a year in the United States is unthinkable for most people. But it is also a long way from living on $300 a year in Uganda. Both are examples of poverty, but the challenges faced are very different. That said, I think the important piece for this analysis is that the biggest challenge they face is the same: a lack of money.
The family in Uganda may use their new cash to dig a well or buy livestock or install a metal roof. The family in the United States may use their cash to move to a new neighborhood or take classes or start a business. The biggest difference between the two seems to be based not in whether money can solve the problem, but how much money is needed to solve the problem.
There is a point where money has severely diminishing returns. We see this in happiness data. More money makes people’s day-to-day life better up to a point, but the correlation dramatically decreases after that point. We’re nowhere near that point in the UBI discussion, though, as this would be over $85,000 in today’s dollars.
There are lots of other concerns with UBI and we will dive into those next week. But one thing is clear: Giving people money works. We’ve got a long way to go and a lot of poverty to eliminate before we have to question that aspect of Universal Basic Income.
Join the Conversation!
Are you convinced of the power of giving people money? Why or why not? What other concerns do you see with UBI that need to be investigated? Let us know in the comments!