[ppereira]

There is quite a bit of anger and little fact in this comment. GiveDirectly has been running UBI programs in East Africa for years now. They have an all-star cast of economists examining their performance. Their pilot programs were very successful, which is why they are scaling out further.

East Africa, and Kenya in particular, was carefully chosen for two reasons. First, Kenya has a mobile phone network that makes it easy to transfer funds electronically to individuals and then for those individuals to get cash from the funds at local stores. I wish we had this in the West, but we don't because of banking regulations. Second, poorer households in certain communities tend to use natural roofing materials rather than aluminum, which makes it easy to target funding by satellite.

The UBI is a political football in the West. The Mincome experiment was cancelled because of a change in government. The potential utility benefits from a UBI are so great that it would be a shame to think that we have to wait for it to be politically palatable in the West before we can get any data on its effectiveness. Also, as far as poverty relief is concerned, money really does help.

[asuffield]

> GiveDirectly has been running UBI programs in East Africa for years now. They have an all-star cast of economists examining their performance. Their pilot programs were very successful, which is why they are scaling out further.

Possibly the most important detail here is one they call out explicitly in the article: this is the first time anybody's run a large scale experiment of this form which was not based on selecting candidates who were likely to benefit highly from receiving the extra money.

I am completely convinced that giving people extra money can cause a permanent, substantial increase in their income and living standards, if I get to pick the tiny fraction of the population who receive it. That's well-established and easy, and also not viable as a long term solution.

I really want to see if it works the same way when you aren't selecting the best candidates, but just distribute the money blindly. No matter what the answer is, this is really important research that will drive economic policy decisions. We're about to get the answer to one of the biggest economic policy questions of our time, and that's awesome.

[alttab]

You seem to know a lot of the detail. How are you affiliated with GiveDirectly?

Also, this post also has little fact when the emotive aspects of it contain phrases like "all star cast", "very successful", "political football", "benefits so great", "would be a shame."

Where's the facts?

[colah3]

I'm not the GP, but I have a similar impression that GiveDirectly has really strong evidence supporting it.

GiveWell -- a separate organization that does in depth analysis of giving opportunities -- has a detailed write up on them: http://www.givewell.org/International/top-charities/give-dir... GiveWell recommends GiveDirectly as one of their top charities.

GiveDirectly's intervention has been tested by multiple randomized controlled trial, including a large-scale one. (See the GiveWell write up for details.) In general, the case of them seems much stronger than most charitable interventions.

(I have no affiliation with GiveDirectly, though I'm close friends with many GiveWell staff.)