3 ms·
It doesn't sound like they picked people are random but rather unemployed people. So it doesn't seem a reasonable test of just giving some random group of peopl
by joe_the_user 9y ago
It doesn't sound like they picked people are random but rather unemployed people. So it doesn't seem a reasonable test of just giving some random group of people money but rather just something looked to everyone involved as a kind of welfare. Thus it doesn't seem like a fair.
That's not saying I think this could work but that it's different from the experiment in India I remember where everyone in a village was given an income.
One problem with universal income I would imagine is it would give landlord a strong incentive to immediate raise rents - IE, how could it not create inflation?
- gwern 9y agoI'm not familiar with this trial specifically, but my guess is that the reason for limiting it that way is to focus on the most plausible kind of welfare program in the group most likely to benefit in a large way (the unemployed) and get as much certainty out of the data as possible. They had limited money (only enough for 2000 people, apparently), so if they split it across a random set of people, or a specific village, they would get a much less precise answer (eg n=200 unemployed, n=1800 employed; or arguably, n=1, at the village level). And if you don't get a clear good result in the unemployed, why would you expect clear or good results in the employed? Clinical trials do a lot of screening of enrolled participants for similar reasons: you want the sickest or most likely to benefit, so you can see effects clearly, and you want to avoid participants with conditions that might override or obscure the benefits, like old age (a couple dropping dead on you at random might cover up a real benefit).
- majewsky 9y agoBut once you actually roll out UBI, everyone gets it. If you only concentrate on UBI's effects on unemployed people, you're missing a huge part of the picture.
- lucozade 9y agoWell yes, but Finland wasn't experimenting with basic income for philosophical reasons. They were experimenting because they have high unemployment and a complex, and expensive, welfare system. So what they wanted to understand was the effect a basic income had on those factors. The next stage of the experiment, that wasn't funded, was to extend the experiment to include the employed. Even there, I would expect they would still focus on low incomes as that's where you would expect to see an effect. To go beyond that, it's not clear what you'd learn without having a very large scale experiment i.e. one that materially affects the level of taxation. I can't imagine they'd be interested in that without seeing a significant, positive effect on the first two categories. And as a side point, I don't believe the idea was ever to test a UBI just a basic income. Unless you're intending to do away with all differentiated welfare, highly unlikely in a social democratic country, the universal bit is probably more harmful than beneficial.
- joe_the_user 9y agoI'm not familiar with this trial specifically, but my guess is that the reason for limiting it that way is to focus on the most plausible kind of welfare program in the group most likely to benefit in a large way (the unemployed) and get as much certainty out of the data as possible. The question is; "is this a welfare program or isn't is??" I mean, if the program is structured to provide benefits for the unemployed and if we're looking how much the unemployed benefit, then it is a welfare/unemployment program. The article mentions that the participants also saw this as an unemployment program and it seems logical that given this, they'd react to it in the fashion that we already know people tend to react to unemployment. There are already some unemployment programs that promise to taper off benefits rather than abruptly ending them and these don't have much difference from other welfare programs. The sell of basic income is - "everyone benefits, everyone maintains an interest in working, the reduced bureaucracy provides more money to give people." Those together aren't saying "this is program to get more people working" and if it just gets reduced to that, the original plan is basically lost. Which again, isn't saying basic income would work. But it seems reasonable to avoid letting the discussion just degenerate to debates on standard welfare. That said, I guess I could accept one argument against basic income - "it will never work because in practice, the waters are always going to be muddied by these studies (done by the same old welfare bureaucracies) until the discussion becomes just a discussion of welfare strategies."