To clarify: The hypothesis on the table is that you can only measure by poverty by income. That is, if you gave a person making 0 per year 30K per year they wo

To clarify: The hypothesis on the table is that you can only measure by poverty by income. That is, if you gave a person making 0 per year 30K per year they would suddenly have an apartment, healthcare, transportation, and all those other things that most other people who make 30K per year have. A certain amount of money over a certain amount of time is the thing to be adjusted (maximized?) To test this, use a substantial amount of money and a non-trivial time period. 15K and six months come to mind because the sum of money is large enough to do quite a few things and the time period is long enough to allow various random things in life to occur. I'd think you could also test at greater income levels. In fact, to do it right you'd use 15K, 30K, and 45K and time periods of 6, 9, and 12 months. So when you come back, you're not looking for poverty to be "solved", whatever that means. You're looking on the previous time period to see if that person lived the same as somebody who was making a similar amount from a job. Did they have access to shelter? Food? Did the money last? And so on. I'm by no means suggesting that poor people are going to blow it all on booze and cocaine (of course, some will), but that you're measuring the wrong thing. This is a classic example of managing to whatever the metrics are, instead of the underlying condition. Put a different way, we shouldn't care how much money anybody makes, but whether they live a comfortable, happy life and have access to opportunity. But whatever my personal bias, an experiment like this would put in sharp relief whether or not the hypothesis had legs or not.