Look, (a) you screwed up the numbers. You should not be comparing, in this hypothetical scenario, 1.33 with 1, but rather 1/(1 - 0.33) with 1.33, which is inst

Look, (a) you screwed up the numbers. You should not be comparing, in this hypothetical scenario, 1.33 with 1, but rather 1/(1 - 0.33) with 1.33, which is instead 0.1625. That is, "these people who don't understand the numbers involved" in the above article includes you. :P And the article above does not make the pretense that this is a uniform problem affecting all of the consumption of the poor; rather just that certain deals may be irrationally treated because of how they are phrased. There is no mythical 33%-across-the-board discount for the poor at stake here; at best it's a 16.25% discount available in very limited circumstances. The real point of the original article was that many people seem unable to determine the difference between 1/(1-0.33) and 1.33, which is only a 16% increased-cost gap (b) It would be very tough to make these economics work out in the present market where even 5% returns seem overoptimistic. Suppose you could reliably convert a poor person into a rich person (and I'll take a modest $200k/year income to define "rich"), by giving them $5k per year and making sure that they spent it on "the right sort of thing," banks could probably do it. I mean, just imagine making this pitch to a poor person: "I'm going to make it so that in 10 years you are earning $200k/year, and then I will take $20k of those every year for 20 years -- so that you only earn $180k/year -- as payment." To a poor person that's an unbelievably sweet deal. They would take that offer and might even offer a much larger cut of that salary to that bank. Suppose that with oversight the cost to the bank is as much as $10k/year, of which $5k goes directly to the person and $5k is administrative losses in making sure people don't misspend the money. From the bank's perspective that's a 10% return on investment, which is not amazing but not too shabby. Even if it only works 50% of the time, in today's economic climate that easily beats treasury bonds as investments go. It's in the bank's financial interests then to lift people out of poverty with this mechanism. I don't see any banks doing this, and while it may be possible I would guess that it takes much more money to do it.