That's the theory, at least. But the $40 billion figure rests on the assumption that we can identify the world's poorest, work out exactly how poor they are, and deliver them the right amount of money to get them to $1.25 a day. We can't. Even the best income surveys are inaccurate, and enough people cycle in and out of absolute poverty that it would be an impossible task to precisely track and target them over time.
Still, we shouldn't overstate the scale of the problem: It isn't that hard to get a reasonably accurate measure of a household's income and wealth. In 1998, economists Lant Pritchett and Deon Filmer found that tracking people's ownership of 23 different assets -- bicycles, land, and flush toilets among them -- was a very reliable guide to their affluence or lack thereof. If you are willing to accept a little more imprecision, even simpler approaches are possible. In Bangladesh, a cash-transfer program kicks in if families meet one of only a few criteria for eligibility: working as day laborers, as sharecroppers, or in one of a few low-paid occupations such as fishing or weaving; belonging to a female-headed household; or owning less than half an acre of land.
The Bangladeshi program is designed to target families in the bottom 40 percent of the country's population with cash transfers. The Primary Education Stipend is given to parents of 4.8 million children from deprived households in return for sending their kids to school, at a rate of about $1.76 per child per month. Six national banks disburse funds to parents with bank-issued identity cards at temporary distribution points set up within a maximum of five kilometers from each school. The program has been analyzed by Bob Baulch of the International Food Policy Research Institute, and the study suggests that even a very poor and very populous country can operate a large-scale targeted cash-transfer mechanism. A little under 30 percent of the poorest fifth of the country's rural households get the stipend compared with around 10 percent of the richest fifth -- far from perfect accuracy, but some evidence that targeting can work. (And considering that the average income in Bangladesh is under $4 a day, even the least-poor recipients of the subsidy are still poor by any reasonable definition.)
Based on the Bangladeshi experience, it's safe to assume that the real price tag of ending absolute poverty in poor countries by 2015 would be a lot higher than the theoretical cost of $40 billion. But it's hard to imagine even a relatively inefficient, bureaucratic, and poorly targeted cash-payment-based program exceeding $100 billion. That's less than the value of current aid flows, which stands at around $129 billion -- and it amounts to only 0.25 percent of the GDP of high-income OECD members. In cash-strapped times and with the effectiveness of traditional aid still widely questioned, many rich countries have been wary of any commitment to increase assistance. But perhaps they could all agree to an additional one-quarter of 1 percent of their GDP going directly to the planet's poorest? For a measure that could end absolute poverty worldwide, it hardly seems like too much to ask.