It's common for aid organizations to define "extreme poverty" as those people who subsist on less than a dollar a day around the world. The U.N. Millennium Project generally claims that there are 1.1 billion people below this threshold, and want to halve the number of "extremely poor" by 2015. Who can argue with that? The problem, though, is that the "dollar a day" metric is terribly imprecise, as David Singh Grewal mentions in a thoughtful essay on globalization:

In fact, for such an assessment [of global poverty], the dollar-a-day measure is "meaningless," as the economist Sanjay Reddy and the philosopher Thomas Pogge have argued in a series of papers, because the figure uses purchasing power parity (PPP) ratios in order to convert foreign purchasing power into U.S. dollars.

The problem is ultimately one of aggregation: to value real purchasing power (the ability to buy goods and services) in one currency in terms of another, economists compare representative baskets of goods and services—cars, movie tickets, doctor visits, heads of lettuce, and so on—purchased in each country for a given amount of currency. But many of the goods and services that enter into these calculations may be irrelevant to the poor, because no poor person consumes them, and so they skew poverty calculations that include them. Reddy and Pogge conclude that it is impossible to construct a meaningful poverty line of the dollar-a-day type that uses PPP ratios.

I'd probably go even further than this. The mere fact that a person lives on $1 a day tells us relatively little about that person's actual condition. After all, a dollar a day will go much further in a stable country with generous social services than in a basket case country like Angola. On the flip side, World Bank studies have charted a decline in "extreme poverty" since the 1980s, ostensibly thanks to free markets and the like. But most of that decline has occurred in China and India, where the amount of money needed to maintain a decent standard of living may well have risen even more dramatically, so that even, say, $3 a day still leaves on in dire straits.

But let's take this further. Consider the people living in extreme poverty—a billion people or so. Many of them live in rural areas. They grow their own food and live in their own homes and have their own communities. To be sure, they're all really fucking poor, and suffer from economic insecurity, varying degrees of political insecurity, and could undoubtedly use better health care and education. But without at all trying to romanticize this sort of existence, living in a farming or fishing community could, conceivably, be preferable to life as a Third World urban-dweller in an overcrowded slum, foraging for trash to sell on the black market, stealing food, and contracting various exotic diseases because of a lack of clean water.

Economic growth, of course, usually uproots rural life. It's been happening for decades in the developing world. Corporations move in and start appropriating natural resources; trade and large agribusinesses put farmers out of business; the young folks are lured by jobs in the city, various processes disrupt traditional communities. Urbanization proceeds apace. Regardless of the merits of this process, though, it's entirely possible that many people could end up worse off in the short term—pushed into a life of urban squalor, with no escape in sight—even if "extreme poverty," the number of those living on less than $1 a day, is declining.

Now lots of developed countries have gone through the changes just described, and even though the transition period can, like puberty, get ugly, it might well be the only way countries can get rich and prosper. I don't think that's true (in fact, I think that urbanization and slum-ification will be much more painful for the Third World than it was for, say, Britain, because there's no "New World" to siphon off all those excess city-dwellers brought about by industrialization), but that's not the point—the point is that official "extreme poverty metrics" could easily obscure much of what's going on, and might even steer experts towards faulty prescriptions in the never-ending quest to "make poverty history".