Buenos Aires — WE don’t know who they are. The people suffering from hunger are not our relatives, friends, co-workers; they probably don’t read this paper. We don’t know them, but at the very least we ought to know how many of them there are, because policy and aid decisions depend on that number; because very often their lives depend on that number. But we don’t know, because the hunger statistics reported by the United Nations Food and Agriculture Organization are flawed.
Every year, around this time, in collaboration with the International Fund for Agricultural Development and the World Food Program, the F.A.O. publishes its report on hunger — or, as it is now called, “food insecurity.” What most people remember are the numbers: whether hunger went up or down, and by how much. But hunger statistics are confusing. It is very hard to calculate with precision how many men and women do not eat enough. Most live in countries where weak states are incapable of accounting for all their citizens, and the international organizations that try to come up with head counts must use statistical calculations instead of detailed census reporting.
The F.A.O. makes an effort: by studying agricultural inventories; food imports and exports; the local uses of food; economic hardship and social inequality. From there it determines the estimated availability of food per capita. The difference between required and available calories gives the F.A.O. its number of undernourished people. This sounds like a sensible method, but it is entirely malleable. And so its results can be adjusted according to the needs of the moment.
This month, the F.A.O. reported jubilantly that the total number of chronically undernourished people went down to 805 million — 209 million fewer than in 1990-92. These numbers carry a particular weight: They will be the last published before 2015, when the United Nations Millennium Development Goals are supposed to be completed. According to the report, these new numbers show that the organization’s aspiration “of halving the proportion of undernourished people in developing countries by 2015 is within reach.”
This sounds like great news, but it’s not so simple. According to the development goals, the rate of hunger in 1990 is the one that’s supposed to be halved. But that 1990 number has been adjusted several times, usually making the current numbers look more favorable by comparison.
It’s a long story. At the World Food Conference in Rome, in 1974, when Henry A. Kissinger famously stated that “within a decade, no child will go to bed hungry,” F.A.O. experts estimated that the number of hungry people in developing regions was close to 460 million, and that in 10 years it could reach 800 million. That prediction was close: In a 1992 report, the F.A.O. stated that there were 786 million hungry people in 1988-90. It was a dramatic increase, a serious blow.
In that report, the F.A.O. revised its previous calculations, saying that its statistical method had been wrong. Now, the F.A.O.’s experts said, they believed that in 1970 there weren’t 460 million hungry people in the developing world, but more than twice that number, 941 million. This, in turn, allowed them to say that the 1989 figure of 786 million did not represent a dramatic increase but, in fact, a decrease of 155 million: quite an achievement.
The changes kept coming. In 2004, the F.A.O. said that the number of undernourished people in developing regions had reached 815 million. This would have seemed like a disappointing increase from the 786 million figure. But in that same report, the F.A.O. revised its 1990 numbers once again, and stated that in 1990 there hadn’t been 786 million but rather 823 million hungry people. So hunger had gone down after all.
In the F.A.O.’s 2011 report, the number of hungry people in the developing world in 1990 was 833 million — 20 percent of the developing world’s population. Then it was 980 million in the 2012 report, after the F.A.O. experts revised their methodology once again. By 2013, those 980 million hungry people had become 995 million — 23.6 percent of the population — who, along with the 20 million undernourished people in the developed world, added up to a staggering total of 1.015 billion.
This is the 1990 number that the F.A.O. now uses to explain how we have won a new battle against hunger. It also means that, to reach the United Nations’ goal of halving the proportion of the hungry in the developing world by 2015, we don’t need to go down to 10 percent of the population (half of that 20 percent), but to “only” 11.8.
It is possible to assume that statistical methods — factoring in new population, caloric and economic data — may be much better now than 30 years ago. It is harder to imagine that they have changed so much in the last three years as to add more than 160 million people who had previously stayed under the radar. We know that there is nothing more variable than the past, but it is unusual to watch it changing so fast, so visibly. You could say they are just numbers, abstractions; they wouldn’t really matter much if they were just bad propaganda figures. The problem is that they are, in fact, canonical figures: the kind that are used to determine funds and priorities.
This is not conscious corruption. It’s a symptom of an institutional culture that has to prove it is achieving important progress. The 1990 change justifies the United Nations’ efforts and jobs, as much as it quiets our consciences. And it has a double economic effect: It convinces donors that their money has been fruitfully invested, and it justifies the reductions of these investments. International food aid, after peaking at $5.5 billion in 2008, decreased to $4 billion in 2012, according to the Organization for Economic Cooperation and Development.
So maybe next time, when we are told that hunger is being defeated, it would be wise to keep asking where and how and whose. After all, these figures help define the lives of hundreds of millions — wait, how many hundreds of millions? — of the victims of persistent hunger, the greatest outrage of our time.