The Blog


11:00 PM, Nov 24, 1996 • By NICHOLAS EBERSTADT
Widget tooltip
Single Page Print Larger Text Smaller Text Alerts

Malnutrition is an affliction suffered by individuals. Short of clinical or biomedical examination, there is no reliable way to determine a person's health or nutritional status. Because they lack this person-by-person information, these studies draw clumsy inferences about individual well-being. They cannot cope with such exacting, but important, issues as whether individuals with lower caloric intake have lower than average caloric requirements; whether individual metabolic efficiency adjusts in response to changes in the nutritional supply; or whether individuals predicted by their models to be undernourished actually suffer from identifiable nutritional afflictions. To pose these questions is not to presuppose an answer to them; it is simply to discharge a basic duty of careful inquiry.

Sometimes the results of these hunger studies can be dismissed after even the most casual inspection. In 1980, the World Bank published a paper purporting to show that three-fourths of the population of the less developed regions suffered from "caloric deficits." This ominous conclusion was reached by a chain of dubious suppositions, the final and most spectacular of which was that anyone receiving less than the average "recommended dietary allowance" was underfed. In truth, about half of any population will need less than the average allowance; that is the meaning of the word "average." Consequently, this model could only generate nonsense numbers: Its computations suggested, among other things, that nearly half the people in prosperous Hong Kong were getting too little food!

To their credit, the World Bank researchers on this particular project recognized that their work failed the "reality test" and went back to the drawing board to improve their product. Unfortunately, others working on the problem have not always met the same standards of intellectual accountability. Lord Boyd-Orr, for example, did not at the time explain the method underlying his now-famous estimate of the prevalence of world hunger. After reviewing contemporary data, one of the leading agricultural experts of the day, Merrill K. Bennett, surmised that this estimate might actually be an elementary computational mistake -- a misreading of the figures in two particular columns of a particular table. The Food and Agricultural Organization, which prepared the figures Lord Boyd-Orr used, never replied to Bennett's inquiry and has never offered substantiating evidence for Boyd- Orr's assertion.

Other estimates about world hunger from the same organization have remained similarly protected against outside inspection: Most of the data and calculations in the first three FAO World Food Surveys, for example, are still unavailable to the public. In its more recent studies, the organization's determination of the number of calories an individual needs for nourishment has been rising steadily over time. Why? These upward revisions do not seem to reflect any obvious changes in the scientific consensus concerning nutritional norms. But they do produce higher totals for any given estimate of the number of hungry people in the world.

If we could only for a moment extricate ourselves from this numerical house of mirrors, we would see that there are indeed meaningful data that bear upon the actual nutritional status of humanity, and that they tell a rather different story.

Household spending patterns in less developed regions, for example, can reveal how the poor assess their own nutritional status. If a family treats food as a "superior good" -- that is to say, if an increase in income raises the overall share of the household budget going to food it renders a telling judgment that its members have had too little to eat. By this criterion, the incidence of serious hunger in the world would be far lower than the Food and Agricultural Organization currently suggests: about two- thirds lower in some years for India, to take one example.

Mortality rates, for their part, offer a direct and unambiguous measure of the material condition of any population. It is clear that the so-called Third World has experienced a revolution in health conditions over the past generation. According to the U.N. Secretariat, life expectancy at birth there rose by an average of almost a decade and a half between the early 1960s and the early 1990s. And over that same period, infant mortality there is estimated to have dropped nearly by half. Can one really imagine that such dramatic gains were entirely unaccompanied by nutritional progress?