How much can farming improve people's health?

The Economist opines on agriculture and micronutrient deficiencies:

Farming ought to be especially good for nutrition. If farmers provide a varied diet to local markets, people seem more likely to eat well. Agricultural growth is one of the best ways to generate income for the poorest, who need the most help buying nutritious food. And in many countries women do most of the farm work. They also have most influence on children’s health. Profitable farming, women’s income and child nutrition should therefore go together. In theory a rise in farm output should boost nutrition by more than a comparable rise in general economic well-being, measured by GDP.

In practice it is another story. A paper* written for the Delhi meeting shows that an increase in agricultural value-added per worker from $200 to $500 a year is associated with a fall in the share of the undernourished population from about 35% to just over 20%. That is not bad. But it is no better than what happens when GDP per head grows by the same amount. So agriculture seems no better at cutting malnutrition than growth in general.

Another paper† confirms this. Agricultural growth reduces the proportion of underweight children, whereas non-agricultural growth does not. But when it comes to stunting (children who do not grow as tall as they should), it is the other way round: GDP growth produces the benefit; agriculture does not. As a way to cut malnutrition, farming seems nothing special.

Why not? Partly because many people in poor countries buy, not grow, their food—especially the higher-value, more nutritious kinds, such as meat and vegetables. So extra income is what counts. Agriculture helps, but not, it seems, by enough.