This was not a sudden crisis. It may be only this spring that food prices have started sparking riots on the streets of Haiti and Egypt, not to mention rice rationing at Wal-Mart's cash-and-carry stores, but food prices have been rising since 2000. The rises accelerated in 2006, when global cereal stocks dropped to levels not seen since the early 1980s. And although the factors driving them are many and various, a good few of them look likely to persist for years to come.

Nor is the crisis unremittingly heinous. Higher food prices, other things being equal, mean higher farm incomes, and there are a lot of poor farmers in the world who could do with such a boost. But although this may suggest benefits for some in the future, the net effect so far has been negative. An interim report released by the World Bank in April says that seven years' poverty reduction has been undone by the past two years of high staple-food prices.

Agriculture has povertybusting powers beyond revenue increases.

The causes of these shortages are not easily undone, and some of them are things no one should want to undo. In China and India there is ever more — and utterly reasonable — demand for a third meal in the day and more meat in the diet. In Australia prolonged drought has had a severe effect on wheat production. High energy prices mean costly fertilizers and insecticides, not to mention making farm machinery more expensive to run. In the United States, more and more corn (maize) goes to making ethanol, raising the price of both corn and other cereals that can substitute for it.

There are various ways in which the fruits of scientific research might have helped ease the suffering that comes from this confluence of factors. But here, too, the harvest is not what it might have been. Public spending on basic agricultural research fell during the 1980s and 1990s in rich countries (see page 8). The proportion of US aid ploughed into agriculture wilted from 25% to 1%, bilateral farming aid from Europe dropped by two-thirds and World Bank lending in the sector slipped from 30% to 8%. The reasons for this included the perceived success of Green Revolution technologies in Asia, and, indeed, some backlash against intensified farming among green groups. The downslide was most pronounced in sub-Saharan Africa, where the cutbacks were still severe even though there had been no Green Revolution comparable to that in Asia. A contributing factor to this decline from the 1990s on was Europe's attitude to genetically modified crops, which both chilled research in the area and reduced incentives for such technologies to be fielded in countries looking to European export markets.

One might assume that such cutbacks in research reflected poor results. Not so; the pay-offs to agricultural research are massive. The World Bank's World Development Report 2008: Agriculture for Development (http://tinyurl.com/2ngyqd) — the first of the annual reports to focus on agriculture for a quarter of a century, the bank noted with self-reproach — cites 700 published estimates of rates of return on investment in agricultural research, development and extension services in developing countries. It reports an average annual return of 43%.

Agriculture has poverty-busting powers beyond straightforward revenue increases. One reason for this is that poor people in poor countries who earn a little extra cash will spend it on basic local goods and services — agricultural growth spurs economic growth from the bottom up. A study of 42 developing countries covering the period from 1981 to 2003 found that growth in gross domestic product (GDP) that originated in agriculture increased spending by the poor two-and-a-half times more than does GDP growth in other sectors.

The past weeks have brought signs that global institutions and donors are beginning to bow, belatedly, to this logic. On 2 April the World Bank announced its intention to double agricultural lending to sub-Saharan Africa over the next year, and bank administrators say that a portion of the new money will go towards basic research. Britain, the International Monetary Fund and the Bill & Melinda Gates Foundation are also opening their coffers. In the case of the Gates's money, much will be channelled through the Alliance for a Green Revolution in Africa led by Kofi Annan.

There are many useful directions for such development; higher yields, drought resistance and reduced requirements for inputs such as fertilizers and pesticides are all promising. But the more pressing problem for poor farmers is not the development of new technologies but access to those already there. Plenty of good agricultural science — such as locally adapted seed varieties and soil surveys — sits unused because it has not been delivered in a form adequately tailored to the end users and their limited means. Resources need to go towards coordinating and strengthening local agricultural extension services as an integral part of revamping and reintegrating the research infrastructure. Agricultural research systems in sub-Saharan Africa are fragmented into almost 400 distinct agencies, eight times the number in the United States and four times the number in India.

Access is not just a matter of seeds and bloodlines and new agronomic know-how. Weather services that rich-world farmers rely on are extraordinarily hard to get hold of in poor countries. The first attempt to evaluate the impact of climate change on agriculture in Africa, for example, had to rely on climate-sensitivity studies carried out in the United States for lack of any data from the continent in question. Better weather and climate data would allow lenders, be they development banks or local sources of microfinance, to create insurance products for farmers of a sort that is almost completely unavailable to the world's poor. These pragmatic solutions should get a large slice of a rapidly expanded pie.