Better estimates of agricultural production in hungry countries are possible. Most of the food consumed in the most food insecure countries is produced inside them. Knowing how much food a country has produced is generally the first stage in identifying incipient food crises. In both good and bad years, key decisions (emergency food needs assessments, official or commercial import targets, pricing decisions, safety net threshold analyses) are driven by these estimates. But incredibly, food production estimates in most food insecure countries have NO known accuracy. Most of them are based only on accreted assumptions. It is likely that bad information is driving key decisions in many countries (a 60% error in area planted has just been found in one of the largest and most food insecure countries in the world). In the agricultural production estimate equation, the amount of area planted is the key weakness. To produce estimates of known statistical accuracy, the technological innovation required is the use of an area frame sample informed by high-resolution imagery. For these purposes, the imagery would be available for free, the field costs would be more moderate than current field-based assessments, the statistical methods are standard, the skills are easily trained, and the financial, commercial and humanitarian benefits of more accuracy would be enormous.
Idea No. 103