Hartwig de Haen
Assistant Director-General, Economic and Social Department
(ES)
Food and Agriculture Organization of the United Nations (FAO)
Rome, Italy
As we come to the close of this Symposium, a number of things have been reconfirmed and many lessons learned. The first of these is that the fulfilment of FAO's mandate to monitor progress in hunger reduction requires accurate, reliable and timely measures of the prevalence of hunger and malnutrition, food insecurity and vulnerability, and how these change overtime. What we do is extremely important, and even if we do not do it perfectly we should do it in the best way possible.
It was confirmed by your discussions that food insecurity is a multifaceted and complex phenomenon. I think I can safely conclude that there is no perfect single measure that captures all aspects of food insecurity. I have heard the term "suite of indicators" being used in your discussions. This refers obviously to the necessity of describing the phenomenon we need to understand through the use of multiple indicators. We know that food insecurity is determined by food availability, access and utilization as well as by individual vulnerability, and each of these determinants may require a separate indicator. We have discussed several of these already: the FAO measure of food availability adjusted for access; poverty or other indicators of food access derived from household surveys; food consumption from individual dietary intake surveys; anthropometric indicators; and also self-assessed or "qualitative" indicators to measure hunger and vulnerability. You have indicated that a suite of indicators could help identify the relative importance of different determinants of food insecurity. This morning, I heard agreement that the data derived from these measurements should help us to understand why people are food insecure, although this is not the primary task. While the primary task is, of course, to measure the extent of food insecurity or hunger, identifying some of the main causes of hunger will enhance the likelihood that this information can lead to better policies.
A comprehensive discussion has taken place during this meeting of which indicators should be chosen. I can only submit to you several criteria to be kept in mind when considering which indicators to use. The first criterion is how well the indicator measures what it claims to measure, that is, its validity and reliability. Another consideration is how helpful the indicator is in identifying the causes of food insecurity. Also important is how quickly the information becomes available to policy-makers and those who work on hunger-related issues. It is clear that timeliness is essential, especially in the assessment of emergency situations. Necessary action to tackle problems of food insecurity may be delayed if we do not relay the information to policy-makers in time. Another vital consideration is whether the indicator measures and differentiates transitory and chronic food insecurity. And of course, one cannot ignore the cost of obtaining, processing and disseminating the information. Collection costs in terms of time, equipment and training of personnel must be acknowledged, and countries and agencies must determine whether the benefits of using a particular indicator are worth the costs.
Lastly, we must consider the link between the measures or indicators provided to policy-makers and the decisions taken on the basis of that information. One aspect of this problem is the level at which the indicators are collected (internationally, regionally, nationally or subnationally). The FAO indicator of chronic undernourishment cannot be disaggregated to subnational levels, so for that reason alone additional indicators are necessary for policy-making at these levels. Another very important issue that has been discussed widely during the Symposium is whether we should make more use of trends over time and less on absolute numbers when measuring progress. The comparison of trends is very important across countries but is equally important for monitoring change over time within individual countries. A related issue is the frequency with which the indicators are released and the appropriate time intervals for data collection.
This Symposium has been instrumental in suggesting how all of us - researchers, international organizations and government officials alike - can proceed to improve the measurements and the methodologies. It has specific implications for the work we do at FAO. We began in 1999 to issue an annual report, The State of Food Insecurity, in close collaboration with FIVIMS, so the task of collecting and reporting multiple indicators is already underway. We have also used this framework in all reports to the Committee on World Food Security. However, it has been made clear that we must try to better explain the methodology, the database and the assumptions underlying our undernourishment measures. In light of your suggestions here, we may also need to consider replacing the term "undernourishment" with a more precise term of what we are measuring, such as "food energy deficiency". We have been made aware of the importance of incorporating data on nutritional outcomes and related concepts into our databases, and we must learn how to use the additional information to improve our indicator. The data on micronutrients and protein, and the measurements of intakes of these nutrients are other areas mentioned for further attention by FAO. We must improve our measures of access to food: this means collecting and analysing more complete information on the distribution of household income, assets and agricultural landholdings, an area that we have not sufficiently covered up to now.
It was suggested in the course of your discussions that we initiate a working group that should strive to improve our understanding of the concepts of risk and vulnerability and to develop indicators that capture these concepts, including further development of "qualitative" measures of hunger, also referred to as "experience-based assessment", direct assessment" or "self-assessment".
Currently, most of the information known about food insecurity is compiled and used by international agencies, but developing countries need to take ownership and begin to develop the capacity to compile and use their own data. Through FIVIMS, we will continue to extend support to national governments to generate their own data on indicators for use in national-level policy and decision-making.
In conclusion, what we have discussed together in these three days is just the beginning of a process. We at FAO must continue our scientific work in partnership with you and other experts. Therefore we will be drawing on you and your expertise again, and I invite you all also to stay in contact with us so that we can continue our collaboration for improving measures of food insecurity.
I want to express my explicit thanks to the members of the Scientific Advisory Committee for their effort, to the Government of The Netherlands for their support of this Symposium, and of course to all of you for your fruitful and active discussion of these important issues and suggestions for further work.
Ladies and Gentlemen, this Symposium is closed.
