Wed, 01 Mar 2006 22:30:26 +0000 To: Subject: Food adequacy in international goals and reports Dear Mr Ziegler Food adequacy in international goals and reports Thank you for your reply. I thought it might be helpful at this stage to provide some more brief details. In the newspaper article of 2003 which I mentioned in my email, there are references to officials in charge of FAO and World Bank statistics who confirmed to me details of the methods. Below are sources for the respective methodologies; some notes on communications from me to staff of the World Bank, FAO and UN, and notes on communications passed to Governors of the World Bank. A: FAO method and food needs Quotation: “The cutoff point is derived by aggregating the sex-age-specific minimum energy requirements using the proportion of the population in the different sex-age groups as weights.” Above extract from: FIVIMS. An inter-agency initiative to promote information and mapping systems on food insecurity and vulnerability. Proceedings. Measurement and Assessment of Food Deprivation and Undernutrition. International Scientific Symposium, Rome, 26-28 June 2002. Symposium convened by the Agriculture and Economic Development Analysis Division. Food and Agriculture Organization of the United Nations, Rome, 2003 Part II: Methods for the measurement of food deprivation and undernutrition. FAO methodology for estimating the prevalence of undernourishment. Loganaden Naiken, FAO (retired), Rome, Italy Source: B: World Bank method The World Bank uses per-person statistics. There is no mention of children’s food needs in the World Bank methodology paper for Millennium Goal Indicator 1. The paper is “How did the World’s Poorest Fare in the 1990s?” by Shaohua Chen and Martin Ravallion. The following version is from August 2000: C: Communications from MB to officials and others Below are some details of past communications to World Bank, UN and FAO staff, and to World Bank governors on the subjects of: a) claims to have aggregated outcomes for individuals without reference to survival rates, and b) claims to have aggregated trends in consumption adequacy without reference to food needs. 3 April 2001: Explained mortality flaw in Millennium Goal 1 (expressed as faster reduction if more children die) to Caroline Anstey, head of Media Relations at World Bank. April 2001: Explained mortality problem to Srikanth Puranam, speechwriter to President of World Bank. Telephone; forwarded above email to him. April 2001: Explained mortality problem to Jeffrey Hammer, a Lead Economist at the World Bank and co-author of “Life and Death among the Poorest” (1999). He expressed interest in an approach taking outcomes into account of those who did not survive. Telephone. 11 April 2001: Explained mortality problem to Eric Swanson, head of World Development Indicators project at the World Bank. Discussed potential impact of AIDS on demography. I explained need to know survival rates in order to know how many crossed poverty line upwards. Telephone. [For reference: same day, 11 April 2001: email to Professor Sachs ( )]. 15 October 2001: Informed Andrew Smith, Member of UK Parliament, of mortality flaw in economics in relation to both a) distributional studies and b) reported trends in poverty. Email. Message mentioned AIDS. Mr Smith informed Britain’s Governor of the World Bank, Clare Short MP. August 2002: Discussed mortality problem with Robert Mayo, FAO statistician. If more die, hunger is less. Telephone. 30 August 2002: Discussed aspects of UN Millennium Goal statistics including mortality flaw with Robert Johnston, Chief of Statistical Services at UN Statistics Division. Telephone. 19 February 2003: Informed James Wolfensohn, President of World Bank, of fact that Research Department of Bank using statistics which a) looked better if poorest die and b) treated adults as needing same as children. In front of an audience at the Sheldonian Theatre, Oxford. Early 2003: Discussed mortality flaw and discrepancy between Bank and FAO assumptions on trends in food needs with Jorge Mernies, senior FAO statistician. He stated: a) FAO considering including mortality in estimates of food inadequacy. b) Mr Mernies aware that World Bank staff not increasing estimates of food needs per person as birth rates fall, while FAO do increase estimates. c) Currently unable to estimate increases in food needs of global hungry because of design of FAO database. 28 March 2003: Confirmed through Shaohua Chen, co-designer of methodology paper for World Bank Millennium Goal Indicator 1, that World Bank do not adjust estimates of global poverty for children’s food needs. Telephone. 25 June 2003: Informed Chairman of Parliamentary Select Committee in UK of mortality flaw and food error. Email. Chairman then wrote to Department for International Development for an answer. Hilary Benn MP replied to Chairman on behalf of the Department. Mr Benn subsequently became British Governor of World Bank. Mr Benn’s reply to the Chairman did not refer to the food error, or to the mortality problem in relation to either AIDS or policy research. 6 November 2003: Informed Britain’s Alternate Governor of the World Bank, Gordon Brown, of: a) mortality problem in poverty reduction strategies where survival rates not known to be increasing, and of b) difference between FAO and Bank assumptions on food needs. Handwritten note acknowledged by Mr Brown who offered thanks. 2005: Again asked Mr Mernies of FAO for rough estimate of increases in per person food needs of global hungry in past years. Email not answered. The above details are from memory. I began raising the mortality problem and the problem of using per capita statistics while birth rates fall with economists in 2000. The list of recipients of communications from me on these matters is quite long. These problems are in addition to a) the problems inherent in using expenditure surveys, which provide the bulk of the survey data on which both the World Bank and FAO methods are based. These problems include the fact that expenditure does not tell a researcher either about food prices or about overall needs for expenditure. b) problems of aggregation: the number of rich people doesn’t tell me how rich they are. c) the general demographic problem of distinguishing between “one baby fewer is born” and “one more person rose out of hunger/poverty”. d) questions about the quality of food (FAO method is in terms of calories). e) problems relating to data gaps (the destitute may not be reachable) and comparability of surveys (different questions in different countries at different times, and different methods for valuing food consumption in money terms). The existence of such problems could be seen as adding weight to arguments to the following effect: It is not possible to judge the adequacy of consumption to meet basic survival needs without estimating survival rates. Best wishes, Matt Berkley