16. The Expert Consultation learned from David STEANE in STAT/LVSTK/4, Use of livestock sector data in policy, that policy did not need to be based on data but could simply reflect ideological views. Many policies for livestock have been based on partial information and have had serious long-term consequences. He pointed out that in Asia, for example, many breeding programmes were not sustainable and required large investments to correct this problem. The predictions of demand for livestock and livestock products were such that they required a livestock revolution if demand was to be met. This increased demand could have serious consequences in a region which already had the largest number of livestock and which had the greatest predicted increase in demand. The potential hazards in terms of pollution, health risks, land degradation and water use for livestock could create serious dangers unless countries achieved more sustainable systems of livestock production.
17. He stated that the dualistic development of high technology, high capital, intensive holdings as well as extensive, resource-based labour-intensive small holdings could create serious difficulties, but mentioned that there were ways in which the two types of development could collaborate for mutual benefit. He noted that this situation created considerable pressure to improve the database to enable planners and policy makers to increase the likelihood of achieving sustainable programmes for livestock production.
18. He observed that FAO statistics, at present, failed to provide adequate data because such information was not provided by the countries. He pointed out that there was an urgent need to provide more information on various aspects including production systems, mature sizes and types of animal, details on reproductive rates, farm sizes, cropping and market accessibility in a form which allowed the different proportions of each to be quantified. It was important that policy makers be made more aware of the potential and of the various programs available for assistance. He suggested that FAO needed to become more pro-active in providing such awareness and in encouraging countries to provide the basic data. Without better data, the likelihood of achieving sustainability was low and could result in serious repercussions for the livestock industry. Such repercussions could mean not only a reduced increase in demand, but that the opportunity for alleviation of poverty and increased food security could be lost.
19. Ayush DEMBEREL described the main uses of livestock sector data in National Accounts in Mongolia in STAT/LVSTK/5, Use of livestock sector data in national accounts. He explained that the primary data sources for Gross Domestic Product (GDP) were the estimates of annual value added of different economic activities including livestock.
20. He noted that a revision had been made in the GDP compilation methodology concerning the extraordinary losses of animals that occurred in 2000-2002 due to a disaster resulting from the cumulative result of a summer drought followed by an extremely harsh and cold winter (called DZUD). He mentioned that the National Statistical Office and the Ministry of Finance and Economy adopted a Joint Resolution in November 2002 approving the new methodology in which distinctions were made between types of animals and between types of losses.
21. He explained that now animals were classified into two categories: a) “work-in-progress” and b) “fixed assets”; and that losses were classified, using the historical data of the last 40 years, as recommended in the EUROSTAT Manual on the Economic Accounts for Agriculture and Forestry (1997 Rev, 1.1), into three categories: a) normal losses, b) accidental losses, and/ or c) catastrophic/ exceptional losses.
22. Losses within one standard deviation from the arithmetic mean were considered as “recurrent or normal losses”, those between one and two standard deviations were considered as “unexpected incidental losses”, and those in excess of two standard deviations were considered as “exceptional or catastrophic losses”. Based on the above criteria, a threshold for each species of animal (camels, horses, cattle, sheep, and goats) had been established to separate exceptional losses from normal and accidental losses.
23. Furthermore, he indicated that, in accordance with recommendations from the “SNA 93”, FAO handbook (1996) System of Economic Accounts for Food and Agriculture United Nations and EUROSTAT Manual on the Economic Accounts for Agriculture and Forestry (1997 Rev, 1.1), the new methodology introduced certain exclusions of such exceptional losses in the national accounts. For animals classified as “fixed assets,” exceptional losses, which were losses in excess of two standard deviations above the historical mean (a fixed threshold), were recorded in a different way to “normal” losses. In contrast to the normal losses (which were deducted from output), the value of such exceptional losses (up to the time of the loss) were not deducted in the calculation of output (as if the loss had not occurred). Production account, generation of income account, allocation of primary income account, gross capital formation and goods and services accounts were compiled with these definitions.
24. He made a number of suggestions for improvement of the sample survey of agriculture, of economic accounts and of estimates in Mongolia, including
- expanding the livestock census to an agriculture census;
- conducting a survey of agricultural households, companies and cooperatives every three to five years on a regular basis;
- developing and testing the methodology for the compilation of value added in agriculture at constant prices using the double deflation method; and
- conducting a special survey of yield rates and producer prices of bi-products slaughtered in households, agricultural companies and cooperatives and slaughter houses on a regular basis.
25. In STAT/LVSTK/6, Use of livestock sector data in food security and nutrition, Biplab NANDI, Senior Food and Nutrition Officer, RAP, presented some issues related to food security and nutrition in the context of livestock statistics as an important source of information for addressing problems of nutrition, including micronutrients.
26. He pointed out the need for human dietary diversification to address micronutrient (minerals and vitamins needed for proper growth, development and function) malnutrition. He emphasized that livestock were a good source of both macro and micronutrients and helped to improve the nutritional quality of the diet. He reminded the Expert Consultation that the daily per capita energy availability from animal foods/products varied widely in the region and that Southeast Asia had a more diverse diet compared to South Asia.
27. The Expert Consultation was informed that the nutrition sector made use of livestock statistics data and that household level data were more important than country data for addressing household food and nutrition security.
28. He mentioned that FAO has promoted nutrition programmes at the country level which recommend home gardens with small livestock such as chickens, ducks, pigs, rabbits, and it was important to have disaggregated data in order to get reliable consumption data.
29. The Experts summarised the surveys and associated methodologies used in selected countries (India, Indonesia, Mongolia, New Zealand, the Philippines, Republic of Korea and Thailand) in order to provide a basis for identification of improved strategies and methodologies. These case studies are found in STAT/LVSTK/7, Existing methodologies for collection of livestock statistics.
30. Mamadou DIEDHIOU reviewed the strengths and weaknesses of some methodologies used in the collection of livestock sector data in STAT/LVSTK/8, Review of strengths/weaknesses in the methodologies. He described livestock data collection methods such as agricultural census, livestock census, periodic surveys, ad-hoc surveys, household consumption surveys. Two main methods for the enumeration of livestock - ground counts and aerial counts - were presented. He believed that aerial counts had not been used in the Asian region but could be a useful tool to consider in future livestock surveys. Commonly used sampling techniques were reviewed along with their advantages and disadvantages.
31. He pointed out the need for better coordination of institutions involved in livestock statistics within each country. It was suggested that relational databases of livestock data be put in place to permit integration of data from these various institutions.
32. In the discussions following the presentation, the Expert Consultation agreed that the use of aerial surveys for areas where accessibility could be a problem was useful, but pointed out that the logistics required (airport, trained team, etc.) for an aerial survey could be a limiting factor.
