Global Livestock Costing Model

1 Report Global Livestock Costing Model Vittis Yiorgos1, Lindner Soeren2 and Obersteiner Michael3 October 2025 [email protected] 2lindn[email protected] [email protected] 2 Table of contents Abstract ................................................................................................................................................... 4 About the authors ................................................................................................................................... 5 Data .................................................................................................................... 6 Data to estimate production costs ......................................................................................................... 6 Livestock sector system typology ........................................................................................................... 8 Data processing and extrapolation: .................................................................................................... 9 Bottom-up cost engineering framework ............................................................... 11 Intensification Factor ............................................................................................................................ 11 Data sources: ..................................................................................................................................... 12 Data Evaluation ................................................................................................. 14 Production cost estimations ................................................................................................................. 14 Input factor validation ....................................................................................................................... 15 Cost of Land ....................................................................................................................................... 17 References ............................................................................................................................................ 19 3 ZVR 524808900 Disclaimer, funding acknowledgment, and copyright information: IIASA Reports report on research carried out at IIASA and have received only limited review. Views or opinions expressed herein do not necessarily represent those of the institute, its Member Organizations, or other organizations supporting the work. The authors gratefully acknowledge funding from the Food and Agriculture Organization of the United Nations (FAO) for the research project "Technical and science-policy engagement services to support country development of sustainable food and land use pathways (FABLE-FAO)" under the Letter of Agreement RRS 5675 with IIASA. This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 International License. For any commercial use please contact [email protected] 4 Abstract This report describes the construction of a detailed bottom-up database quantifying production costs of nine input factors to livestock production systems distinguished in two main classifications for globally dominant production systems: grass-based and mixed feed-grass systems. The resulting livestock costing model uses national-scale information to estimate on-farm production costs covering three animal species: cattle, sheep and goat. Production costs were sourced from national household surveys and global agricultural databases. For countries with no data availability cost input factors were extrapolated based on case-study countries representing different global production regions each with unique biophysical, agroecolocial and socio-economic characteristics. The costing data presented here allows for flexible integration and linkage with broader agroeconomic assessment frameworks, or large-scale land based production sector models such as the FABLE calculator. The cost data can be downscaled to a spatially explicit grid level following a method that uses intensification factors. It serves as baseline representing costs of production in a given base year and has the purpose to assist in addressing a variety of research questions on economic cost implications of sustainable transition pathways of livestock production. 5 About the authors Yiorgos Vittis is a research Fellow in the Exploratory Modeling of Human-natural Systems (EM) program at IIASA. (Contact: [email protected]) Soeren Lindner is a research Scholar in the Exploratory Modeling of Human-natural Systems (EM) program at IIASA. (Contact: [email protected]) Michael Obersteiner is the head of the Environmental Change Institute (ECI) at Oxford University. (Contact: [email protected]) 6 Data Data to estimate production costs Production cost data for cattle, sheep, and goats are sourced from household surveys and agricultural databases, including SIMLESA (Nyagumbo et al. 2025), LSMS (D’Alessio et al. 2025), PRHPS (IFPRI & IDS, 2018, p. 2), KIDS (IFPRI, 2000, pp. 1993–1998), MLA (Meat and Livestock Australia, 2020), FADN (2020), FBS (Farm Business Survey, 2020), and the Chinese Bureau of Statistics (Table 1). Additionally, empirical data on production costs are derived from grey literature sources (such as Farmers Weekly) containing expert’s opinion on input requirements and costs. Data surveys including SIMLESA, LSMS, PRHPS and KIDS do not distinguish between livestock farm types and thus, to derive detailed commodity-specific production costs, data was extracted from farms recording only one type of livestock. FADN distinguishes farm types on the basis of the majority of outputs (> 2/3 of farm’s total output) and therefore commodity-specific costs is obtained directly from the database. Finally, FBS and the Chinese Statistics Bureau provides cost information explicitly for the production of each of the livestock commodities. These costing datasets were used to develop a cost engineering framework that contains information on livestock production costs, disaggregated in 9 direct and overhead cost elements following the cost accounting system of the Farm Accountancy Data Network (FADN) and the Agri-benchmark network (Table 2). Table 1: Household and agricultural data surveys Organisation Survey Abbreviation Description Countries included Source CIMMYT SIMLESA Pathways to sustainable intensification in Eastern and Southern Africa - Ethiopia 2010 Ethiopia https://data.cimmyt.org/data set.xhtml?persistentId=hdl:11 529/10746 ILRY IMPACTLite The Integrated Modelling Platform for Mixed Animal Crop systems Southeast Asia sites (Nepal) https://data.ilri.org/portal/da taset/implite-sa World Bank LSMS Living Standards Measurement Study Malawi https://microdata.worldbank. org/index.php/catalog/1003/ study-description#metadataidentification IFPRI PRHPS The Pakistan Rural Household Panel Survey Pakistan https://dataverse.harvard.ed u/dataset.xhtml?persistentId =doi:10.7910/DVN/LT631P IFPRI KIDS KwaZulu-Natal Income Dynamics Study South Africa https://dataverse.harvard.ed u/dataset.xhtml?persistentId =doi:10.7910/DVN/SKACYP 7 Australian Government MLA Meat and Livestock Australia Australia http://apps.agriculture.gov.a u/mla/mla.asp European Commission FADN Farm Accountancy Data Network EU countries (Bulgaria, Finland, France, Greece, Ireland) https://agridata.ec.europa.eu /extensions/DashboardFarmE conomyFocusLivestock/Dash boardFarmEconomyFocusLive stock.html UK Government, DEFRA FBS Farm Business Survey United Kingdom http://www.farmbusinesssur vey.co.uk/benchmarking/Def ault.aspx?module=GrossMarg ins China, Bureau of Statistics Compilation of National Agricultural Product Costs and Benefits China http://www.stats.gov.cn/tjsj/ tjcbw/202008/t20200824_17 85455.html Farmers Weekly FW South Africa Farming & Agricultural Industry South Africa https://www.farmersweekly. co.za/ Table 2: Description of cost components for livestock production Cost variable code Cost Variables Description FDC Feeding stuffs costs Costs for homegrown and purchased feeds GRC Grassland (forage) costs Seed, fertiliser, plant protection, silage wrap and twine VMC Veterinarian and medicine costs Veterinary fees and medicines OLC Other livestock costs Bedding, ear tags, haulage TLAC Labour costs Paid labour Unpaid labour TFLC Fuel, lubricant and water costs TFIN Financing Insurance Taxes Capital costs INFC Infrastructure (Buildings) Heating, lighting TMAC Machinery Repairs Depreciation Interest 8 Livestock sector system typology For many countries financial data on livestock production is either not publicly available or impossible to obtain for the interested researcher. To fill this data gap, four classes were adopted from a country typology for livestock decision making by Herrero et al. (2020) to form country clusters of similar production characteristics. They combine livestock-related and socio-economic factors including livestock production and consumption, cropland and grassland cover, income and expenditure (table 3 for details of the cluster characterisation and figure 1 for a spatial representation). The chosen typology divides countries into four global livestock system clusters: High income countries with high inputs of production and high levels of consumption (HIHC), Lowand middle-income countries with low inputs of production and low levels of productivity and consumption (LILC), Low-, middleand highincome countries with higher livestock consumption and extensive grazing areas (EXHC) and Lowand middle-income countries with varying livestock consumption and emerging importance of livestock (EILI). This cluster typology allows to extrapolate financial information from the case-study countries with good cost data availability for those countries without information in the livestock sector. For this, countries with no costs were further grouped into 22 geographical subregions according to the United Nations geoscheme, and use of a purchasing power parity (PPP) index allowed to extrapolate specific data on production cost variables from case study countries of the same geo-region and system typology. Table 3: Livestock sector typology: cluster characterisation by income, consumption and productivity LILC EXHC EILI HIHC Lowand middleincome countries Low, middleand high income countries - Transition countries Lowand middle income countries High-income countries Low livestock consumption Higher livestock consumption Varying livestock consumption High livestock consumption Low productivity Ag-based Large grazing areas High productivity Industry High productivity livestock industry LILC: Low input, low consumption. EXHC: Extensive input, high consumption. EILI: Emerging importance of livestock. HIHC: High input, high consumption 9 Fig. 1: Livestock systems cluster typology. Clustering of countries based on similar attributes relating to the role of livestock, society and the country’s economy (adapted from Herrero et al. 2020). Table 4 summarizes the data sources for cost information by livestock species, livestock typology cluster and livestock system type used for the baseline countries used to extrapolate costing information within the cluster and livestock production system for the remaining countries. In occasions where some cost elements are missing from the baseline data survey, costs were estimated using the respective cost analogue from a different country, within the same cluster and livestock production system that has the available information. Data processing and extrapolation: 1. Baseline case-country data has been taken into the code 2. To convert all country-case study cost records from the various years to a common baseline year we introduce the Consumer Price Index (CPI) with which we inflate/deflate wages to 2000. This year was selected because the phyiscal dataset by Herrero et al. was developed for the year 2000. Data gaps for CPI at the country level were filled by subregion-scale averages. Regions correspond to the world subregional classification by the United Nations 1 that includes 22 groups. 3. To offer the option of conversion from LCU to Purchasing Power Parity dollars (PPP$) and US dollars (US$) we import respective data from the World Bank for the year 2000. 1 https://unstats.un.org/unsd/methodology/m49/ 16 farms in these EU countries. Based on empirical data, these cost items represent the most significant shares of inputs, in the respective livestock category and thus, are of primary importance in this examination as their magnitude drives systems change. Across these countries, observed feeding stuffs costs (including homegrown and purchased feeds) in cattle producing systems correspond to 43% and financing costs (including interest and financial charges, depreciation of capital assets and insurance) to 20% of total inputs. For feed costs in cattle production systems, it was found that 9 out of 15 countries have a difference between observed and extrapolated shares of less than 13%, 2 countries have no difference as they use information from FADN directly and 4 countries have difference between 20% and 37% (Figure 4a). In the latter cases, while the difference in the share seems to be larger, our estimated share of feed costs varies between 28% and 43% which is very close to the regional average (43%). The average difference in the shares of inputs for feeding stuffs between cost estimations and FADN empirical data is 14%. For financing costs, 11 countries have a difference in shares of less than 17% and Austria is the only country exceeding this percentage with a difference of 28% (Figure 4b). Similarly to the feed costs, a larger difference was identified for Austria, but the overall estimated share of financing costs is 19% across country averages while the respective regional mean based on empirical data is 20%. The average difference in the shares of inputs for financing costs across the 15 EU countries is 7%. Fig. 4: Cross validation of production input shares for cattle production in EU countries. Comparison of the shares of a. feeding stuffs (purchased and homegrown) and b. financing costs (interest and financial charges, depreciation of capital assets and insurance). Bars in plots demonstrate the share of the respective modelled and observed (FADN empirical data) cost item. a 17 b Cost of Land The incorporation of land costs (actual land costs and rent charges) would represent an important additional cost factor. However, such information along with corresponding tenure data remain very limited at global scales and the explicit calculation of land costs is out of the scope of the cost data framework presented here. However, the relative importance of land costs across the different system types and clusters compared to the nine production cost factors can be derived using empirical data from MLA and FADN on rent charges for farm land and buildings (Table 5). The average share of rent costs to total costs are relatively small in this sample of countries (𝑛 = 20) in both grazing-based and mixed systems, with the highest value emerging from mixed systems in cluster HIHC, being close to 6%. It was found that on average, grazing-based systems have smaller land rent charges than mixed systems for both sheep and goat (3.3% and 4.1% respectively) and cattle farms (3.3% and 5.4% respectively). Yet, their difference is small and thus, the relative cost advantage we give to extensive systems by ignoring land costs is minimal (1% for sheep and goats and 2% for cattle). 18 Table 1: Share of land costs (rent charges) to total costs using FADN empirical data Cluster System Countries Average rent share in sheep and goat farms Average rent share in cattle farms LILC Livestock only Estonia, Finland 3.3% 3.9% LILC Mixed Lithuania 2.4% 4.1% EXHC Livestock only Australia, Greece, Slovenia 2.7% 1.9% EXHC Mixed Bulgaria, Romania, Cyprus 4.3% 6.9% EILI Livestock only Not covered by FADN EILI Mixed Not covered by FADN HIHC Livestock only Austria, France, Ireland, Italy, Slovakia, Sweden, Portugal 3.8% 4.1% HIHC Mixed UK, Germany, Spain, Denmark 5.7% 5.3% All clusters Livestock only system average 3.3% 3.3% All clusters Mixed system average 4.1% 5.4% 19 References D’Alessio, G., Toma, I. : Measuring Wealth in Household Surveys in Lowand Middle-Income Countries. An introduction to the World Bank Guidelines. Statistical Journal of the IAOS, Volume 41 p.: 846 -857. https://doi.org/10.1177/18747655251342644 . 2025 Nyagumbo I. et al.: SIMLESA. On-station and on-farm agronomy data from 2010 to 2019. CIMMYT Research Data & Software Repository Network https://hdl.handle.net/11529/2223085. 2025 FPRI/IDS (International Food Policy Research Institute/Innovative Development Strategies). Pakistan Rural Household Panel Survey (PRHPS) 2014, Round 3. Washington, D.C./Islamabad, Pakistan: IFPRI/IDS. 2016 Herrero, M., Mason, D., McMillan, L., Dennis, G. & Palmer, J. Metrics for Decision-Making in the Livestock Sector: Project Summary Report 49. 2020. Herrero, M. Havlík, P. Valin, H. Notenbaert, A. Rufino, M. Thornton, P. Blümmel, M. Weiss, F. Grace, D. Obersteiner, M. Biomass use, production, feed efficiencies, and greenhouse gas emissions from global livestock systems. Proceedings of the National Academy of Sciences of the United States of America.Volume: 110, P.: 20888 -20893. https://doi.org/10.1073/pnas.1308149110. 2013 Sheahan, M. Barrett, C. Food loss and waste in Sub-Saharan Africa. Food Policy. Volume 79, page 1-12. 2017 Alston, J. The benefits from agricultural research and development, innovation, and productivity growth. OECD Food, Agriculture and Fisheries Papers, No. 31, OECD Publishing, Paris. http://dx.doi.org/10.1787/5km91nfsnkwg-en. 2010 Robinson, T. et al.: Global livestock production systems. Food Agricultural Organisation. U. N. (2011). https://openknowledge.fao.org/handle/20.500.14283/i2414e