PPLPI Mid-Term Review March 2005 FAO, Rome PPLPI Output 3: Effective systems for livestock policy information, analysis, decision-support and monitoring and evaluation PPLPI Mid-Term Review March 2005 FAO, Rome

Overview Baseline Information Analysis Decision support tools Information dissemination Monitoring and evaluation

Overview IGAD Network of Data Providers and Users Djibouti Eritrea Ethiopia Kenya Somalia Sudan Uganda Policies Trade Biophysical Socio-economic Livestock Decision Support Tools Livestock Policy Information System Policy Analysis Policy Recommendations

1. Baseline Information Quantitative Qualitative GIS data Living Standards Measurement Surveys (LSMS) data Global Livestock Information System Qualitative policy; legislation; institutions networks of livestock policy stakeholders documentation, reports etc.

Global livestock information system Overview Data collection Livestock modelling Information dissemination Applications of global livestock distributions

Global livestock information system SALB / UNCS OGC - compatible ISO metadata FAOSTAT GeoNetwork (KIDS) Agro-MAPS EMPRES-I Sub-national livestock data Sub-national admin. boundaries Livestock disease data Livestock prod. coefficients Oracle Database GLiPHA Livestock modelling Livestock sector reports Ad hoc queries and reports

Data collection Sub-national administrative boundaries Standards and codes (UNCS, SALB, FAO-SDRN) Sub-national livestock data Global network of data providers Statistical year books, project contacts, Internet Data verification - link to FAOSTAT Livestock disease data OIE standards Livestock production coefficients Regional estimates

Livestock modelling Sub-national numbers from database Area suitable for livestock (slope, land-use etc.) Sub-national “Observed” densities Environmental predictor variables (5 km rasters) Ecological zonation (cluster analysis) Multivariate models for each zone “Predicted” density of livestock Adjust to observed totals where available Adjust to FAOSTAT 2000 national totals

Cattle in Africa - Observed

Cattle in Africa - Predicted

Information dissemination Database query Livestock sector reports Ad hoc queries and reports Web-based information dissemination GLiPHA GeoNetwork Output to other FAO applications (planned?) EMPRES-I Agro-MAPS FAOSTAT (2)

Applications Livestock projections Livestock production system classification Livestock production estimates Environmental risk mapping Livestock disease risk mapping Public health risk mapping Poverty analysis Policy analysis

2. Analysis Living Standards Measurement Surveys (LSMS) GIS analysis Agricultural production systems mapping Poverty mapping

GIS analysis Vietnam Population (population census 1999) Poverty (VLSS 1997/98 & VHLSS 2002) Environmental data Accessibility Local spatially weighted regression

Poverty incidence and density

Environmental data Topography Land Cover

Environmental data Forest Soil Type

Environmental data Rainfall Temperature

Infrastructure and accessibility

livestock holdings Regional variation in livestock holding North-South decline in the share of livestock-derived income in total income Dominance of pig-derived income in all regions, but less so in poorest regions Cattle relatively important in CH and SCC

Spatial aspects of livestock holding

Local spatially weighted regression Local regression coefficients Local indicators of goodness of fit

Local spatially weighted regression Based on traditional regression framework Produces all standard regression output at local level (R2, b, t, P) Reveal local variation in spatial pattern Parameters can be quantified for sub-regions Parameters and the reliability of estimates can be mapped Divergence from global model can be tested

Local spatially weighted regression Vietnam Household Living Standards Survey 2002 over 2200 locational data points nearly 45,000 households Dependant variables Income, Expenditure Poverty Independent variables Income composition environmental variables accessibility

Local spatially weighted regression Income source PIG POULTRY Per capita income vs. income composition

Local spatially weighted regression Per capita income – pig and poultry Clear relationship where the share of income from pig or poultry declines with rising per capita income This relationship tends to exhibit the strongest negative coefficient in proximity to urban centers There are two sub-urban areas near Hanoi and HCMC where the share of income derived from poultry tends to rise with rising per capita income

Local spatially weighted regression Income source PIG POULTRY Accessibility vs. income composition

Local spatially weighted regression Road access – pig and poultry Share of income derived from pig and poultry tends to increase with improved road access in areas where accessibility is generally low, whereas an inverse relationship can be observed in areas where accessibility to roads is generally better This pattern of spatial variation and change of sign in the relationship is stronger for poultry - easier to transport ?

Conclusions Locate poor livestock keepers Visualize, communicate and disseminate spatial dimensions of patterns and trends in livestock holdings Importance of different livestock types across space Identify and quantify relationships among poverty, livestock, environment and infrastructure Analyze spatial variation in these relationships Generate information to analyze policy impacts and to inform policy development

3. Decision support tools EXTRAPOLATE Integrated Poverty Assessment for Livestock Promotion (IPALP) Dairy household model Policy menus Livestock policy analysis tables

EXTRAPOLATE Participatory decision support tool Uganda Model for pro-poor stakeholder analysis Stakeholder analysis Field testing in five regions

Model for pro-poor policy analysis Actual beneficiaries Groups of potential beneficiaries Constraints Opportunity Policy change Key Stakeholders Trade-offs Constraints

Stakeholder analysis – Uganda Autonomous – DDA, COCTU NAGRC&DB Research – NARO, Universities, Int. Organizations NGOs – Panos, Oxfam, HPI, Send-a-Cow, Land O’ Lakes Private Sector – Services, Traders .... Government – Local Govt., Other Ministries, Bureaus Farmers’ Organizations – UNFU, Local Organizations MTIP Parliament Legislation Finance PEAP Policy UPPAP PMA MAAIF International – NEPAD-CAADP Donors (DfID, USAID, DANIDA, EC, etc...) Banks (WB, ADB)

EXTRAPOLATE

EXTRAPOLATE Identify development opportunity / commodity Identify beneficiary groups – determine livelihood status Determine constraints faced by different groups Determine outcomes – the measurable effects of relaxing constraints Estimate the impact of each outcome on the livelihood status of each beneficiary group Estimate the impact of policy changes with respect to each constraint

Ex-ante analysis of policy change

Ex-ante analysis of policy change

Field testing and development Uganda – small holder dairy Senegal – small ruminants (Tabaski feast) Peru – camelid fibre India (AP) – small ruminants Viet Nam – small holder pork

IPALP Vietnam Northern Mountain Region Social Accounting Matrix Calculable General Equilibrium (CGE) model Poverty (VLSS 1997/98 & VHLSS 2002) Two policy scenarios

CGE modeling Analytical tool to estimate impacts of economic policies on production structure, household income, poverty rates Aimed at policy makers and analysts Improve visibility for policy makers by identifying resource constraints, tradeoffs and adjustment challenges Elucidate complex direct and indirect linkages that determine the costs and benefits of policy implications

Schematic modeling facility Social Accounting Matrix Household Survey Data Policy Scenarios (tariffs, taxes, etc.) CGE model Sectoral Results (production, exports-imports) Household Results (income, expenditures, occupation) Poverty Mapping (incidence and inequality)

Northern Vietnam: CGE model Prototype dynamic CGE micro-macro model developed for Northern Mountain Region of Vietnam 14 rural provinces and aggregated urban households Two scenarios for the 2004-2010 period WTO accession Livestock productivity growth (livestock promotion policy) coupled with WTO accession

Northern Vietnam: preliminary results Scenario 1 (WTO only) Most benefits to urban populations where significant poverty reduction occurs (21%) Some gains propagated to the rural poor but to a much lesser degree (1%) in some provinces poverty even increases in the first two years. Scenario 2 (WTO plus production increase) Poverty reduction is less in urban areas (7%) more significant gains in rural provinces where poverty is reduced more than 30 percent Over 27% of population in the region escape poverty

Scenario 2 (WTO + production increase) 90 100 110 120 130 140 150 160 170 2004 2005 2006 2007 2008 2009 2010 Index (2004 = 100) Ha giang Cao bang Lao cai Bac can Lang son Tuyen quang Yen bai Thai nguyen Phu tho Bac giang Quang ninh Lai chau Son la Hoa binh Urban Households Dynamic of household income growth

4. Information dissemination PPLPI web site GLiPHA GeoNetwork

PPLPI web site Current static pages updated monthly Developing a news and events front page Developing an interactive resource searching facility based on EIMS

PPLPI web site

GLiPHA

Data Display Tables Graphs Text Maps Bar Line Pie Metadata Links Chloropleth Vector Point Raster

GeoNetwork better slide Claudia ? FAO’s portal for hosting and sharing georeferenced thematic information Provides a common metadata structure and search facility through which geographic (map) outputs can be searched for and downloaded by users Used to disseminate livestock population and production data

5. Monitoring and Evaluation Indicators - Framework Time-frame: Level Short 1-2 yrs Medium 5 yrs Long 10+ yrs PPLPI activities OVIs Livestock policy Policy documents Impact on PRSPs Legislation Investment Projects Programmes Livestock development Wealthier livestock keepers Contribution of livestock to GDP Rural development Rural poverty Contribution of agric. to GDP National development National poverty Trade GDP

Thank You