CAPRI

This page is a knowledge crystal of subtype method. The page identifier is Op_en3164
Moderator:Heta (see all)
Upload data

CAPRI(Common Agricultural Policy Regional Impact Assessment) is an economic modelling system developed to assist in analysing impacts of the Common Agricultural Policy (CAP), of European agri-environmental measures or of agricultural trade policies from regional to global scale. It combines the representation of regional agricultural policies and production with political and economic conditions on regional, Member State, EU and global agricultural commodity markets. An activity based approach allows the direct implementation of relevant CAP policy measures and the technological definition of appropriate environmental indicators related to the agricultural production activities. Product and activity coverage are in line with the Economic Accounts for Agriculture (EAA) by a combined top down / bottom up approach.

The model is split up into a supply and a market component. An iterative process between the supply and the market component establishes a comparative static equilibrium and returns market clearing prices for tradable agricultural products and young animals.

Result

Typical Model Applications:

The CAPRI model is typically used for simulating EU- 25 wide impacts of the Common Agricultural Policy (CAP) on production, income, markets, and the environment. Typical Applications have been

• the simulation of Agenda 2000 and MTR scenarios
• modelling agricultural reform under new WTO proposals
• sugar and dairy market reform options
• developing / modelling of passive environmental indicators for the European Agriculture
• assessing environmental impact of agricultural policies
• impact of agricultural policies on specific markets
• tradable permits for Global Warming emissions from agricultural
• effects of a compulsory insurance scheme for food and mouth disease

Standard Model Specification:

A setup which could combine regionalisation with complete coverage of the EU-agricultural sector was chosen in order to simultaneously analyse the effect of commodity market and policy developments on agriculture in the individual regions as well as the feedback from the regions to EU and world markets. The model system is conceptually split up into a supply and a market component. The supply module consists of individual programming models for about 200 NUTS II regions or about 1000 farm types consistent with NUTS II statistics, where the farm type definition is derived from the Farm Accounting Data Network. The market module follows the tradition of spatial multi-commodity models. Based on aggregated supply quantities from the regional models, the market model returns market clearing prices. A specific module deals with clearing markets for young animals. An iterative process between the supply and market components ultimately achieves a comparative static equilibrium.

Supply module

The supply module consists of independent aggregate regional programming models which represent a high degree of activity differentiation and relevant farm policy measures (e.g. premiums, setaside obligations) and ensure simulation results consistent with general resource constraints. The objective functions maximise the aggregated gross value added including CAP premiums minus a quadratic cost function related to Positive Mathematical Programming (PMP). The choice of the optimal production mix is restricted by a relative small number of constraints:

• availability of arable and permanent grass land,
• selling quotas for milk and sugarbeets,
• set-aside obligations,
• base area related premium reductions,
• and upper bounds for voluntary set-aside according to CAP regulations.

Feed costs are minimised endogenously by determining the optimal mix of a limited number of aggregated marketable (e.g. "cereals") and non-tradable feedingstuffs (e.g. "hay") subject to requirement constraints, ensuring a technologically plausible mix. Nutrient requirements of crops can be covered either by mineral or organic fertilisers, the latter restricted to the amount produced by the regional herds. Constraints ensure that a crop specific percentage of the nutrient need is covered by mineral fertiliser.

Since elasticities were not available for the differentiated set of crop production activities, the team explored the possibility to estimate multi-output quadratic cost functions based on a cross-sectional sample. The supply part includes the coupled and decoupled premiums schemes of the CAP in a very rich detail, as closely as possible to current legislation.

Market module

The market module is a spatial multi-commodity models and covers about 35 primary (different cereals and oilseeds, pulses, potatoes, textile crops, tomatoes and other vegetables, apples/pears/peaches, citrus, other fruits, olive oil, pig meat, beef meat, poultry meat, sheep and goat meat, eggs) and processed agricultural products (different oils and cakes, different dairy products). Flexible functional forms (normalised quadratic profit functions for supply, feed and processing; Generalised Leontief Expenditures functions for human consumption) are applied for the behavioural equations with clear regional markets, where the Armington assumptions determines the composition of demand from domestic sales and imports from different trading partners. The parameters for these functions are calibrated to elasticities found in literature.

The model represents bi-lateral specific and ad valorem tariffs, where necessary depending on the Tariff Trade Quota mechanism. For the EU, market interventions and subsidised exports are modelled endogenously depending on market and administrative prices.

Processing of oilseeds is modelled explicitly assuming fixed extraction rates for cakes and oils from crushing. In the case of dairy products, constraints equilibrate fat and protein content of processed quantities of raw milk and with the processed products. The price of raw milk and processed milk products is derived from fat and protein prices weighted with their contents plus a processing margin.

The model distinguishes between producer and consumer prices, where the margin between the two is endogenously determined for dairy products, oils and cakes, and fixed in case of all other products.

Main Model Results:

Regional and aggregate impacts of the CAP on production, income, markets including trade flows and the environment (detailed nutrient balances for N,P,K; climate relevant gas emission from agriculture according to IPCC guideline; waster scarcity indicator)

Welfare analyse at EU25 member state level (agricultural income, profit of dairies and oil seed crushing industry, money metric for consumer, budget impact), and for the different countries/country blocks of the market model

Time Horizon

CAPRI is a comparative static modelling system for medium term analysis (about 10 years), a recursive-dynamic version is under development.

Required technical infrastructure:

PC: Specific software including GAMS. For the reporting tools, which are accessible via the internet, JAVA is needed (interactive WB based maps) and a XLST/XLM transformation tool (interactive report tables).

Exogenous variables and parameters:

A lot of necessary parameters are taken from the CAPRI data base which was developed together with the model.

Data Sources:

The data rely on different domains from EUROSTAT including Economic Accounts of Agriculture (EAA), the REGIO domain of EUROSTAT, FADN for input cost estimations and FAOSTAT (Supply Utilization Accounts, trade flows).

Model Extensions:

A farm type module was developed for DG-ENV some years back. An update and extension of the farm type module is planned in the context of the "Integrated projects SEAMLESS" from 2005-2008.

In the context of the CAPRI-Dynaspat project (2004-2007), the recursive dynamic version is in development, together with a link to GIS (25000 homogenous mapping units at EU25) which allows linkage to biophysical models and landscape indicators. The same project provides the expansion from EU15 to EU10 (already operational) and an energy indicator for agricultural activities.

There was a model extension splitting the supply side into conventional and biological production, which was however not updated since 2000. A further extension introduced stochastic simulations of yields for major crops, again not updated for some years.

Links to other Models, Projects, Networks:

SEAMLESS, SENSOR, CAFE, RAINS

Regional Scope:

Supply Model: EU-25, Norway: Production activities, I/O coefficients, and political variables on NUTS 2 level (about 300 administrative regions for EU25). Agricultural activity levels, yields and certain inputs will be mapped down to 25.000 homogenous mapping units from 2005 onwards. The regional scope and detail are to a large extent determined by the availability of EUROSTAT statistical data as comprised in the CAPRI database.

Market model: Globally closed, about 20 regions (USA, Canada, Mexico, China, India, Japan, Algeria, Morocco, Tunisia, Egypt, Turkey, Israel, Bolivia, Chile, Argentina, Brazilia, Uruguay, Paraguay, Rest of South America, LDCs, Non LDC ACP countries, Rest of World)

See also

• IA TOOLS
• More information

References

Frank van Tongeren: Brief Review of the CAPRI modelling system (pfd-file) An external review of the system, status spring 2004 (see: [2])

Perez I. and Wieck C. (2004) Welfare distribution between EU member states through different national decoupling options - Implications for Spain, Paper to be presented at the V Congreso Nacional de Economía Agraria, Santiago de Compostela (Spain), September 15-17, 2004

Adenaeuer, M.; Louhichi, K.; de Frahan, B.H., Witzke, H.P. (2004). Impacts of the "Everything but Arms" initiative on the EU sugar sub-sector. Selected paper prepared for presentation at the International Conference on Policy Modelling (EcoMod2004), June 30 - July 2, Paris

Bäckstrand, A.; Jansson, T.; Rabinowicz, E. (2003). Effekter av enhetspremie på jordbruket i EU och Sverige (Effects of standard subsidy for the agriculture in EU and Sweden). Sveriges Utsädesförenings Tidskrift (Journal of the Swedish Seed Association) 1: 44-49.

Britz W., Perez I., Wieck C. (2003): Mid-Term Review Proposal Impact Analysis with the CAPRI Modelling System, In: Mid-Term Review of the Common Agricultural Policy. July 2002 Proposals Impact Analyses, European Commission, Directorate-General for Agriculture ed., pp. 111-140, Brussels (pdf)

Wieck, C. Perez, I. and Britz W. (2003): Expectativas a Medio Plazo del Mercado Lácteo Europeo tras las Propuestas de Agenda 2000 y Revisión Intermedia - Un Análisis cuantitativo y regional del Sector Lácteo Español con el Modelo CAPRI, AEEA Seminar (3/13/2003)

Wieck, C., Perez, I. and Britz.W: (2003) New challenges for the European Agriculture: Modelling agricultural reform under the new WTO proposals, Conference: "Agricultural policy reform and the WTO: where are we heading?", CAPRI, Italy, June 2003

Perez, I. and Britz W.: (2003) Modelling of passive environmental indicators for the European Agriculture: the role of marginal abatement costs, 12th Annual Conference of the EAERE, Bilbao (Spain), June 28-30, 2003

Britz W., Wieck, C., Perez, I. and Jansson, T. (2003): Auswirkungen der MTR-Beschlüsse auf die Landnutzung in den europäischen Regionen und Betrieben - Eine quantitative Analyse mit CAPRI Modellsystem, Gewisola Tagung Hohenheim 2003 , Hohenheim, September 2003

Perez, I. and Britz W.: (2003) Europaweite Reduktion des Ausstoßes klimarelevanter Emissionen durch handelbare Emissionsrechte - eine Analyse mit dem regionalisierten Agrarsektormodell CAPRI, Gewisola Tagung Hohenheim 2003, September 2003

Jansson, T. (2002). Analys av enhetliga arealstöd i EU. SLI Skrift 1, Swedish Institute for Food and Agricultural Economics. Lund (Sweden): SLI. ISSN 1651-4742 Download or order from www.sli.lu.se/publ.asp

Jansson, T. (2002). Mjölkproduktion utan gränser - Europas bönder på en avreglerad mjölkmarknad. SLI Rapport 5, Swedish Institute for Food and Agricultural Economics. Lund (Sweden): SLI. ISSN 1650-0105 Download or order from www.sli.lu.se/publ.asp

Jansson, T.L and Britz W.: (2002) Experiences of using a quadratic programming model to simulate removal of milk quotas. 10th EAAE congress, organised session on milk quotas, August 28 - 31, 2002, Zaragoza (Spain).

Wieck C., García Alvarez-Coque J.M. (2000): Análisis de la Convergencia de las Rentas Agrarias Regionales mediante los Sistemas SPEL y CAPRI. Convenio de Colaboracíon entre el Ministerio de Agricultura Espanola y la Universidad Politécnica de Valencia, Valencia (Spain)

Heckelei T. and Britz W.: (2000) Positive Mathematical Programming with Multiple Data Points: A Cross-Sectional Estimation Procedure in Cahiers d'Economie et Sociologie Rurales Nr 57, 4e trimestre 2000

Britz W. and Heckelei T. (2000): Effekte von Agenda 2000 auf den deutschen und französischen Agrarsektor - Anwendung des regionalisierten Modellsystems CAPRI Paper presented at "Gemeinsame Tagung von SFER und Gewisola, Strasbourg, 11-12.10.2000"

Heckelei T. and Britz W. (2000): A Cross Sectional Maximum Entropy Estimation of Cost Functions for Regional Programming Paper presented at "XXIV. IAAE Conference, August 13-18, 2000, Berlin"

Britz W. and Heckelei T. (2000): Concept and Explorative Application of an EU-wide, Regional Agricultural Sector Model (CAPRI-Project) Paper presented at "65th EAAE seminar, Agricultural Sector Modelling and Policy Information Systems, Bonn (Germany), March, 29-31, 2000"