In order to implement an agricultural policy based on tools for rational decision making, a partial equilibrium model has been implemented on behalf of the authorities in charge of agriculture. The objective is to stimulate demand and supply of major agricultural products. The partial equilibrium model constructed is decomposed according to supply and demand of main agricultural commodities and is implemented with a series of statistical data on the period 1980-2008. The estimates were made using the two-stage least square method for solving simultaneous equations systems. Three equations constitute the model: an equation of agricultural supply, an equation of demand for agricultural products and an equation of equilibrium (the equilibrium condition states that supply equals demand).

The main variables involved are prices, quantities produced or demanded, exports, imports, historical yields, rainfall, exchange rates, growth rates, population, and other macroeconomic variables. To take account of expectations of economic agents (consumers, producers and state) lagged variables were considered: thus, the delayed price, quantity demanded delayed, the gross domestic product and investment have been used as instrumental variables. Under the assumption that consumers and producers are price takers in the various markets concerned, the equilibrium on each of these markets is obtained by making endogenous the price or the quantity supplied or demanded.

Different commodities have been modeled in three different fields: Animal field, vegetable and fishing industry sector and scenarios were explored. It comes out from these econometric estimates that the prices of the products play a key role as a principal tool of regulation of the various agricultural markets. Thus, for the prices, partial equilibrium, the quantities supplied and demanded are significantly influenced, which has a tangible impact on the surplus of each of the agents involved.

The food crisis occurred in 2008, following the surge in prices of essential commodities has highlighted the pivotal role of certain crop sectors (maize, rice). Indeed, the rise in prices is nothing but the result of an imbalance between supply and demand of agricultural products. The availability of a forecasting tool as this forecasting model allows therefore a rational decision making.

However, a good model requires reliable and current statistical data. But the statistics used in this model are not always complete, on time or data quality does not always meet expectations. It is therefore imperative to review the system of collection of agricultural data. Unfortunately, the difficulties encountered with the National Agricultural Census, a first in Benin, failed to have some relevant statistical information. A major effort must be made at the national system of agricultural statistics to make available to the team modeling the relevant and reliable data needed. This is why recommendations are made in this direction.

Keywords: Agriculture; partial equilibrium model; census and data quality; forecast prices

Biography: Mr Accrombessy is Statistician-Macroeconomist and Development Project Analyst born in Benin in 1971, Married with 4 daughters. He is the Chief of the Unit of Informatics and Statistics at the National Agency For Employment (ANPE) and has devloped expertise in statistical system coordination and harmonization, in implementation of indicators and databases, in analysis of socio-economic statistics, in monitoring-evaluation of the MDGs and poverty analysis. High skills in program/project leading and economic policy management.