This paper explores the capacity of computable general equilibrium (CGE) models to track down policy induced economic changes and their ability to generate contrastable data for an economy. Starting from an empirically built regional Social Accounting Matrix (SAM), a first stage CGE calibrated model is constructed. The model is then perturbed with a set of policy shocks related to European Union Structural Funds 2000-2005 invested into the region of Andalusia in the south of Spain. The counterfactual equilibrium is translated into a virtual SAM, conformal with the initial one, which is in turn reused to calibrate the next stage in the CGE modeling. And so on until the last stage is reached and all European funds yearly invested have been absorbed by the economy. Since at the end of the process another empirical SAM is available, it can be compared with the terminally produced virtual SAM. The comparison shows the sequence of SAMs to provide a very good fit to the actual data in the empirical SAM. Regional GDP and unemployment rates are two examples of the close approximation. With this novel approach, the projection capabilities of CGE modeling are evaluated from the methodological viewpoint, and at the same time an empirical assessment of the said European policies is provided.
