1. Introduction
The Structural Fund (SF) interventions play a crucial role in improving the social and
economic cohesion of the EU. A particular focus of the structural funds is on those
regions that lag behind to the extent that their GDP per capita is below 75 per cent of
the EU average. These regions are classified as Objective 1 and make up a significant
part of the EU. In 1999 these regions accounted for 25 per cent of total EU
population, and in general they are poorly endowed in a number of areas, such as
infrastructure, human capital, and modern high productivity industries and services.
As a consequence, they tend to have higher rates of unemployment.
The amount of investment that is funded though the structural funds by the EU is
substantial. For the Objective 1 regions for the period 1994 to 1999, this amounted to
some €103 billion, which was allocated to investment in 11 separate EU Member
States. Given the size and significance of the EU aid package, legislation in the form
of the Council Regulation No. 1260 of 26.06.99 requires the appraisal of the structural
funds as well as a regular reporting on the economic and social cohesion in the EU.
However, while systematic monitoring and evaluation frameworks are available at the
national level, a rigorous and systematic method for quantifying the socio-economic
impacts of structural fund interventions on the regional economies has not been
developed to the same extent. One problem at the regional level is that policy-makers
seldom have access to accumulated research on the macroeconomic and macro-
sectoral performance at a regional (NUTS II) level, which would allow them to assess
the overall impact of the structural funds.
Furthermore the estimation of the long-run impact of the Structural Funds is more
important than the estimation of their simple Keynesian demand side impact, since the
Structural Funds aim at changing the economic potential of a region over the long run
rather than to provide a short run cash injection. This limits the number of potential
methodologies since some are not capable of capturing these long-run effects.
Another important limiting factor is that one model does not fit all regions. In other
words even the application of a common modelling framework, which is desirable in
order to yield comparable results requires that the models should be adapted to each