Within the framework of European Union (EU) Cohesion Policy, Latvia as an EU member state has received and will continue to receive substantial funds, aimed at increasing the production potential of its economy. The funds thus received were invested in various segments of the economy: fixed assets; infrastructure development; human capital; and technology. The evaluation of the impact of such investment is a complicated task for two main reasons. Such investment has both a long-term effect, which manifests itself in a greater productive capacity of the economy, and a short-term effect, which acts through the demand side of the economy and may benefit both the sector in which investment is made and other sectors. Additionally, investment in one sector may indirectly affect other sectors even in the long term. For example, the construction of a new road increases the output of the construction sector, which directly affects the demand for labour and the wage levels in that sector. At the same time, a better road network potentially increases productivity (including labour productivity) in all sectors of the economy, and it can therefore be expected that an indirect effect of better road infrastructure is, for example, higher overall wage levels, which would lead to more demand both for goods produced domestically and for goods imported.
When modelling the impact of funds investment, taking account of these mechanisms and of the interrelationships between the various segments of the economy in a consistent manner requires an appropriate economic model. There is a considerable literature on the evaluation of EU structural policies based on macroeconomic models (an overview of simulation models used in the ex-ante evaluation of EU funds is provided by Lolos (2011)). For Latvia, an evaluation of the impact of EU structural policy using macroeconomic modelling was first done in 2000, when the HERMIN model was developed for modelling the ex-ante effect of pre-accession EU funds (Bradley et al, 2000). In 2007-2008, the LATFUN model was developed for the ex-ante and ex-post evaluation of pre-accession funds and EU funds that Latvia received or expected to receive after 2004 (BICEPS, 2008a and 2008b).
The macroeconomic model developed within the framework of this project has a number of advantages compared with the earlier HERMIN and LATFUN models. First of all, the model equations are fully based on econometric estimations, without calibration of short-term parameters; this provides better model in-sample fit. Secondly, the sectorial interrelationships have been estimated econometrically and rather than using input-output tables as in the HERMIN and LATFUN models. Input-output tables are available with a very long time-lag and for an economy such as Latvia which has experienced substantial structural changes in the past few years, it seems more appropriate to use an approach that is based on an empirical estimation. Lastly, this model minimises the use of dummy variables as possible, even though the equation parameters were estimated on data for the period including the 2008 crisis. This increases the robustness of the model. We tested model robustness as part of the project itself: initially, we estimated equation parameters for the period up to the fourth quarter of 2010, but in the second stage of the project we re-estimated the equations for the period up to the first quarter of 2011, and this had no significant effect on model stability.
 In the current programming period Latvia receives funds under the Convergence objective of EU Cohesion Policy and in the post-2013 period will receive funding a less developed region i.e. as a result of having GDP per capita at less the 75% of the EU average.
Access the full report.