Intermittent Renewable Generation and Network Congestion: An Empirical Analysis of Italian Power Market.

Summary The literature demonstrates the likely reduction of wholesale electricity prices due to a larger penetration of renewable energy sources (RES). When markets are organized as two or more inter-connected sub-markets within a larger power market the final impact of increasing RES production may be less straightforward given the presence of network constraints. We tests this phenomenon by analyzing the impact of RES production on the probability of congestion and on the size of congestion cost in Italy. Using a database with hourly observations for a five year period we estimate two econometric models on five zonal pairings: a multinomial logit model for the occurrence and direction of congestion and a three stage least square model for the size of congestion costs. The analysis suggests that the e ffect of a larger local wind and solar supply is to decrease the probability of suff ering congestion in entry and to increase the probability of causing a congestion in exit compared to no congestion case. Increasing hydroelectric production has a similar eff ect. These results hold for both importing and exporting regions, but importing regions are less likely to cause congestion in exit, therefore the installation of new RES capacity in these zones may have a positive eff ects in terms of flow balance between regions. Concerning the cost level, a larger local RES supply seems to push the congestion cost towards negative values as it decreases the marginal cost for balancing the system. This is true for all zones in the case of explicit congestion cost, but it is only verified in importing regions in the case of implicit congestion cost. This result suggests that the increase of RES production should be promoted in importing zones, but the overall growth should be controlled in order to avoid congestion in the opposite direction.