Division Rationale and description of the econometric analysis

The econometric analysis undertaken by the Commission aims at creating benchmark prices for the contracts under investigation, which would result from a regression analysis carried out on contracts in the dataset, which are not under investigation, by using features of these contracts. In a first step, the regression analysis allows constructing a benchmark price as a function of the contract characteristics of the dataset (‘in-sample predictions’). In a second step, the results of the regression analysis are used to predict a benchmark price for the contracts investigated taking into account their contract characteristics (‘out-of-sample predictions’). In the regression analysis, the variation in prices among the contracts in the dataset is explained using the following characteristics: quantity purchased and year dummies^f1.

The econometric analysis relies on the principle that there are a number of drivers of prices, such as quantities. It would be misleading to compare the prices across different contracts without considering these drivers. The rationale of this quantitative exercise is, therefore that, once some external factors are taken into account, the prices across different contracts become more comparable. In the absence of normalisation, only perfectly identical contracts could be meaningfully compared.

This empirical exercise does not aim at estimating a causal relationship between prices and some external factors. For example, estimating a causal relationship between some factors and the prices would require dealing with the risk of endogeneity, that is, the risk that a causal variable (e.g. the quantities) is itself influenced by the explained variable (e.g. the price), due to omitted variables or simultaneity bias. The purpose of the quantitative exercise is to ‘normalise’ the prices across different contracts to make them more comparable to each other. This normalisation is necessary in the absence of perfectly identical contracts and supply characteristics.

The regression analysis captures the main features of the bilateral contracts at stake:

• including the quantity variable in the regression captures that prices are in general lower when quantities purchased are higher^f2,

• including year dummies captures the time dimension and the possible change in market conditions between different years.

As regards the first step of the empirical analysis, the results of the regression analysis on the dataset contracts are reported in Table 1 below.

As regards the second step of the empirical analysis, the Commission has identified a benchmark for every year and then tested the position of the contracts relative to this benchmark in order to conclude whether the prices charged by Hidroelectrica were lower or higher than the modelled benchmark price. The following steps describe in detail the methodology used to determine the benchmark:

• Firstly, for each contract under investigation, it is calculated if and how far the actual price every year departs from its corresponding benchmark price computed using the regression and the contract characteristics;

• Secondly, the most-upward-diverging contract (‘MUD’) is identified^f3; this is the dataset contract with the observed price which departs the most above its own corresponding benchmark price (in absolute terms). The choice of MUD which provides a range of variation above the central estimate of benchmark price, whilst being conservative, is justified; firstly, the econometric model does not explain 100 % of the observed price in the dataset and the price benchmark single estimate is provided within an interval of confidence and a margin of error above or below the estimate; secondly, price deviations from a single possible price exist on the real market; the MUD, which stems from market-based contracts (see recitals 42 to 45) brings quantified information about the possible extent of such deviations and provides a market-based range around the calculated benchmark price;

• Thirdly, the price difference from the MUD contract is used to separate observed prices above benchmark price from contracts below benchmark price:

  • if a contract has an observed price above its corresponding benchmark price and if the price difference of this contract is higher than the MUD price difference^f4, then this contract is considered prima facie not market conform,

  • otherwise the contract should be deemed to be market conform;

The table below presents detailed results from the regression analysis on the dataset. The regression explains 36 % of the variations in the data. The coefficient estimates presented in the table below are used in a second stage to predict the ‘benchmark’ price for the contracts under investigation (out-of-sample predictions), assuming they would also be retail contracts, like contracts in the dataset.