Electricity Prices as Signals for the Evaluation of Reforms: An Empirical Analysis of Four European Countries

Abstract

This paper looks at price trends as signals for the evaluation of utility reforms. A specific example is considered: electricity prices in four countries, namely France, Germany, Italy and UK. These countries offer a natural experiment in different patterns of public/private ownership and liberalisation of electricity industry. Electricity prices are mainly influenced by the mix of energy inputs, their costs, and by consumption per capita. Under different institutional settings, prices for business users are often more cost‐reflective than prices for residential users. Beyond these common features, the evidence does not support the view that there is clear dominance of one industry pattern in terms of welfare change for the representative consumer. This conclusion tends to question the widely held idea that one specific ‘orthodox’ reform should be preferred: privatisation with liberalisation and vertical disintegration. Utility reforms should be flexible and country‐specific.

Key Words:

• Electricity industry
• tariffs
• public services
• utilities reforms

Jel Classification:

• L33
• L43
• L94

Notes

1. In 2001 the ‘Pool’ was replaced by the New Electricity Trading Arrangements (NETA). This ended the system of uniform pricing and introduced pay‐as‐bid pricing.

2. The various types were: standard tariff, day/night tariff, day/night controlled circuit tariff and tariffs with only a variable quota. The tariffs for free users varied according to whether the annual consumption was higher or lower than 60 MWh. If the consumption was lower than the stated threshold there were four tariffs (standard, day/night, evening and week‐end). If it was higher then there were seasonal maximum demand tariffs. In fact, these tariff arrangements varied by companies, and in some cases survived the exit from the market of the supplier. Subsequently flat rate tariffs have been introduced, which are independent of consumption, but based on the number of components of the household and the size of the home.

3. Following liberalisation there have been some consolidation of the industry: for example in 1998 PowerGen (a generating company) acquired EME (East Midlands Electricity), the third largest regional electricity company in England and Wales. In March 1999 National Power (a generating company) acquired the British Gas company Calortex, etc. By 2004 there was further consolidation in the residential retail market, with just six major suppliers. Many new entrants did not survive more than a few years. As we have already mentioned, the panorama is in continuous evolution and an update on the structural and regulatory scenario can be found on the website of the new sector authority, the Gas and Electricity Markets Authority (www.ofgem.gov.uk).

4. However there is some public concern about the hidden cost of disposal of nuclear waste.

5. Tariffs are divided according to the power supplied: blue tariffs (from 3 to 36 kW); yellow tariffs (from 36 to 250 kW); green tariffs (over 250 kW). All tariffs also comprise a fixed quota (which depends on the power installed) and a variable amount (consumption and period of use during the day and during the year).

6. In Italy the quota of conventional fossil fuels remained stable at around 80% and, with the absence of nuclear plants, thus so also did the share of hydroelectric and renewable energy sources (20%). There was a sizeable readjustment in Italy when the nuclear plants were abandoned: between 1986 and 1989, in addition to an increase of 10% in total production of electricity, there was an increase in the quota of fossil fuels (from 69 to 80%) and a reduction in the hydroelectric (from 25 to 20%).

7. It was also decided to unify distribution in the urban areas, in cases where the presence of a municipalised company that had survived nationalisation had created dualistic solutions (as in the case of AEM in Milan).

8. In addition to Enel, the major suppliers who already, or soon will, operate in Italy include: AEM Milano, AEM Torino, Dalmine Energy, Edison, Enipower, Energy, Merloni Progetto Energy, Sondel, Verbund, ENBW and EDF. Documents updating the situation in Italy can be found on the authority’s website (www.autorita.energy.it). These four natural examples are far from being representative of the electricity industry across Europe. Particularly, the Scandinavian electricity market has distinctive features (see Amundsen et al. 1998).

9. The data source is the Annual Energy Review by the European Commission. The data refer to the years 1985, 1988, 1990 and from 1992 to 1997. In order to obtain the missing data we carried out a simple linear interpolation between the observations available immediately before and after.

10. This was probably due to special protection offered by law to some independent generators, using non‐conventional energy sources, including gas for the co‐generation.

11. The data referring to the weight of the inputs were taken from the Annual Energy Review (1999). Values are available for the years 1985, 1988, 1990 and from 1995 to 1997. The missing data were obtained by a linear interpolation of the closest available data.

12. Data were not available in a comparable way for coal prices in Italy.

13. The analysis carried out so far can be subjected to two criticisms. The first refers to the sample, which, from the point of view of a correlation analysis, could be scarcely representative. The second, to the possibility that the use of linear interpolation for the unavailable data artificially increases the value of the correlation indices or, in any case, invalidates the results by altering the relative position of the productive inputs in terms of capacity to explain the performance of electricity prices. To answer the first question, one would need a larger data set. Steiner (2000) has similar data limitations. We hope in future to be able to build a more comprehensive data set by a new international project recently approved by the European Commission under the 6th Framework Program (http://www.feem.it/Feem/Pub/Programmes/Privatisation+Regulation+Antitrust/2006 02-UPP.html) (last accessed 8 November 2006). With regard to the second question posed above, we have repeated the previous calculations with the unavailable data calculated as missing values (thus eliminating the linear interpolation) with some minor differences in results. We find that, for domestic users, there was no change in the relative positions of the different productive factors in any country. It should be noted, however, that there was a net relative improvement in both the UK and Germany in the correlation between the weighted average price of input and the price of electricity. In fact, in these countries this value jumps from third and second positions, respectively, to first in the comparison of single productive factors. We may conclude that the interpolation of the missing data did not invalidate the previous conclusions. Details are available with the author.

14. The data are those from the OECD International Regulation Database also used in Steiner (2000) and which can be found on the Internet at http://www.oecd.org/document/1/0,2340,en_2649_34323_2367297_1_1_1_1,00.html (last accessed 8 November 2006). For a more detailed description of them see Steiner (2000, section 3.3).

15. We note that in one country (Italy) the quantity of electricity produced by nuclear plants over the period studied is zero and that it is therefore opportune to divide the study into two parts: with and without Italy, according to whether or not we include the quantity of electricity generated by nuclear plants.

16. A complication arises in the long run when we have demand shifts, e.g. because of exogenous income change or because changes of consumers’ preferences. We ignore this point here.