A cluster analysis study based on profitability and financial indicators in the Italian gas retail market

ARTICLE IN PRESS Energy Policy 38 (2010) 3394–3402

Contents lists available at ScienceDirect

Energy Policy journal homepage: www.elsevier.com/locate/enpol

A cluster analysis study based on profitability and financial indicators in the Italian gas retail market Guendalina Capece a, Livio Cricelli b, Francesca Di Pillo a,n, Nathan Levialdi a a b

Dipartimento di Ingegneria dell’Impresa, Universita di Roma ‘‘Tor Vergata’’, Via del Politecnico 1, 00133 Roma, Italy Dipartimento di Meccanica, Strutture, Ambiente e Territorio, Universita di Cassino, Via G. Di Biasio 43, 03043 Cassino (FR), Italy

a r t i c l e in f o

a b s t r a c t

Article history: Received 9 November 2009 Accepted 4 February 2010 Available online 20 February 2010

In the European Union, the natural gas market is increasingly being liberalized. The liberalization process is aimed at leading to lower prices and higher volumes, and hence higher consumer welfare. This paper focuses on the changes in performance in the natural gas retail market by analyzing the profit and financial position of the companies concerned over the first three years following the market liberalization. The balance sheets of 105 Italian companies in this sector are analyzed, after which a cluster analysis is performed employing the most significant performance indexes. The companies are then analyzed within each cluster with respect to age, size, geographical location and business diversification. The results of our analysis show that the majority of companies attained a high level of performance, although this positive outcome was mitigated by the gradual decrease of the average values of performance indicators during the period concerned. The companies that achieve the best performances belong to longstanding business groups, are medium-large sized and are located in the north of the country. Regarding business diversification, in the first two years, the specialised companies outperformed the diversified companies. & 2010 Elsevier Ltd. All rights reserved.

Keywords: Cluster analysis Company performance Italian gas market liberalization

1. Introduction The demand for natural gas, as a replacement for more expensive, less environmentally friendly resources, has significantly increased in Europe over the last 15 years (Reymond, 2007). The growth in demand and the contemporary process of liberalization have given rise to the need for regulation to protect the consumers and develop the competition. With the objective of creating a fully operational internal gas market, in which fair competition prevails, the European Community decided to regulate the natural gas market. It was in this regard that on 22 June 1998 the European Parliament and the European Council, by means of Directive 98/30/CE (European Parliament and Council, 1998), began the liberalization of the European natural gas market, which had previously been characterized by vertical integration and public monopoly (Guldmann, 1985). This process continued in 2003 with the Gas Directive 2003/55/EC (European Parliament and Council, 2003) and it ended in 2009 with the Gas Directive 2009/73/EC (European Parliament and Council, 2009), which is the European gas legislation in force at present.

n

Corresponding author. Tel.: + 39 06 72597802; fax: + 39 06 72597951. E-mail address: [email protected] (F. Di Pillo).

0301-4215/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.enpol.2010.02.013

The European Gas directive proposes three innovative aspects: 1. The unbundling of transport and trading: i.e. all gas operators belonging to the member states must separate their gas transportation and trading functions into separate companies. 2. The regulated TPA (third party access): potential network users must be able to request and obtain access to gas facilities under transparent, objective and non-discriminatory conditions. 3. The concept of ‘eligible customers’, which is a category of customers who have the freedom to choose their supplier and the right to acquire gas at a competitive price. The European Gas Directive was transposed into Italian law by Legislative Decree no. 164/2000, known as the Letta Decree (Ministry of Economic Development, 2000), which laid down important guidelines concerning the definition of the eligible customers, competition, and conditions of reciprocity. The decree aimed to liberalize the internal market of natural gas and to design an organic framework of the reform of the entire sector, which was under the monopoly of the Eni group. One of the most significant changes brought about by this legislation was that the Letta Decree imposed the unbundling of the distribution companies from those in retail, and thus allowed the latter to operate in a more competitive market. The system is thus divided between companies that have the raw material (producers, importers,

ARTICLE IN PRESS G. Capece et al. / Energy Policy 38 (2010) 3394–3402

wholesalers and retailers) and companies that provide infrastructure and services to the system (transporters, distributors, LNG plant operators and storage). Following the liberalization process, the Italian competitive context has radically changed and companies have implemented new strategies to maintain their market share and winning new customers. Considering these profound changes, it is interesting to analyze the company performances to verify whether the implemented strategies have led to positive results in economic and financial aspects. Extensive research has previously been conducted in order to analyze one specific aspect of performance in the natural gas market, such as the efficiency or the productivity. Examples of such studies are Aivazian et al. (1987), Granderson and Linvill (1996), Herbert and Kreil (1996), Jamasb et al. (2008), Murry and Zhu (2008) and Sickles and Streitwieser (1998), who explore the natural gas industry in the USA; Carrington et al. (2002) and Rushdi (1994) discuss the Australian gas utilities; Price and Weyman-Jones (1996) who study the United Kingdom’s natural gas distribution sector; and Lee et al. (1999) estimate the total factor productivity based on an international comparison. Similar researches were also carried out in other utility sectors such as, for example, telecommunications and electricity (Farsi and Filippini, 2009; Gorini de Oliveira and Tolmasquim, 2004; Granderson, 2006; Yu et al., 2004; Wang et al., 2007). To date, very little research regarding the natural gas market has been carried out in relation to the simultaneous effects of the various aspects of performance, such as financial, liquidity, and profitability indicators. An evaluation of the gas sector with regard to economic performance was carried out by Kim et al. (1999), although their work concentrates on the transportation segment. In connection with the Italian natural gas market, an analysis regarding the consequences of the introduction of competition was performed by Dorigoni and Portatadino (2009), whilst a performance analysis was carried out by Erbetta and Fraquelli (2003). These studies both focused on the distribution segment. In contrast, the present paper concentrates on the retail segment, which, having undergone major transformations in recent years, is the only sector to be opened to free market competition and is therefore deemed more suitable for a comparative analysis of competitiveness. The aim of the analysis is to answer the following research questions:

 What are the performances of the gas retail companies in  

terms of profit and financial results in the period following the market liberalization? Is there a relationship between company performances and their main characteristics, such as age, size, geographical location and degree of business diversification? If this relationship exists, does it remain constant or differ during the period covered?

To answer these research questions, the methodology employed here is divided into two consecutive stages: 1. the evaluation of the main financial, liquidity, and profitability indicators; 2. the application of a cluster analysis utilising the indicators from stage 1.

3395

This research paper is organized as follows: Section 2 provides a description of the Italian natural gas context; Section 3 describes the methodology specifying the data set and the cluster analysis; Section 4 illustrates the composition of the clusters; Section 5 presents an analysis of the results with respect to the following characteristics of the companies: size, geographical position, age, and business diversification; Section 6 describes the impact of retail tariff regulation on profitability of companies; and Section 7 concludes.

2. The context In Italy, in order to ensure the protection and development of competition, the legislature (Legislative Decree no. 164/2000) has imposed the following constraints:

 The rules on anticompetitive practices, abuse of dominant





position and economic concentrations of Law no. 287 of October 10th, 1990 (Italian Parliament, 1990) are applied to natural gas companies. From January 2003 to December 2010, a single operator cannot sell, directly or through subsidiaries, parent companies or companies controlled by the same parent company, to end customers over 50% of national natural gas consumption on an annual basis. From January 2002 to December 2010, a single operator cannot put gas (imported or produced in Italy) into the national pipeline, for sale into Italy, directly or through subsidiaries, parent companies or companies controlled by the same parent company, for quantities over 75% of national natural gas consumption on an annual basis. That rate is reduced by two percentage points for each subsequent year until 2002 to reach 61%.

The last two conditions were imposed to limit the monopoly position that Eni had before the competition opening in the supply sector. Despite market share limits and the growth in the number of importers after liberalization, the market for gas imports still remains highly concentrated (Cavaliere, 2007). Indeed, in terms of supply, production is virtually all under the ownership of the Eni group, that selling part of its gas abroad is able to overcome the national market shares limits. The supply situation is reflected in the retail market, since incumbent firm supplies gas to its competitors for resale in the domestic market, exploiting its access to cheap gas abroad and imposing a mark-up that reduces the profit margins of new entrants. In light of the competitive weaknesses in the upstream supply phase, it is necessary to adopt certain measures to encourage the entry of new operators on competitive conditions, such as the creation of new gas supply infrastructures and the development of a sufficiently liquid gas market. As concerns the sale to end customers, the Italian liberalization law has imposed the legal unbundling between the distribution system operator and the retail company. Furthermore, from January 2003, the retailers must obtain the permission by the Ministry of Economic Development. This permission is required to ensure the fulfillment of the following requirements:

 ownership of technical skills and financial resources relevant to market share;

The latter allows the sample to be subdivided into homogeneous groups, which are then analyzed in relation to the main characteristics of the company.

 guarantees about the natural gas origin and reliability of the transmission system;

 availability of a modulation service (tailored to the needs of

ARTICLE IN PRESS 3396

G. Capece et al. / Energy Policy 38 (2010) 3394–3402

supplies) and of a gas storage capacity, in national territory. In 2006, the companies authorized by the Ministry of Economic Development to sell gas in the retail market were 414, of which approximately 53% remained inactive. The number of firms on the retail market has considerably decreased in recent years. Indeed, until 2000, this sector had a very rugged structure, characterized by a predominance of firms located in a municipal area. Between 2000 and 2003, a process of agreements, alliances and acquisitions began among retailer companies, affecting, in particular, a large number of former municipal firms. Therefore, an intense process of industrial concentration has been observed, which had as a consequence the gradual reduction in the number of operators, from over 700 to just over 400. More generally, in the latest years, the whole energy sector has shown an increasing activism on financial markets, and companies concluded M&As with the objective to merge gas and electricity businesses (convergence mergers) (Verde, 2008). These operations are explained by several reasons, including the creation of an entity with sufficient bargaining power to agree to favourable terms with strong European suppliers and competitors and the re-establishment of some cost-savings resulting from the exploitation of the economies of scope. Despite this intense process of concentration, a series of critical observations has been made both by the Antitrust Authority (Autorita Garante della Concorrenza e del Mercato or AGCM) and by the Regulatory Authority for Electricity and Gas (Autorita per l’Energia Elettrica e il Gas or AEEG) in relation to the Italian gas market (AGCM and AEEG, 2005):

 the development of the competitive context stems from an



acquisition of market share through external growth operations and not winning new customers as a result of competitive actions. The concentration and merging processes are more oriented to the creation of local monopolies, even on provincial or regional scale, and they seem more like the outcome of the unbundling obligation, rather than the development of competition in the retail market; it should be noted that these concentration processes have affected mainly the northern and central regions of Italy, while in the south, the market is still largely characterized by small to medium-sized companies that preside over specific areas.

Given this critical context, it is desirable to continue the concentration process among the retail companies. A further reorganization of the sector would be appropriate to ensure efficiency gains in the retail market and to benefit from possible economies of scale and scope, with a consequent positive impact on final prices. This outcome can be achieved only if there are real possibilities for expansion of market share and availability of raw material that can overcome the logic of market segmentation.

3. Methodology In this section, we explain the methodology used in order to answer the research questions of our paper. More in detail, we specify the data set and we describe the technique and the variables employed for the cluster analysis.

3.1. Data set The data set, supplied by the Unione Italiana delle Camere di Commercio (Italian Union of the Chambers of Commerce),

comprises data relating to 105 companies operating in Italy, including the balance sheet for each company. The present analysis refers to the three-year period from January 2004 to December 2006. Most of the retail companies in the sample were formed in 2003 and were consequently still in the start-up phase during the three-year period considered here. Thus it was not possible to analyze historical data dating back prior to 2004. On 1st January 2006 there was a total of 414 companies authorized by the Ministry of Economic Development to practice in the retail market. However, according to research undertaken by the AEEG (2007), only 194 of these companies appeared to be active. Data regarding the three-year period considered here were available for 105 of the active companies. Therefore, these 105 companies became our sample representing a significant proportion of the total number of retail companies operating in Italy (approximately 54%). Regarding the geographical distribution, 60% of the companies are located in northern Italy, 27% in central Italy and the remaining 13% in southern Italy. Most of the companies taken into consideration (75%) offer their services at the local level, 12% of the firms carry out their activities in a regional area and the remaining 13% are national. In relation to company size, the analyzed companies have been divided into three groups according to their revenue: small, medium and large sized companies. ‘Small’ companies are considered to be those with an income of less than h10 million; ‘medium’ are those with an income of between h10 million and h50 million; and ‘large’ are those that have an income greater than h50 million. Our sample comprises 48 small companies (46%), 33 medium companies (31%) and 24 large companies (23%). Approximately 78% of the firms in the sample are members of longstanding business groups, while the remaining 22% is made up of newly formed companies. Finally, with regard to the choice of strategy, whether to diversify or to specialize, 75 firms (71%) are single-business companies (specialized), while 30 (29%) are multi-business ones (diversified).

3.2. Cluster analysis In order to analyze company performance, cluster analysis is applied (Anderberg, 1973; Everitt et al., 2001; Kaufman and Rousseeuw, 1990; Tryon and Bailey, 1970). This analysis is intended to create groups that have the maximum cohesion internally and the maximum separation externally. Among the various clusterisation methods, the technique employed by Ward (1963) was chosen as it generates a classification hierarchy while minimizing the variance within each of the groups. The cluster analysis is carried out using SPSS (statistical package for the social sciences): this allows subjects (the companies) to be grouped in relation to the probability of certain values being scored by a set of variables (grouping variables). The grouping variables used in this study are the following profitability indicators: return on equity (ROE), return on investment (ROI), return on sale (ROS), and the following liquidity and financial indicators: cash flow (CF) and leverage ratio (LR). The latter provides a view of the company’s overall debt situation and it is calculated by dividing net debt to equity. The above indicators have been chosen since they are the most important and most utilized for profitability, liquidity and financial analyses. Before performing the analysis we verified that the indicators have a low level of inter-correlation, with a correlation coefficient of less than 0.5.

ARTICLE IN PRESS G. Capece et al. / Energy Policy 38 (2010) 3394–3402

3397

Table 1 Performance indicator ranges. Range

Performance

1 2 3 4

Poor Mediocre Good Excellent

ROI

ROE

ROS

CF

LR

xo0 0r xo 8 8r xo 10 x Z10

xo 0 0r xo 5 5 r xo 8 x Z8

xo0 0r xo 6 6 rx o 8 xZ 8

x o72 000 72 000 r xo 345 000 345 000r xo 945 000 xZ 945 000

x 43 2o x r3 1o x r2 xr 1

Table 1 shows the value ranges chosen for the various performance indicators utilized in the cluster analysis. The statistical analysis is performed for each year of the threeyear period (2004–2006). The agglomerative hierarchical algorithm starts with n clusters, where n is the number of observations. The Euclidean distance between observations is calculated and the two closest points are merged into a single cluster. The process is repeated until all observations are included in one cluster. Four clusters emerge from the statistical analysis (as seen in the 2006 resulting dendogram in Fig. 1) in which cluster I contains the companies with the worst performance and cluster IV those with the best results. The results of the Ward method are compared to those of a cluster solution produced by K-means method (McQueen, 1967), which are shown to be similar confirming their validity.

Vice versa, a smaller number of companies shows a good result, going from worst to best clusters. Analyzing more in detail the financial statements of these firms it could be observed that the good performance is due to careful management, which has focused particularly on the following points:

 attention and continuous monitoring of the market that has led to   

an increase in companies’ commercial offerings, thereby meeting the multiple needs of clients and ensuring high profitability; control over procurement policies; monitoring of structural costs, decreasing the incidence proportion of personnel and management costs; development of sales, which registered a significant increase in terms of volume and range of choice.

4. Description of clusters During the three-year period, the average values of all indicators worsen except the cash flow pattern, which can be considered constant, as shown in Fig. 2. The reasons for these negative results can be inferred from the analysis of financial statements and other corporate documents. It may be noted that there has been a general worsening of the financial situation due to increased borrowing. Also the income situation has deteriorated due to the combined effect of the reduction in unit sales caused by the mild heating season and the fall in the margins deriving from the new economic conditions of supply imposed by AEEG (2004a). From the analysis of Fig. 3, which represents the average values of all indicators for each cluster and each year, it can be observed how the firms belonging to the first cluster obtain the worst performance, while the companies grouped in the fourth cluster obtain the better performance. Indeed, analyzing the average level of single indicators, we can highlight how companies achieve excellent results in the fourth cluster, in relation to the ordinary operations and equity and with regard to financial position. Moreover, they have a good performance in the profitability of sales. In the second cluster, firms’ performances are mediocre in terms of return on investment, on sales and on equity. In the third cluster, the company performances improve, reaching levels on average good. More generally, it can be underlined that the mean values of each cluster worsen from 2004 to 2005, whilst they remain fairly constant from 2005 to 2006, except for the third cluster in which the results decrease. A summary of the distribution of the clusters is shown in Table 2. Most of the companies attain a high level of performance and are found in clusters III and IV throughout the three-year period. However, this positive performance is attenuated by the progressive reduction in the average values of the indicators. On analyzing Table 2, we can highlight how the major changes in the composition of the clusters occur in 2006. In particular, 17 companies move from the best clusters (III and IV) to the first, with a significant worsening. These companies are those that have suffered most of the introduction of the regulatory measures adopted by the AEEG (2004a).

5. Analysis of the results In this section we analyze the clusters obtained in relation to some critical factors such as geographical location, size, age and business diversification. With regard to the age of the company, there are two company typologies: those that entered the market after the liberalization of the sector and those that stem from the unbundling of larger companies required by the Letta Decree, which came into effect on 1st January 2002. Despite being a newly registered company, any firm of the latter typology should be considered a wellestablished enterprise in terms of know-how inherited from the company group to which it belonged. In Table 3, ‘new’ denotes the newly formed companies that entered the market after liberalization, whereas ‘established’ refers to those that originated from previously established groups. On further examination of the sample in Section 2, the mean of each cluster is compared to the overall mean for each of the critical factors. Company characteristics that have a greater than average concentration higher than or equal to the 20% within each cluster are shown in Table 3. In 2004 the geographical position is seen to be related to the performance of the companies. There is a prevalence of firms from the north in the best cluster, whilst the worst cluster mainly consists of companies from the south. The companies from the north tend to perform better due to the fact that they have to compete against a larger number of competitors and consequently they must contain their operating costs in order to attain the higher levels of efficiency required in a more competitive market. Firms operating in the northern regions account for the greatest percentage of Italian retail gas companies. These areas are the most attractive from an industrial viewpoint and are also those in which the climate is colder and hence there is a greater demand in terms of volume of gas. Furthermore, firms are able to access capital more easily in the north than in the south. This, in conjunction with the other factors stated above, provides the new companies with strong incentives to operate in the north and to leave the south to the older incumbent retail companies.

ARTICLE IN PRESS 3398

G. Capece et al. / Energy Policy 38 (2010) 3394–3402

Fig. 1. The resulting dendogram from the statistical analysis in 2006.

With regard to company size, the smaller companies are found in the first cluster. In particular, most of the small-sized companies are from the south. The best performances are obtained by the large companies belonging to the old established group. This result shows that the economies of scale and learning are critical factors in achieving high performance.

Regarding the age of the companies, the newly formed firms are mainly distributed in the first two clusters, because they are unable to achieve satisfactory results due to nonamortization of start-up costs and the absence of economies of experience. Vice versa, the best cluster is primarily composed of wellestablished companies; therefore, the group background and the

ARTICLE IN PRESS G. Capece et al. / Energy Policy 38 (2010) 3394–3402

4 3.5 3 2.5 2 ROI ROE ROS

1.5

CF LR

1 2004

2005

2006

Fig. 2. Trend of average values of indicators.

3.5

2004 2005

3

2006

2.5 2 1.5 1 I Cluster

II Cluster

III Cluster

IV Cluster

Fig. 3. The average value of all indicators for each cluster and each year.

Table 2 Percentage distribution of firms within the cluster.

I Cluster II Cluster III Cluster IV Cluster

2004 (%)

2005 (%)

2006 (%)

15.5 14.5 34.0 36.0

13.3 11.4 38.1 37.2

26.7 16.2 24.8 32.3

3399

ability to maintain stable relationships with clients and suppliers are indicative of success. Regarding business diversification we can see how single-business companies are in majority in all clusters, with the exception of cluster III, which has a prevalence of multi-business companies. It is interesting to note that the worst clusters are characterized by small single-business companies, while in the best clusters there are medium-large sized single-business firms. In the third cluster, there is a prevalence of multi-business companies characterized by a history of external growth through mergers and acquisitions with other multi-utility. The good results obtained from these companies and the subsequent placement in the third cluster are derived mainly from two factors. First of all, these companies have achieved cost reductions through the exploitation of economies of scale obtained through mergers and rationalization of staff. Moreover, the diversification of business portfolio allows reducing exposure to risks, changes in scenery and climate effects. In 2005, as concerns the geographical distribution, the composition of the clusters does not undergo any significant changes from the previous year. In terms of firm size, several changes can be observed. In particular, most of the small companies are split between the first and the third cluster, with the difference that in the worst cluster comprises companies mainly from the south, while in the third they are largely from the central-north. Large firms maintain the positive trend of 2004, positioning themselves in the best cluster in 2005. The medium-sized companies are located in the second and fourth cluster, differing from geographical location: prevalence of central-south in the second cluster and predominantly of northern in the fourth. Moreover, in cluster IV many medium-sized firms have a regional catchment area, while in the second cluster they operate locally. Regarding the age of the companies, the newly formed firms are mainly distributed in the worst cluster while the best cluster is primarily composed of well-established companies. With regard to business diversification, we can highlight how single-business companies are to be found in both cluster I and cluster IV, mostly small-sized and with a local catchment area. The difference between the two clusters is the geographical location, because single-business firms belonging to cluster I are mainly from the south, while those of cluster IV are equally distributed. Multi-business firms consolidate their position in the third cluster. In 2006 the changes in the composition of the clusters are more significant. In relation to the geographical position the northern companies are to be found in both cluster I and cluster

Table 3 Company characteristics that have a greater than average concentration within each cluster. Geographical location

Company size

Age of company

Business diversification

2004 I Cluster II Cluster III Cluster IV Cluster

South Central Average North

Small Average Average Large

New New Average Established

Single Single Multi Single

2005 I Cluster II Cluster III Cluster IV Cluster

South Central Average North

Small Medium Small Medium/Large

New Average Average Established

Single Average Multi Single

2006 I Cluster II Cluster III Cluster IV Cluster

North Average Central North

Average Large Small Medium/Large

Average New New Established

Average Average Single Average

ARTICLE IN PRESS 3400

G. Capece et al. / Energy Policy 38 (2010) 3394–3402

IV. In particular, those in cluster IV are mostly large sized, while those in cluster I are equally distributed in relation to size. Firms from the center get a good improvement, moving from second to third cluster in 2006. From an analysis of financial statements, we can underline that such improvement is mainly due to increased revenues, through the consolidation of their position in the local market and the selling on other territorial areas. In addition, some companies from the center have succeeded in significantly reducing the level of indebtedness. With regard to company size, small firms (mainly in centralnorth) strengthen their position in the third cluster, mediumsized ones are placed in the fourth cluster and the large divide between the second and fourth cluster with the difference that in the fourth cluster all firms (except one) are in the north, while in the second, trend reflects an average distribution. As concerns the company age, we can show an improvement of newly formed firms moving from the worst cluster to intermediate clusters (II and III), exploiting the economies of learning. The best performances are obtained by well-established firms, which continue to position themselves in the most high-performance cluster. Regarding business diversification, we can see how the best cluster is characterized by the presence of both single and multibusiness companies. This position shows an improvement of diversified firms, which have as a strength the joint offer of more services. In particular, these companies carry out joint offers differentiated by market segment and standardized in order to simplify the marketing and management. To achieve these results, companies have implemented policies to maintain and expand the customer base, investing in business plans, promoting dual fuel offers, improving call centers and launching initiatives to support loyalty club. Since in the cluster analysis the different indicators are considered jointly, it is not possible to isolate the effect of regulatory policies on profitability of companies. To this end, in the next section, we analyze the trend of return on equity in relation to regulatory policies adopted.

6. Impact of the tariff regulation on profitability In order to analyze the impact of the tariff regulation on profitability of companies, it is necessary to make an assumption about the difference between the protected market and the free market. Within the protected market of end customers who have not switched operator after liberalization, the AEEG has maintained control over prices. Consequently, an important difference between the protected market and the free market is that the revenues of retail companies, operating in the protected market, derive from the regulated tariff set by AEEG, whilst in the free market, companies can propose their commercial offerings without constraints, except those relating to information and transparency. Therefore, the tariff regulation affects more on small companies with sales of less than 100 M (m3) that focus their activities on the protected market. During the period examined in our study, the AEEG has adopted a series of resolutions designed to regulate the economic conditions of sale. In 2004, the Resolution no. 138 of December 4th, 2003 was in force (AEEG, 2003), which identified the methodology of calculation of all components of cost of service, attributing them to the end customer. Given a trend in world prices of petroleum products in substantial and continuing rise since 2003, the Authority has adopted the Resolution no. 248 of December 29th, 2004

(AEEG, 2004a), introducing a mechanism for protecting the end customers. The revision consists, primarily, in changing the methodology of updating of the raw material component, and correction of the variable fee of wholesale trade (Corrispettivo variabile relativo alla Commercializzazione all’Ingrosso or CCI). In particular, the Authority introduced a protection clause to curb gas price as a function of the prices of petroleum products in order to soften the impact of such products on gas prices. In addition, the Authority has established that, from 1st October 2005, a reduction in the variable fee of wholesale trade (CCI) is applied in order to support the negotiation of import prices in line with the average European price. This resolution has generated widespread discussion and debate. Resolution no. 248/04 has been challenged by various actors (wholesalers, retailers and trade associations) in the Regional Administrative Court of Lombardy (Tribunale Amministrativo Regionale per la Lombardia or T.A.R.), which has ordered the cancellation, through judgments nos. 3716/05 and 3718/05 (T.A.R. Lombardia, 2005a,b). Subsequently, on 14th October 2005, the Council of State (Consiglio di Stato, 2005), through the Order no. 4921/05, accepted the application submitted by the Authority, by suspending the enforceability of the Lombardy Regional Administrative Court ruling. Accordingly, the Authority, by Resolution no. 298 of December 20th, 2005 (AEEG, 2005), has updated the economic conditions of sale for the first quarter (January to March) 2006, by applying the methodology established by Resolution no. 248/04 and by providing a refund for end customers (Resolution no. 65 of March 27th, 2006), due to partial compensation of the amounts paid by them in 2005, when Resolution no. 248/04 was pending before the administrative justice (AEEG, 2006). The reduction in margins for the wholesale commercialization and the repayment of the refund have caused a negative effect on profitability of the operators, especially small and new entrants. This negative result is shown in Figs. 4 and 5, where we can observe a significant reduction in the ROE of small companies and new entrants, that is, the categories that have suffered most the introduction of the Resolution no. 284/04 and of the successive Resolutions nos. 298/05 and 65/06. Over the years 2005 and 2006, these two categories of companies suffered a major reduction in return on equity, whilst for medium and large firms and for wellestablished companies we can study a trend of ROE, which has a small increase in 2005 and is slightly decreasing in 2006,

Fig. 4. The average value of ROE range of small, medium and large companies.

ARTICLE IN PRESS G. Capece et al. / Energy Policy 38 (2010) 3394–3402

Fig. 5. The average value of ROE range of new and established companies.

demonstrating that the impact of regulatory policies on the profitability of these companies is not very significant.

7. Conclusions This paper responds point by point to the research questions raised in the introduction, addressing the following lines of study:

 The analysis of the performances of companies involved in gas  

retail over the three years following the liberalization of the sector, i.e., 2004–2006. The identification of a relationship between the performances of companies and their main characteristics such as age, size, geographical location and degree of business diversification. The analysis of this relationship during the period concerned.

In order to demonstrate these points, a sample of 105 firms is analyzed, representing approximately 54% of the companies operating in the sector. Firstly an analysis of the balance sheets is carried out calculating the main profitability, financial and liquidity indicators (ROI, ROE, ROS, LR and CF), which are then utilized as variables for cluster analysis. Four clusters are obtained from the statistical analysis, which is carried out using the Ward method. Cluster I contains the companies with the worst performance and cluster IV those that achieve the best results. It may be observed from the cluster analysis that throughout the three years in question most of the companies perform well and thus are to be found in clusters III and IV. However, this positive performance is attenuated by the progressive reduction in the average values of the indicators. This decline is mainly due to two factors: firstly the economic trend recorded over the three-year period has been strongly influenced by the regulatory and tariff related aspects of the gas market; and secondly the uncharacteristically mild climatic trends at the time significantly reduced the volume of sales. A second analysis is carried out by considering the distribution of the companies throughout the clusters in relation to a number of critical factors: the geographical location, size, age and business diversification. The main findings are summarised below. Throughout the three-year period it is observed that with respect to the age of the companies, in the best clusters there is a prevalence of companies that belong to longstanding business

3401

groups, whereas in the worst clusters there are more businesses that were new to the market. Thus the know-how of personnel, strong ties with suppliers, and trust building with clients would appear to be factors that are critical to a competitive advantage. In relation to company size there is a prevalence of mediumlarge sized companies in the best cluster throughout the threeyear period. Firms of this size are therefore able to take advantage of economies of scale (as opposed to smaller firms). It should be noted, however, that some small central-north companies have improved their performance thanks to the loyalty policies of their consumers that allowed them to maintain their market share. With regard to geographical location, the companies in the north of the country achieve the best performance throughout the three-year period. This is due to the greater degree of competition, which induces companies to reduce their operating costs and thus attain a higher level of efficiency. Most Italian operators in gas retail are based in northern Italy because it is the most attractive area from an industrial standpoint, and also since there is a greater demand for gas as the climate is colder there than in the rest of the country. As regards business diversification, the specialised companies outperform the diversified companies throughout 2004 and 2005. This result may be due to the fact that the multi-business companies were still in the start-up phase. Actually, in 2006 newly diversified companies improve, exploiting the economies of scope. A further analysis in our paper focuses on the impact of the tariff regulation on profitability of companies. Since in the cluster analysis, the different indicators are considered jointly, it is necessary to isolate the effect of regulatory policies on profitability, by analyzing the only return on equity of all companies. The results show that the profitability of small companies and new entrants is significantly worse by a change in regulation occurring through the Resolution no. 284/04 and the successive Resolutions nos. 298/05 and 65/06. From all the analyses carried out in this paper, it can be seen that, in the first three years following the market liberalization, firms that obtained the best performance are the incumbent operators, namely those belonging to longstanding business groups of medium-large size. Consequently, the opportunity arises for the regulatory authorities to implement policies aimed at improving competition, encouraging businesses start-ups and supporting the growth of small companies. In order to increase the competition in the retail market, regulatory interventions are necessary mainly in the upstream supply phase, facilitating the entry of new operators independent of Eni, which can supply gas on competitive terms and conditions. Firstly, in a context of exclusive use of take or pay contracts, for increasing competition, the necessary condition is a sufficient flexibility of supply to fluctuations in demand, that must be ensured by an adequate transport capacity and be able to trigger effective competition for the conquest of market shares (AGCM and AEEG, 2005). Therefore, regulatory policies are necessary to encourage the implementation of new projects in gas supply, such as the construction of LNG terminals and the expansion of the existing pipelines. Furthermore, it is necessary to develop a centralized spot market for gas exchanges, able to relax the competitive constraint represented by the supply agreements under take or pay clauses. To this end, in 2004, the Authority, through Resolution no. 22 (AEEG, 2004b) has identified the phases that lead to the launching of a centralized market for natural gas. On 15th August 2009, Law no. 99, of July 23th (Italian Parliament, 2009), came into force, by which the company for managing the energy markets (Gestore dei Mercati Energetici or GME) is the only one responsible for organizing and economically managing the natural gas market under principles of neutrality, transparency, objectivity and competition.

ARTICLE IN PRESS 3402

G. Capece et al. / Energy Policy 38 (2010) 3394–3402

The development of a centralized market for the exchange of natural gas must run concurrently with the creation of conditions for a greater plurality in the supply phase. Therefore, it is evident that the critical points to consider are both the volume of gas traded on the spot market and the market power exercised by the supply operators.

References AEEG, 2003. Delibera 4 dicembre 2003, no. 138/03. Criteri per la determinazione delle condizioni economiche di fornitura del gas naturale ai clienti finali e disposizioni in materia di tariffe per l’attivita di distribuzione, Milan, Italy. /http://www.autorita.energia.it/it/docs/03/138-03.htmS. AEEG, 2004a. Delibera 29 dicembre 2004a, no. 248/04. Modalita di aggiornamento della componente materia prima delle condizioni economiche di fornitura del gas naturale e revisione del corrispettivo variabile relativo alla commercializzazione all’ingrosso, Milan, Italy. /http://www.autorita.energia.it/it/docs/ 04/248-04.htmS. AEEG, 2004b. Delibera 26 febbraio 2004b, no. 22/04. Disposizioni in materia di mercato regolamentato delle capacita e del gas di cui all’articolo 13 della deliberazione dell’Autorita per l’energia elettrica e il gas 17 luglio 2002, no. 137/02, Milan, Italy. /http://www.autorita.energia.it/it/docs/04/22-04.htmS. AEEG, 2005, Delibera 29 dicembre 2005, no. 298/05. Aggiornamento per il trimestre gennaio-marzo 2006 delle condizioni economiche di fornitura del gas naturale, ai sensi della deliberazione dell’Autorita per l’energia elettrica e il gas 29 dicembre 2004, n. 248/04, Milan, Italy. /http://www.autorita.energia. it/it/docs/05/298-05.htmS. AEEG, 2006, Delibera 28 marzo 2006, no. 65/06. Avvio di procedimento per la determinazione delle condizioni economiche di fornitura del gas naturale e contestuale previsione di un parziale conguaglio di cui al punto 2 della deliberazione 29 dicembre 2005, n. 298/05, Milan, Italy. /http://www. autorita.energia.it/it/docs/06/65-06.htmS. AEEG, 2007. Relazione annuale sullo stato dei servizi e sull’attivita svolta, Regulatory Authority for Electricity and Gas, Milan, Italy. /http://www. autorita.energia.it/it/relaz_ann/07/07.htmS. AGCM, A.E.E.G., 2005. Indagine conoscitiva sullo stato della liberalizzazione dei settori dell’energia elettrica e del gas naturale (IC22), Rome, Italy. /http:// www.agcm.it/S. Aivazian, V.J., Callen, J.L., Chan, M.W.L., Mountain, D.C., 1987. Economics of scale versus technological change in the natural gas transmission industry. The Review of Economics and Statistics 69 (3), 556–561. Anderberg, M.R., 1973. Cluster Analysis for Applications. Academic Press, New York, USA. Carrington, R., Coelli, T., Groom, E., 2002. International benchmarking for monopoly price regulation: the case of Australian gas distribution. Journal of Regulatory Economics 21 (2), 191–216. Cavaliere, A., 2007. The liberalization of natural gas markets: regulatory reform and competition failures in Italy. Working paper—NG20. Oxford Institute for Energy Studies. /http://www.oxfordenergy.org/gasresearch.phpS. Consiglio di Stato, 2005. ordinanza del 14 ottobre no. 4921/05, Rome, Italy. /http://www.giustizia-amministrativa.it/S. Dorigoni, S., Portatadino, S., 2009. Natural gas distribution in Italy: when competition does not help the market. Utilities Policy 17, 245–257. Erbetta, F., Fraquelli, G., 2003. Produttivita e redditivita nella distribuzione locale di gas naturale in Italia: proprieta , diversificazione e scala operativa. L’Industria 4, 745–768. European Parliament and Council, 1998. Directive 98/30/EC concerning common rules for the internal market in natural gas, Bruxelles, Belgium. /http://www. europarl.europa.eu/S. European Parliament and Council, 2003. Directive 2003/55/EC concerning common rules for the internal market in natural gas and repealing Directive 98/30/EC, Bruxelles, Belgium. /http://www.europarl.europa.eu/S.

European Parliament and Council, 2009. Directive 2009/73/EC concerning common rules for the internal market in natural gas and repealing Directive 2003/55/EC gas, Bruxelles, Belgium. /http://www.europarl.europa.eu/S. Everitt, B.S., Landau, S., Leese, M., 2001. Cluster Analysis. Arnold, London, UK. Farsi, M., Filippini, M., 2009. An analysis of cost efficiency in Swiss multi-utilities. Energy Economics 31, 306–315. Gorini de Oliveira, R., Tolmasquim, M.T., 2004. Regulatory performance analysis case study: Britain’s electricity industry. Energy Policy 32, 1261–1276. Granderson, G., 2006. Externalities, efficiency, regulation, and productivity growth in the US electric utility industry. Journal of Productivity Analysis 26, 269–287. Granderson, G., Linvill, C., 1996. The impact of regulation on productivity growth: an application to the transmission sector of the interstate natural gas industry. Journal of Regulatory Economics 10, 291–306. Guldmann, J.M., 1985. Economies of scale and natural monopoly in urban utilities: the case of natural gas distribution. Geographical Analysis 17, 302–317. Herbert, J.H., Kreil, E., 1996. US natural gas market. How efficient are they? Energy Policy 24, 1–5 Italian Parliament, 1990. Legge 10 ottobre no. 287. Norme per la tutela della concorrenza e del mercato, Rome, Italy. /http://www.parlamento.it/S. Italian Parliament, 2009, Legge 23 luglio no. 99. Disposizioni per lo sviluppo e l’internazionalizzazione delle imprese, nonche´ in materia di energia, Rome, Italy. /http://www.parlamento.it/S. Jamasb, T., Pollitt, M., Triebs, T., 2008. Productivity and efficiency of US gas transmission companies: a European regulatory perspective. Energy Policy 36, 3398–3412. Kaufman, L., Rousseeuw, P.J., 1990. Finding Groups in Data an Introduction to Cluster Analysis. John Wiley & Sons, New York, USA. Kim, T.-Y., Lee, J.-D., Park, S.-B., 1999. Profit, productivity, and price differential: an international comparison of the natural gas transportation industry. Energy Policy 27, 679–689. Lee, J.-D., Oh, K.J., Kim, T.-Y., 1999. Productivity growth, capacity utilization, and technological progress in the natural gas industry. Utilities Policy 8, 109–119. McQueen, J., 1967. Some methods for classification and analysis of multivariate observations. In: Proceedings of the 5th Berkeley Symposium on Mathematical Statistics and Probability, pp. 281–297. Ministry of Economic Development, 2000. Legislative Decree no. 164, concerning the implementation of the Directive 98/30/EC and the common rules for the internal market in natural gas, Rome, Italy. /http://www.sviluppoeconomico. gov.it/S. Murry, D., Zhu, Z., 2008. Asymmetric price responses, market integration and market power: a study of the US natural gas market. Energy Economics 30, 748–765. Price, C.W., Weyman-Jones, T., 1996. Malmquist indices of productivity change in the UK gas industry before and after privatisation. Applied Economics 28 (1), 29–39. Reymond, M., 2007. European key issues concerning natural gas: dependence and vulnerability. Energy Policy 35, 4169–4176. Rushdi, A., 1994. Productivity changes in the gas and fuel corporation of Victoria. Energy Economics 16, 36–45. Sickles, R.C., Streitwieser, M.L., 1998. An analysis of technology, productivity, and regulatory distortion in the interstate natural gas transmission industry: 1977–1985. Journal of Applied Econometrics 13 (4), 377–395. T.A.R. Lombardia, sentenza del 6 ottobre 2005a, no. 3716, Milan, Italy. /http://www.giustizia-amministrativa.itS. T.A.R. Lombardia, sentenza del 6 ottobre 2005b, no. 3718, Milan, Italy. /http://www.giustizia-amministrativa.itS. Tryon, R.C., Bailey, D.E., 1970. Cluster Analysis. McGraw-Hill, New York, USA. Verde, S., 2008. Everybody merges with somebody — the wave of M&As in the energy industry and the EU merger policy. Energy Policy 36, 1125–1133. Wang, J.H., Ngan, H.W., Engriwan, W., Lo, K.L., 2007. Performance based regulation of the electricity supply industry in Hong Kong: an empirical efficiency analysis approach. Energy Policy 35, 609–615. Ward, J.H., 1963. Hierarchical grouping to optimize an objective function. Journal of the American Statistical Association 58, 236–244. Yu, L., Berg, S., Guo, Q., 2004. Market performance of Chinese telecommunications: new regulatory policies. Telecommunications Policy 28, 715–732.