The Fourth Railway Package:
the Governance Structure and the
Unsolved Problem of Vertical Integration
di Monica Delsignore
The Fourth Railway Package is a set of six legislative texts designed to finalize the establishment of the Single European Railway Area. Since 2001, the Eu-ropean Commission has enacted a series of «Railway Packages» with the aim of enabling privately owned train operators to access the European rail market on a competitive basis. The separation of train operators and infra-structure managers is often considered to be a prerequisite for the authentic competition in passenger rail operations. However, the latest step in the EU legislative process, the Fourth Railway Package, does not make the final leap of requiring full separation between train operators and infrastructure managers. Moreover, the Government of the UK, which has one of the most liberalized rail markets in Europe, has announced the possible combination of train operations and infrastructure management known as «vertical integration». The paper will deal with these problematic issues, even focusing on comparative experiences in different Member States.
Sommario: 1. The long path to EU railway liberalization. – 2. Modes of competing in the railway sector and governance structure of railway operators. – 3. The Court of Justice and its jurisprudence on the structure of governance. – 4. The new provisions in the Fourth Package: a compromise solution. – 5. The predictable effects of the Fourth Package. – 6. The separation of infrastructure manager and service providers: suggestions from different experience in other network markets.
1. The long path to EU railway liberalization
Introducing competition into European railway transport is an ongoing and challenging issue.
Since 2002, the European Union has started the path of liberalization . The First Railway Package laid the foundations by requiring operational separation of the functions of infrastructure management, on one side, and of train operations, on the other. The Second Railway Package (2004) and the Third Railway Package (2007) made incremental gains towards full liberalization in freight and international passengers transport.
Specifically, the Second Railway Package provided for the creation of a dedicated Authority, the European Union Agency for Railways (ERA), in the belief that the simultaneous pursuit of the goals of railway safety and interoperability requires substantial technical work led by a specialised body. The objective of the Agency «shall be to contribute, on technical matters, to the implementation of the Community legislation aimed at improving the competitive position of the railway sector by enhancing the level of interoperability of railway systems and at developing a common approach to safety on the European railway system, in order to contribute to creating a European railway area without frontiers and guaranteeing a high level of safety» (art. 1 Reg. EC no. 881/2004, repealed and replaced by Reg. (EU) 2016/796).
The intervention of European law, through the first three Packages, moves gradually to the open market in the railway sector.
Despite this framework, the majority of rail passenger services in Europe have, until recently, remained largely in the hands of state-owned operators. Many continue to be so. This has led the European Commission to litigate with Member States for infringing the Directives in the three Packages.
The subsequent several Court of Justice’s judgements  did not clarify the general framework and required for new provisions on the State duties and obligations, which finally lead to the adoption of a new package, the Fourth Railway Package.
The Fourth Railway Package  contains important changes designed to improve the functioning of the single European railway area through amendments by way of recast to Directive 2004/49/EC and Directive 2008/57/EC, both of which are directly linked to the tasks of the Agency. Those Directives provide in particular for the performance of tasks relating to the issuing of vehicle authorisations and safety certificates at Union level. This will involve a greater role for the European Union Agency for Railways (ERA). Furthermore, the Fourth Package aims to open the whole rail transport market to competition.
Currently only freight and international passengers transport are open to competition. National passenger transport can still be a monopoly in all Member States. A different trend is developing for high speed lines . The difference in the market is due to the fact that high speed lines constitute rather a new market, which often exploits a dedicate infrastructure.
The Fourth Railway package intends to open the whole rail transport market to competition by 2026. By eliminating national barriers, the Fourth Railway Package will promote competition between large railway undertakings. The homogenization of infrastructures across countries should create a single market between numerous pre-existing big national companies rather than only much smaller and younger companies.
2. Modes of competing in the railway sector and governance structure of railway operators
As clarified , three possible modes of competition can exist in the railway sector:
– Competition «in» the market between vertically integrated companies. In this case, at least two competitive infrastructures exist (i.e. two rail lines with similar paths). This kind of competition has historically existed in Europe  and is the predominant form of competition in the US;
– Competition «in» the market between competing train operating companies using the same infrastructure. The final consumer can choose between different options for a similar service. This kind of competition, also called «open access competition», exists today in the airlines and in the telecommunication industries. Another important feature of railway open access competition is the fact that service providers would do not receive any State subsidies and would be free to determine their train schedule;
– Competition «for» the market. In this case, different train operating companies tender for service provision. The winning company is then entitled to a monopoly for a time-limited franchise defined for a specific area and /or a specific service. This kind of competition can be found in other industries such as water distribution or garbage collection.
In the European context, the Fourth Package plans the opening of the market for regional transport around franchises (competition «for» the market). On the contrary, competition for long distance services – at least for profitable services – should be «open access» (or competition «in» the market).
Specifically, the regulation of access alone is often not enough to avoid opportunistic or discriminatory behaviour of the network operator. This is especially true when the network operator is included in a group together with companies involved in the rail service distribution. The provisions on unbundling, c.d. unpacking, aim to avoid possible conflicts of interest between the managing body and the supplying companies.
European Union Packages should entail the overcoming of a vertically integrated structure (which concentrates network management and transportation service into one subject), in favour of a new model with three poles: State, exclusive manager of the network infrastructures, companies competing for the railway transportation services.
This separation of train operators and infrastructure managers is often considered to be a prerequisite to the true opening to competition of passenger rail operations. The separation ensures that access to rail infrastructure is not restricted by discriminatory measures imposed by infrastructure managers: for example, by granting preferential train paths to train operators within its corporate group. This separation has been instrumental in liberalizing (or, in some cases, part-liberalizing) passenger rail operations markets in a number of EU Member States, although the degree of liberalization so far actually implemented varies significantly between Member States. The separation of train operators and infrastructure manager has been one of the issues challenged by the Court of Justice in the previous cited judgements.
3. The Court of Justice and its jurisprudence on the structure of governance
Since 2008 European Commission opened multiple infringement procedures against different Member States on the correct implementation of Railway Packages.
One of the most tough issues at stage was the structure of governance of the service operator, particularly concerning the role reserved to the infrastructure owner.
The Court stated that the previous Directive required «only that the accounts for business relating to the provision of transport services by railway undertakings be kept separate from the accounts for business relating to the management of the rail infrastructure, since business relating to the provision of transport services by railway undertakings and business relating to the management of rail infrastructure may be kept separate by means of the organisation of distinct divisions within a single undertaking, as is the case with a holding company … in order to guarantee equitable and non-discriminatory access to rail infrastructure, Member States are to take the measures necessary to ensure that the essential functions … are entrusted to bodies or firms that do not themselves provide any rail transport services and that this objective must be shown to have been achieved, regardless of the organisational structures established» (C-556/10, judgement par. 55).
The holding company model, and therefore the presence of a parent company and its subsidiaries, is not per se condemned by the Court: it must not be presumed that the holding company effectively exercises economic control over its subsidiary, the infrastructure manager entrusted with essential functions. «The mere fact that the holding company owns a majority or all of the shares or voting rights in the entity entrusted with essential functions is not sufficient, otherwise the ‘normal’ holding company system would be outlawed» (C-555/10, opinion par. 88).
In the reasoning of the Court the Railway Packages provisions let most of the railway operators still have strict connection and be influenced by the railway owners.
Therefore, the European Commission, even in the light of this jurisprudence, started a new dialogue for emending, above all, the governance structure in the packages in order to find the best solutions for the implementation of a single railway market.
4. The new provisions in the Fourth Package: a compromise solution
The latest step in the EU legislative liberalization process, the Fourth Railway Package, opted not to require full separation between train operators and infrastructure managers, as many were expecting.
Specifically, the amendments to the 2012 Directive retain the concept that network capacity (i.e. train path) allocation and responsibility for setting the charges for access to infrastructure must be kept separate, but expressly acknowledge that the required separation can be achieved within a vertically integrated structure. This concession was the product of a hard battle by a number of Member States, given that the abundant literature on the subject is still inconclusive regarding the impact of separation on the overall system.
Vertical integration is therefore permitted rather than outlawed, but with a clear set of safeguards. In fact, the Directives introduce the so called «Chinese wall» arrangements , prescribing strict safeguards to protect the independence of the infrastructure manager with an oversight process enacted by the Commission to ensure that a genuine level playing field for all railway undertakings is put in place.
The key question is how much regulation and Chinese wall can be effective in the presence of such a strong public property of infrastructures and of such a frequent admixture of infrastructure manager in railway undertaking, considering this is the usual framework in most Member States.
The Fourth Package deals with this question but finally offers a compromise solution.
As previously noted, the Directive EU 2016/2370, moving from the recalled previous jurisprudence, introduces specific further requirements to ensure the independence of the infrastructure manager.
Member States should be free to choose between different organisational models, ranging from full structural separation to vertical integration, subject to appropriate safeguards to ensure the impartiality of the infrastructure manager as regards the essential functions, traffic management and maintenance planning.
The Directive offers a specific definition of «vertically integrated undertaking», which means «an undertaking where, within the meaning of Council Regulation (EC) No 139/2004 (*): (a) an infrastructure manager is controlled by an undertaking which at the same time controls one or several railway undertakings that operate rail services on the infrastructure manager's network; (b) an infrastructure manager is controlled by one or several railway undertakings that operate rail services on the infrastructure manager's network; or (c) one or several railway undertakings that operate rail services on the infrastructure manager's network are controlled by an infrastructure manager. It also means an undertaking consisting of distinct divisions, including an infrastructure manager and one or several divisions providing transport services that do not have a distinct legal personality. Where an infrastructure manager and a railway undertaking are fully independent of each other, but both are controlled directly by a Member State without an intermediary entity, they are not considered to constitute a vertically integrated undertaking for the purposes of this Directive».
Furthermore, Member States should ensure that, within the limits of the established charging and allocation frameworks, the infrastructure manager enjoys organisational and decision-making independence as regards the essential functions. Safeguards should apply in vertically integrated undertakings to ensure that other legal entities within those undertakings «do not have a decisive influence on appointments and dismissals of persons in charge of taking decisions on the essential functions. In this context, Member States should ensure that there are complaints procedures in place. Member States should put in place a national framework for the assessment of conflict of interests. Within this framework, the regulatory body should take into account any personal financial, economic or professional interests which could improperly influence the impartiality of the infrastructure manager. Where an infrastructure manager and a railway undertaking are independent of one another the fact that they are directly controlled by the same Member State authority should not be considered to give rise to a conflict of interest within the meaning of the Directive» (par.9).
Decision-making by infrastructure managers with respect to train path allocation and decision-making with respect to infrastructure charging are essential functions that are vital for ensuring equitable and non-discriminatory access to rail infrastructure. Stringent safeguards should be put in place to avoid any undue influence being brought to bear on decisions taken by the infrastructure manager relating to such functions. Those safeguards should be adapted to take into account the different governance structures of railway entities.
Notwitstanding the previous jurisprudence, the new Directive still recognizes high discretion to the choices of Member States. The absence of «decisive influence on appointments and dismissals of persons in charge of taking decisions on the essential functions» together with the requirement of organisational and decision-making independence (as regards the essential functions of the manager infrastructure) recall ambiguous terminology which may leave too much space to Member States to implement the liberalisation and finally to evade the real question.
5. The predictable effects of the Fourth Package
Accordingly, the impact of the Fourth Railway Package should differ across the EU, depending in part upon the current state of liberalisation in the relevant Member State and in part upon the different choices of the relevant Member State in the implementation of those separation requirements.
In Italy  article 37 of law no. 214 /2011 established ART, Autorità di Regolazione dei Trasporti, a new independent technical agency with specific tasks in defining the conditions for access to railway infrastructure . As well known, in Italy the infrastructure manager is RFI, an undertaking controlled directly by the State and vertically integrated with FS, a railway undertaking of total public property. Since July 2018 RFI has incorporated Cento Stazioni, the owner of the station infrastructure. This framework has helped in creating a competitive market for high speed rail . Nevertheless, the same framework has been criticised in the past and the holding company FS has been fined by the Italian Competition Authority for the abuse of its dominant position in the distribution of access to the railway infrastructure . The vertically integrated structure of FS will stay still with no new adjustments due to the Fourth Package.
The new independent Authority should be able to introduce new stringent parameters in order not only to open the infrastructure to new undertaking, but even to improve in quality and frequency of rail services, and finally ensure greater overall passenger satisfaction.
The Italian model recalls the German Framework.
The infrastructure manager in Germany remains a subsidiary of Deutsche Bahn, a position which the German Government has been keen to protect. The infrastructure manager is split into two parts: one responsible for the rails, the other for the station infrastructure. With over 33,000 kilometres of track under its control, DB Netz AG owns and manages the longest rail network in Europe. Meanwhile, DB Station & Service AG operates the station infrastructure and some of the station buildings at almost all of the passenger stations in Germany. Access to the rail infrastructure on a non-discriminatory basis is supervised by the German Federal Network Agency (Bundesnetzagentur). DB Netz AG and DB Station & Service AG are subsidiaries of state-owned Deutsche Bahn AG, but they do not operate passenger transport services. Within the Deutsche Bahn group, passenger transport services are operated by separate group subsidiaries, such as DB Fernverkehr AG and DB Regio AG.
France has implemented the series of EU directives aimed at liberalising the European railway market. However, the national railway services operator, SNCF, has a monopoly on national and regional passenger services. The French rail regulator Authority, Arafer, has delivered a negative opinion on the government’s investment and performance contract with SNCF Network. This will only change with the implementation of the Fourth Railway Package, which will require the opening to competition of long distance commercial passenger services in 2020 on an open access basis, and of national and regional services from 2023/2024 on a franchise/ concession basis. In this model the infrastructure manager is independent but delegates several functions to the state-owned incumbent. The monopolistic operator SNFC is deep rooted and this make difficult to change its structure .
The UK model stands in Europe as the example of open market.
As other have noted , the UK has one of the most liberalised rail markets in Europe, with a number of operators owned by foreign entities competing successfully for franchise awards alongside fully domestic operators. Separation between infrastructure managers and train operators has been in place since the privatisation of the rail network in the 1990s, well before EU legal requirements to do so had been enshrined. The main infrastructure manager, Network Rail, is a separate government body which is independent of the train operators.
Even before Brexit, the impact of the Fourth Railway Package in the UK should therefore be less significant than in other Member States, such as Italy or Germany, where there is currently less separation. However, proposals recently announced by the Transport Minister, Chris Grayling, for a fully privatised rail line from Oxford to Cambridge signal a possible move away from the principles of separation. In December 2016 Mr Grayling set out plans for a new entity to be responsible for track and infrastructure, as well as the operation of train services, on a new Oxford to Cambridge line. If these plans are implemented, it will be the first integrated heavy rail operation in Britain for many years .
The government of the most liberalised rail system in Europe, which has lived with this separation of wheel and rail for the longest time, is now keen to point out the technical and operational difficulties which can arise from such separation: «When things go wrong, a lack of a joined up approach can make things much worse for the passenger ... Our railway is much better run by one joined up team of people. They don’t have to work for the same company. They do have to work in the same team».
If the European Commission’s aim with the Fourth Package was to provide independent infrastructure management, then the new framework may not be adequate to reach the target and sort the desired effects. The new Directives and the Regulations did not make a clear choice in favour of separation so hesitant Member States are free to opt for the favourite system of vertical integration.
The implementation of EU regulatory reform is always complex and even more when reform objectives are ambiguous . This allow sufficient discretion for Member States for them to delay and relax the legal transposition, as well as the technical harmonization.
6. The separation of infrastructure manager and service providers: suggestions from different experience in other network markets
Other network markets can offer previous and different experience regarding the issue of separation.
The electricity and ICT are open and competitive markets with transparent governance framework and non-discriminatory rules and both these sectors have faced similar concerns as railways . In all of these sectors the presence of a tradition of State monopolies raised issues about the preferable framework of the new liberalise market in terms of network ownership.
Large and integrated firms can often enjoy considerable economies of scope, of scale or of coordination depending on the degree of conglomerate (e.g., multi-product or multi-service), horizontal or vertical integration. This (supposed) pursuit of efficiency may sometimes be at the expense of competition, in that large firms are likely to acquire a strong if not dominant market position. Energy sectors display considerable problems of this type, especially as most of the times their current set-up has its origin in a long tradition of State monopolies, where horizontal as well as vertical integration were the rule. Therefore, as in the railway sector, the access to the transport and distribution infrastructures is a fundamental piece in the design of liberalization in energy markets. Overall, the need of access regulation seems a long run necessary solution for energy markets. Without proprietary separation the network owner has very high incentives to preclude, or at least limit, the access of competitors in the downstream market, vanishing the perspectives of liberalization. Once proprietary unbundling is introduced, the incentives are diminished, and in some cases an ex-post antitrust intervention might be sufficient, but still the possibility of foreclosure remains high .
The former de facto monopolistic structure of the Italian electricity and gas markets, which were managed by ENEL and ENI respectively, underwent significant changes following the liberalisation process and the introduction of unbundling obligations on vertically integrated energy operators at the beginning of 2000. In transmission and dispatching, ENEL transferred the ownership of the national grid to a newly established company, Terna, and the management of the grid to another newly established one, Gestore della Rete di Trasmissione Nazionale (GRDN, later renamed Gestore dei Servizi Elettrici) that would guarantee access to the network to any third party. Distribution and retail at the local level would be opened up to competitive tender offers in each municipal area. The ownership unbundling model requires a full separation of transport activities (including both the ownership and management of electricity transportation infrastructures) from the production and sale of electricity. Even though the separation, the liberalisation and re-regulation of the electricity industry did not result in significant effects on tariffs, which are relatively quite high in comparison to EU countries .
As for telecommunications, the initial phase of liberalisation supported competition within service providers. Nevertheless, in a short time technological dependence on the network hindered competition. On this basis, the EU has moved to also promote competition between networks. In ICT the Commission envisaged an obligation to functionally separate network operators with significant market power. One of the main obstacles to the development of a true level playing field for access seekers to electronic communication networks is the preferential treatment of the downstream businesses, for example the retail arm, of a vertically integrated operator with significant market power (SMP operator) through price and non-price discrimination (for example, discrimination regarding quality of service, access to information, delaying tactics, undue requirements and the strategic design of essential product characteristics) .
Therefore, even in this sector separation has always been a key issue, as for railways.
National regulatory authorities may impose an obligation on vertically integrated undertaking to place activities related to the wholesale provision of access products in an independently operating business unit. Furthermore, Dir 2009/140/CE allows voluntary separation of network operators.
Still, as highlighted , implementing a functional separation can be far more costly and disruptive than introducing behavioural remedies. It remains dubious which economic benefits are to be expected from a separation of network operators. Vertical integration of networks and services is an essential element of the sector. It fosters innovation and investment as network operators adapt networks to requirements of new services. Vertical integration has many economic efficiencies which would be eliminated by forced separation. Forced functional separation risks to create severe economic inefficiencies and to harm investments in infrastructure.
Likewise, OECD doubted on the effectiveness of the separation of the infrastructure, which implies very high costs, without ensuring clear benefits.
Admitting separation as just one of the possible solutions, ICT offers two different models in order to ensure fair and equivalent access whenever there is vertical integration of network operator and service provider: the equivalence of output and the equivalence of input.
The Commission considers that equivalence of inputs (EoI) is in principle the surest way to achieve effective protection from discrimination as access seekers will be able to compete with the downstream business of the vertically integrated SMP operator using exactly the same set of regulated wholesale products, at the same prices and using the same transactional processes. In addition, and contrary to an Equivalence of Output (EoO) concept, EoI is better equipped to deliver transparency and address the problem of information asymmetries . EoI though has very high potential costs. Therefore, where national regulatory authorities conclude that an obligation to provide regulated wholesale inputs on an EoI basis is disproportionate, an EoO model should be applied. This model ensures that the wholesale inputs provided to alternative operators – while not using the same systems and processes – are comparable, in terms of functionality and price, to those the vertically integrated SMP operator consumes its.
At present time in most of Member States there is still a vertical integration between network owner and ICT service provider (Italy is nowadays moving through the separation), even if liberalization in this sector has sensibly increased number of competitors and variety of services.
The different effects of liberalisation in these two sectors, anyway, certainly do not derive exclusively from the choice about separation or vertical integration between the network owner and the service providers.
The issue of access to network, whatever is the choice, highlights the key role of regulation and of national regulatory authorities in the liberalised sectors .
Regulation can be a much stricter instrument than competition law, and severely influence the business decision of the market players. The issue is to find the right fine-tuning is not trivial. If there is no or insufficient possibility for new entrant operators to get access, market will never exists. If access is granted at too low costs, in the long run the company with access obligation will have only limited or no incentive at all to invest and innovate the infrastructure .
These problems do not have a distinct answers referring to the nature of the owner.
The ongoing debate since the separation of train operators and infrastructure managers has still time and space to carry on and it could be very interesting to open the comparison to most Member States and find collaboration between law and economics in order to find the best solution. This solution, in the end, may be a compromise solution as in the Fourth Package, but only when this should be the outcome of a complex and extensive reasoning on the relevant issues.
For a general framework see P. Pellegrini-J. Rodriguez, Single European Sky and Single European Railway Area: A system level analysis of air and rail transportation, in Transportation Research Part A 57 (2013) pp. 64-86, K. Barrow, The Fourth Railway Package – Magic bullet or missed opportunity?, in International Railway Journal, 2018 and Study on the prices and quality of rail passenger services, European Commission, Directorate General for Mobility and Transport, Final Report April 2016. Suggestions on the different models could be found in L. Lorenzoni, Centralization or fragmentation of competences in the regulation of rail transport: a comparison between institutional models, Accentramento o frammentazione delle competenze in materia di regolazione del trasporto ferroviario: modelli istituzionali a confronto, in Munus, 2018, fasc. 1, p. 1.
 Considerations on this jurisprudence in Italian cases can be found in M. Delsignore, L’attuazione della liberalizzazione del servizio ferroviario al vaglio della Corte di Giustizia: qualche considerazione in attesa della decisione nei confronti dell’Italia, website www.associazione
deicostituzionalisti.it; M. Delsignore, Il sistema ferroviario e la liberalizzazione incompiuta, in Riv. ita. dir. pubbl. com., 2014, p. 437.
The Fourth Railway Package’s Technical Pillar, comprising the ERA Regulation 2016/796, the Interoperability Directive 2016/797 and the Rail Safety Directive 2016/798, was published in the Official Journal of the European Union (OJEU L138) on 26 May 2016. It legally entered into force on 15 June 2016. The Market Pillar was published in the Official Journal (OJEU L352) on 23 December 2016. The Governance Directive 2016/2370 entered into force the day after. Member States must transpose the Directive into national legislation by the end of 2018. Regarding the PSO Regulation 2016/2338, the EP and Council agreed on the need to have a transition period after the entry into force of the new Regulation, during which the direct award of contracts should still be possible under the rules of current Article 5(6). After that, the new rules (Article 5(4a), i.e. direct award with efficiency criteria built into the contract) will kick in. Article 5(6) will cease to apply and Article 5(4a) will be switched on as of 6 years after the entry into force of the Regulation. The Regulation entered into force on 23 December 2017, so that the passage between Article 5(6) and Article 5(4a) will foreseeably take place in 2023. For a brief comment on the proposal M. Petr, Fourth Railway Package (Almost) Adopted, in 4 Eur. Networks L. & Reg. Q. 88 (2016) and C. Cataldi, Il «quarto pacchetto ferroviario»: la proposta per uno spazio ferroviario europeo unico e liberalizzato – The EU fourth railway package. Proposal for a single and open EU railway area, in Munus, 2016, p. 143. For considerations on the impacts of the Fourth Package see P. Perennes, Open Access for Rail Passenger Services in Europe: Lesson Learnt from Forerunner Countries, in Transportation Research Procedia 25 (2017) 358-367. On the role of railway transportation in the EU environmental policy see C. Feliziani, Il trasporto ferroviario nel quadro della politica ambientale europea, in Nomos, 2018, p. 17. For a general overview of transport regulation in Italy see F. Bassan (ed.), La regolazione dei trasporti in Italia, Giuffrè, Milano, 2015.
The Italian opening to competition in high speed is commented by A.S. Bergantino-C. Capozza-M. Capurso, The impact of open access on intre- and inter-mosal rail competition. A national level analysis in Italy, in Working Paper SIET 2015; C. Desmaris, High Speed Rail Competition in Italy. A Major Railway Reform with a «Win-Win Game»?, in Discussion Paper No. 2016-11. Paper was presented to the Working Group (Public Transport Market O., 2016, p. 24).
M. Finger-N. Bert-K. Bouchard-D. Kupfer, Competition in Passenger Railways in Europe, in European Transport Regulation Observer 2017/24.
 For the Italian case see B. Tonoletti, Beni pubblici e concessioni, Padova; Cedam, 2008, spec. p. 31, V. Parisio, Beni ferroviari, servizio di trasporto ferroviario, demanio e patrimonio, in A. Police (ed.), I beni pubblici: tutela, valorizzazione e gestione, Giuffrè, Milano, 2008, p. 197 and M. Renna, La regolazione amministrativa dei beni a destinazione pubblica, Giuffrè, Milano, 2004, spec. p. 250.
D. Maresca, The Chinese walls between infrastructure managers and rail operators as a compromise between railway liberalization and State monopolies, in this Review 2014/1.
 For a general description of the Italian framework see A. Police, Il servizio pubblico di trasporto in Italia: regolazione e mercato nel trasporto ferroviario – Public transport service in Italy: regulations and market of rail transport, in Nuove Autonomie, 2015, p. 321, P. Chirulli, Il trasporto ferroviario tra sviluppo sostenibile, concorrenza e tutela del diritto alla mobilità, in Concorrenza, regolazione e tutela dei diritti nel trasporto ferroviario, Editoriale scientifica, Napoli, 2016, p. 7 and M. Bignardello-E.G. Rosafio, Il settore ferroviario, in La regolazione dei trasporti in Italia, cit.
Suggestions on the recent opinion of this Authority and its influence in the railway market see L. Saltari, Anomalie di una segnalazione «normativa», in Giorn. dir. amm., 2018, p. 256.
 A.S. Bergantino-C. Capozza-M. Capurso, The impact of open access on intre- and inter-mosal rail competition. A national level analysis in Italy, cit.; C. Desmaris, High Speed Rail Competition in Italy. A Major Railway Reform with a «Win-Win Game»?, cit.
 A. Candido-A. Canepa-V. Carfì, La tutela della concorrenza nei mercati regolati: le decisioni dell’AGCM in materia di servizi di interesse economico generale, The protection of competition in regulated markets: the AGCM decisions on services of general economic interest, in Conc. merc., 2013, p. 49, A. Boitani-F. Ramella, Arenaways e altre storie ferroviarie (poco edificanti) – Arenaways and other (non-edifying) railway stories, in Merc. conc. reg., 2012, p. 97 and A. Palmieri, Vettore ferroviario, pratiche commerciali scorrette e sanzione pecuniaria, ammenda, inflitta dall’Autorità Garante della concorrenza e del mercato, in Foro it., 2010, p. 652.
B. Tchofo Emmanuel-Y. Crozet, Beyond the «bundling vs unbundling» controversy: What is at stake for the French railway?, in Research in Transportation Economics 48 (2014), pp. 393-400; H. Dyrhauge, EU Railway Policy-making. On track?, Palgrave Macmillan, London, 2013.
S. Temple, Open Access Long Distance Passenger Rail Services in the United Kingdom: The Grand Central Experience, in Transportation Research Procedia 8 (2015), pp. 114-124.
See F. De Francesco-G. Castro, Beyond legal transposition: regulatory agencies and de facto convergence of EU rail liberalization, in Journal of European Public Policy, vol. 25 (2018), no. 3, pp. 369-388.
See A. Asquer, Liberalization and regulatory reform of network industries: A comparative analysis of Italian public utilities, in Utilities Policy, vol. 19 (2011), p. 172, J.A. Gómez Ibánez, Regulating Infrastructure: Monopoly, Contracts, and Discretion, Harvard University Press, Boston, 2003.
M. Polo-C. Scarpa, The liberalization of energy markets in Europe and Italy, in IGIER. Working Paper 230 (2002).
A. Asquer, Liberalization and regulatory reform of network industries: A comparative analysis of Italian public utilities, cit., p.176.
 In this sense F. Bassan, L’evoluzione della struttura istituzionale nelle comunicazioni elettroniche: una rete non ha bisogno di un centro, in F. Bassan (ed.), Diritto delle comunicazioni elettroniche, Giuffrè, Milano, 2010, p. 33, specifically p. 49.
Commission Recommendation of 11.09.2013 on consistent non-discrimination obligations and costing methodologies to promote competition and enhance the broadband investment environment.
 F. Dalle Nogare, Regolazione delle comunicazioni elettroniche. La storia, la governance delle regole e il nuovo Codice europeo, Giappichelli, Torino, 2019, and for an institutional perspective European Regulators Group (2007) ERG Common Position on Best Practice in Remedies Imposed as a Consequence of a Position of Significant Market Power in the Relevant Markets for Wholesale Leased Lines, erg (O7) 54 final 080331.
K.E. Winkler-Glenn Baumgarten, The Framework for Network Access and Interconnection, in C. Koenig-A. Bartosch-J. Braun-M. Romes (ed.), EC Competition and Telecommunications Law, Kluwer Law International, 2009, p. 421, here p. 469. For scepticism in the USA system see R.W. Crandall-J.A. Eisenach-R.E. Litan, Vertical Separation of Telecommunications Networks: Evidence from Five Countries, in Federal Communication law journal, vol. 62 (2010), p. 494
OECD (2001), Structural Separation in Regulated Industries; OECD (2003), The Benefits and Costs of Structural Separation of the Local Loop.
Commission Recommendation of 11.09.2013 on consistent non-discrimination obligations and costing methodologies to promote competition and enhance the broadband investment environment.
See F. De Francesco-G. Castro, Beyond legal transposition: regulatory agencies and de facto convergence of EU rail liberalization, cit.
K.E. Winkler-G. Baumgarten, The Framework for Network Access and Interconnection, in EC Competition and Telecommunications Law, cit.