In a fluid state? Competition policy in the water sector

The Water Act 2014[1] includes provisions to open up the England and Wales water sector to greater competition, most notably by introducing retail competition for non-domestic customers (in England) from April 2017, but also through other measures aimed at facilitating more bulk supplies between companies, and, potentially, upstream entry.

Under the new regime, Ofwat, the economic regulator of the water industry in England and Wales, will develop access-charging rules in the light of government guidance, and these provisions will replace the ‘costs principle’ which, through its application (a specific form of ‘retail-minus’ approach), is regarded as having created an inefficient barrier to competition. At the same time, following Ofwat guidance, companies have already developed retail default tariffs for non-domestic customers, which are based on wholesale charges plus retail costs and an allowed margin. Going forward, therefore, in applying its sector-specific powers, Ofwat wants companies to move towards a bottom-up approach to retail access to wholesale services, as opposed to the retail-minus approach used to date.[2]

However, as well as these sector rules, water companies need to ensure compliance with more general competition law. Ofwat and the Competition and Markets Authority (CMA) have concurrent powers to enforce the Competition Act 1998, in relation to anticompetitive agreements (Chapter I prohibition) and abuse of a dominant market position (Chapter II prohibition). A number of recent cases in the sector shed some light on the key areas of concern, and highlight the importance of competition law as Ofwat seeks to develop a level playing field.

Recent competition issues

Given the current structure of the industry, which involves regional monopolies and vertical integration, a key concern is that of a potential abuse of a dominant position. Three practices are of particular relevance.

• Margin squeeze—a margin squeeze occurs when a vertically integrated operator, which is dominant in an upstream market and provides an essential input to entrants in the contestable downstream market, sets its wholesale access (upstream) charges ‘too high’—and/or its (downstream) retail charges ‘too low’—so as to ‘squeeze’ the margin available to efficient entrants, excluding them from the downstream market. The benchmark that the European Commission generally relies on to determine the costs of an ‘equally efficient’ downstream competitor is the long-run average incremental cost (LRAIC) of the downstream division of the integrated dominant undertaking, although other benchmarks are also available.[3]

• Discrimination with anticompetitive effects—price discrimination is not an abuse under European competition law per se, as the practice may be welfare-improving. However, an abuse can occur if a company holding a dominant position applies ‘dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage’.[4] Here, the issues need to be considered on a case-by-case basis.

• Predation—predatory pricing involves a dominant firm setting prices ‘too low’ in order to drive competitors out of the market. The Commission’s enforcement guidance states that pricing below average avoidable cost (AAC) indicates that the dominant undertaking is sacrificing profits in the short term, and that an equally efficient competitor cannot serve the targeted customers without incurring a loss. Even if this test has been passed, failure to cover LRAIC may indicate that an equally efficient competitor could be foreclosed from the market.[5]

Margin squeeze

The Albion Water/Welsh Water case is a seminal case in the water sector, which was reviewed by the Competition Appeal Tribunal (CAT)[6] and Court of Appeal.[7] In this case, under the inset appointment regime (a form of entry into the sector), Albion Water took over the retail supply to Shotton Paper Mill from Welsh Water, and subsequently sought to buy water upstream from United Utilities, while paying a ‘common carriage’ network access charge to Welsh Water.

The arrangement left Albion Water with no effective margin. In 2001, Albion Water complained to Ofwat that the access price was excessive and gave rise to margin squeeze.[8] Ofwat disagreed, arguing that the Water Industry Act 1991 mandated the use of the ‘costs principle’, which in turn required a retail-minus approach to access pricing—specifically, one in which access prices are calculated by taking the retail price of the network owner and subtracting any avoided costs to the network owner (Welsh Water) from providing access (the ARROW methodology). In the regulator’s view, since there were no avoided retail costs in this case (Albion Water would simply replace Shotton Paper as the interface with Welsh Water), there was no margin squeeze.

The CAT, however, argued that Ofwat had not assessed the alleged margin squeeze correctly. It stated that the retail-minus approach to access pricing in this instance was unsound, and that Welsh Water’s access price was excessive in relation to the value of the service provided. It argued that the retail-minus approach applied, which subtracted avoidable costs only, meant that an entrant would need to support the incumbent’s overheads (and loss in revenues) as well as its own overheads. This needed a new entrant to be ‘super-efficient’, rather than just ‘efficient’.[9] Moreover, the CAT argued that the subtraction of ‘short-term’ avoided costs only (associated with one customer switching), as applied in Ofwat’s Decision, was not sound, since this eliminated existing competition from Albion Water and prevented market entry. Indeed, as the CAT case progressed, Ofwat stated that in the medium to longer term all retail costs could become avoidable, although the CAT noted that there were difficulties in calculating those elements that are avoidable.

The CAT similarly took the view that, in applying a test for margin squeeze, avoided costs (and, by extension, short-term avoidable costs) were not an appropriate benchmark.[10] Rather, the relevant imputation tests, set out in established cases, were the ‘as efficient competitor’ test (based on the incumbent’s own downstream costs, as per the Deutsche Telekom case[11]), and the ‘reasonably efficient competitor’ test (based on an entrant’s downstream costs). As regards the former, the CAT argued that Ofwat’s ‘failure’ to consider the costs of a notional Welsh downstream business, which would have placed Welsh Water and Albion Water on an equal footing, was ‘an error of analysis’. The CAT ruled that there had been a margin squeeze,[12] a Decision upheld by the Court of Appeal.[13]

The Albion Water case is likely to influence how Ofwat approaches margin squeeze cases going forward, including how it approaches the imputation test to explore whether an incumbent’s upstream and downstream prices are justified.[14]

In fact, the regulator is currently presiding[15] over an allegation of a margin squeeze, in a case dating back to November 2009. This arose from a complaint from Independent Water Networks Limited (IWNL) about an agreement in 2008 between Anglian Water and a developer concerning the supply to a new development at Fairfields, Milton Keynes. Under the new appointments and variations (NAV) regime (previously the inset regime), new entrants (such as IWNL) need to obtain an upstream off-site bulk supply from the surrounding incumbent (in this case, Anglian Water) to enter the market.

While Ofwat is continuing to investigate, it has said in its Statement of Objections that Anglian Water’s pricing for the Fairfields site ‘may have resulted in a margin squeeze and excluded competition’[16] and, subsequently, that ‘an equally efficient competitor could not have matched Anglian’s offer’.[17] It therefore appears that Ofwat has undertaken some form of ‘equally efficient’ imputation test, although the precise cost benchmarks used are unclear. For its part, Anglian Water ‘strongly refutes’ the claim of any wrongdoing, citing problems with Ofwat’s market definition, and the returns assumed in assessing whether a margin squeeze had occurred.[18] It is not clear when a Decision on this case will be made.

Leveraging through discrimination

The Bristol Water case, initiated by Ofwat in March 2013,[19] stemmed from two separate complaints made by Aquamain and Energetics into the price and non-price terms offered by Bristol Water for services enabling the provision of new infrastructure by self-lay organisations (SLOs) to new development sites. While some works are ‘non-contestable’, and must be completed by the incumbent water company, ‘contestable’ works can be supplied either by the incumbent water company (requisition option) or by SLOs (self-lay option).[20]

Importantly, Ofwat noted the wider ‘strategic significance’ of the case for market opening in the sector, in that it concerned potential discriminatory behaviour by a vertically integrated company that could have an adverse effect on competition in a prospectively competitive market.[21] In this case, Ofwat reached tentative conclusions on market definition and dominance.[22] The regulator also set out four prongs to how the leveraging of a dominant position in one market into another might occur (through applying dissimilar conditions to equivalent transactions).

• Vertical integration—according to Ofwat, providing the ability and incentive for Bristol Water to leverage its dominant position in upstream services (services that SLOs need to compete) to foreclose effective competition in downstream (contestable) services.

Vertical integration made possible the following practices.

• Lower requisition option prices—stemming from differences in the treatment of costs in the calculation by Bristol Water of self-lay and requisition quotations, which could potentially mean that the amount presented to the customer as being recovered by Bristol Water is less for the requisition than for the self-lay option.

• Higher self-lay option prices—stemming from the additional charges that Bristol Water required of SLOs, and concerns regarding their transparency and rationale, their level, and differences in their application between SLOs.[23] This had the potential to exclude ‘equally efficient’ competitors.

• Non-price terms—Bristol Water’s interactions with SLOs, in terms of communicating information to SLOs and developers on the processes and requirements they needed to satisfy to access the upstream services provided by Bristol Water. This had the potential to prevent the entry and expansion of equally efficient competitors.

Ofwat did not reach a final Decision on these issues since, at an early stage in the case, Bristol Water notified Ofwat that it was interested in offering a comprehensive package of commitments to address the potential concerns.[24] Part of Bristol Water’s commitments were process measures, in terms of policies and procedures, transparency in the calculation of costs (for requisition versus self-lay), and non-discrimination in terms of charges and interactions with SLOs. However, the main commitments were structural:

• Bristol Water proposed functional separation of its upstream (wholesale) business from its downstream (retail) business for providing contestable developer services;

• those parts of the upstream organisation providing services to developers and SLOs for self-lay queries would be separate from those letting and managing Bristol Water’s term contract for delivering its competing requisition option;

• there would be greater transparency in information flows between the wholesale and retail functions.

Ofwat noted that ‘the proposed structure would reduce Bristol Water’s potential ability (although not the potential incentive) to leverage its dominant position’.[25] The package of commitments would apply for three years (beyond which new market codes would be developed as part of the Water Act 2014 reforms).

While the case provides some precedent on market definition, market power and competition concerns, it appears to have stopped short of considering in more detail relevant cost benchmarks and returns, or undertaking further analysis of foreclosure effects. While Ofwat is minded to accept the commitments, the industry body for SLOs has expressed concerns regarding their subsequent implementation.[26]

The structural remedy of functional separation was put into place voluntarily, in dialogue with the regulator, and, given the ‘strategic significance’ of the case, it will be interesting to see what this means for wider market opening in the sector, such as retail competition.

Leveraging through predation

In a prior case, Ofwat formally accepted the commitments proposed—the first time Ofwat accepted commitments under the Competition Act. These went further than the Bristol Water case, in that they involved complete divestment.

In April 2010 Ofwat received a complaint from ALcontrol Laboratories, alleging that Severn Trent Laboratories had won contracts to supply water analysis services to South Staffordshire Water and Yorkshire Water by pricing below ‘any relevant measure of cost’—i.e. predation. It was alleged that this below-cost pricing had been funded by cross-subsidy, with higher levels of pricing by Severn Trent Laboratories to Severn Trent Water, Severn Trent Laboratories’ sister company, and that the result had been the complete withdrawal of a competitor from the water analysis market.[27]

Ofwat considered two scenarios of predation:

• whether Severn Trent Laboratories itself was dominant in the market in which the predation occurred;

• whether Severn Trent PLC might have leveraged its dominance from the market of its core regulated business (Severn Trent Water) into the more competitive water analysis market.

Ofwat’s main concern was with the latter, and that the structural link between Severn Trent Water and Severn Trent Laboratories might have facilitated this leveraging.

As per the Bristol Water case, Ofwat received commitments that brought the case to a halt—in this case the divestment of Severn Trent Laboratories.[28] This meant that Ofwat did not establish whether Severn Trent Laboratories or Severn Trent Water engaged in predatory pricing during the course of the investigation. There had also been no detailed work on market definition or dominance.

Ofwat acknowledged that predation, as a ‘rule of thumb’, was a ‘serious’ abuse, and that the (then) UK Office of Fair Trading (OFT) would not normally accept binding commitments in cases involving serious abuse of a dominant position. However, in its final Decision, it stated:

Ofwat has had regard to this guidance in reaching our decision. It is important to note that as a concurrent competition authority the decision to accept commitments is at our discretion. In this case we carefully considered whether pursuing the case would be of benefit and have concluded that there are special features of this case which means that it is appropriate for us to use our discretion and to take a different decision to the guidance that is offered by the OFT.[29]

In Ofwat’s view, the commitments offered were sufficiently ‘exceptional’ to justify closing the case, in that they permanently removed the potential to leverage market power between contestable and non-contestable markets. Ofwat also found no evidence of intent to exclude competitors. However, ALcontrol disagreed with the strength of commitments offered, in that they did not address Severn Trent Laboratories’ dominant position in water analysis, or restore competition to the relevant market. ALcontrol also argued that the circumstances of the case were not ‘exceptional’, and that halting the case could undermine the deterrence effect of the Competition Act. Ofwat’s view was that it was beyond the remit of the commitments to ‘restore’ competition in the market, and the regulator also disagreed with the other points made.[30]

Taking stock, and the future?

Across the Anglian Water, Bristol Water and Severn Trent cases, Ofwat has now undertaken some work on market definition and dominance, but it is less clear as to what work has been undertaken on relevant cost benchmarks or on the effects of the alleged conduct on competition. To explore competition concerns, the regulator appears to be leaning towards an equally efficient competitor test (as per the Anglian Water and Bristol Water cases), although the precise benchmarks used are as yet unclear. Interestingly, in a recent Decision using its sector-specific powers in relation to bulk supply pricing, Ofwat stated that it had considered whether an equally efficient operator could earn a sufficient margin.[31] Ofwat has similarly stated that companies’ special agreements with non-household customers should allow an as-efficient entrant to enter the market.[32]

Ofwat’s recent approach has also been to secure commitments from the alleged infringing party, to deal with the concerns raised, rather than conducting a full investigation into whether an infringement has occurred. On a forward-looking basis, this may lead to an end-solution more quickly (and with less expense), although it does not impose a direct financial punishment on the alleged infringer for past behaviour (notwithstanding that requiring some form of separation does indirectly impose a cost).

As noted above, the Water Act 2014 will usher in several changes to the industry, in particular retail competition for non-household customers in April 2017. As part of the recent 2014 periodic review process, companies have been set separate price controls for wholesale and retail. At the retail level, companies have needed to set out default tariffs for non-household customers. In this regard, Ofwat has placed the onus on companies to be compliant with competition law:

For the avoidance of doubt, companies are responsible for ensuring that they are compliant with their duties and obligations (including competition law). In offering alternative tariffs, companies will need to further consider whether their proposals create any legal risks.[33]

Ofwat’s approach has echoes of the Deutsche Telekom case, in which it was ruled that a company was not exempt from competition law merely because the regulatory body had approved its charges.[34] Therefore, companies will need to ensure that their retail default tariffs for each customer class are not predatory, in the sense that they are priced below the relevant cost benchmark; and that their retail default tariffs and wholesale charges do not generate a margin squeeze that forecloses the market (which might occur if retail charges are too low and/or wholesale charges are too high).

Contact: Alan Horncastle