This post is the third in a series on the European Commission’s initial assessment of the package of measures by which the UK Government proposes to provide financial support for the proposed new nuclear generating station at Hinkley Point (HPC) by NNB Generation Company Limited (NNBG).  In this post we focus on the Commission’s analysis of the UK Government’s arguments that its support for HPC does not constitute state aid within the meaning of Article 107(1) of the Treaty on the Functioning of the European Union and that HPC would be performing a service of general economic interest (SGEI), effectively meaning that it fell outside the state aid rules (for a summary of the overall framework of the Commission’s appraisal, click here).

As noted in the previous post, any “no aid” decision, or categorisation of the HPC as an SGEI, effectively turns on the application of the so-called Altmark criteria.  The quality of the Commission’s arguments in this strategically important area is variable. 

The Commission begins by making the point that it sees a service of general economic interest (SGEI), such as the Government claims would be provided by NNBG, as a service which an undertaking would not supply if it were considering its own commercial interest, and which serves a general economic interest.  In the context of HPC, the Commission’s starting point is that NNBG’s service would be to supply (baseload) electricity; yet that, the Commission says, is “normally considered a commercial activity and a market in which competition takes place”.  It suggests that nuclear generation is no exception to this principle, noting the “nuclear plants which are operated commercially” in the UK by NNBG’s parent EDF, and the “UK’s own assessment” that “private investors [would]…invest in nuclear energy in the UK by 2030 at the latest.  Finally, if the service which would not be provided without aid is the construction of HPC by an earlier date than the private sector would otherwise build new nuclear capacity, the Commission suggests that the UK has not made a convincing case for such early construction being in the general economic interest on security of supply or decarbonisation grounds.  

Almost every assertion that the Commission makes in the two pages or so which it takes to reach these provisional conclusions on “the existence of a SGEI” is questionable in terms of its accuracy or its relevance.  Electricity generation is indeed a commercial activity.  That does not mean that the construction of a new nuclear reactor is a service that will be provided without state aid.  Nor does the existence of the UK’s legacy nuclear fleet help the Commission’s case, constructed as it was by the CEGB in the days of nationalisation.  The Commission’s dismissal of security of supply and decarbonisation as interests served by the putative service of constructing and operating HPC is similarly one-sided.  For example, it effectively denies that there is any benefit in securing decarbonisation sooner if you think the market will decarbonise a few years later, and it ignores the effects on both security of supply and decarbonisation in both the longer and the shorter term which assurance about the viability of HPC (in the form of state aid clearance) could have.

The first Altmark criterion (which is also key to any attempt to justify a measure under Article 106(2)) is that the beneficiary be entrusted with a public service obligation (PSO).   The Commission argues that provisions of the CfD which limit the return which NNBG can make on its investment in HPC or penalise it for late delivery of the project are not capable of being PSOs.  The best claim that the CfD has to being regarded as placing NNBG under an obligation is that if it does not build HPC (or delivers it late), it will receive no money (or less money) under the CfD.  The Commission appears to be suggesting that in order to be a PSO, an obligation (e.g. to commission HPC by a certain date) has to be “enforceable” by some means other than the payment or non-payment of aid.  If the Commission is right about this, it may have implications for the design of the CfD contract terms more generally.  However, the Commission only engages very briefly with the case of Fred Olsen, which appears to offer some support to the UK Government’s view.  In that case, which concerned ferry services, the Court of First Instance remarked that the fact that an operator “unilaterally abandoned or altered the conditions for the operation of some maritime routes indicates at most” that it “failed to honour some of the obligations imposed on it by the provisional arrangements”, and seems to have found that not even the fact that an operator was subsidised at its own request prevented it from satisfying the first criterion.   

Looking beyond the particular circumstances of HPC, what the Commission seems to be saying here could have implications for the financing of other CfD-subsidised schemes.  If the Altmark criteria do truly require the state to have the means of enforcing compliance with requirements, such as the construction of HPC, that go beyond the stimulus provided by the absence of CfD revenues if no electricity is generated, it may not be possible to construct bankable CfDs which satisfy those criteria.  Elsewhere, in the analysis of Article 106(2) arguments, the Commission suggests that the absence of a true PSO is what excuses the UK from having to comply with the public procurement rules in respect of letting a CfD in respect of HPC, and that, conversely, if the requirements imposed on NNBG could be shown to constitute a PSO, the UK Government would have failed in its alleged obligation to follow the public procurement rules.

The Commission broadly accepts that the second Altmark criterion is satisfied – i.e. that the parameters on the basis of which the compensation is calculated are established in advance in an objective and transparent manner.  However, when it comes to the third criterion, that the compensation cannot exceed what is necessary to cover the costs incurred in the discharge of the PSO, its assessment is much less favourable.  Moreover, some of the arguments which emerge here also read across into the Article 106(2) and Article 107(3) analysis.

The Commission is concerned, firstly, that the Government does not appear to have a firm view of what the costs of discharging the PSO are (making the level of compensation by definition hard to assess); secondly, that the level of profit that NNBG can expect to earn over the lifetime of the CfD was negotiated with NNBG rather than being “established by reference to the rate of return on capital that would be required by a typical undertaking considering whether or not to provide the alleged SGEI”; and thirdly, that because the 35 year lifetime of the CfD is shorter than the 60 year lifetime of HPC, NNBG could earn super-normal profits in years 36 to 60. 

It is hard to comment on the first two of these points as far as HPC is concerned without access to the UK’s submissions to the Commission, although in response to the second one might ask: what is a “typical undertaking” considering whether or not to build HPC, let alone (as the Commission goes on to elaborate) “the average cost structure of efficient and comparable undertakings in the sector under consideration” – none of which have been built under exactly the same regulatory regimes as HPC would be built and operate under?  Moreover, for much of the period during which the Government was negotiating with NNBG, it was simply the only undertaking willing to contemplate any form of investment in new nuclear build in the UK.  On the other hand, prospective recipients of aid under enduring CfD regime for renewables in mind regime may take some comfort in this context from the fact that their strike prices will not be the result of bilateral negotiations. 

But the Commission’s point about the duration of the comparative lifetimes of the CfD and the generating station is something on which we can comment in the HPC context.  The strike price, we are told, has been set at a level which is designed to ensure that NNBG covers the costs of construction and operation and makes a return of 9.87% on the project as a whole over its lifetime (in post-tax, nominal terms).  Yet, as the Commission points out, once the CfD expires, the profitability of the plant is uncertain because the level of revenue accruing to the operator from the sale of electricity is no longer controlled by the strike price mechanism.  This makes it harder for the Commission to rule out the possibility of overcompensation during the post-CfD period of the plant’s operation.  The Commission suggests two ways of dealing with this problem: making the CfD coterminous with the life of the plant, or providing some means for the state to recover any overcompensation within the CfD itself (effectively a gain-share provision for the period when the strike price mechanism no longer applies).  One problem with the first of these, if taken in isolation, is that it is not possible to predict the lifetime of a plant with certainty when the strike price is initially calculated.  

In principle, it would seem that this arguments is not unique to the case of HPC and could be applied to the wider CfD regime.  The differences are that the periods of time involved – both CfD durations and plant lifetimes – are shorter for non-nuclear projects, so that the calculations are less dependent on very long range predictions of electricity prices; and that there is more comparative data on which to assess technology costs.  Whether the Commission will consider these differences to be sufficient for it to take a more favourable view of this aspect of the wider CfD regime than it has so far in the case of the HPC package remains to be seen.  In this context it is curious that the Commission states that “nuclear production, which requires very high levels of capital for the investment in the construction and hence before revenues can be generated, while also being characterised by a relatively low level of operating costs once the plant has been built, has few, if any, equivalents in commercial activities”: the CfD regime as a whole is surely predicated on the assumption that all the technologies it covers (renewable, nuclear, CCS) have in this sense a similar cost profile.

The fourth Altmark criterion is that where the undertaking which is to discharge a PSO is not chosen through a public procurement process, the level of compensation must be determined on the basis of an analysis of the costs which a typical, well run, undertaking would have incurred.  Here again, the problem is in finding the appropriate comparator.  Unsurprisingly, the Government has commissioned a review of NNBG’s cost estimates to determine whether they are “reasonable”.  The Commission says that this is not what the Altmark criterion requires.

The final sections of the Commission’s analysis of the UK’s “no aid” arguments deal with the credit guarantee and the proposal to compensate NNBG in the event of a “political shutdown” of HPC.  On the credit guarantee, the Commission essentially reserves judgment owing to the lack of detail available.  However,  it does lay down a marker when it observes that the guarantee “seems to differ from ordinary debt guarantees in that it would be drawn before equity, apart from equity already spent…It would therefore appear that [it] might diminish the risks borne by equity holders”.  The Commission appears prepared at this stage to accept the UK’s argument that political shutdown proposals do not constitute state aid, subject to the provision of more information “on whether this compensation…would also be available to other market operators placed in a similar situation”.  This is intriguing.  It is presumably possible that the UK Government would be prepared to offer a similar deal on political shutdown to another nuclear operator, but such a deal is clearly not on the table for operators of renewable technologies, for example, and whilst a political shutdown of UK wind farms may be a more remote possibility than something like the German reaction to Fukushima, will that point be sufficient to satisfy the Commission that the enduring regime for renewables should not in this respect be “levelled up”, to confer on its beneficiaries the additional protection offered to NNBG?

It is clear that the Commission is highly reluctant to reach a finding that there is “no aid” in the HPC package, or to find that there is an SGEI within the meaning of Article 106(2).  It does not want to treat nuclear power as a special case.  Yet unless it is prepared to recognise that nuclear power is not just another source of baseload electricity, how could the Commission find that there is no aid (or that there is a SGEI) in a CfD negotiated directly between a Government and the beneficiary undertaking which includes a generous strike price, a 35 year term and investor protection in the event of political shutdown - and still realise its ambition of cutting back on subsidies for renewables? 

To go by the evidence of the Commission’s initial assessment, the Government would – rightly or wrongly – have to do a lot more work both in terms of scheme design (including changing some features of the currently proposed CfD arrangements) and in terms of arguing its corner with the Commission if it is to persuade the Commission that there is “no aid” to NNBG.  It is possible that the Court of Justice might be more sympathetic to the Government on some of these points, but EU litigation would not help the timeliness of the delivery of EMR objectives.