Charging is a key part of the Reformed National Market

The decision to reform the national market has been made, and we have seen a high-level view on the possible direction of travel for DESNZ.

The building blocks of the Reformed National Market (RNM) broadly aligned with the details that we had trailed in our previous article.

However, we now have some more detail on the potential direction of travel in the key area of network charging reform. With zonal pricing ruled out, network charging will be a key tool (along with the Balancing Mechanism) through which price signals are transmitted to influence locational siting decisions by investors. It is therefore welcome that Ofgem has set out its initial views on the role of transmission charges in the RNM and some of the principles likely to underpin their design going forward.

It is fair to say that the document is not extensive. It focuses on setting out the principles that Ofgem expects to follow, leaving lots still to think about and significant uncertainty as to where charges may end up.

In terms of the process from here, we know that Ofgem plans for the reforms to be delivered by 2029 and that it will lead on their design and delivery. DESNZ has already indicated that it plans to introduce primary legislation to give the necessary powers for both Ofgem and Secretary of State to amend codes and licenses to implement the reforms. What this means for the progress of the raft of existing proposals for transmission charging (TNUoS) reforms already in the modification process is unclear. While, the onus will be on Ofgem to decide on the way forward, rather than responding to industry proposals, we would expect Ofgem to review and rationalise the various proposals to be consistent with the new direction of travel, and then hope it continues to keep industry views as a key input into that decision process.

Here are the key things that stood out to us from Ofgem’s letter.

The role of price signals versus central planning

In our earlier post on the RNM, we highlighted that it would be important to understand the role of network charging alongside the step-up in strategic planning (i.e. SSEP and Gate 2 connections process etc.).

Historically, the argument has always been that you send investors locational price signals so that, in their choice of location, they internalise the cost they impose on the network. However, in a world of strategic network planning it feels valid to ask if network costs should be considered variable or sunk (i.e. because the network will be built out as planned in all scenarios, individual investor decisions will not affect network costs). The nearer we are to believing future network costs are sunk, the weaker is the argument for locational price signals.

Ofgem clearly sees a role for network charging alongside other non-price levers (which could include SSEP, CNSP, connections reform, and Capacity market or CfD capacity limits by location). However, Ofgem wants to better understand how important that role should be. It says they will work “with Government and NESO to explore the best combination of locational levers that can achieve an appropriate balance between coordination and competition. This will help us to determine the appropriate weight between price and non-price levers in guiding locational investment decisions”.

What Ofgem might conclude in terms of this balance is as yet unclear. A key consideration is likely to be the extent to which Ofgem believes it is right to rely principally on the non-market based processes to optimise the system. Ofgem might conclude that the move towards greater planning means individual locational choices (which are constrained to be within the envelope of those plans) do not materially impact network costs. Alternatively, they might believe that choice at the margin within the plan envelope can still impact network costs, or that there are risks around the effectiveness of the plans. This latter belief is more likely to be consistent with an ongoing important role for network charging signals.

In any case, as Ofgem notes, the case for sending locational signals to demand and storage may be stronger if they are not subject to spatial planning to the same degree. Although, we note that while it is clear demand is not subject to spatial planning, storage is included in the SSEP, so it is unclear why Ofgem believes it will not be subject to the same spatial planning as generation.

Charging signals should more clearly reflect ‘planned network spare capacity’

In terms of the design of network charges, while Ofgem has not given much detail on its plans, the principles that they have set out could be considered quite significant. Ofgem wants network charges to incentivise generation, demand, and storage to locate according to need as set out in the spatial plans (i.e. the plans and charges need to be working in tandem) so that long-term constraint costs are reduced. By linking the charges more closely to congestion, Ofgem believes the charges would be more “cost reflective”. In practice, this appears to mean two things in terms of network charge design:

First, charges should take account of the ‘availability of grid capacity’ (“spare capacity”) on the network; and
Second, charges at the point of connection should reflect the ‘expected state of the network and the energy system over time’ i.e. as opposed to a specific point in time as is currently the case.

With regard to the first point, taking account of spare capacity would represent a significant change to the methodology. Under the current methodology for calculating charges the network is assumed to have no spare capacity, and charges are calculated on the basis of incremental flows over the network from a connecting generator or load triggering a certain amount of new investment depending on where they are located. This is the so-called Investment Cost Related Pricing (ICRP) approach to charging.

Taking account of spare capacity implies something quite different. Charges would be low on those parts of the network with spare capacity, but as more capacity locates there (presumably consistent with the spatial plans) and spare capacity declines, charges should gradually rise. At the point new investment is triggered, charges would fall recognising the new spare capacity (See Figure 1).

Stylised representation of charges which recognise ‘spare capacity’

Source: Frontier Economics

Taken in isolation, this approach could imply more volatile charges as charges adjust to reflect the constantly changing amount of spare capacity. Although, it is worth noting that a system with more anticipatory network investment (e.g. as a result of the CSNP) would likely imply less volatility than otherwise might be the case. As an aside, conceptually, this type of volatility is not dissimilar to the locational signal that would be expected in a zonal market.

However, considered in conjunction with Ofgem’s second point, which is to make charges at the point of connection reflect the available capacity ‘over time’, this might imply charges that reflect the expected amount of spare capacity in a location over an extended period. What that period might be is unclear, but it could relate to a more reasonable investment horizon such as 10 years, or something closer to an asset’s lifetime.

Whatever the period, this alternative approach would suggest the need to forecast what spare capacity might be (based on the spatial plans) and set a charge which is somehow averaged over time. Ofgem is right to note that this is in contrast to the current approach which estimates annual charges based solely on the configuration of the current network.

This would be a very different approach to charging. However, what the effect might be on the level of charges is more difficult to determine. Estimating charges that account for spare capacity over time, is likely to smooth out the volatile “saw-tooth” charges in Figure 1.

However, conceptually, that may not be significantly different from the current approach. Whether charges go up or down is likely to depend on the specific approach to detailed design. While that is clearly uncertain, one interpretation of Ofgem’s rejection of the cap and floor proposal, is that Ofgem may believe that truly cost reflective charges set on this basis could be higher than current levels, or at least higher than the proposed cap.

It is also worth noting, that a practical example of how such a methodology could be implemented already exists in ScottishPower’s “OpTIC” proposal for TNUoS reform (CMP433). The principle behind this approach is to estimate charges based on the difference in revenues that a generator would expect to earn in a national market compared those expected in a zonal market, assuming a particular future “optimised” plan for the network. This is a fundamentally different approach to charging, but one which would take account of both the planned evolution of the network (i.e. it could incorporate the SSEP), and would reflect the evolution of spare capacity over time, in line with Ofgem’s principles.

Charges should be more predictable

If the changes we have discussed so far can broadly be considered as changes to improve cost reflectivity of charges, Ofgem also wants to make the charges more predictable. This could involve:

‘fixing TNUoS at the point of investment for a defined period of time’; or
sending the signal ‘through the connection charges to achieve a similar effect’ on the grounds that they are set at the point of investment and would therefore remain stable.

Whether Ofgem chooses to improve predictability by fixing TNUoS, or a deeper connection charge will have significant implications for implementation. However, irrespective of the route through which it is done, in principle, fixing charges in this way will typically imply a trade-off with the cost reflectivity of charges. If charges are fixed for an extended period of time, then by definition they cannot adjust to the evolving network conditions related to congestion. However, even if charges can change over time, if they are not predicable their effectiveness at driving efficient locational decisions will be limited and the unpredictability of charges will push up the cost of capital. If charges were predictable, fixing may still not be a problem for new investment if the charges are cost reflective at the point of investment. However, it could imply some distortion to closure decisions if charges do not change toward the end of an asset’s life.

While Ofgem is clearly showing willingness to face into any trade-offs, it must be considered carefully. Ofgem’s recent rejection of the proposed cap and floor (CMP444) demonstrated the importance of making sure that any potential loss in cost reflectivity is more than outweighed by tangible improvements to competition e.g. by improving certainty for investors. Ofgem judged this not to be the case for CMP444.

There will be a need for transitional arrangements

Ofgem expects transitional arrangements may be needed, noting that any reformed charges may be materially different to those assumed by investors at the point of investment. As a result, Ofgem might consider some transitional period of ‘phasing’ to the new charges, or ‘fixing the charges’ for existing assets to ‘a certain level post the implementation of the reformed charges’.

While it is not easy to understand the direction of any impact, the conceptual changes that Ofgem is considering are material enough to believe that these arrangements will be important. However, any arrangements will rely on an understanding of the extent to which existing investors are actually impacted i.e. Ofgem would need to take a view on what TNUoS charges would have been in the absence of reform. However, this is a challenging exercise. The value that is assumed for TNUoS (absent reform) would be critical. But it would depend on subjective judgments and would likely be difficult to relate to actual investor expectations at the time of investment.

It is also worth noting what this might mean for AR7 investors. They are unlikely to receive further clarifications on TNUoS ahead of the auction, and there is not enough detail in the Ofgem document to take any kind of view on the level of reformed charges post-2029. Therefore, they may need to trust in any transitional arrangements to make them whole relative to their expectations of TNUoS (absent reform). Or, if they believe this is likely to be imperfect of not possible at all, they will have to reflect the uncertainty in the bid price.