C(2026)126 Future Proof Network Charges Guidelines

Commission Notice on Guidelines on future proof network charges for reduced energy system costs, C/2026/126, OJ C series, 9 January 2026.

Non-binding guidance to national regulatory authorities on designing electricity network tariff methodologies. It implements Action 1(a) of the Affordable Energy Action Plan (COM(2025) 79) and builds explicitly on ACER’s biennial network tariff practices report (26 March 2025).

The Notice does not create obligations: tariff competence sits with the NRA under Art. 59(1)(a) of the Electricity Directive, and the guidance repeatedly frames itself as supporting NRAs while respecting national specificity. Its force is as the Commission’s stated view of what Art. 18 of the Electricity Regulation now requires.

The legal hook — Art. 18 as amended

Art. 18 of the Electricity Regulation requires network charges to be cost-reflective, transparent, to take account of network security and flexibility, and to reflect actual efficient costs. The 2024 amendment (Regulation 2024/1747) added the requirement that tariffs “consider both capital and operational expenditure” — a TOTEX approach — to give operators short- and long-term incentives including anticipatory investment.

The Notice reads this as a deliberate move against the CAPEX bias: tariff methodologies should encourage “holistic thinking about system needs”, letting NRAs incentivise investment in flexibility and optimal grid use “rather than a more traditional approach of building out the grid”. Art. 18(2) is cited for incentives toward renewables integration, flexibility, grid-optimisation solutions and demand response. See Flexibility › The CAPEX bias problem and RP5 Revenue Cap Methodology (2028–2031).

Art. 6a is treated as the connection-side counterpart: where a flexible connection does not offer unrestricted access, the network charges applied should reflect that reduced access. See Flexible Connection Agreements, Villkorade Avtal.

The scale argument

• Flexibility requirements will more than double by 2030 and be seven times current levels by 2050 (JRC 2023).
• 60–90% of new connections are at distribution level (Clean Energy for all Europeans Package Impact Assessment, 2016); 70% of renewables are expected to be distribution-connected by 2030 (Eurelectric, Grids for Speed, 2024).
• Distribution charges dominate: ~EUR 60 bn against ~EUR 20 bn transmission across the EU in 2023.
• Average low-voltage network utilisation across the EU is 2–21% — the headroom that time-of-use signals are meant to unlock (JRC 2018 DSO Observatory).
• Netherlands: better network use could cut cumulative investment needs by up to EUR 22.5 bn to 2040.
• Germany (Agora Energiewende): dynamic tariffs including dynamic grid charges could shift over 100 TWh of load by 2035 — over 10% of annual consumption, about half of household consumption — and almost halve the grid-expansion cost of integrating flexible demand, from EUR 10.5 bn to EUR 5.8 bn.

Six design elements

a. Transmission/distribution interplay. Cost cascading (top-down, higher to lower voltage) is applied in all member states and made sense when power flowed downward. With generation increasingly distribution-connected, NRAs “may wish to consider altering how they apply the cost cascading principle”, and should coordinate with each other on new approaches.

b. Reducing peak load through capacity charges. NRAs should add a capacity element reflecting peak load, combined with a time-of-use energy element. Evidence: Belgian transmission capacity charges cut the industrial synchronous peak 4% between 2019 and 2024 while withdrawal volumes rose 10%; Flanders’ 2023 distribution capacity tariff produced a 1–3% peak reduction, mostly from changed EV charging.

c. Time-of-use elements. Should be included to correlate cost allocation with peak network usage. Spain’s six-period structure (75% capacity-based at transmission, 84.6% at distribution) cut transmission and distribution charges 5.6% between 2019 and 2020, with even non-responding low-voltage consumers seeing 0.6%.

d. Locational elements. NRAs should promote locational signals to steer siting of generation, demand and non-fossil flexibility toward available capacity. Practice examples from Denmark, Portugal and Romania.

e. Specific calculation methods for user categories. Special regimes for energy-intensive users, prosumers or energy communities are permitted but must be justified against cost-reflectivity: the NRA should show, from consumption profiles, that the category genuinely creates lower network cost per MWh. Pure volume discounts or unjustified exemptions may fall foul of Art. 18. Special regimes should be paired with conditions requiring the user to actually lower grid impact.

f. Storage. Tariff methodologies “should most importantly be designed so as not to hinder the development and roll-out of storage”, and in particular should prevent double-charging — storage paying both generation and consumption charges. But NRAs should equally consider whether full exemption is justified, since exempt storage contributes nothing to cost recovery. Charging and discharging that ignores congestion status would make storage part of the problem.

Flexible connection agreements — the Dutch three-model example. Fully flexible (congested areas only; user pays for monthly peak, not contracted capacity); minimal availability (85% guaranteed capacity, cannot increase actual peak); timeslot agreements (contracted access windows; pays part contracted-capacity tariff plus monthly peak).

Transition management

The Notice is unusually explicit that design quality is not enough. Tariffs must be understandable and predictable enough for consumers to react; ability to respond varies by user profile; where capability is limited (no smart meter), methodologies should allow opt-ins or a gradual, well-communicated shift. Peak periods and price levels should be defined ex ante with sufficient anticipation. Aggregators are named as the mechanism that reduces perceived complexity for consumers and speeds automated response.

Sweden in the Notice — and why it is already out of date

Sweden appears in two member-state practice boxes, both as a positive example:

• Under capacity charges: “the NRA announced in April 2025 comprehensive plans to introduce a new grid tariff consisting of four components, including a new time-differentiated capacity charge, determined by peak load consumption.”
• Under time-of-use: “all DSOs will move to dynamic time-of-use network tariffs from 2027”, with the flexibility potential quantified as 1.2 million of 4.7 million households on heat pumps plus rising EV penetration. Transmission-level tariffs already carry a time-differentiated energy component based on actual hourly prices per bidding zone.

Both describe the EIFS 2022:1 regime. Ei repealed EIFS 2022:1 on 17 June 2026, five months after this Notice, removing the binding requirement that tariffs include an effektavgift from 1 January 2027, and also repealing ställningstagande Ei2025:06 which rested on it. A replacement effektavgift model is due 12 April 2027. (Source - Ei Effektavgifter webb (2026), Source - Ei Effektavgifter Uppdrag (2026))

So the Commission’s showcase example of a member state adopting exactly the design this guidance recommends was withdrawn by that member state within months of publication. Ei’s stated reason — that fragmented DSO-level models confuse customers and obstruct automatic steering services — is not a rejection of time-differentiation in principle, and the April 2027 model may reinstate it in a more uniform form. But between June 2026 and that model, Sweden has weaker tariff-side flexibility signals than the Notice describes, governed only by the general standards in ellagen and Art. 18 itself. See Swedish DSO Tariff Reform — Three Parallel Tracks (2025–2027).

Relevance to this wiki

• The clearest EU-level statement yet that TOTEX and anticipatory investment are what Art. 18 now requires, which is the frame for RP5 Revenue Cap Methodology (2028–2031) and the lösningsneutralitet debate.
• Supplies the cost-reflectivity test for special tariff regimes — relevant to any Swedish prosumer, energy-community or energy-intensive exemption.
• The storage double-charging position bears on Swedish battery connection economics; see Energy Storage.
• Gives Flexible Connection Agreements an explicit charging principle: reduced access should mean reduced charges.
• Its Sweden examples are a dated snapshot that the vault should not repeat uncritically.
• Named as the tariff-design companion instrument in Source - COM(2026)850 Retail Flexibility Report, which covers the retail-contract side of the same coordination question.

Data gaps

• Whether Ei’s April 2027 effektavgift model reinstates time-differentiation in the uniform form this Notice recommends, or moves further from it
• Whether Sweden’s regionnät/lokalnät split affects how the cost-cascading reconsideration in element (a) would apply, given the three-level Swedish grid structure now under review by Dir. 2026:83

Follow-up in legislation

The Commission’s 17 July 2026 proposal COM(2026) 600 would replace Art. 18 of the Electricity Regulation and empower a delegated act on a harmonised tariff methodology (Art. 61(5a)), so part of what this Notice sets out as non-binding guidance would become binding; the EU DSO Entity opposes the harmonisation step (DSO Entity’s reaction).