Flexibility

The ability to adjust electricity generation, consumption, or storage in response to grid needs — whether for balancing supply and demand, managing congestion, or maintaining power quality. Flexibility is the central concept of this wiki.

EU legal definition

Regulation (EU) 2024/1747, Art. 2(79) provides the first codified EU legal definition of flexibility:

“The ability of an electricity system to adjust to the variability of generation and consumption patterns and to grid availability, across relevant market timeframes.”

This definition was introduced by the Electricity Market Design Reform 2024 and entered into force 16 July 2024. It emphasises three dimensions: (1) system-level adjustment (not just individual assets), (2) variability in both generation and consumption, and (3) relevance across all market timeframes — from real-time balancing to seasonal planning. (Source - Electricity Market Design Reform Regulation (EU 2024-1747))

For comparison, Eurelectric (2025) defines flexibility as “the ability of an energy system to adjust both power generation and consumption in response to signals from the grid, or the market, to ensure security of supply and avoid blackouts” — closely aligned but from an industry/operational perspective. (Source - What is Flexibility in the Power Sector (Eurelectric 2025))

Why flexibility matters

The Electric Power Transmission system has a fundamental constraint: electricity must be generated at the same rate it is consumed. Historically, this balance was maintained by dispatching large centralized generators up and down. Two trends are breaking this model:

1. Variable renewables (wind, solar) produce when conditions allow, not when demand requires. Supply becomes less controllable.
2. Electrification (EVs, heat pumps) adds large new loads at the Electric Power Distribution level, creating local capacity constraints.

Flexibility is the set of mechanisms that maintain grid stability and efficiency despite these changes.

Svk’s long-term market analysis LMA2026 (June 2026) makes the adequacy case concretely: in high-demand scenarios for 2050, if price-responsive demand for hydrogen and e-fuel production is removed from the simulation, a significant number of shortage hours appears — underscoring that developing incentives for flexible electricity use is essential to resource adequacy, not optional. The same analysis shows rising reserve needs and balancing costs as variable generation grows. (Source - Svk LMA2026 Långsiktig Marknadsanalys, Long-Term Market Analysis)

Four-category flexibility taxonomy

Sweden’s original three-category framework (energi / balansering / överföring) comes from the earlier government assignment’s joint final report (Ei R2023:18, 2023). A fourth category — beredskap (preparedness) — was added by Power Circle’s 2024 factsheet “Vad är flexibilitet?”, reflecting growing total-defence/security framing in Swedish energy policy. This is the actual origin point: Energimyndigheten’s ER 2026:14 (the 2026 annual indicator report) traces the addition explicitly to that 2024 factsheet, not to either of the two 2025 reports that popularized it. (Source - Energiindikatorer 2026 (ER 2026-14))

Both reports that popularized the four-category framework in 2025 built on that 2024 origin rather than reaching it independently: FlexAbility Delrapport 1 (Power Circle et al., published October 2025) restates it mapped onto the four components of “trygg elförsörjning” from Elmarknadsutredningen (SOU 2025:47) (Source - FlexAbility Delrapport 1 (2025)); Energimyndigheten’s ER 2025:20 (published May 2025 — earlier than FlexAbility’s report, but later than the 2024 factsheet) uses the near-identical label flexibilitet för energiberedskap (Source - Energimyndigheten ER 2025-20 Flexibilitet Målgruppsanpassad Information (2025)). Energimyndigheten’s own 2026 indicator report only tracks two of the four categories (överföring and balansering) with actual indicators so far, judging energi and beredskap not yet well-enough bounded for measurement — see Official flexibility indicators (ER 2026:14) below.

Two dimensions of flexibility

Temporal (when)

• Real-time / seconds: frequency regulation (FCR) — maintaining 50 Hz
• Minutes: automatic reserves (aFRR, mFRR) — restoring frequency after disturbances
• Hours: day-ahead and intraday market optimization
• Seasonal: long-duration storage, cross-border trading

Spatial (where)

• System-level: TSO needs — frequency balance, system adequacy
• Local/distribution-level: DSO needs — Congestion Management, voltage quality
• Behind-the-meter: customer-level optimization

Flexibility mechanisms

Rules-based vs market-based

Two approaches to procuring flexibility, often complementary:

Rules-based (implicit): flexibility mandated or incentivized through regulation and tariff design. Examples: dynamic network tariffs, connection requirements, grid codes requiring DER to provide voltage support, Villkorade Avtal (conditional connection agreements where curtailment is a contractual obligation). Lower transaction costs, less granular.

Market-based (explicit): flexibility procured through competitive markets. Examples: local flexibility markets, ancillary service markets open to demand-side resources, aggregator-dispatched portfolios. More efficient price discovery, higher transaction costs.

The EU Clean Energy Package makes this explicit in law. The Electricity Market Directive Art. 32 requires DSOs to procure flexibility via transparent, non-discriminatory, market-based procedures — rules-based approaches are permitted only where market-based procurement is uneconomic or would cause distortions. The Electricity Market Regulation Art. 3(j) requires that generation, storage, and demand response participate on equal footing. DSOs are expected to become neutral market facilitators (Directive Art. 31(5)).

The DSO flexibility challenge

This is where the wiki’s focus lies. DSOs (elnätsföretag in Sweden) face a new challenge: managing an increasingly active distribution grid with bidirectional power flows, local congestion, and voltage quality issues — all while remaining a regulated neutral party.

Key questions:

• How should DSOs procure flexibility? (markets vs. contracts vs. tariffs)
• How do DSO flexibility needs interact with TSO flexibility needs? (coordination, priority)
• What digital infrastructure is needed? (platforms, data standards, APIs)
• What regulatory framework enables this? (Ei rules, EU transposition)

Quantified flexibility potentials (2030)

FlexAbility (2025) quantified realistic technical maximum potentials across 14 resource types at hourly timescale for 2030. Seasonal totals at hourly timescale:

Short timescales (seconds–hours): dominated by Demand Response and Energy Storage. Long timescales (days–seasons): dominated by hydro (and gas turbines as reserve).

Investment slowdown risk: during the FlexAbility project period (2023–2025), wind, hydrogen, and industrial electrification forecasts were all revised downward — the 2024 preliminary potentials were higher. The 2030 numbers carry meaningful uncertainty, with the risk on the downside. (Source - FlexAbility Delrapport 1 (2025))

Generation-side regulability by technology

FlexAbility’s per-resource potentials above answer how much; Energimyndigheten’s ER 2025:20 answers why for each generation technology — the specific technical, economic, and regulatory factors behind each number:

• Vattenkraft — the historical backbone of Swedish flexibility. Flexibility is concentrated in ~255 “klass 1” plants, which contributed >98% of the regulation contribution across 365-day/28-day/1-day timescales (2009–2014 data). Fleet-wide ramp rate up to 3 GW/h. Expansion is capped by protected nationalälvar, so growth comes from uprating existing plants: Sweco (2016) estimated 3–4 GW technical potential, AFRY (2024) ~1.3 GW realisable by 2035.
• Kärnkraft — technically capable of load-following (practised in Sweden in the 80s/90s, still used in France/Germany today), with ramp rates regulated under SvKFS 2005:2 (PWR 5%/min, BWR 10%/min, both in the 60–90%-of-rated-power range). Constrained by PCI, xenon poisoning, and minimum-load levels that rise with fuel burnup. Theoretical upregulation potential at an assumed 65% minimum load: 2.4 GW — technical only, not economic; four cost categories (capital, O&M, fuel economics, personnel) all rise with flexible operation, and no new nuclear is modelled before 2035.
• Solkraft — near-zero marginal cost means near-zero incentive to regulate down except near/below-zero spot prices (Jan 2025: only 30 MW prequalified on FCR-D down). Growth is slowing after the announced 2026 removal of the household tax deduction.
• Vindkraft — already the third-largest prequalified technology on Svk’s balancing markets after hydro and storage. Theoretical up-regulation potential (at an assumed 5% reglerbar-kapacitet rate) is 0.5–1 GW but “helt beroende på ekonomiska förutsättningar” (entirely dependent on economic conditions).
• Kraftvärme — production is tied to heat demand, which peaks exactly when electricity-flexibility is least available (heat demand maxed in winter). Facing a structural headwind: rising biomass prices (EU fossil-substitute demand competing for the same biomass) can create a perverse incentive to conserve fuel for heat rather than flexible peak-price electricity generation. Capacity is expected to decline, not grow, across Energimyndigheten’s scenarios.
• Gasturbiner — ~20 units, 1.3 GW, used in Svk’s överbelastningshantering; low capital cost but high running cost keeps them at a handful of hours’ use per year (toppload only).

Aggregate: Energimyndigheten’s own estimate is 6–10 GW of theoretical additional flexibility from new generation capacity by 2035 (out of 20–32 GW total new capacity across its long-term scenarios), conditional on nuclear reinvesting in load-following capability. (Source - Energimyndigheten ER 2025-20 Flexibilitet Målgruppsanpassad Information (2025))

An industry counterweight: theoretical potential vs. practically realisable

Svenskt Näringsliv’s 2026 flexibility policy program pushes back explicitly on how figures like the above get used in debate: Energimyndigheten’s commonly-cited 5–15% industrial demand-flexibility potential is theoretical, and what’s practically realisable given order books, technical constraints, and plant economics is much harder to establish — in its assessment, “näringslivets efterfrågeflexibilitet har regelbundet överskattats” (industry’s demand flexibility has been regularly overestimated). Swedish law and regulation, in this view, are not actually designed to release whatever flexibility potential does exist — a structural gap, not just a measurement one. (Source - Svenskt Naringsliv Startprogram Flexibilitet (2026))

Official flexibility indicators (ER 2026:14)

Energimyndigheten’s May 2026 annual indicator report is the first to compile official EIFS 2022:5 data (Ei’s mandatory smart-grid reporting) into a flexibility indicator series — covering only two of the four taxonomy categories so far (överföring and balansering; energi and beredskap are judged not yet well-enough bounded for measurement).

The headline adoption number: across local, regional, and transmission grid levels combined, only 7 Swedish grid companies report ever having called on a flexibility market for up- or down-regulation, in either 2023 or 2024. Bilateral-agreement counts grew faster (22→32 network-company↔producer agreements, 233→393 network-company↔user agreements, 2023→2024) than flexibility-market usage did — bilateral deals, not market participation, are where the growth is actually happening. Connected storage capacity roughly doubled or more at every grid level from 2023 to 2024 (e.g. local-grid directly-connected storage: 128→531 MW), and pre-qualified balancing-market storage grew 25× in two years (240 MW → 6,060 MW, Jan 2024 → Jan 2026) — by far the fastest-growing pre-qualified technology category.

A new causal mechanism for 2024–2025 balancing prices: flow-based capacity calculation (introduced October 2024) left less inter-area transfer capacity available for balancing, forcing more frequent, more local activation — which combined with the March 2025 automated mFRR-EAM launch (raising each bidding area’s own-capacity requirement) to push up local mFRR prices sharply, especially in SE4. Separately, Sweden is structurally a price-taker on the shared Nordic aFRR capacity market: only 7.5% of Sweden’s own aFRR-up requirement was met by Swedish bids in December 2025 (62% for aFRR-down), the rest imported from more competitively-priced southern Norway and Finland — a self-sufficiency gap not previously quantified in the wiki. Sweden joins the European PICASSO aFRR platform end of 2027. (Source - Energiindikatorer 2026 (ER 2026-14))

TSO flexibility needs

It’s not only DSOs that need flexibility. Svenska kraftnät faces a structural north-south transmission bottleneck across the four Bidding Areas (SE1–SE4), with the connection queue (>175 GW applied) far outpacing grid expansion (SEK 225 billion over 2025–2035). The NordSyd initiative will increase physical capacity over 10–15 years, but flexibility is essential in the interim. Svk explored conditional connection agreements at TSO level (analogous to Villkorade Avtal at DSO level) but found “managing full-scale implementation proved to be impossible with the systems and tools available today” — underscoring that the digital infrastructure gap exists at both TSO and DSO levels. (Source - Svk Network Development Plan 2026-2035)

The Network Code on Demand Response

The Clean Energy Package sets the principles; the Network Code on Demand Response (NC DR) will operationalize them into binding rules. The forthcoming regulation will establish: a standardized pathway from resource registration to market participation (CU → SP → SPU/SPG → prequalification), mandatory national flexibility registers, harmonized product attributes for local services, and detailed TSO-DSO coordination frameworks. ACER’s recommendation (March 2025) substantially revised the original ENTSO-E/EU DSO Entity proposal, redistributing provisions across four legal instruments. Key unresolved issues include minimum bid size (0.1–1 MW), the pace of EU harmonisation vs national flexibility, and the practical burden on small DSOs.

The CAPEX bias problem

A structural barrier to flexibility in Sweden: the intäktsramsreglering (revenue cap regulation) creates a CAPEX bias that incentivizes DSOs to invest in grid expansion over purchasing flexibility services, even when flexibility would be more cost-effective:

• Capital investments earn regulated return on the capital base — attractive for DSOs
• Flexibility procurement is classified as operating expenditure (löpande påverkbara kostnader) subject to efficiency requirements and historical-period lag
• Result: DSOs prefer building infrastructure over buying services, distorting the choice between grid expansion and flexibility

(Source - Ei Flexibility in Distribution Grids (2023))

The reform: TOTEX and lösningsneutralitet (from RP5, 2028)

Ei’s methodology for the 2028–2031 supervisory period (RP5) introduces a TOTEX (Total Expenditure) approach to the cost efficiency incentive, directly addressing this bias:

• TOTEX benchmarking includes both capital costs (CAPEX) and operating costs (OPEX) in a single efficiency framework
• A company that achieves congestion management through flexibility procurement (opex) scores identically to one that achieves it through grid investment (capex) — this is lösningsneutralitet (solution neutrality)
• Result: the regulatory incentive no longer systematically favors building over buying

Parallel changes reinforce this: switching from kapacitetsbevarande (market-valued assets) to förmögenhetsbevarande (acquisition-cost-valued assets) removes the upward drift in the capital base that made CAPEX returns especially attractive; and a new connection fee deduction ensures customer-funded investments don’t earn a double return.

Ei confirmed enstegsmetoden as the chosen TOTEX application method (December 2025). In May 2026, three further design decisions were announced: (1) no general efficiency requirement — actual cost outturns already embed industry-level productivity improvement; (2) full cost coverage at the third quartile (Q3) — companies at Q3 efficiency get 100% cost coverage, better performers get a revenue frame increase, below-Q3 companies get a deduction; the Q3 threshold is relative (moves as the sector improves), following UK practice; (3) 8-year realiseringstid (realization time) retained. Two parameters remain under investigation in spring 2026: adjustment for heterogeneous company conditions (electricity prices across bidding areas, ground conditions) and the maximum cap on the incentive’s revenue impact. Regulations enter into force H1 2027; revenue frame decisions by October 2027. (Source - Ei Inriktning intäktsramar 2028-2031 (2025), Source - Ei Effektiviseringsincitament Webb (2026-05-12))

International research reaches the same diagnosis. A 2026 ISGAN and BRIDGE discussion paper (ISGAN and BRIDGE discussion paper) finds that 7 of 11 surveyed jurisdictions use cost-of-service or hybrid remuneration that acts as a barrier to flexibility for investment deferral, that 6 of 11 have not yet included flexibility in remuneration, and that support often lapses when pilots end.

The elmarknadshubb

A central electricity market hub (Elmarknadshubb) is widely identified as critical missing infrastructure for flexibility: it would give all market actors equal access to customer metering data, enable data portability, and lower entry barriers for aggregators. The original 2015 Svk mandate was paused in 2020 and formally cancelled in September 2025; the concept has been reframed as a centralt datahanteringsverktyg, now the subject of a joint Ei and Svk proposal delivered in September 2026 (government decision pending). The Network Code on Demand Response‘s FIS requirement is a direct driver. See Elmarknadshubb for full history and current status. (Source - Uppdrag Centralt Datahanteringsverktyg (2025))

A partial solution within the existing infrastructure is the Berättigad part mechanism, operational from 1 June 2025. It gives aggregators and energy service companies a standardized right to access consumer metering data at 15-minute resolution via PRODAT messages through each DSO — replacing the prior power-of-attorney (fullmakt) handling of meter-value reporting (Berättigad part). The limitation is that bilateral agreements with each of Sweden’s 168 DSOs remain necessary, keeping the transaction cost high for entrants with geographically scattered customers (Distribution System Operator). DHV/FIS is intended to reduce this friction. (1 June 2025 operational date and 15-minute PRODAT resolution per Source - Svensk Elmarknadshandbok 26A (2026))

Flexibility for preparedness

The fourth category of the taxonomy above — beredskapsflexibilitet, resources activated only under crisis conditions — including the four system operating states, ö-drift/black-start/synthetic-inertia capabilities, the compensation-mechanism landscape, the försörjningsanalys planning methodology (ResiliENt Syd), and connected-DER cybersecurity risk, is covered in Beredskapsflexibilitet (split out 2026-09-05 as this had grown into its own coherent sub-topic).

Data gaps

• Art. 31.3 implementing regulation — a follow-up search (August 2026) found requirement text matching the gap’s description (capacity information published within 3 months of a connection request, then at least quarterly until final decision) alongside Ei’s confirmation that updated nätutvecklingsplan regulations (EIFS 2026:4) entered force 30 April 2026 — but the two weren’t confirmed as the same instrument; needs direct verification against EIFS 2026:4’s actual text