Source - Sweco Kartläggning av lokala flexibilitetsmarknader (Ei, 2025)

Document: Kartläggning av lokala flexibilitetsmarknader (Mapping of Local Flexibility Markets) Author: Sweco (consultant team) Client: Energimarknadsinspektionen (Ei) Date: 2025-04-07 Pages: 59 Type: Commissioned consulting report — follow-up to Sweco’s 2022 mapping

A second sweep of all Swedish local flexibility markets: current and historical market landscape, actor and liquidity analysis, interview study with nine actor groups, and ten prioritized recommendations for the national NC DR implementation.

Context and method

Ei commissioned Sweco to follow up on the 2022 mapping as markets had evolved significantly. Sweco’s method: literature review of official reports (CoordiNet D4.7.2, sthlmflex season 3, E.ON BeFlexible D5.1), direct data collection from E.ON/NODES on market outcomes, and semi-structured interviews with representatives from: network companies (market owners/buyers), platform providers (SWITCH, NODES), resource owners, aggregators/service providers. The report uses ACER’s September 2024 NC DR consultation draft as regulatory reference — not the final text.

Market landscape (as of April 2025)

Seven flexibility markets have existed in Sweden. Three active, four closed.

Active:

1. E.ON Energidistribution‘s flex markets — 9 areas at report date (now 12 in SWITCH as of V2025/26); Hässleholm, Södra Skåne, and Stor-Stockholm areas via the EU-funded BeFlexible project; others developed organically post-CoordiNet
2. Effekthandel Väst — Göteborg Energi Elnät + Mölndal Energi Elnät; Gothenburg + Mölndal with sub-areas; 4th season as of report date (V2024/25); uses NODES platform
3. Götene Flex — operated by Kinnekulle Energi; V2023/24 onwards; no public data available

Closed:

1. CoordiNet — Vattenfall Eldistribution + E.ON + Svk; four areas; 2019–2022; time-limited from the start
2. UppFlex — Vattenfall Eldistribution; continuation of CoordiNet’s Uppsala market; V2023/24 only; closed when capacity situation changed
3. sthlmflex — Svk + Ellevio + Vattenfall Eldistribution; Stockholm region; 4 seasons (2020/21–2023/24); closed: warm winters + high energy prices + TSO subscription grants meant insufficient congestion need to justify continuation
4. JämtFlex — Jämtkraft Elnät; V2023/24 only; no public data available

Closure reasons: markets close when either (a) the project was time-limited from inception, or (b) the capacity challenge that motivated it disappeared — typically because the grid was reinforced (e.g., Södra Skåne 400 kV upgrade in October 2024, adding 600 MW) or because the underlying congestion driver changed.

Key fact: No Swedish local flexibility market has yet managed an acute capacity shortage as its primary sharp tool. All markets have demonstrated the concept (load shifting is possible) and some have reduced subscription violations — but villkorade avtal remain the tool providing actual operational certainty.

Two platforms

All Swedish market trade flows through one of two platforms:

• SWITCH — developed and operated by E.ON Energidistribution; used by CoordiNet and all E.ON markets
• NODES — NODES AS (Norwegian company, Lysaker); independent third-party provider; used by sthlmflex, JämtFlex, and Effekthandel Väst

Both platforms support the three core product types. The existence of two competing platforms has produced broadly similar product designs across markets — a de facto industry standard has emerged organically.

Products traded

All markets trade three product variants:

Notable product evolution:

• CoordiNet started with only day-ahead bids (V2019/20); added intraday and seasonal availability (V2020/21); added Veckoflex/medium-term (V2021/22)
• Effekthandel Väst (NODES) has a product called MaxUsage (V2023/24): customer commits to capping consumption at a pre-agreed level (e.g., 75 kW) during designated high-load hours rather than reducing from an unknown baseline. Eliminates baseline calculation problem. Works well for season 1; breaks down in season 2+ if energy efficiency has already occurred (the “baseline” has structurally shifted)
• E.ON BeFlexible markets changed product design to comply with Swedish public procurement rules (LOU): blind bidding (providers submit bids without seeing the published need) was found non-compliant; changed to need-first design (DSO publishes need → providers bid)

Price and volume data

Sweco compiled all available market data into Tabell 6. Notable findings:

SE4 (Skåne) markets command much higher prices than SE3:

• Hässleholm: ~14,672 SEK/MWh avg (highest observed)
• E.ON Södra Skåne, Effekthandel Väst: ~3,000 SEK/MWh
• CoordiNet Skåne (V2021/22): ~2,285 SEK/MWh

SE3 markets have much lower prices:

• CoordiNet Uppland: ~248 SEK/MWh
• E.ON Vaxholm: ~1,650 SEK/MWh (highest SE3 market observed)

Price spread between SE3 and SE4 reflects the structural congestion difference: SE4 has persistent grid bottlenecks that create high willingness to pay.

Volume variability: CoordiNet Uppland shows year-to-year variance driven by temperature and energy price:

• V2019/20 (warm): 3.3 GWh, 140 subscription-violation-hours avoided
• V2020/21 (cold spell in February): 6.6 GWh, 329 hours avoided
• V2021/22 (coldest, but high energy prices): 0.1 GWh — high spot prices meant cheap heating load providers left the market

Actor composition

Five actor roles are distinguished: market owner, resource owner, service provider/aggregator, flexibility buyer, platform provider.

Actor concentration analysis (from CoordiNet and E.ON market data):

• Aggregators: have the most units (most FSPs, most resources registered) but smallest total capacity
• Energy companies (e.g., district heating): have few units but large capacity — dominate activated volumes
• A few large actors are decisive for market liquidity → concentration risk and market power concern

This matters for market design: if 1–2 large energy companies leave (e.g., because Svk’s balancing market pays better), a local market can instantly lose most of its clearing capacity.

Key barriers identified

A: Prerequisites for flex

• Trust problem (bilateral): DSOs need confidence that flex will be available long-term before reducing grid investment; FSPs need confidence that flex will be activated (that avrop will happen) before investing in measurement/control tech
• Geographic mismatch: capacity problems are in cities; large industrial flex resources (data centers, wind park batteries, industrial loads) are often connected to regionnät, not lokalnät
• Villkorade avtal preferred: DSOs often see villkorade avtal as simpler, cheaper, and more reliable than flex markets — regulatory compliance forces market-first, but delivery reliability for flex is lower (~80% of contracted flex actually delivered, per interview; villkorade avtal seen as near-100%)
• CAPEX bias: present in current regulation but addressed from RP5 (2028); some actors feel the 2024–2027 period already improved neutrality

B: Roles and responsibility

• Unclear who must initiate: no legal obligation specifying who must evaluate whether a flex market should be established, or who bears the cost
• Grid hierarchy ambiguity: flex resources connected to lokalnät can solve congestion at regionnät level, and vice versa — but cost and responsibility allocation across levels is unresolved
• 190+ DSOs, equal obligations: NC DR treats all network companies as system operators with equal rights and duties — challenging for Sweden’s many small DSOs (some serving only a few thousand customers)
• “Chicken race”: DSOs at different grid levels wait for the overlying/underlying operator to procure flex first, to avoid incurring the cost themselves

C: Profitability and market design

• Price competition from Svk balancing markets: when Svk offers high availability payments for FCR/aFRR/mFRR products, resources leave flex markets. Unsynchronized gate closure times compound this
• Liquidity too thin for investment: flex need is often only a few hours per season → activation revenue alone cannot justify investment in metering/control tech → availability contracts needed but only partially solve the problem
• Blind bidding was procurement-rule violation: resolved in BeFlexible markets
• Taxation: cross-sector energy transfers (electricity → heat for fjärrvärme companies) attract “double consumption taxation” on both the electricity input and the heat output — reduces profitability for the largest flex providers

D: Processes, IT, routines

• Forecasting capability gap: many small DSOs lack tools and data for flex need forecasting; grid topology maps and historical data not always digitized
• Baseline complexity: establishing what consumption would have been without dispatch is technically and commercially complex; no harmonized method; MaxUsage partially sidesteps this
• Prequalification burden: each market has its own process; FSPs that want to bid in multiple markets must prequalify separately
• Automation: mostly manual today; E.ON has the highest degree of automation; others activate manually

NC DR analysis

Sweco’s analysis uses the September 2024 ACER consultation draft (not the final recommendation). Key observations:

• NC DR is “extra complicated” because it bridges monopoly (grid) and competitive (market) territory simultaneously
• DSOs are named as system operators for the first time in a network code — unprecedented scope for DSOs in rule-writing
• Sweden-specific complexity: Sweden has an extra grid level (regionnät, between stamnät and lokalnät) that most EU countries lack; this complicates the assignment of “system operator” responsibility for local markets
• 190 DSOs: far more DSOs than most EU countries; code is not designed for this level of fragmentation
• National conditions must be developed across: baseline methodology, service provider terms, flexibility information system, TSO-DSO and DSO-DSO coordination
• Centralized flex register: Norway is building a centralized register (hub-connected) vs NC DR’s assumption that each DSO has its own FIS; Sweco recommends Sweden consider the Norwegian approach

Ten recommendations

AP1 (Philosophy and fundamentals — led by Ei)

Rec 1: Define redispatching and the tool hierarchy Standardize what “omdirigering” means and when each tool (flex market, villkorade avtal, bilateral agreements, tariff pricing) is permitted. Sweco’s proposed definition: flexibility procured after day-ahead market gate closure = redispatching; before gate closure = preventive or directed. Sweco identifies a “traffic light” model: green (normal) → market-based required; amber (risk of stress) → reduced market requirements; red (imminent interruption) → non-market tools allowed.

Rec 2: Revenue regulation and incentives for flex Clarify and harmonize how “socioeconomic efficiency” should be evaluated when comparing flex vs grid investment. The RP5 TOTEX reform moves in the right direction; Sweco recommends tracking RP4 (2024–2027) neutrality claims in practice and developing TOTEX further.

Rec 3: Responsibility for capacity problems and cost allocation Establish methodology for attributing responsibility: which operator level, geographically and in the grid hierarchy, bears the obligation and cost for each type of capacity constraint. Without this, the “chicken race” between grid levels will persist.

Rec 4: Grid hierarchy regionnät/lokalnät Investigate how flexibility should optimally be procured at each grid level. EVs and heat pumps (lokalnät resources) can relieve regionnät constraints; industrial loads (regionnät resources) can relieve lokalnät constraints. Cross-level coordination rules are needed.

Rec 5: “System operator” in Swedish context Investigate and document how the NC DR’s SO concept should be interpreted for Sweden, given the extra grid level and extreme DSO size range. A consequence analysis is needed covering: operational security implications, flex market efficiency, and competence requirements on small DSOs.

AP2 (NC DR implementation — led by system operators)

Rec 6: Definition and criteria for market-based solutions Ei’s existing guidance on what constitutes a “market-based mechanism” (from Ei R2023:08) is seen as insufficient by actors. Develop clearer criteria, including: what investigations must DSOs perform before concluding no market exists, and what triggers Art. 13(3) exception applicability.

Rec 7: Market coordination Initiate a study (Ei + Svk) on how bid forwarding between markets should work — including interoperability standards, consequence analysis, and implementation path. Value stacking requires synchronized gate closures and standardized forwarding protocols.

Rec 8: Standardization and harmonization Standardize products, prequalification processes, and baseline methods across markets. This should build toward API/integration standards that enable automation, reducing administrative costs for both DSOs and FSPs. Include vision for resource-centric (vs DSO-centric) market design.

Rec 9: Transparent flex needs and flex resources Ei should develop guidelines for how DSOs must produce flex need forecasts (flexutredningar), update frequency, and how the forecasts are published. Mutual trust between buyers and sellers requires visible, credible demand signals.

Rec 10: Cost and complexity reduction Ei should appoint a steering group to lead standardization of processes, APIs, data exchange, and IT requirements. Compared to the “Elmarknadshandboken” (a handbook developed post-1996 deregulation for market participant processes) — a similar industry-level guide is needed for flexibility markets. Centralized FIS should be evaluated.

What this source adds to the wiki

• Complete seven-market landscape (first source to cover all Swedish markets including Effekthandel Väst, Götene, JämtFlex, UppFlex)
• Introduces NODES as the competing platform to SWITCH (40%+ of active market area)
• Introduces Effekthandel Väst as an entity
• Explains Södra Skåne market trajectory: October 2024 grid expansion (400 kV line, +600 MW) as the capacity context
• Actor concentration data: aggregators have units, energy companies have capacity
• ~80% flex delivery rate vs near-100% for villkorade avtal — explicit interview finding
• BeFlexible public procurement compliance fix
• SE4/SE3 price spread with data
• MaxUsage product design and its limitations
• Confirms sällanköpsmarknad problem is universal, not specific to SWITCH markets
• “No market has yet managed actual capacity shortage as primary tool” — important calibration

Relationship to other sources: Updates and extends Source - Ei Flexibility in Distribution Grids (2023) (which covered only CoordiNet, sthlmflex, and Effekthandel Väst through 2022). Complements Source - SWITCH Marknadsdata (info.switchmarket.se, 2026) (which provides post-2023 quantitative SWITCH data). Complements Source - CoordiNet D4.7.2 Swedish Demonstration (2022) with actor-interview perspective on what has and hasn’t progressed since 2022.