Demand Response

A change in electricity consumption patterns in response to supply conditions, price signals, or explicit requests. Instead of only adjusting supply to meet demand (the traditional approach), demand is adjusted to match supply. One of the core mechanisms for implementing Flexibility.

The fundamental insight

The Electric Power Transmission system has no significant energy buffer — generation must match consumption in real time. Traditionally this was achieved entirely on the supply side: dispatching generators up or down. Demand response adds the other side of the equation: making demand itself adjustable.

This matters more as variable renewables grow, because supply becomes less controllable. If you can’t always adjust generation to meet demand, you must be able to adjust demand to meet generation.

Implicit vs explicit

This distinction is central to EU/Swedish flexibility strategy. The Clean Energy Package mandates both pathways in law: the Electricity Market Directive requires that consumers have access to dynamic pricing (Art. 11 — implicit), and that independent aggregators can participate in markets without supplier consent (Art. 13 — explicit). The EU legal definition of demand response (Directive Art. 2(20)) explicitly covers both: response to “market signals, including time-variable electricity prices or incentive payments” and bids “to sell demand reduction or increase at a price in an organised market.”

The Network Code on Demand Response

Demand response is getting its own EU network code — the Network Code on Demand Response (NC DR), the first EU regulation specifically focused on demand-side market participation. The NC DR will standardize how DR resources register, qualify, and participate across EU markets, establish mandatory flexibility registers, and define rules for aggregation, baselining, and settlement. Key debates include the minimum bid size (0.1–1 MW — critical for whether household-level resources can participate directly) and baseline methodology. Entry into force is expected in 2026 — note however that as of May 2026, the Commission’s final text had not yet been published and the NC DR was listed as “delayed” on Svk’s market roadmap. (Source - NC DR Proposal (ENTSO-E and EU DSO Entity, 2024), Source - ACER Recommendation 01-2025 on NC DR, Source - Elmarknadsrådet Meeting 1 February 2026)

Swedish post-EIF obligations: after the NC DR enters into force, Svenska kraftnät and the DSOs must jointly develop and submit four sets of terms and conditions to Ei for approval. This triggers a legally defined but operationally tight development cycle:

1. FIS terms and conditions — flexibility information system registration and data exchange rules
2. TSO-DSO and DSO-DSO coordination framework — formal grid prequalification procedures, temporary limits, and multi-level coordination architecture
3. Market-based procurement terms — LFM-h/p/e product specifications, bid formats, merit-order rules
4. Near-real-time activation terms — baseline methodology, activation protocols, settlement

The timeline for developing these four deliverables is described as “tight” given the complexity of the coordination across ~155 Swedish DSOs and Svk. Industry broadly supports standardization and market-based procurement as a principle; concerns focus on IT/information exchange complexity. (Source - Elmarknadsrådet Meeting 1 February 2026)

Parallel track — NC DC 2.0: The Network Code on Demand Connection (NC DC, EU 2016/1388) is being simultaneously amended via ACER Recommendation 03-2023. While NC DR governs how demand-side resources participate in markets, NC DC governs the connection requirements for demand facilities. NC DC 2.0 explicitly brings EVs, EVSE, heat pumps, and power-to-gas under harmonized EU connection requirements for the first time — creating the connection-side regulatory foundation that NC DR then builds on for market participation. (Source - ACER Recommendation 03-2023 NC RfG DC (2023))

Three product categories

1. Emergency DR — activated to prevent blackouts during supply scarcity. Last resort.
2. Economic DR — activated when market price exceeds value of consumption. Continuous optimization.
3. Ancillary services DR — demand-side resources providing grid services: frequency regulation (FCR, aFRR, mFRR), contingency reserves. Competes directly with generator-provided services.

In the Nordic Balancing Markets, category 3 is increasingly important. As inverter-based generation (solar, wind) replaces synchronous generators, mechanical inertia decreases and demand-side ancillary services become critical for frequency stability.

Technologies and enablers

• Smart meters — foundation for dynamic pricing and measurement/verification
• Building automation / HEMS — automated response without manual intervention
• EV smart charging — large flexible load; aggregated EVs as demand response resource
• Industrial process flexibility — aluminum smelters, data centers, cold storage
• Aggregation platforms — aggregate many small resources into market-participant-sized portfolios
• Short-term load forecasting (STLF) — the operational data layer enabling market participation; BRPs cannot submit day-ahead plans and FSPs cannot construct bids without a baseline load forecast. Unknown demand = no bid. See Load Forecasting for the full method landscape and its role as an enabling condition for explicit DR.

Non-linear price effects

Small demand reductions create outsized price effects because peak generation is the most expensive:

• 5% peak demand reduction → estimated 50% price reduction during crisis periods
• 1% peak demand shift → 3.9% system cost savings (PJM study)
• 10–20% of total electricity costs driven by just ~100 peak hours per year

This makes demand response economically powerful even at modest participation levels.

VRE integration costs and the DR value case

OIES EL36 (2019) provides a systematic decomposition of the costs that variable renewable energy imposes on the power system — and identifies which DR mechanisms address each: (Source - OIES EL36 Electricity Market Design for Decentralized Flexibility (2019))

This framing explains why the business case for DR strengthens as VRE penetration grows: DR doesn’t only provide ancillary services — it reduces the total system cost of VRE integration across all five categories.

Relationship to other flexibility mechanisms

Demand response is one tool in the broader Flexibility toolkit:

• Energy Storage — shifts energy in time (batteries, pumped hydro, thermal)
• Demand response — shifts or curtails consumption
• Distributed generation — local supply adjustment
• Grid reconfiguration — topology changes to manage congestion
• Cross-border trading — spatial flexibility via Electric Power Transmission interconnectors

These mechanisms complement each other. A Virtual Power Plant typically combines DR, storage, and DG into a single dispatchable portfolio.

Demand flexibility as a system necessity — LMA2024 finding

LMA2024 provides the strongest quantitative case for demand response as a structural system necessity rather than a market supplement. In Svk’s high-electrification scenarios (EP: ~343 TWh/year with new nuclear; EF: ~342 TWh/year with large-scale renewables — both by 2045):

• Without flexible EV charging and hydrogen electrolyzer dispatch, the system accumulates hundreds to over a thousand shortage hours annually by 2045 — system adequacy fails
• With active flexibility (EVs charging off-peak, electrolyzers providing virtual battery dispatch via H2 storage), shortage hours drop to manageable levels
• The difference is not marginal — flexibility is a binary requirement for adequacy in high-demand pathways

The electrolysis-hydrogen pathway is particularly significant: large-scale electrolyzers operating as dispatchable loads can absorb surplus variable renewable energy and respond to system needs, functioning as multi-GWh virtual batteries via the hydrogen storage buffer.

The finding contextualizes why Svk explicitly includes demand flexibility in its system capability requirements and why the Balancing Markets › Strategic reserve — first procurement failure (autumn 2025) and BSP framework development are not just market design refinements — they are prerequisites for the 2045 scenarios that underpin Sweden’s electrification strategy.

Swedish DR potentials to 2030

FlexAbility (2025) quantifies realistic technical maximum potentials for Swedish demand response resources at hourly timescale for 2030:

The heat pump gap is the most striking: only 300 MW prequalified today vs 5,750 MW potential — a 19× gap driven primarily by the absence of smart control hardware at the household level.

Heat pump flexibility: temperature dependence

Field data from the E.ON Simris microgrid pilot (2015–2018) provides the most operationally concrete description of heat pump flexibility availability as a function of outdoor temperature (Source - InterFlex Simris Microgrid (2018)):

The coldest periods — when grid stress is highest — are precisely when heat pumps are at maximum output and cannot be further increased: they can only be curtailed (downward). Flexibility in the −10 to 0°C range is bidirectional and most abundant. Control systems should implement an indoor comfort boundary: DSR is disabled if indoor temperature deviates by more than ±1°C from the customer-set setpoint.

The Simris DSR platform issued steering signals as a percentage of maximum capacity (−100% to +100%), derived from the central battery’s state of charge — not in kW — allowing the platform to adapt the setpoint without knowing each heat pump’s nameplate capacity.

Electric boilers represent an acute policy failure: 975 MW of technically available, cost-competitive demand response sitting idle because the energy tax structure eliminates the incentive. The analysis explicitly assumes the tax will be reformed by 2030 — without that, the potential is effectively zero.

Industrial demand response has a structured cost profile: a first tranche (~350 MW) at 100 SEK/MWh reflects easy-to-shift loads; a second tranche (~850 MW) at 2,000 SEK/MWh requires process adjustments; a third tranche (~100 MW) at 4,000 SEK/MWh represents near-core-process flexibility. (Source - FlexAbility Delrapport 1 (2025))

Swedish household demand response

Swedish household DR adoption is characterised by high informal willingness to act but very low formal market participation — the 42% → 2.8% conversion gap identified by Ramboll (2024). Across all surveys, economic incentives dominate: aggregator delegation is driven by earning potential, not awareness. Lack of knowledge is not a statistically significant barrier (FlexAbility 2025). The primary binding constraint is structural — tariff fragmentation, absent standardised service interfaces, and the underdeveloped BSP/BRP market. Only ~12–13% of households face actual hourly price signals despite nominally higher timprisavtal penetration rates.

For the full analysis — aggregator delegation surveys (FlexAbility, n=2,872), adoption baselines (IVL, n=10,328; Ramboll, n=1,173), consumer service landscape (AFRY: 44 services, 172 actors), and the 2024 snapshot on EV smart charging, effektavgift knowledge, and market penetration — see Swedish Household Demand Response — Consumer Adoption and Barriers.

Effekttariffer — the double-edged price signal

Capacity-based effekttariffer (demand tariffs) were mandated for all Swedish DSOs by January 2027 under EIFS 2022:1 — but the government has since tasked Ei to repeal EIFS 2022:1 by 30 June 2026 and develop a new effektavgift model (proposal due 12 April 2027). Effektavgifter remain permissible after repeal under Art. 18 EU Electricity Market Regulation and ellagen, but are no longer mandatory. (Source - Ei Effektavgifter webb (2026))

Construction diversity: The Elmarknadshandbok §9.1.4 confirms the range of constructions deployed in practice: measured monthly peak; mean of the highest N measurements in a month or year; time-varying rates by hour or season; and an additional högbelastningsavgift applicable on weekdays November–March. This diversity is precisely what Ei has identified as the obstacle to standardized automated steering services — each DSO’s unique construction requires bespoke integration by any aggregator or building management platform. (Source - Svensk Elmarknadshandbok 26A (2026))

These tariffs correctly incentivize customers to flatten their demand profiles and avoid expensive peaks. But they create a structural conflict with spot price signals:

When spot prices are low (e.g., midday solar peak), the economically correct signal is to charge EVs and batteries — load is cheap and beneficial for the system. But the demand tariff still penalizes any peak demand event, regardless of what the spot price is doing. The two signals point in opposite directions.

This double-edge undermines the coordination between implicit DR (responding to price) and capacity management (avoiding tariff peaks). Göteborg Energi Nät is testing a dual-tariff model to separate the two incentive structures. (Source - FlexAbility Delrapport 5 (2025)) The double edge is one instance of a more general structural problem — a single DER receiving multiple uncoordinated price signals with no defined arbitration — analysed in The Signal Stack — Price Signal Collisions at the Customer DER.

Deployment status — Ei R2026:02 (2024)

Ei‘s first biennial smart grid monitoring report (Ei R2026:02, December 2025) confirms national deployment as of 2024: several DSOs have deployed time-differentiated tariffs but many customers are still not covered. No significant change between 2023 and 2024. All DSOs are legally obliged to time-differentiate their tariffs by 1 January 2027 — Ei expects the indicator to reach 100% at the 2028 reporting cycle. (Source - Ei R2026-02 Utvecklingen av Smarta Elnät (2025))

Ei’s SGI microdata (Sh_63A_tdtariffs, 2024, n=111 REL reporters) provides company-level granularity:

• System average: 18.2% of customers ≤63A have a time-differentiated grid tariff
• 37 of 111 companies report any coverage (66% of lokalnät companies have none at all)
• 11 companies at 100%: Falkenberg Energi, Skövde Elnät, Skånska Energi, Sollentuna Energi, Växjö Energi Elnät, Västra Orusts Kraft, Kraftringen Nät, Telge Nät, Kungälv Energi, Trelleborgs Energi, Jukkasjärvi

For RER/RET reporters: Öresundskraft, Svk, and Skellefteå Kraft are at 100%; E.ON at 72%; remaining companies at 0%. (Source - Ei SGI Data 2023-2024)

Tidsindelade effektavgifter — first behavioral evidence

Göteborg Energi Nät introduced tidsindelade effektavgifter (time-divided demand tariffs) on a voluntary basis from 5 February 2025. As of June 2025, more than 1,100 customers had enrolled. First-season behavioral outcomes: (Source - Göteborg Energi Elektrifieringsrapporten nr 1 (2025))

The +16% rebound at 20:00 is non-negligible — DSOs planning tidsindelade effektavgifter must model both effects, as the rebound may stress transformer capacity at evening transition.

For demand response’s role as the named-but-undelivered adequacy mitigation — and why the strategic reserve’s CONE cap was benchmarked to household DR — see Capacity Adequacy and Flexibility as the Missing Reserve. The pattern recurs in real time: when a two-fault disturbance exhausted the SE3/SE4 mFRR bid stack on 8 June 2026, the fast firm capacity that answered was gas turbines and Baltic imports — demand response was absent from the response. (Source - Svk Driftstörningar 8 Juni 2026)

The effektavgift complaint surge (Ei konsekvensutredning, 2026)

The impact assessment behind EIFS 2026:8 quantifies the consumer reaction to the effektavgift rollout. Complaints/questions to Ei about the grid (elnät) jumped from 67% of all complaints in 2024 to 80% in 2025 (2,426 of 3,053), and of the 2025 elnät complaints, 64% (837) were specifically about tariff design / effektavgifter. As of spring 2025, ~13% of households had received an effektavgift, and by early 2026 drygt 30 DSOs applied one to customers with säkring up to 25 A. Ei names the core comprehension problem directly: users cannot reconcile the effektavgift (grid) signal with the spotprice (energy) signal — the two are separate and not necessarily aligned, which is the double-edge above experienced from the consumer side. (Source - EIFS 2026-8 Nätföretags Information till Elanvändare (2026))

Regulatory push for consumer DR — EIFS 2026:8 (in force 2027-01-01)

Ei‘s EIFS 2026:8 (nätföretags information till elanvändare, decided 2026-05-21, in force 1 January 2027) turns consumer-side DR awareness into a legal obligation on DSOs — the DSO counterpart to the supplier-side EIFS 2024:2. From 2027, every DSO must (4 kap.):

• inform users (website/app) about flexible electricity use via manual control and via automated control, and that there are market services on the electricity market for this — the first explicit regulatory requirement to tell households that automated steering/aggregation services exist;
• place a mandatory link to Ei’s efterfrågeflexibilitet portal ei.se/kundflex on the invoice and website (and to Energimyndigheten’s energiochklimatradgivningen.se);
• publish the meter’s öppet kundgränssnitt (HAN-port-type interface) and supported dataprotokoll, and show on mina sidor whether the interface is active — enabling third-party realtidsmätare and steering services (see Submetering);
• if the DSO trades on a local flexibility market, inform affected users in the trading area on mina sidor, name the market, and explain how to participate — but may not name or recommend specific services or providers (DSO neutrality). This is a recruitment channel into SWITCH, Effekthandel Väst, and similar markets.

The föreskrift also forces effektavgift transparency (purpose/design explained; effektavgiftsgrundande mätvärden shown on mina sidor with a calculation explanation; weighted-average pricing on invoices for dynamic per-kvart designs). Read alongside the Tommy Johansson automation-barrier statement, EIFS 2026:8 attacks the information barrier to consumer DR while the effektavgift redesign attacks the tariff-fragmentation barrier. (Source - EIFS 2026-8 Nätföretags Information till Elanvändare (2026))

Grid risks of demand response at scale

Implicit DR — households and devices responding to the same price signal simultaneously — creates a systemic balancing risk when participation is large enough. When many resources respond at once, BRPs cannot forecast the aggregate effect accurately and Svenska kraftnät faces large, unpredictable imbalances. A 2013 Elforsk study estimates the breakpoint at 100,000–700,000 simultaneously responding households — Sweden is approaching this range as smart metering and time-of-use pricing scale up. (Source - FlexAbility Delrapport 5 (2025))

15-minute pricing and synchronisation patterns

The EU Single Day-Ahead Coupling transitioned to 15-minute market time units on September 30, 2025 (96 price points per day vs. 24 previously), providing finer signals for flexible loads. The effect on synchronisation is not yet known: finer intervals may soften abrupt simultaneous starts, or may create new synchronisation patterns. Explicit monitoring is flagged as necessary. (Source - EC 15-Min MTU Day-Ahead Market (2025), Source - Energimyndigheten ER 2025-35 Förbättra Flexibiliteten (2025))

Random startup delay

A technical mitigation for EV charging synchronisation: a short random delay (typically 0–100 seconds) applied before a charger starts, distributing aggregate load across time. UK regulation (Electric Vehicles (Smart Charge Points) Regulations 2021) mandates delays up to 600 seconds for all private charge points sold after June 2022.

Compatibility requirement (critical): random delay must be opt-out compatible for grid balancing service participation. Svenska kraftnät requires activation within minutes for FCR/FFR — a charge point dispatched for FCR cannot apply a delay that prevents immediate response. Any Swedish mandate would need the same carve-out. No Nordic country has yet mandated random startup delay.

NC DR staggered activation

The Network Code on Demand Response addresses synchronisation at the aggregation layer: TSOs and aggregators are required to NOT send simultaneous activation commands to all managed devices — staggered dispatch must be implemented. This is the explicit DR system-level equivalent of random startup delay. (Source - NC DR Proposal (ENTSO-E and EU DSO Entity, 2024)) Price-signal synchronization is the non-malicious twin of a coordinated DER cyberattack — both drive the same physics; see Security and Resilience of the Digitalized Flexible Grid.

Emerging solar+battery synchronisation risk

After the tax reduction for microproduction solar sales ended in Sweden (January 2026), many solar+battery systems may switch to self-consumption mode. When the spot price hits approximately 60 öre/kWh, large numbers of these systems may simultaneously stop exporting and start self-consuming or charging batteries — creating a large coordinated load shift whose grid impact is unknown.

The two-market design

A theoretical market design proposed by Malcolm Keay (OIES EL17, 2016) and analysed by VTT (2016) offers a structural approach to embedding persistent DR incentives rather than relying on occasional price spikes. (Source - VTT-R-04621-16 Electricity Market Designs and Flexibility (2016))

Core idea: split electricity into two products:

• Firm electricity: available on demand; conventional dispatchable generators; higher price
• Non-firm electricity: available when weather allows; VRE generators; lower price; consumer bears imbalance costs if consumption doesn’t match allocated non-firm supply

Consumers who opt into the non-firm market pay the lower VRE-linked price but must match supply variability to avoid imbalance penalties. This creates a permanent everyday financial incentive to flex demand — unlike spot-price responsiveness, which requires consumer attention and yields intermittent signals.

VTT’s simulation confirmed the design restores price stability for dispatchable generators (solving the missing money problem) and reduces peak demand variability when consumer flexibility is active. Limitations: the model is too simplified to capture full dynamic ramp costs; the design has not been adopted in EU/Nordic policy. The incentive logic is visible in LFM product evolution: the LFM-h/p availability-based structure creates ongoing flexibility payments rather than event-only payments.

EVs and Vehicle-to-Grid as demand response

Electric vehicles are among the largest and fastest-growing flexible demand resources in Sweden. An EV’s 60–90 kWh battery, drawing 7–22 kW when charging, can be a significant DR asset if the charging session timing is controllable.

Smart charging (G2V only): shifting or modulating the charging rate in response to price signals, grid operator dispatch, or aggregator control. FlexAbility estimates 1,600–1,700 MW upward and 5,200 MW downward smart charging potential from light EVs by 2030, assuming 85% non-public smart charging penetration. (Source - FlexAbility Delrapport 1 (2025))

Vehicle-to-Grid (V2G): bidirectional operation where the EV battery also discharges back to the grid. Adds up to 5,000 MW potential by 2030 — but largely theoretical due to regulatory, technical, and commercial barriers. See Vehicle-to-Grid for the complete picture.

Adoption gap: Flexläget 2026 (December 2024 data) shows only 29% of EV owners with home charging use smart charging services — and 81% of apartment EV chargers are on shared parking where smart charging services are underdeveloped.

V2G as ancillary services DR: aggregated V2G fleets can in principle participate in FCR-N, FCR-D, aFRR, and mFRR through the BSP framework. The first Swedish V2G delivery to a local flexibility market was four Volvo Cars EVs delivering 111 kWh to Effekthandel Väst (March 2025, via CheckWatt). (Aggregation › CheckWatt — multi-market VPP aggregation at Nordic scale)

Key Swedish barriers specific to EV/V2G DR participation: double taxation (dubbelbeskattning) for cross-area discharge; Svenska kraftnät‘s physical address registration requirement; and classification ambiguity (EV as microproduction vs mobile injection point). (Source - Power Circle V2X Synthesis 2024, Source - KTH Thesis V2G Sweden 2024)

Data gaps

• Swedish DR landscape: which aggregators operate, what products are available (beyond CheckWatt — Flower, Ingrid Capacity, Capalo AI not yet documented)
• Real-world DR volumes in Sweden/Nordics — comprehensive cross-aggregator data
• Targeted energy tax reform for district-heating electric boilers — general electricity tax reduced 7.9 öre to 36.0 öre/kWh from January 2026, but the specific elpanna/värmepump targeted reduction (Energiföretagen proposal: ~1.2 GW flexibility, ~500 MSEK/yr cost) has not been approved; Energiföretagen filed a formal begäran om ändring March 2025; advocacy continuing as of 2026
• Measured effect of 15-minute pricing (from September 30, 2025) on synchronisation patterns and grid stress

(Source - Demand response (Wikipedia))