The Flexibility Services Market in Poland

The report, "Rynek usług elastyczności w Polsce — analiza regulacyjna" ("The flexibility services market in Poland: a regulatory analysis"), was prepared by the Centrum Transformacji Energetycznej (CTE), a Polish energy transformation think tank. Its legal state is October 2026: the first version was completed in September 2026 and supplemented with a correction (errata) in October 2026 after a CTE seminar at the University of Warsaw's Faculty of Law and Administration (WPiA UW) on 22 September 2026.

What "flexibility services" means: the ability of a resource (generation or consumption) to change its operating point on an external signal, used to manage constraints in the distribution grid. As generation shifts from large system power plants to distribution-connected sources, system management mechanisms need to adapt, and the EU favours market-based mechanisms for this.

Core research question: does Poland's current path toward building a flexibility services market need to change, given the experience of other EU countries and the requirements of the draft Network Code on Demand Response (NC DR) (ACER Recommendation 01/2025)?

Main conclusion: yes, it needs to change.

• Poland's current approach is a fully decentralized model, where each Distribution System Operator (DSO, "OSD") independently defines, registers and purchases its own flexibility services. The report judges this model unjustified going forward.
• Instead, the report recommends a hybrid model: DSOs still define their own products, but within common minimum standards and coordination rules with the balancing market. A fully centralized model (a single product catalogue set in law) is proposed only as a fallback, in case the regulator (Prezes URE, the President of the Energy Regulatory Office) cannot be sufficiently strengthened.
• Conditions for this to work: regulating the aggregation model and dedicated measurement devices (DMD), standardizing the reference consumption profile ("baseline"), strengthening Prezes URE's capacity, and active Polish participation in NC DR comitology (the EU committee process).
• Importantly, after the 22 September 2026 seminar, most stakeholders shifted their position toward starting with a centralized model, with a planned evolution toward the hybrid model later; the report's roadmap (Table 6) was updated to reflect this.

How the report builds this conclusion (its internal logic):

1. Lessons from mature EU markets, UK, the Netherlands, Belgium (section 1);
2. A map of Polish market actors and their interests (section 2);
3. An inventory of today's fragmented, uncoordinated Polish practices (section 3);
4. A gap analysis against the ten key NC DR requirements (section 4);
5. The model recommendation itself (section 5);
6. Required legal changes in Polish energy law (section 6);
7. An NC DR implementation timeline (section 7);
8. A roadmap of concrete next steps per actor (section 8).

This is a broad report covering several distinct themes, country comparisons, the current Polish market, legal gaps, the recommended model, required legal changes, deadlines, and actor-specific recommendations. Would you like me to go deeper into any particular thread, for example the country comparisons, the legal changes needed, the NC DR timeline, or what this means for a specific market participant (DSO, aggregator, supplier, etc.)?

Short answer: The report's original recommendation (September 2026 version) is a hybrid model: distribution system operators (OSD) keep defining their own flexibility products, but subject to common minimum standards and coordination rules with the balancing market. The centralized model (a single, statutory catalogue of products for all OSD) is proposed as a fallback. However, following the CTE WPiA UW seminar of 22 September 2026, the report's October 2026 supplement updates this position: most stakeholders shifted toward recommending the centralized model as the starting point, with a planned evolution toward the hybrid model over time.

Why hybrid (original rationale), from section 5 (Optimal model for Poland):

• OSD differ significantly in network characteristics, ambitions and capabilities; imposing a single rigid product catalogue at this early, still largely experimental stage of the market would be premature.
• The draft system regulation (22 May 2026) already leaves too much discretion to bilateral contracts, pushing key decisions, like the baseline methodology, outside regulatory oversight; this risks operators imposing overly restrictive service conditions.
• A hybrid model would address this by keeping common minimum standards (while OSD still define their own products), but it increases regulatory complexity and requires strengthening the energy regulator, Prezes URE (President of the Energy Regulatory Office), with more resources.
• The report draws on lessons from the Netherlands (section 1): the Dutch GOPACS platform, a shared coordination tool run jointly by the TSO TenneT and major DSOs, shows the value of one common platform and coordinated OSP-OSD (transmission-distribution operator) action without full centralization of product definitions.
• Poland's own market already resembles the Dutch model: several large OSD have begun implementing services independently, and the largest market players appear to be moving toward a shared platform echoing GOPACS.

Why the update toward centralized-as-starting-point (seminar conclusions, October 2026 supplement):

• Greater standardization lowers entry thresholds for new suppliers (especially aggregators) and eases OSD-TSO coordination, not just cost regulation of OSD.
• A majority of seminar participants on barrier 4 (restrictive requirements for congestion management services) favored shifting toward the centralized model.
• The discussion also flagged that the legal definitions and conceptual framework need revision, particularly that the regulation should not narrow the statutory definition of flexibility services by tying it to remuneration, something not addressed in the original report because it took end-July 2026 legal framework as a boundary condition.

Comparative table (Table 4) highlights trade-offs: the centralized model offers slower but more durable implementation, easier harmonization with Article 16 NC DR (Network Code on Demand Response) equivalence tables, and simpler alignment with smart metering (AMI) rollout; the hybrid model allows faster, more autonomous local market development and better use of dedicated measurement devices (DMD) where AMI is lacking.

This recommendation chain follows the report's logic: lessons from UK/Netherlands/Belgium (section 1) → current fragmented Polish state (section 3) → gaps versus the NC DR draft (section 4) → recommendation (section 5) → required legal changes (section 6) → roadmap actions (section 8).

Would you like me to expand on the regulatory changes needed to implement this (section 6), or the updated roadmap for specific actors (section 8)?

Flexibility services (usługi elastyczności) are defined in the report as services that resolve constraints in the electricity network by changing the power fed into or drawn from the grid in response to an external signal from the system. Legally, under Article 3(11k) of the Polish Energy Law, they are narrower: services provided to a distribution system operator (DSO) by an aggregator or by active consumers, generators, or energy storage holders connected to the distribution network (excluding the coordinated 110 kV network), aimed at ensuring security and improving the efficiency of distribution network development, including managing network constraints. Under the draft "system regulation" (22 May 2026), they consist of changing the active power drawn from or fed into the distribution grid relative to a reference profile (baseline), and the draft explicitly provides for remuneration to the service provider.

How they differ from balancing services

The report places flexibility services within the broader set of services operators purchase from system users, and draws a clear institutional line:

• Balancing services (usługi bilansujące), along with non-frequency system services (voltage regulation, black start), network services, and frequency-related intervention services (IRP/IZP), are procured by the transmission system operator (TSO, PSE).
• Flexibility services and non-frequency system (voltage) services are the only two categories procured by DSOs.

Functionally, the report concludes that flexibility services bought by DSOs are essentially the DSO-level equivalent of network services bought by the TSO; it considers this categorical split in EU and Polish law to be more a matter of legislative convention than a deliberate conceptual distinction.

A related but distinct concept is redispatching. Flexibility services (strictly defined) contain elements equivalent to market-based redispatching, especially regarding active power. In Poland, however, non-market-based redispatching (mainly curtailment of renewable generation, ordered unilaterally by the operator rather than offered voluntarily) dominates in practice, while market-based redispatching remains the exception. Both tools ultimately serve the same purpose, managing network constraints, which is why one of the report's stated aims for a flexibility market is to reduce reliance on non-market redispatching.

Current legal practice also matters: under present rules, the same resource generally cannot simultaneously provide multiple service categories, though a resource can be qualified for balancing services while also providing flexibility services, and IRP/IZP activity may potentially be combined with flexibility service provision. Also notable: services for PSE (TSO), such as IRP/IZP, fall outside the statutory definition of "flexibility services" since that definition only covers services to DSOs, even though the report treats these PSE-level experiences as valuable groundwork for the flexibility market (section 3.4, Energy market capacity mechanism and IRP/IZP experience).

One practical consequence: because the draft regulation requires remuneration for flexibility services, some services already implemented by Polish DSOs, such as Enea Operator's IDC (Interventional Active Power Delivery), which is settled via a fee exemption rather than direct payment, would not meet the formal definition if the regulation enters into force as drafted.

Note this reflects the report's legal state as of October 2026; the system regulation referenced is still a draft.

Would you like more detail on how this distinction plays out for a specific actor (e.g., DSOs vs. PSE), or on the NC DR's treatment of these categories?

FIS (Flexibility Information System) is the central registry foreseen by the draft NC DR (Network Code on Demand Response, ACER Recommendation 01/2025). It consists of two modules: the SP module (Service Providers) and the CU module (Controllable Units). Its role is to register, store and make available data on who provides which flexibility service to which operator, using which resources (NC DR art. 2, points 15-17). The report stresses an important distinction: FIS is not itself a trading platform – it is not a place where services are offered and purchased. By analogy, the Dutch platform GOPACS matches supply and demand, but using it requires prior registration in the operators' own systems, which is the kind of function FIS performs.

Who should run it

The NC DR draft defaults to one or more system operators (TSO/DSOs) as responsible for running FIS (art. 24-28). However, art. 8 allows operators to delegate tasks to third parties while retaining legal responsibility (art. 8 (1)-(3)), and lets member states assign responsibility for specific tasks to other entities if those entities demonstrate capability (art. 8 (4)-(5)).

The critical condition, which the report highlights as the central issue, concerns conflict of interest. Running SP or CU registers gives privileged access to commercially valuable data, so NC DR art. 26(4) requires that any third party entrusted with FIS tasks maintain proper "business separation" (unbundling) from parties with a commercial interest in the services registered there, this applies not only to service providers but also to any operator of a market platform.

The report's own recommendation for Poland (Section 6, "Who has legal access to what data") builds directly on this:

• The entity running FIS should not participate in market processes other than as a buyer (operator).
• In vertically integrated companies, the parts of the business operating on competitive markets must be excluded from access to confidential commercial data held in FIS.
• Market platform operators should also be kept separate from FIS access.

This is directly relevant to the report's discussion of TGE (Towarowa Giełda Energii, the Polish energy exchange), which is a candidate to operate a national flexibility market platform. Per the report's analysis (not the media-based elements on TGE's agreements, which this service does not treat as a source), the logic of art. 26(4) and section 6.7 implies that whoever operates a market platform should be kept organizationally separate from whoever runs FIS, to avoid the conflict of interest the code is designed to prevent.

Current state: per the report's implementation-status assessment (section 4.4), FIS and the CU registry are not implemented anywhere yet in Poland, this is one of the open gaps versus NC DR requirements.

Would you like me to go deeper into how this connects to the TGE platform question, or to the broader NC DR data-access and unbundling requirements in section 6?

Based on the report's analysis (section 5, "Optymalny model dla Polski"), the two models differ mainly in who defines the products and in the resulting governance, speed, and permanence of the market.

Core difference

• Centralised model: a single, uniform catalogue of flexibility service products set out in legal regulation ("przepisy prawa"), applicable to all distribution system operators (DSOs). The system operator's role is limited to purchasing, activating and settling services; it does not define the products itself.
• Hybrid model: different DSOs can create and run their own, potentially different, products, but these must comply with common minimum standards and rules coordinating them with balancing-market services. Here the operator has an additional role: it also defines the products, within boundaries set by regulation.

In both models the other participant roles (flexibility service provider, system user, balance responsible party, energy supplier, independent aggregator, possible platform operator) stay the same.

Other differences highlighted in the comparison (Table 4):

• AMI metering penetration: centralised assumes eventual 100% smart-metering coverage (any use of dedicated measurement devices, DMD, would also need standardising); hybrid allows DSOs to move faster with DMD in areas lacking smart meters, adapting to local capabilities.
• Market structure: centralised concentrates competence with the Minister of Energy and the President of the Energy Regulatory Office (URE), forcing standardisation, slower rollout but greater durability; hybrid gives DSOs more autonomy, potentially allowing faster development of local markets.
• NC DR (Network Code on Demand Response, draft code) timeline: centralised supports early alignment with the code's harmonisation direction, simplifying later implementation of the "equivalence table" (art. 16 NC DR); hybrid allows early gathering of experience in product design within the freedom the draft code allows, with harmonisation added later.
• European experience: centralised draws on the Dutch GOPACS platform as good practice (joint TSO-DSO coordination tool), while noting it is too early in Poland to consider regulated prices for flexibility services; hybrid draws on Belgian and British experience showing the need to standardise products across different DSOs, also without price regulation at the first stage.
• Financing: a centralised catalogue makes it structurally easier to socialise the cost of purchasing services (via network tariffs), since costs are uniform; under the hybrid model each operator must individually justify to the regulator that its defined products are cost-optimal.
• Governance burden: the hybrid model shifts more regulatory weight onto the President of URE, who would need strengthened resources to supervise a more complex, decentralised standard-setting process; the centralised model concentrates this burden instead on the legislator and the regulator at the rule-making stage.

The report's recommendation The report recommends the hybrid model as more suitable for a large, diverse system like Poland's, since DSOs differ significantly in network characteristics and capabilities, and a rigid centralised catalogue would be premature at this still largely experimental stage. However, it flags that the draft system regulation (22 May 2026) currently leaves too much to individual contracts (e.g. baseline methodology), outside regulatory oversight, so common standards are still needed. The centralised model is treated as a fallback option, particularly if the regulator cannot be given sufficient resources for the more complex hybrid oversight.

Notably, after the CTE seminar of 22 September 2026, most stakeholders shifted toward favouring the centralised model as the starting point, with an eventual evolution toward the hybrid model, mainly because greater standardisation would lower entry barriers for new providers and ease TSO-DSO coordination. Would you like more detail on this post-seminar update, or on how either model would affect a specific market participant (e.g. aggregators, DSOs, PSE)?

Based on the report, barriers fall into several overlapping categories, summarized most explicitly in Table 4a (section 5.5, Recommendation), which cross-references the official Flexibility Needs Assessment (FNA, carried out by the Polish TSO body NCAE) with the authors' own analysis and the conclusions of the 22 September 2026 seminar.

1. Inadequate legal framework for new and small participants. Current regulations are incomplete, overly complex, or lag behind a fast-changing market; there is a need for simple, transparent rules that do not create unnecessary barriers to entry.

2. Lack of tools and incentives to provide flexibility. Participants, including prosumers, lack material incentives to offer flexibility services. There is ongoing disagreement among stakeholders about what exactly should be remunerated, in particular whether to pay for mere availability ("payment for readiness") and not only for activation (relevant to art. 35(3)(a) and (c) of the draft NC DR).

3. Restrictive requirements for balancing services. The FNA flagged the lack of a requirement for smooth power-point change (ramping); separately, trading companies at the seminar pointed to a long and opaque prequalification process and overly restrictive technical requirements from operators, which makes it hard to gain experience with new types of resources. This is the one point where the report's authors diverge from the official NCAE diagnosis.

4. Restrictive requirements for congestion-management (network) services. There is a need to standardize products across all system operators (e.g. via the national regulator); after the seminar, most stakeholders actually shifted toward favouring a more centralized model as a way to address this.

5. No regulatory incentives for operators to consider non-wire alternatives. The absence of regulation, standardization and methodologies makes purchasing flexibility services risky for distribution system operators (DSOs), and there are no regulatory incentives for DSOs to prefer flexibility over grid expansion (what the report elsewhere calls "CAPEX bias").

Beyond these five, the report (section 3.7, Aggregators) identifies specific barriers affecting aggregators, who are meant to be a cornerstone of the market but are largely blocked in practice:

• Technological: end customers often lack the automation (control systems) needed, lengthening sales cycles beyond 6-12 months.
• Regulatory: independent aggregation (acting without the balance-responsible party's involvement) is only legally possible in the balancing market; DSO contract templates from Energa-Operator and Tauron Dystrybucja require the service provider to be directly connected to the DSO network and exclude subcontracting, effectively excluding independent aggregators.
• Economic: revenue from existing services (interventional demand response for the TSO, capacity market) covers only 5-15% of a large plant's annual energy costs, often too low for SMEs to justify investment.
• Market structure: the aggregation market is an oligopoly of 4-5 large players (e.g. Enel X, Enspirion).
• Disproportionate regulatory burden: broad legal definitions of aggregation trigger burdensome obligations (e.g. registration), whose informational value is unclear as long as the independent-aggregation question remains unresolved.

Underlying all this, the report's broader diagnosis (section 5.5) is that the current draft system regulation gives DSOs too much freedom, pushing key product decisions such as the baseline methodology into individual contracts, outside regulatory oversight, which risks fragmented and overly restrictive local rules. Addressing this is why the report recommends a hybrid model with common minimum standards, which in turn requires strengthening the Energy Regulatory Office President (Prezes URE) with more resources.

Would you like me to expand on the regulatory fixes the report proposes (section 6) to address these barriers, or on the aggregator-specific recommendations?