Battery storage and the 2,500-hour rule – What should be considered?
Jul 11, 2025
The 2,500-hour rule in the electricity sector is a special tariff regulation for grid fees that primarily affects larger electricity consumers. It defines a threshold of 2,500 annual usage hours, beyond which the calculation of grid costs changes significantly. The following explains what this rule is about and why it exists. After that, we will look at the impact that the installation of a battery storage system (e.g., for peak shaving or optimizing self-consumption) can have on this rule – including an example scenario with 500,000 kWh annual consumption and about 200 kW peak load. Finally, we will discuss criticisms of the rule and how the system of grid fees could change in the future.
What does the 2,500-hour rule state and why does it exist?
Usage hours indicate how many hours per year a consumer would theoretically make full use of their maximum capacity. Mathematically, it is the annual work (kWh) divided by the highest consumed power (kW). This ratio serves as an indicator of the regularity of network demand: An operation with constant load has high usage hours, while an operation with pronounced load peaks and otherwise low consumption has low usage hours.
The graphic shows the annual electricity costs depending on the peak load for a constant annual consumption of 500,000 kWh. The prices correspond to a medium voltage in the Bayernwerk network. At the 2,500h threshold, the tariff structure changes: Below that, there is a high working price and low capacity price; above it, the reverse is true. This leads to a significant kink in the slope of the cost curve – even if the absolute jump in costs remains minimal (here: €8). The optimization around this point is therefore economically sensitive.
The 2,500-hour rule represents a kink point in the calculation of grid fees. Up to 2,500 usage hours, a relatively low capacity price (€/kW) and a high working price (ct/kWh) apply; beyond 2,500 hours, the ratio reverses – a high capacity price and a low working price apply. In other words:
Less than 2,500 h/a: The network operator classifies the customer as a low-user. The annual capacity component (for the maximum kW peak) is cheaper, but the working component (for each kWh consumed) is more expensive. This model therefore favors customers with irregular consumption and rare peaks – they pay relatively little for the provision of connection capacity but much per kilowatt-hour consumed.
More than 2,500 h/a: The customer is considered a high-user with a steady load. Here, a high capacity price applies, while the working price per kWh decreases. Thus, a consistent high electricity consumption is rewarded with low working prices, but more must be paid for the constantly demanded network capacity (high kW fee).
This duality ensures that at exactly 2,500 hours, both tariff variants yield approximately equal grid charges. The threshold of 2,500 hours was deliberately set in the German Electricity Network Charge Ordinance (StromNEV), based on empirical values, to achieve a cost-sharing based on causation [1]. Network operators therefore structure their fees with four pricing components: working price and capacity price each for <2,500 h and for >2,500 h. Small consumers with standard load profiles (without registered load metering) are often unaffected and pay flat-rate working prices. However, for larger consumers with load profile measurement (RLM customers), this 2,500-hour system is automatically applied in most network areas.
Why does this rule exist? The goal behind it is the causation-based distribution of grid costs and the establishment of incentives for more efficient network use. Electricity networks have high fixed costs, particularly to cover peak loads. Consumers with highly fluctuating or irregular consumption burden the network with high peaks and long periods of low usage. Therefore, they should contribute more to the costs by paying higher working prices. In contrast, steady consumers are relieved – those who keep the network continuously utilized receive cheaper kWh prices but pay a higher base price for the provided capacity. Overall, this regulation aims to prevent peak-load generators from benefiting at the public's expense and instead rewards network stability and steady utilization.
Battery storage, peak shaving, and the impacts on the 2,500-hour rule
Battery storage can significantly influence the load profile of a company – especially through peak shaving and optimization of self-consumption. This leads to a dual effect on usage hours:
Peak shaving increases the usage duration: If a battery catches peak loads, the maximum grid power consumption (kW) decreases. The annual consumption from the grid remains approximately the same (aside from storage losses), resulting in an increased quotient consumption/peak. High peak loads shorten the usage duration, and their capping increases it accordingly. A company that previously had e.g., 2,400 hours of usage can potentially exceed 2,500 hours with sufficient peak shaving. This would switch it to the category with a higher capacity price and a lower working price.
Self-consumption optimization lowers the usage duration: If less load is drawn from the grid by charging the battery instead of using PV injection, the annual work (kWh) decreases, and thus do the usage hours.
Warning: However, the 2,500-hour rule can have undesired effects on the cost-effectiveness of a battery storage system. One must closely analyze which category one slips into with the storage system and how the tariff structure changes:
If a company was just below 2,500 hours, meaning it had a low capacity price and high working price, peak shaving could push it over the threshold. Then, the capacity costs suddenly increase, and the working prices decrease. Without limiting the usage hours, for example, installing a battery that significantly reduces the peak could result in the usage hours rising from ~2400 to perhaps 3300 – grid fees would be recalculated and could even increase. In our example (500,000 kWh, ~205 kW peak originally), the company paid ~28,000 € in grid fees (as calculated above). Assuming the storage lowers the peak to 150 kW (usage duration ~3,333 hours). Then the expensive capacity price tariff applies: even if the working price drops significantly, the high capacity price could negate the savings. The savings on grid fees from the storage in this case would be low – or in the worst case negative, if the high base fee results in paying more than the savings on kWh costs.
Therefore, in practice, there is often a focus on not unfavorably exceeding the 2,500-hour threshold. For instance, one could control the storage so that the peak is reduced but not too much – so that the usage hours remain just below 2,500. One prefers to retain a small residual peak load and stay in the “cheaper” tariff (with low capacity price), rather than smoothing everything out and slipping into the expensive capacity price. In other words: A certain degree of peak shaving can be sensible, but too much smoothing flips the cost ratio around. This optimization requires careful simulation of the cost structure in advance, as is possible with Lumera.
In summary: A battery storage system can optimize grid fees by reducing peak loads – but one must keep an eye on the 2,500-hour effect. For project developers of batteries in commerce and industry, this means: Before installation, both tariff states should be calculated. It needs to be determined whether the storage system moves the company into a different fee category and whether this ultimately saves costs or incurs additional costs.
Criticism of the 2,500-hour rule and outlook on future changes
The 2,500-hour rule is logically justified from a historical perspective but is currently also under criticism. Critics particularly point out:
Impairment of flexibility: The rigid separation between working price and capacity price rewards a steady electricity consumption, but punishes flexible consumers. In times of the energy transition, it would be desirable if industry and commerce could flexibly adjust their consumption to the electricity supply from renewable energies.
Rigid thresholds: Fixed limits (like 2,500 hours or even 7,000 hours for other privileges) lead to abrupt cost changes once one is just below or above. These jumps are seen as unjust and economically suboptimal because small changes in consumption behavior can lead to disproportionately large tariff changes.
Regional differences and lack of reality reference: The grid fees vary significantly depending on the network area, partly for historical reasons, which means that the same consumption at different locations can incur different costs – not always proportional to the actual use of the network. The 2,500-hour rule itself applies everywhere, but the exact value of the working and capacity prices is determined locally.
In light of these criticisms, it is foreseeable that the grid fee system – and thus the 2,500-hour rule – will be reformed. In fact, the Federal Network Agency presented a position paper for new industrial grid fees in July 2024. It outlines a transition from rigid to flexible incentives [2]. In the future, industry and commerce will receive reduced grid fees if they consume more during high electricity supply and less during electricity shortages.
Excursus: Atypical network usage and the option at <2,500 h
A special case arises with the so-called atypical network usage (§ 19 Abs. 2 S. 1 StromNEV). In this case, a consumer whose annual peak load lies outside the high-load time windows defined by the network operator can apply for an individual grid fee. In this case, the capacity price is no longer based on the annual peak load, but is only calculated based on the simultaneous load in the high-load time window – which can even theoretically be 0 kW for controllable systems like battery storage.
Important: The usage hours are still calculated traditionally – that is, as
Usage hours = annual work (kWh)/annual peak load (kW)
even if the actual peak load is outside the high-load time window. This can lead to inconsistencies in grid fee models: A consumer with few usage hours would normally pay the high working price, even though they hardly burden the grid – for example, by intentionally charging outside the critical time windows.
Therefore, the Federal Network Agency allows a choice: Customers with <2,500 hours may voluntarily choose the tariff structure
above the threshold – that is, lower working price, higher capacity price. However, since the capacity price is calculated based on the (very low) simultaneous peak load, this can lead to a significantly lower total fee.
In practice, the optimal decision – whether to “stay below 2,500 hours” or “choose the option” – depends on several parameters: height of the annual peak load, load distribution, storage control, network area, high-load time window. A suitable simulation tool like Lumera can automatically compare these options and select the economically best variant. Only in this way can it be ensured that battery storage operators do not accidentally slip into an unfavorable grid fee model, but instead realize the full savings potential.
Conclusion and recommendations:
For project developers of battery storage systems in commerce/industry, this means that the planning of current projects must necessarily take the 2,500-hour rule into account – but one should also keep an eye on the future. Currently, the rule is: Optimize grid fees = smooth load profile, but do not fall into the cost trap. Companies should be informed why their storage may not bring the expected savings (keyword: 2,500-hour threshold). At the same time, one should emphasize that the framework conditions may change. During the transitional period until then, it is all the more important to maximize all saving potentials (including existing special fees under §19 StromNEV) and to possibly adapt business concepts to new regulations. The 2,500-hour rule was a step towards fair cost allocation and has created incentives for energy efficiency and load management. However, with the increasing production of renewables and new technologies, it is reaching its limits. A modernization towards more flexibility is emerging – which ultimately should benefit both grid operators and battery storage operators.
[2] https://www.bundesnetzagentur.de/SharedDocs/Pressemitteilungen/DE/2024/20240724_IndustrieNE.html
