A Toll Order

Several Models for Marketing a Battery

First, an overview of the family of contracts. Virtually all models can be traced back to three basic types, distinguished above all by who bears the market price risk:

Merchant. The optimiser dispatches the battery across the balancing, day-ahead and intraday markets and receives a share of the revenue. The operator bears the full market price risk – and the optimiser’s performance risk on top.

Tolling. The operator “leases” the capacity of the battery to an offtaker and receives a fixed tolling fee. The offtaker dispatches the battery and keeps all the marketing revenue. Market price risk thus shifts entirely to the offtaker. Tenors of five to ten years are currently market standard.

An important distinction runs between the physical toll and the virtual toll (also called financial toll). Under a physical toll, the offtaker integrates the battery into its own balancing group: the offtaker sends dispatch schedules, the operator executes them, and the offtaker carries balancing-group responsibility.

Under a virtual toll, the battery remains operationally with the operator (or its service provider) and settlement is financial, as if the offtaker had optimised the battery itself – typically against an index. Structurally, this is often a swap.

Tolling is the “gold standard” for bankability because it provides a complete hedge against market price risk (provided the offtaker is sufficiently creditworthy).

Against the background of declining revenues in the coming years, pure merchant models are barely bankable for larger projects unless they reach the grid in the very near term. Merchant nonetheless remains relevant as the “merchant nose” and “merchant tail” – the phases before and after a hedge period – and as a component of hybrid and floor structures.

Floor. The offtaker guarantees the operator a minimum revenue. Revenues above the floor are shared, often on an 80/20 basis. The floor secures the downside but leaves the operator with a substantial share of the upside. From an investor perspective, this has an another advantage: the operator stays close to the asset and retains stronger incentives to optimise well than under a pure toll.

Portfolio and Hybrid Structures. Alongside the basic models there are hybrids – virtual slicing across portfolios of several batteries, tolling on a portion of capacity with merchant marketing of the rest, and insurance products structured as financial swaps that protect the operator against the risk that market volatility collapses and arbitrage opportunities shrink.

In co-located projects – battery and PV or wind the same grid connection – further questions arise because the assets must share connection capacity. Who gets to feed in when the sun shines and the wind blows: the battery, which could capture a high price right now, or the renewables asset, which enjoys priority dispatch under the German Renewable Energy Sources Act (EEG)?

And when, if at all, can the battery actually charge to capture the evening peak – if the cable is already taken up by PV feed-in during the day? The key here is the optimiser’s “grid headroom forecast”: its continuous assessment of how much free connection capacity is actually available at any given moment – that is, whether the battery can charge or discharge without colliding with the PV power. A poor forecast translates into lost revenue or imbalance costs.

Merchant Topics: Where the Profit Share Quietly Crumbles

Which costs may reduce the revenue? Everything in a profit-share model turns on the definition of “revenue” on which the share is calculated. Rising or newly introduced costs – new grid fees, taxes, higher maintenance costs – can shift this base materially. From the operator’s perspective, these are costs of doing business that reduce profit. From the optimiser’s perspective, these are burdens that fall within the asset’s risk sphere – and therefore on the operator.

The answer typically lies in a clean list: which cost categories are deductible, which are not, and what happens with new burdens that no one knows about today? This is the interface to the change-in-law clause, and the two must be aligned.

Redispatch compensation as “revenue”? A small but expensive question: when the network operator curtails the battery and the asset is entitled to compensation, does that compensation flow into the revenue base on which the profit share is calculated? Optimisers answer “yes” – after all, they take on the work of managing redispatch claims. Operators see the redispatch compensation more as damages for a service that could not be rendered.

The pragmatic solution is usually a tiered mechanism: the compensation counts as revenue, but the optimiser receives a separately stated processing fee for it – and no profit share on the bare compensation.

Audit, Non-Discrimination, Benchmarking. Economically, this is the trickiest point of all: how, exactly, does one check whether the optimiser is marketing the battery well? Optimisers trade at balancing-group level; the profit share refers to the individual battery. How can it be ensured that the good trades are not allocated to the optimiser’s own book while the bad ones are dropped onto the asset? And a further discrimination issue: if the optimiser earns more under a different marketing contract, will it dispatch the other battery preferentially?

Open-book processes and audit rights can help, but are not easy to implement in practice, because the optimiser’s trades and marketing strategies are technically hard to verify. Beyond that, the practice falls back on representations and warranties, and termination rights triggered if certain economic benchmarks are missed. In the end, the only effective remedy often remains terminating the contract and switching to another optimiser.

Allocation of Imbalance Costs. The operator must report planned and unplanned unavailabilities. If it fails to do so, or does so too late, imbalance costs may arise. These are costs the optimiser does not want to bear itself.

The clean solution allocates by causation: faulty or delayed reports by the operator are its problem (with the follow-on question whether the operator can in turn recover from its O&M contractor); imbalance costs caused by faulty forecasting, IT failures at the optimiser or by a risky marketing strategy fall to the optimiser. Simple in theory, intensely negotiated in practice – above all on the question of where “risky marketing” begins.

What Matters in the Tolling Agreement

With tolling agreements too, the details matter. Below we look at a few important topics.

Dispatch Restrictions. High volatility in the intraday market tempts operators to dispatch the battery aggressively – three or four cycles per day look commercially attractive. But the manufacturer’s warranty typically covers only one or two cycles. Anyone who discharges too early, too often or too deeply risks not only faster degradation, but the loss of the warranty altogether.

The typical contractual answer combines two mechanisms. First, an Annual Energy Throughput Limit: the agreement caps the maximum annual MWh that the battery may cycle. Second, a Marginal Cycle Cost: if the offtaker wants to dispatch more, it pays a surcharge, typically credited to an Augmentation Fund – a budget that later finances the early replacement of components. This is complemented by guaranteed round-trip efficiency values (which decline over the asset’s life).

Technical Availability. Availability clauses look simple – until you look more closely. What matters is not only which availability is guaranteed, but also whether the operation & maintenance contract provides enough headroom for the operator to meet its contractual undertakings to the offtaker. This is a classic back-to-back issue: if the operation & maintenance contract grants the service provider more downtime than the tolling agreement allows the operator, the operator finds itself caught in the middle.

Grid Availability. What happens when the network operator curtails the battery (redispatch), or the network simply fails? Operators argue that this is force majeure or must be treated as deemed availability – the battery counts as available. After all, the operator has no influence over the grid. Offtakers counter that location risk lies with the operator. Pragmatic solutions usually meet in the middle here as well – for example, treating redispatch (with compensation) differently from grid failures (without compensation).

Change in Law: Who Wins, Who Loses at Germany’s Regulatory Roulette?

No corner of Germany’s energy sector has shifted as quickly over the last 24 months as the law governing battery storage. The reform of storage grid fees being negotiated in the Federal Network Agency’s AgNeS process and the looming loss of the grid-fee exemption under § 118(6) of the German Energy Industry Act (EnWG); the phase-down of avoided grid fees set in motion by the Federal Network Agency’s February 2026 ruling; a draft bill from the Federal Ministry of Economic Affairs that would impose a “redispatch reservation” on new assets in “capacity-constrained areas”; the unresolved treatment of multi-use batteries at the intersection of the MiSpeL proceeding, AgNeS, and § 14a EnWG; on top of that, an ongoing debate about splitting the German bidding zone – the list is long. Anyone signing a long-term tolling agreement does so in a shifting regulatory landscape.

The German Civil Code (BGB) sets a high bar for unilateral contract adjustments based on changed circumstances. As a starting point, you stick with the contract you signed. § 275 BGB releases a party from its obligation to perform when performance becomes impossible — a threshold rarely reached.

§ 313 BGB recognises the doctrine of frustration of contract (“Wegfall der Geschäftsgrundlage“), but requires a serious, unforeseeable change that makes adherence to the unaltered contract unreasonable. “Unforeseeable” is tricky: the industry is already aware of the main reform pathways. “Unreasonable” is trickier still: a mere shift in margin is not enough.

In practice, the EFET standard for power purchase agreements provides a common reference point. § 16(1) EFET distinguishes two cases: a change in law may render performance of contractual obligations impossible or – at a lower threshold than § 313 BGB – materially and adversely affect the benefit of the agreement. The legal consequences are negotiable.

§ 16(2) EFET is worth noting: an adjustment of the contractual remuneration may optionally be excluded as a legal consequence – an important lever for tolling agreements, where the bank is relying precisely on a stable tolling fee. And § 16(5) EFET clarifies that mere market price fluctuations, absent a change in law, do not justify any adjustment.

Bankability: What the Bank Wants to See

From a bank’s perspective, tolling agreements are preferable because they deliver exactly what non-recourse project financing requires: a contractually secured cash flow. The bank’s only security is the asset itself; it has to hedge the downside (it does not share in the upside). For precisely that reason, the hedge structure – tolling or floor – is indispensable.

That, however, is only the basic structure. For a route-to-market agreement to be bankable, several building blocks need to fit together.

Offtaker Creditworthiness. Banks require an investment-grade offtaker or adequate security. What happens if the rating deteriorates during the term? The usual answer is a contractual obligation to post alternative security – a bank guarantee, for example.

Direct Agreement with Step-In Rights. If the operator can no longer perform, the bank wants time to install a replacement operator, rather than being squeezed out of the contract immediately. A direct agreement between operator, offtaker and bank governs this. During the step-in decision period, the bank wants the offtaker to keep paying, even though there would otherwise be grounds for termination.

The typical compromise: the offtaker keeps paying, provided the operator (or the replacement chosen by the bank) keeps performing.

Right of First Offer. Some offtakers insist on a right of first offer in the event of early termination – they want to prevent the operator from manufacturing termination grounds in order to find a new offtaker on better market terms. Banks naturally dislike the right of first offer, because it constrains their enforcement options. With clean drafting, it is nonetheless bankable.

Change of Control. Offtakers want to prevent the operator from turning into an incompetent player – or, worse, a competitor – through a change of ownership. Banks want to keep their realisation options open. The typical compromise: the offtaker may not object to a change of control, provided that the battery continues to be operated properly after the ownership change, no direct competitor of the offtaker gains control, and compliance and KYC requirements are met.

There are further matters to consider at the edges of a route-to-market contract: a well-designed warranty stack between the manufacturer’s warranty, the EPC contract and the O&M contract that allows no warranty leakage; an insurance package that adequately covers lithium-specific risks.

And the list does not end here. A route-to-market agreement looks straightforward enough at first sight – but in the details, it turns out to be a bit complicated after all.