Agora puts fossil power off-take contracts at the center of the coal transition

The most important insight in Agora Energiewende's new study on coal contracts in South and Southeast Asia is not about coal plants. It is about delivery obligations. Once you see the delivery obligation as the unit of analysis, not the plant, not the policy target, not the headline phase-down commitment, a different set of questions becomes possible.

Including one that Agora's reform agenda does not quite reach: whether an instrument can be designed that works within an existing coal contract, requires no comprehensive renegotiation, and can be activated with marginal policy adjustment rather than systemic overhaul, allowing clean energy to begin substituting contracted coal delivery now, deal by deal, while the longer reform process plays out around it.

That question is what this post explores. But to understand why it matters, it helps to understand precisely what Agora has found, and where their analysis stops.

What the contracts actually say

Coal PPAs in this region were built for a world of baseload operations. A utility needed reliable, continuous power. A developer needed bankable revenue to finance the plant. The contract bridged those interests by guaranteeing a minimum volume of energy purchases, a capacity payment regardless of dispatch, and a fuel supply agreement that locked in coal offtake at volumes consistent with high utilisation.

When renewables entered these markets, they did so on a separate track: their own procurement programmes, their own support structures, their own PPAs. They were layered on top of the system, not substituted into it. The coal contracts did not adapt. A cheaper, cleaner alternative can exist on the same grid and still fail to displace the contracted coal output.

Agora identifies three specific mechanisms that now act as structural barriers. Minimum energy offtake obligations require system operators to purchase a defined volume of coal power regardless of whether cheaper alternatives are available. Capacity payments, in integrated utility structures, create internal incentives to maximise coal plant utilisation even when it is economically inefficient. Coal lifting obligations in fuel supply agreements impose financial penalties on generators who reduce consumption below contracted volumes. Together, they effectively veto flexibility. This is what Agora means when it says the barrier is contractual.

The repurpose, reserve, retire framework

Agora's central recommendation is a differentiated treatment of the existing coal fleet, grouped by asset age, technical capability, and system role. Young, technically capable plants shift from baseload into a supporting role for renewables integration: filling gaps, providing backup when solar falls at dusk, ramping down when midday output peaks, offering the frequency response that a high-renewables grid depends on. The payment structure changes to match, from flat capacity fees toward a three-part tariff covering flexible capacity, time-differentiated energy, and ancillary services.

Mid-life plants move into strategic reserve, available for adequacy during extreme events but not dispatched in ordinary operations. Older, less efficient units retire, making space for renewable capacity that does the same job at lower cost and lower emissions.

What makes this framework analytically useful is its rejection of the single-pathway approach. Much of the policy conversation has treated coal assets as interchangeable, as though a retirement mandate could be applied uniformly across the fleet. Agora's argument is that this misses the system management problem that emerges during the transition itself. Retiring a young, technically flexible plant before the grid has sufficient storage and demand response may solve an emissions accounting problem while creating a reliability one. Sequencing and differentiation matter as much as direction of travel.

Why renegotiation is slow, even when it is available

Agora challenges the conventional assumption that existing contracts are essentially fixed. Most PPAs in the region already contain provisions that allow for renegotiation in response to regulatory change: change-in-law clauses, tariff adjustment mechanisms, and the evolving legal framework around states' climate obligations all provide a basis for governments to introduce new operating requirements. The legal space for structured renegotiation is larger than is typically assumed.

Agora is equally clear about what that renegotiation requires in practice: it must restore economic equilibrium for investors, respect due process, engage lenders, and be structured to avoid arbitration. A process that requires regulatory mandate, utility agreement, lender consent, investor consultation, and contractual amendment across dozens of individual PPAs, with sophisticated counterparties in multiple legal jurisdictions, will not run quickly. Contract reform at scale is a viable long-term trajectory. It is not a near-term implementation pathway for individual transitions.

The study does not claim otherwise. Its scope is the policy and system architecture. The question of how a specific delivery obligation can be fulfilled by clean generation inside an existing contract, before the broader reform has arrived, is a separate problem that sits downstream of what Agora is doing here.

A different implementation layer

Where Agora establishes the policy case for reforming coal contracts, the question of how to act inside an existing contract before reform arrives sits at a transactional level that the study is not designed to reach. That is where the work SPARC is doing picks up.

SPARC is a proposed market-based mechanism that does not require the PPA to be renegotiated as a starting point. It does not require the coal plant to exit the system, assume new flexibility obligations, or accept a different payment structure from the utility. What it requires is that a defined delivery band within an existing coal contract is registered as a SPARC stream, and that eligible renewable generation fulfills that stream.

When a SPARC stream is fulfilled, a coal-backed contractual MWh is called for delivery. But instead of coal, a renewable energy asset now fulfills that obligation, delivering a megawatt hour of green electricity evidenced through metered delivery data and a linked REC or equivalent energy-attribute certificate, and producing a SPARC record. Where the verified emissions of that delivered MWh are lower than the embedded emission right carried by the stream, an Avoided-emissions Record (ARC) is issued. The ARC records the unused emission right and is evidence of verified impact rather than a modelled counterfactual.

The commercial logic is that the coal operator retains margin through the avoided fuel cost and transferred delivery risk of the SPARC arrangement, while the renewable operator gains a contractual delivery channel that would otherwise require building an independent route to market from scratch. ARC monetisation is a further potential revenue layer, not the primary driver of deal economics. The ARC value is real, but it is the residual gap-closer, not the transaction thesis.

Building on the diagnosis

There is genuine alignment between Agora's diagnosis and SPARC's design. Both focus on the delivery obligation as the real unit of analysis. Both reject the idea that the problem is primarily technical: operators are not holding out because they lack engineering capacity. The constraint is in the contract.

Agora's answer is systemic reform: restructure the payment architecture, differentiate assets by function, give governments the regulatory tools to reopen PPAs without triggering arbitration. That reform agenda, if implemented, changes the rules of the market.

SPARC's entry point is narrower and closer to the transaction. It does not wait for the PPA to be renegotiated. A registered SPARC stream can operate within an existing contract, with a renewable operator fulfilling a defined coal delivery band today, before the broader architecture has changed. The sequencing matters: where policy reform works at the level of the system over years, SPARC works at the level of the delivery obligation, deal by deal, as the reform environment develops.