Re-linking the NEM: could fungible ESEM ‘bulk energy’ contracts be bankable?

Could a renewable revenue contract be both bankable and fungible? For the Electricity Services Entry Mechanism (ESEM) – proposed to replace Australia’s current Capacity Investment Scheme (CIS) – the answer will be critical in determining whether the scheme delivers new investment. 

The economic case for fungibility is compelling. Standardised, tradeable contracts represent an opportunity to re-link medium- and long-term energy markets in a manner envisioned at the National Electricity Market’s (NEM) inception almost three decades ago. However, the implications for project bankability remain somewhat less certain.

Wind and solar farms have traditionally been financed using bespoke power purchase agreements (PPAs), and project debt raised against a tradeable contract remains effectively untested in the NEM.

Three recent works from Griffith University’s Centre for Applied Energy Economics & Policy Research suggest that bankability and fungibility need not be mutually exclusive.

The industry-led Working Group drafting the contract appears to be arriving at a similar position, with recommendations due in November and a first tender anticipated for late 2027.

The problem: incomplete markets in a weather-dependent NEM

It is well known that the NEM relies on a secondary derivative market for forward hedging and price signals. Energy derivatives can reliably facilitate “medium-term” risk management for both buyers and sellers up to around three years in advance.

Sellers use derivatives to hedge and manage price risk positions, while buyers (including retailers and large energy users) use them to manage wholesale cost exposure on behalf of customers.

However, as the NEM transitions to variable renewable energy (VRE), our existing products are no longer fit for purpose. A baseload swap cannot be plausibly defended by stochastic, weather-dependent renewable output. Indeed, the emergence of VRE as the dominant new-entrant technology has created incompleteness in hedging markets.

Price signals weaken with low market liquidity, a phenomenon that economic literature often refers to as the problem of “missing” or “incomplete” markets. Incompleteness carries real costs, inflating the cost of capital through suboptimal risk allocation and, in turn, depressing investment below socially optimal levels.

Of course, imperfect hedging is not new – thermal generators have long sold flat swaps and caps against plant that trips, ramps and requires minimum load, while carrying the residual mismatch. VRE is simply the next, and more acute, evolution of the problem.

Historically, VRE developers have navigated incompleteness by signing long-term run-of-plant (RoP) power purchase agreements (PPAs) to mitigate price risk.

Securing a RoP PPA has generally been required for project finance. Yet liquidity in the long-term market is weak, with a collision of forces eroding buyside appetite for bespoke PPAs. The result is the so-called “tenor gap” between the revenue certainty project finance requires and the far shorter horizons over which buyers are willing to contract.

Governments have filled that gap with off-market underwriting through the CIS and state-based schemes. These policies elicit commitments, but at scale they can draw hedge capacity out of secondary markets, which compounds incompleteness. Asset-linked settlement can also disconnect dispatch from real-time market signals, distorting the merit order – an amalgam of issues which have been described as “incentive incompatibility”.

The ESEM: re-linking medium- and long-term markets

Unlike other entry support policies, the ESEM has been conceived to directly address these live issues regarding secondary market liquidity, while also bridging the tenor gap. In other words, it addresses gaps in both the medium- and long-term markets.

The ESEM’s mechanics are straightforward in principle. A central buyer would procure standardised contracts for bulk energy, shaping and firming through reverse auctions, hold them, and recycle them into secondary markets as they come within medium-term liquidity windows. Draft legislation and rules went out for consultation in September, with submissions closing on 13 October, and ministers are expected to make final decisions in December.

However, to achieve these outcomes, contract fungibility becomes a non-negotiable. Contracts used to support VRE entry must be mechanically aligned with those traded in secondary markets to ensure that resale by the central buyer remains a straightforward administrative exercise. And so, the puzzle of designing a defendable VRE contract, which balances the needs of both buyers and sellers (and, importantly, their capital providers), takes centre stage.

Given the risk appetite of the typical VRE capital stack, and the reliance on debt finance in facilitating affordable entry costs, revenue “bankability” becomes a threshold question. Project lenders are asymmetrically exposed to downside outcomes, making them chief monitors of risk. A minimum level of revenue certainty is paramount. 

Risk allocation: who carries the shape risk?

VRE revenues are exposed to three broad categories of risk: price risk (that energy is worth less than expected), volume risk (that the plant produces less than expected) and shape risk (that the plant dispatches during periods that are less valuable than expected).

Putting aside broader impacts on energy markets, a RoP (or “pay as produce”) PPA design inherently mitigates shape risk by transferring it from seller to buyer. In other words, if a VRE plant operates during windows of the day that capture a lower dispatch-weighted price than anticipated, contract settlement ensures the seller receives the same effective fixed strike price. That transfer extends to both site-specific (“idiosyncratic”) risks and those common to other VRE sites in a given market (“systematic” risks).

A fungible contract cannot, by definition, achieve a perfect transfer of site-specific risks. This is because contract settlement payments are not linked to the site itself. Naturally, a lack of site-specific risk protection raises bankability questions. 

And so, the ESEM bulk energy designs have focused on finding a “middle ground”: transferring systematic risk in a manner familiar to VRE buyers and sellers, while maintaining adequate protections for capital investors. The design path has progressed from the dispatch-weighted price swap and time-of-day block trialled in the 2025 pilot co-design process to the Working Group’s current preference, a “Regional Reference PPA”. A reference revenue swap, which exchanges the reference fleet’s revenue for a fixed payment, was also considered. The table below summarises where each design leaves the risk.

Contract design	Price risk	Volume risk	Systematic (market-wide) shape risk	Idiosyncratic (site specific) shape risk
Run-of-plant PPA	Buyer	Seller	Buyer	Buyer
DWA swap (fixed volume)	Buyer	Seller	Shared (price via index, volume mismatch with seller)	Seller
Reference revenue swap	Buyer	Buyer (systematic)	Buyer	Seller
Time-of-day block	Buyer	Seller	Seller	Seller
Regional Reference PPA	Buyer	Seller (volume tracks the fleet)	Buyer	Seller

Indicative only; the final allocation will depend on contract terms. Both the DWA swap and the Regional Reference PPA settle against an ex-post dispatch-weighted price; the difference is that the latter also scales contract volume with the fleet, moving the systematic volume-price mismatch from seller to buyer.

And so, the threshold question becomes one of materiality – namely, how significant is idiosyncratic shape risk for a typical stand-alone VRE asset? If a site’s deviation from its regional fleet were too significant, projects with ESEM contracts may still be manifestly un-bankable.

The analysis

Three recent works – using different datasets and models across multiple NEM regions – now address the materiality question. 

The first paper, now published in the Australian Journal of Agricultural and Resource Economics, tested two fungible designs from the 2025 pilot – a dispatch-weighted price swap and a four-hour time-of-day block – against a RoP PPA across 100 years of synthetic, stochastic South Australian market data. Under historical dispatch behaviour, site-specific shape risk proved material for poorly located sites, with the weakest site’s median cash flow falling 19 per cent below a RoP equivalent. Yet once projects stopped generating when prices turned negative, the swap supported debt service cover ratios (the cashflow buffer required by lenders) and gearing broadly equivalent to a RoP PPA. Put simply, aligning project dispatch with spot market signals appears adequate to counteract shape risk within the confines of a typical project finance structure. By contrast, the fixed-shape time of day block was un-bankable unless the seller’s exposure was capped.

A second paper tested an “Asian wind swap” – a swap against the wind fleet’s dispatch-weighted average price, of the same family as the DWA swap – using 10,000 bootstrapped years of South Australian data and stochastic project finance modelling across several wind farms. With hedge quantities optimised and strike prices calibrated to entry costs, debt was effectively sized to around 70 per cent of the capital structure, broadly equivalent to a PPA, and revenue variability was virtually identical to that of a RoP PPA, albeit for different reasons. Basis and quantity risk remain, but can be identified, measured, priced and managed.

A third paper tested earnings risk rather than debt capacity across 85 historical weather years of a projected 2033 NSW market. Contracts settled against a regional wind index materially reduced earnings risk, with a revenue swap providing protection comparable to a RoP PPA for two of the three projects studied. Hedge sizing was also better framed as an “earnings hedge ratio” than as physical coverage.

To be clear, none of the three papers tested the Regional Reference PPA directly, although the fixed-volume swap examined in two of them was, on volume risk at least, arguably the more demanding case for sellers. Of course, no hedge is truly perfect. But the message across all three is consistent: site specific shape risk is real and present, yet bankability and fungibility do not appear to be mutually exclusive.

The way forward

The Working Group appears to be reaching similar conclusions. At its 12 August meeting, it moved its preferred bulk energy structure from the DWA swap to the Regional Reference PPA. According to the minutes, the DWA swap’s fixed volume left sellers exposed to both idiosyncratic and systematic volume risk (a potential bankability hurdle), while a revenue swap placed substantial volume risk on buyers, devaluing the contract as a risk management tool. The Regional Reference PPA instead settles on the price and volume of the regional reference fleet, with the floating price potentially capped at $600/MWh to limit sellers’ exposure to extreme prices. In effect, it is a RoP PPA written on the fleet rather than the individual plant: systematic risk sits with the buyer and idiosyncratic risk with the seller, who is incentivised to outperform the fleet.

Perhaps the most consequential open question remains the treatment of negative prices. Three options are being considered: no provision, a $0 floor, or a volume “knock-out” that suspends settlement during negative price intervals. All incentivise sellers to curtail. Through the lens of the analysis, however, much of the bankability gain arises from sellers continuing to receive contract revenue while curtailed; a knock-out would forgo part of that benefit, while a floor would partly preserve it. Settlement during negative prices also bears on the integrity of the index itself – in that volume from plants running below marginal cost, whether due to legacy PPA positions or otherwise, could flow into the reference fleet’s settlement unless excluded.

How negative price periods are treated in the settlement calculation will therefore have implications for both index integrity and bankability. Force majeure protections, which the Working Group has agreed cannot sit within a fungible contract, also remain to be resolved. Though here, fungibility is arguably the solution rather than the problem – just as a coal generator that has sold a baseload swap can buy cover back during an outage, an ESEM seller could do the same in a liquid secondary market, with the central buyer potentially acting as a seller of last resort. In other words, neither issue appears insurmountable.

Takeaways for sellers and lenders

The ESEM encourages a rethink of the way risk is hedged in Australia’s NEM, in a subtle but significant way. Its objectives are ambitious, but well founded and appear tractable. The re-linking of medium- and long-term markets is an opportunity that should not be taken lightly.

For sellers, the task becomes to assess how well an ESEM contract hedges their specific project. Any delta in site-specific shape risk will need to be accounted for in tender bidding – though this also inherently promotes projects in stronger locations. Dispatch capability, including economic curtailment and flexible bidding, is part of the contract’s value rather than an operational afterthought.

For lenders, a sharper focus on dispatch behaviour and site-specific risk becomes essential. Downside cases will need to model dispatch strategy as well as weather, acknowledging that cover ratios comparable to RoP financings appear achievable where dispatch is rational and tail exposure is capped. For policymakers, the negative price provision and the cap are, at heart, bankability decisions.

According to current modelling, a well-designed fungible contract ought to be bankable. Almost three decades after its inception, the NEM has a rare chance to re-link the markets its designers envisioned – and on current evidence, bankability need not be the price of that ambition.

Nicholas Gohdes is a PhD candidate at Griffith University’s Centre for Applied Energy Economics & Policy Research. “Alternative Contract Design: Analysing the Bankability of Fungible Derivative Contracts in Energy-Only Markets”, co-authored with Tim Nelson and Philip Hirschhorn, was published in the Australian Journal of Agricultural and Resource Economics in 2026.