Yesterday we published an updated look at the Capacity Investment Scheme (CIS) project pipeline, following on from an article series we ran last year titled Are CIS-ters doin’ it for themselves?.

As mentioned in yesterday’s article, a growing share of the industry now appears to concede that the CIS hasn’t proven itself yet, and the subsequent lack of progress for projects that won in early tender rounds looks particularly poor when compared to other state-based policies.

In order to have a better understanding of how the scheme differs under the bonnet, in this article we will look at some of the main design differences between the CIS and the state auction schemes that came before and alongside it. Our focus is on the two most directly comparable — New South Wales’ LTESA scheme and South Australia’s FERM — with lighter reference to the earlier Victorian VRET auctions and the ACT’s reverse auctions. We will close the article with suggested improvements we’ve heard from other members of the industry, mapped back to those key differences.

Key differences between the CIS and state-based counterparts

Although not an exhaustive list, we’ve grouped these differences into five key points about the design and execution of the scheme.

1. Contract mechanics

In simple terms, the main mechanism of how the CIS supports projects is through revenue collars — it underwrites the majority of a project’s net revenue against an annual floor and splits the upside above a ceiling, for a term of up to 15 years. The collars take into account a total revenue position, including revenue earned under eligible wholesale contracts, rather than being based on a fixed $/MWh. As we highlighted on one of our slides during a presentation to CEC members last November, it’s important to note that there are limitations in the type of contracting a project can enter into as part of their CIS agreements. This distinction regarding wholesale revenue is part of why the treatment of curtailment has been difficult to pin down in contract wording, as we explored in Part 3 of the earlier series.

The state-based schemes mostly use simpler instruments, closer to a traditional Contract-for-Difference (CfD) structure. The NSW LTESA gives an option — it gives the operator a series of rights to enter fixed-price cash-settled swaps, exercised at the generator’s election, over a term of up to 20 years. Victoria’s VRET Round 2 awarded capped two-way CfDs settled per MWh sent out, and the ACT’s feed-in tariff entitlements are CfDs against a fixed $/MWh. The mechanics of South Australia’s FERM scheme resemble the CIS the most, underwriting the shortfall below an agreed revenue floor and clawing back the upside above it, though it is narrowly targeted at long-duration dispatchable capacity.

2. Timing and process of announcing successful projects

Curiously, as the Department of Climate Change, Energy, the Environment and Water (DCCEEW) notes at the top of its Closed CIS Tenders page, successful projects are announced at the close of each tender, before or while contracts are negotiated with the project proponents:

The announcement of successful projects is made at the close of each tender. A successful project is still subject to the signing of a Capacity Investment Scheme Agreement (CISA) with the Australian Government.

From Tender 5 onward, DCCEEW introduced several reforms related to this process: successful proponents were given a deadline to execute a CISA, with the Government free to discontinue negotiations if it is not met; and commercial departures were dropped as a merit criterion, and are now permitted only where strictly necessary. The Government has also reserved itself the discretion to place runners-up from each round on a reserve list, and to invite them to contract should a winner fail to execute.’

DCCEEW’s stated rationale for the deadline — ensuring the rapid execution of CISAs — implies to me that negotiations in the earlier rounds did not proceed smoothly. It also raises a question about leverage: how much bargaining power did a proponent hold in contract negotiations once it had already been announced as a winner?

By comparison, both the FERM and LTESA schemes press bids toward execution-ready terms before any award is made. The FERM scheme scores commercial departures as a criterion in its own right, weighted at 20%, with the agreement offered substantially “as is”; the LTESA scheme assesses the materiality of departures from the pro forma Project Documents.

3. Bid bonds and security

The LTESA and FERM schemes both require a bond from bidders. In NSW, the Consumer Trustee has taken a Security for Process Bond at the Financial Value Bid stage to ensure a project reaches contract close, sized round by round in the tender guidelines. The FERM scheme currently sets a bond at $4,000 per MW for conventional generation or $500 per MWh of storage for bidirectional facilities, capped at $800,000, in the form of an unconditional bank guarantee drawable on demand — returned only to a proponent that is unsuccessful, withdraws, or signs. Both schemes take a second tranche at execution — in NSW, bonding provided on execution of the LTESA secures against both financial close and the commercial operation date.

The CIS requires no equivalent bid bond. Its only security applies after execution, and at bid stage a proponent need only evidence its capacity to provide that security in due course. It therefore appears that a developer can win a CIS tender, decline to sign, and walk away having forfeited nothing other than their time and the consulting fees associated with applying for the scheme.

4. The line between government’s role and ASL’s role

Under the CIS, ASL administers the tender and recommend bids, but the energy minister is the ultimate decision-maker on which projects are successful, so government ultimately and literally does “pick the winners”.

By comparison, the state-based schemes place government further away from the award decision. NSW put the recommendation itself outside the department: ASL acts as an independent Consumer Trustee under a statutory duty to act in the long-term financial interests of NSW electricity customers, with financial value the primary consideration by law, and a separate Scheme Financial Vehicle signs. FERM splits the roles a third way again, with ASL as Scheme Administrator and the AER and ESCOSA as Scheme Regulators. A large part of the reasoning for this is that consumers pay for those two schemes directly, through DNSP contributions or a scheme fund, while the CIS sits on the Commonwealth budget and has no equivalent constituency.

5. Transparency after the fact

For the CIS, LTESA and FERM, the auction process is commercial-in-confidence and very little is published beyond the winning projects — nothing about strike prices or contract terms.

The ACT reverse auctions were also run on a commercial-in-confidence basis, but the feed-in tariff prices were publicly disclosed, and the ACT Government has published the monthly settlement flows under those contracts. Victoria’s first VRET auction sat somewhere in between: the CfD strike prices were published, but the fixed payment component and each project’s bid payment cap were not, so the state’s total exposure couldn’t be worked out from the outside.

Practical suggestions for improvement from industry

Many of those differences come down to specific design decisions, most of which were likely made to balance competing objectives or to avoid unintended consequences elsewhere. Rather than prescribe “fixes” to the CIS ourselves, below is a list of some of the suggestions we’ve picked up from our reading, from discussion on the conference circuit, and from the Nelson Review panel.

• Add in bid bonds. This has been the most widely suggested improvement we’ve heard — including by Brad Hopkins (then working for ASL) during the early-stage consultation when the scheme was being designed.
• Publish some level of detail about signed contract prices after each round. Some have argued that greater transparency and reporting of strike prices would help future bidders calibrate their project economics prior to applying for the scheme. The Nelson Review’s proposed Electricity Services Entry Mechanism (ESEM) recommends publishing clearing prices.
• Make room for long-lead technologies. During the long-duration storage session at the Australian Clean Energy Summit (ACES), Reece Stephens from EDF Australia commented that the CIS has not assisted longer-duration technologies in any way, as the policy was singularly focused on the 2030 renewable energy target — and there was next to no chance long-lead-time firming assets would be ready by then.
• Deal with wind on its own terms. Origin’s Frank Calabria said at a recent CEDA event that even with CIS support, Yanco Delta Wind Farm was still not in a position to take FID: “The problem is economics – and it is getting worse, not better.” The Nelson Review contract working group is discussing the use of ex-post dispatch-weighted average (DWA) swaps and time-of-day blocks settled against a reference wind or solar shape, though the CEIG argues those would shift more risk onto wind investors, not less.
• Hold the auctions at a consistent cadence. At ACES, Tim Nelson himself suggested that running auctions on a regular cadence — annually, for instance — would help developers plan their businesses around a known schedule and would settle market expectations more broadly.

Key takeaways

The CIS was always likely to be a more hands-on scheme than its state counterparts. It was the vehicle chosen to deliver a specific 2030 target, and that choice was political as much as it was technical. It was also designed at a far larger scale than the state schemes, and it sits on the Commonwealth budget rather than being paid for directly by electricity consumers. Those circumstances probably account for a fair few of the original design decisions we’ve explored above.

With CIS auctions winding down in 2027 in any case, the question that matters now is which of these lessons get applied to whatever comes next — the ESEM, or something else again.