Why using a capacity mechanism to prop up coal and gas generators is a really bad idea

Paying power generators for their available capacity – on top of the energy they sell to the grid – could lock in high emissions from coal and gas-fired power plants, deter new entrants and add substantially to already high household power bills.

So-called “capacity mechanism” payments are not the solution to supply and cost issues affecting the National Electricity Market (NEM).

The mechanism would pay generators based on availability in periods where the system is “at risk”.

Issues with the capacity payment proposal

The government wants the mechanism to prioritize renewables, new entrants and storage yet the model in its current form does not do this. It proposes to pay all forms of generation capacity – both new and existing – and includes payments to coal and gas.

Coal-fired power stations in the NEM are already confronting outages due to age and financial difficulties which are likely to increase due to extra competition from the large influx of renewable energy supply.

A capacity mechanism would essentially be ‘propping up’ those old coal generators at a huge cost to energy users while acting as a deterrent to new entrants.

Paying existing generators to stay in the system longer is likely to lead to underinvestment in new renewables and storage – leaving the NEM even more exposed to future unplanned coal outages.

The capacity mechanism does not provide certainty around when coal generator exits will occur, but rather delays the exits. This will deliver uncertainty for investors in new replacement capacity.

The proposed capacity mechanism is also ill-suited to valuing flexibility in generation as it pays capacity regardless of how flexible it is.

A study found that capacity mechanisms typically favor low fixed cost resources with higher operating costs – because they can bid more competitively in capacity auctions. Therefore, capacity mechanisms favor the cost profile of fully depreciated, high fuel cost resources like gas and coal generators, rather than low-carbon resources – the most scalable of which have high fixed costs and near-zero operating costs.

Capacity mechanisms also come at a cost to consumers. The ESB has not costed the capacity mechanism proposal, however experience from Western Australia can provide a guide.

In the Western Australian electricity market, the capacity payment each year ranges from a low of $78,573/megawatts (MW) to a high of $186,001/MW.

Such a payment to cover the NEM’s forecast 2022-2023 one-in-two-year (POE50) peak in demand of 37,161MW (per AEMO ESOO 2020) would entail capacity payments of between $2.9 billion and $6.9 billion per year. This could mean a cost per household of $182-$430 per year. This is more than the impact the carbon price would have had, yet at least the carbon price raised revenue that enabled reductions in income taxes and encouraged emission reductions.

This could be a big cost to impose on energy users, who are already facing high bills due to the massive increase in international coal and gas costs filtering through into the Australian market.

Capacity market case studies from around the world

Capacity markets have been implemented in other countries providing valuable learnings for the NEM.

Poland implemented a capacity market, and research showed that “the primary beneficiaries of the capacity market in Poland have been the existing units (including the refurbishing ones) responsible for more than 80% of capacity obligation volumes contracted for 2021–2025”.

Further, while coal was excluded from the mechanism for delivery year 2025 onwards due to high emissions, a study found that though coal-fired units were being phased out in Poland, they were mainly replaced with natural gas. And that “the introduction of a capacity market delays the decarbonisation of the power system and has a negative impact on carbon neutrality”.

The UK also excludes coal from its capacity mechanism as it imposes emissions limits on qualifying capacity. The UK government undertook a review of the capacity market and found:

“Whilst the Capacity Market has seen growing participation in recent years from low carbon forms of generation such as wind and solar renewables, electricity storage, and some types of Demand Side Response (DSR), it has historically secured predominantly carbon intensive forms of generation, particularly unabated gas-fired generation. For example, about two thirds of capacity with agreements for Delivery Year 2024/25 is gas fueled.”

Other nations are in the midst of an energy transition yet do not have a capacity market, including Denmark which has 61% wind and solar in the grid, and Germany and Alberta Canada, which both decided against implementing a capacity mechanism after consultation with stakeholders and experts.

What could be used to manage the energy transition, instead of a capacity mechanism?

Instead of a capacity payment that is likely to be costly and delay decarbonization, Energy Ministers and the Industry should explore other proposals to deliver certainty in managing the exit of coal generators and incentivizing the entry of new low emissions capacity. Improving reserves could also be explored.

Options for providing certainty around coal exits include government agreements to close coal plants once new capacity is built, auctions for closure, regulate closures through emissions performance standards (like the UK), or strengthening penalties for not providing adequate notice of closure.

Options to drive replacement capacity into the system include renewable electricity storage targets (like the REST recommended by VEPC) and government underwriting (like the NSW Electricity Infrastructure Roadmap).

Potential options to improve reserves including capacity reserve, operating reserve and jurisdictional strategic reserve proposals.

There are many alternative routes that can provide better price, reliability and emissions outcomes than the capacity payment proposal.

Johanna Bowyer is an electricity analyst for IEEFA

Leave a Comment

Your email address will not be published.