New Energy Expert Insights: A Brave New Post-5MS World

This article is a follow-up to our two-part series (Part I and Part II) on the landmark shift in the National Energy Market (NEM) from 30-minute settlements (30MS) to 5-minute settlements (5MS).

We again sat down with Quintas Energy to take stock of what has happened over the last few weeks as the market has responded to the new conditions of the post-5MS world.

What have been the short-term effects in the market since the commencement of 5MS?

In the weeks immediately following the commencement of 5MS, there has been significant price volatility. However, it is not entirely clear whether this has been due to 5MS, other factors, or a combination of both. Continued periods of negative pricing have led to renewable generators bidding such that they fail to clear the market and curtail their generation. However, the market has also seen positive prices occurring in close proximity to negative prices. This price volatility has introduced a significant amount of uncertainty for market players.

Renewable generators typically refer to the local dispatch price or the connection point dispatch price in their bidding strategy as these prices inform the generator’s optimal bidding strategy to determine whether the asset is dispatched or constrained off. These local constraints have taken on a larger role in the post-5MS world as a number of renewable generators have switched their default bid from the price floor (-$1000/MWh) to their financial break-even price (which is typically “negative LGC”, or around -$40/MWh). Whereas under 30MS, most renewable generators were bidding low and raising bids to avoid negative prices, under 5MS, many renewable generators have changed strategies, and are now bidding at their break-even price, and lowering bids to mitigate the impact of constraints.

Short run price volatility coupled with more dynamic bidding and the increased relevance of local constraints under 5MS have all highlighted the importance of automated trading and bidding. These factors all amount to an increased volume of information and data to assess in calibrating an optimal bidding strategy.

So far as local constraints go, “this has upped the ante for our clients who have never invested in auto trading software”, Joanna Leigh, Quintas Energy’s Country Manager in Australia says. “While this space is constantly evolving, what is clear is that the post-5MS world is irretrievably different and more complex, requiring market players to be more adaptive and proactive.”

What mitigation techniques have renewable generators been using to offset any of these short-term fluctuations?

It has now become apparent that the key mitigation technique in the post-5MS world will be automated algorithm-based bidding strategies. Further, the algorithms themselves will have to continually evolve and learn from the more frequent bidding intervals under 5MS to remain fit-for-purpose.

“As the market transitions, so too will the algorithms that forecast and predict the optimal bidding price and strategy”, Joanna says. “All indications are that this is not merely a mitigation technique – temporary in nature – but is here to stay as a critical part of any viable bidding strategy in the NEM”.

What patterns are emerging and do our original predictions still hold true?

While price volatility is the major feature of the introduction of 5MS, what is now becoming more and more apparent is the heightened importance of automated bidding for renewable generators.

In our previous article, we ultimately concluded that ‘set and forget’ was no longer a viable bidding strategy in the post-5MS world. This prediction has been borne out in the subsequent experiences of the market since the introduction of 5MS. Automated trading is now not merely an ingredient for success but a baseline requirement to remain competitive in the market. Previously, market players were able to get away with doing the bare minimum in due diligence to ascertain whether they had the best algorithms to forecast prices with high accuracy, relative to their competitors.

We also predicted that the dynamic bidding environment of the post-5MS world would incentivise the use of batteries coupled with renewable energy projects. The experience of the market so far post-5MS is that price forecasting will need to be dynamic and made on a rolling basis. Batteries are uniquely placed to be able to ‘fill the gaps’ at short notice and this shift will become more apparent in the coming months and years.

What does the future look like?

It is often said that it is difficult to make predictions, especially about the future. However, what is clear is that market players will need to deploy automated, algorithm-reliant trading strategies in the post-5MS world. Matt Baumgurtel, lead partner in Hamilton Locke’s New Energy team, says that “there is a clear parallel in this regard with the early evolution of the financial stock markets which adopted highly sophisticated frequency trading based on algorithms and it appears that lessons learned from that may be applicable in the energy markets”.

A key consideration post-5MS is that a renewable generator’s decisions to bid up or down will have to be made on a constant, rolling basis. The 5MS situation is far more complex than under 30MS; while there was a clearer start and end of the bidding period under 30MS, the 5MS world is more dynamic and fluid. A solar farm, for example, faces ramp up and down rates, which factor significantly in its ability to ramp output up or down in response to price signals. This complexity is compounded by the fact that 5MS requires more frequent and more accurate price forecasting. A decision to turn off in a negative price period might mean losses are avoided, but the solar farm might incur opportunity costs during higher prices in a subsequent dispatch period due to its ramp up rate.

Joanna Leigh says that “renewable energy suppliers in this situation face a complex trade-off, amplified by 5MS, between losses avoided versus gains foregone.”

This all shows that the field of play in the energy market will increasingly move towards intense competition in computing power and software complexity. The winners will be generators who are able to deploy the best algorithms in order to accurately forecast prices. This will involve machine learning and continuous minute-by-minute updating of algorithms.

Matt Baumgurtel says that “the real power in the market is going to be with the people who have the best algorithms.”

In the case of batteries especially, algorithm-based trading will be critical. These storage technologies are nimble and able to adjust strategies quickly, but have limited energy to trade in the market – batteries must choose their opportunities wisely. Due to the near instant computation and response times required, it will prove near impossible for manual, human operated bidding strategies involving batteries to be effective in the post-5MS world. However, the commercial opportunities are real: accurate price forecasting from a dynamic algorithm coupled with the responsiveness of batteries would make these energy sources ideal in the post-5MS world.

5MS has introduced significant complexity in the NEM. Market players that fail to adopt automated, algorithm-reliant bidding at the heart of their trading strategies will be left behind. However, with uncertainty and complexity comes an opportunity for those who embrace change and the technologies best suited to navigate it.

The Hamilton Locke team advises across the energy project life cycle – from project development, grid connection, financing, and construction, including the buying and selling of development and operating projects. For more information, please contact Matt Baumgurtel.