We propose to design, implement and trial a weekly cleared price auction for a volume of frequency response. The aim of the trial is to test the hypothesis that closer to real-time procurement of frequency response will lower overall procurement costs by increasing liquidity and transparency in the market, and deliver a stable market price for the relevant products.
Objectives
At the end of the project, the ESO will understand the impact of closer-to-real-time procurement on:
- Market liquidity
- Diversity of providers
- Ability for renewables and DSR to bid
- Market and price transparency
- Market and price stability
- Cost of frequency response vs counter-factual
- Cost of administering frequency response market vs counter factual
- Internal systems and processes
Learnings
Outcomes
- In July 2020, we published an initial evaluation report conducted by an independent consultancy. The report covered the following auction aspects: design; algorithm; submission process; participation; results; and EPEX SPOT platform, as well as suggested further improvements. The document can be accessed here.
- As part of the evaluation, we conducted a market survey and received 18 responses in total. One of the questions explored whether the Auction Trial was successful in lowering barriers to entry. The participants rated the importance of this route to market as 8.6 out of 10. The scores were particularly high among the demand-side response providers, demonstrating that the closer-to real-time procurement lowers barriers to entry for participants who are unable to forecast their long-term availability.
“Weekly auctions allow for the dynamic nature of some DSR sites, where operational patterns cannot be planned a month in advance. It allows for participation of more capacity, and therefore a more competitive and lower cost market”
- The survey feedback identified that the standardised design of the DLH product prevented providers with asymmetrical capabilities to participate in the market. As a result, we have decided to separate future procurement of DC into low and high frequency with an option to link bids by product. This improvement increases flexibility for both ESO and market participants by allowing providers to tailor bids as most beneficial to them, i.e. they can provide a one-sided frequency service (either xMW of DCL or yMW of DCH) or a two-sided service but with asymmetrical capability (xMW of DCL and yMW of DCH). This functionality is expected to go live once Dynamic Containment high frequency is launched at the end of 2021.
- The weekly operational processes, such as order submission and auction clearing, have proven to be robust and as referenced above, these processes form the basis of our improvements to how we currently procure DC. One provider noted that further improvements could be made to an error checking process at the submission stage. Following this request, we are introducing an automated assessment process that provides real-time order validation for DC procurement. This function sends immediate feedback to providers who then can modify their orders up to gate closure, hence reducing the number of non-compliant orders.
- Over two thirds of the respondents indicated they would be interested in moving to day-ahead procurement for the new response services but expressed their concern that this might become more admin intensive, especially with weekend submissions. To address this query, we extended the DC procurement window to 14 days ahead of auction, removing the need to work weekends and on holidays.
- We have listened to the participants’ feedback and published a document on the paradoxically rejected orders, which could be accessed on our website. This document received positive feedback from providers, who are also learning by doing. Whilst EFA blocks aren’t the desired end state for ESO procurement, moving closer to real-time provides increased transparency to industry.
Lessons Learnt
There are several processes that ESO and market providers follow in the auction trial which could be improved from a time saving perspective. We have used the weekly auction trial process as a basis for designing the daily DC PAC procurement that is due to launch later this year.
- We are feeding in our experience from the weekly auction trial into the enduring auction platform project, which we have committed to deliver in our RIIO-2 business plan.
- We know participants are looking for API functionality, which reduces manual effort and data errors. This requirement will be addressed in the Single Market Platform programme.