Electricity market 2024: Wholesale market developments

Electricity market 2024: Wholesale market developments

While year 2023 was still plagued by the consequences of the energy crisis, in 2024 we were finally out of the woods. Although judging by the latest events last November and December, maybe the type of woods just changed. More on this below.

Generally speaking, 2024 recorded substantially lower day-ahead market prices across Europe compared to 2022-23 but – and this is a big BUT – they never got back to the pre-crisis levels. At the same time, price volatility as grown.


Hourly average price evolution in the Single Day-Ahead Market Coupling (2019-2024)*

Next to overall volatility, both the frequency and the magnitude of negative prices increased across the EU (+Norway). Absolute “winners” by the number of hours with negative prices are Finland and the 4 Swedish price zones. There negative prices were observed about 8% of all hours. ?By the number highly negative prices, -50€/MWh and below, the Netherlands and Belgium are leading the way. Interestingly, negative prices showed face in the countries where none were observed before, such as in Greece and Bulgaria.


Frequency of negative prices per BZ in 2024 (01/01/2024 – 22/12/2024)*

Looking at the regions, we can observe a few noteworthy events in the second half of the year:

1)??? Price spikes in Southern and Central Eastern Europe over summer and a June blackout

2)??? Go-live of Nordic flow-based market coupling and

3)??? Dunkelflaute events in Germany and its neighbors in November.

Summer price spikes in Southern and Central Eastern Europe

The countries and the Balkan region and their neighbors in Central Easter Europe, such as Hungary, Slovakia or Romania and Bulgaria, experienced extreme price hikes this summer with prices in some hours scratching 1000€/MWh in some of the countries. Looking at the weekly averages, the prices between June and September 2024 were 1.5 to 2 times higher than the SDAC average over multiple weeks.

?

Weekly SDAC average vs. SEE bidding zones’ average price evolution over June until September 2024*

What appears to be the reason for these events? Two main reasons, climate change and limited market integration in the area. More specifically, high peak demand in the summer months resulted from extended heat waves. Periods of severe drought affected the availability of domestic hydro generation, which was exacerbated by lower availability of nuclear in the region due to outages. That would have been not all that bad if more supply could have been imported from elsewhere. This is where the second reason comes in. Located, so to speak, at the periphery of the market-coupled area and bordering non-coupled (non-EU) Balkan states, the situation with imports can get rather tight, e.g. due to import constraints on the Austrian Southern and Eastern borders. Then, simultaneous events of tight supply in the region and beyond create negative chain effects. These events prompted some sharply criticize the flow-based market coupling algorithm, which the Greek prime minister called “an incomprehensible black box” while the Romanian Energy Minister event sought price compensation from the EU.

Nordic flow-based go-live

One of the major wholesale market changes is arguably the go-live of flow-based capacity calculation in the Nordic region on November 1, 2024. The “incomprehensible-black-box” sentiment extended to this region, the (sometimes non-intuitive) FBMC results caused quite a stir.

According to the results of a simulation performed by the Danish TSO, Energinet, the overall welfare in the region was expected to increase by more than 180 million euros, overall net exports to increase by about 5%. But at the same time, the prices were projected to increase by 4-6% on average. This does not sound like much but sufficient to raise eyebrows in the region where low electricity prices are essentially seen as an inalienable consumer right whereas the share of dynamic price contracts pegged to DA prices (so-called Nordic System Price) is the highest in Europe.

To take a step back, the overall motivation for introducing FBMC in the Nordics (following the Cre Capacity Calculation Region), is to This shift to enhance accuracy in capacity allocation and increase overall cross-border flows thus creating high overall welfare gains. This, in turn, introduces complexities in comparing current Nordic System Price and FB-based results. The Nordic System Price acts as a reference price for future contracts and refers to an unconstrained market-clearing price calculated using anonymized order books from Nordic bidding zones and hourly planned import/export flows with neighboring regions. According to Nordpool, FB capacity calculation can alter electricity flows on the nine external borders between the Nordic region and the rest of Europe. This can lead to differences between the system price and the production system price, primarily due to changes in import-export flows and the different ways capacities are handled.

Dunkelflaute events in Germany and its neighbors

Dunkelflaute, one of notoriously famous German words meaning conditions of low light and low wind limiting (eliminating) generation from RES, is the Boogeyman of the energy transition. No matter what the discourse about RES is, one can always retort with “But-the-Dunkelflaute!” scare.

But last year, the Boogeyman, in fact showed up, more specifically on November 6, 2024, in Dunkelflaute’s motherland. This led to massive – yet still insufficient – imports from Austria, Czechia and France. Result? Prices skyrocketed to up to 800€/MWh. The effect rippled towards Germany’s neighbors with prices reaching 720€/MWh in Austria, 770€/MWh in Czechia, 550€/MWh in the Netherlands and 360€/MWh in Belgium in the same hour. Main reasons for this event are unsurprising: a temperature drop in the region drove up the demand while very low wind conditions were registered across most of CWE. Cold November caused higher-than-average prices of over 100€/MWh in the region.?

Net power generation in Germany in November 2024 **

To make the matters worse, Dunkelflaute showed up again in December, namely on the 12th and 28th. At least this time around the notorious Merit Order is not to blame! Outcome? More political turmoil and finger-pointing. But what is the learning? Dunkleflaute as a natural phenomenon always existed. With RES expansion is actually became tangible. Despite the fearmongering, stubborn statistics shows it is far from frequent. However, exactly in those specific hours it does put security of supply and system reliability as risk. Big time. Thus diversification of supply and flexibility matter. The matters can get much worse, if there is insufficient interconnection between bidding zones. It is exactly in these moments that it becomes particularly evident how much of a blessing an interconnected European power grid and its integrated markets are. It also reminds me of the Texas power crisis of February 2021, where a cold snap caused rolling outages leaving millions to freeze in their homes without power. Main reason? Texas is an energy island.

?Moving closer to real time, what are the main highlights from the intraday market?

The year 2024 was quite eventful on the intraday market as well. The key changes implemented in Single Intraday Market Coupling (SIDC), are:

1)??? Introduction of Core flow-based intraday capacity calculation (IDCC) in May and June, 2024.

2)??? Go-live of intraday auctions (IDAs) in June 2024.

IDCC(a-b) go-live

The FB IDCC process provides cross-zonal transmission capacities to the Single Intraday Coupling (SIDC) in Europe, aiming to enhance market integration and improve electricity price convergence across the Core region. The first steps involved capacity updates delivered to the market by D-1 15:00 (IDCC(a)) and D-1 22:00 (IDCC(b)). Future stages will include additional intraday capacity calculations to align with the Core region’s target model pursuant to the requirement of CACM Article 20(2). This evolution is aimed to ensure more accurate and efficient allocation of cross-border capacities in the ID timeframe.

A key distinction between IDCC(a) and subsequent processes (IDCC(b-e)) consists in the main approach: IDCC(a) updates capacities remaining after the Day-Ahead Market without re-computing flows (so-called DA leftovers), while IDCC(b-e) involve a full flow-based calculation, which differing in input grid models and timeframes, as clarified below.

More detailed steps per IDCC are described by ENTSO-E.

SIDC introducing a new product line: IDAs

IDAs were introduced to complement (not substitute) continuous ID trading in the Single Intraday Coupling (SIDC) by pricing cross-border capacities in the intraday market. Unlike continuous trading, which allocates capacity on a first-come, first-served basis, IDAs harmonize capacity calculation and allocation to handle scarce transmission resources. By pricing intraday cross-border capacities based on their availability, IDAs aim to reflect real-time scarcity and provide clear price signals to market participants. This implementation is in line with ACER’s Decision 01/2019 on the pricing of XZ capacity in the intraday timeframe.

IDAs operate as implicit auctions (similar to SDAC), simultaneously matching orders and allocating cross-zonal capacities for various bidding zone borders using the EUPHEMIA algorithm. Gate closure times for IDAs are set at: IDA1 (D-1, 15h), IDA2 (D-1, 22h), and IDA3 (D, 10h). The algorithm supports hourly, half-hourly, and quarter-hourly products, with additional features like block orders.?

Continuous ID and IDA timeline ***

There is then a clear link between IDCC and IDAs. During IDAs, cross-zonal capacity allocation for continuous trading is suspended for a period of 40 minutes (20’ before and 20’ after the GCT of an IDA) to avoid double-allocation. These capacities are calculated somewhat differently across CCRs. Using the example of Core CCR:

  • For IDA1, cross-zonal capacities (CZC) are updated based on the Intraday Capacity Calculation (IDCC(a)) methodology. This involves extracting Intraday Available Transfer Capacities (ATCs) from the D-2 Common Grid Model (CGM) using Day-Ahead Market Clearing Prices (DA MCP) as a reference. However, the parameters for virtual capacities may differ, and complete removal of these capacities is permitted. Specific exceptions apply to certain borders, as detailed in accompanying documentation.
  • For IDA2, the updated CZC is determined using the IDCC(b) methodology. This calculation is based on the D-1 CGM and incorporates the most recent information on cross-zonal exchanges (referred to as the market point).
  • For IDA3, until March 26, 2024, the capacities available for continuous trading in the XBID platform at 09:40 are also utilized for IDA3. After this date, the updated IDCC(d) methodology is expected to be implemented. Under IDCC(d), CZC will be calculated based on the Intraday CGM and the latest information on cross-zonal exchanges (source).

It is important to note, however, that these again refer to capacities for cross-zonal ID trade. By definition, intrazonal trades are not affected.

?What are the results so far?

IDAs volume over the first six months since implementation in June 2024 kept creeping upwards with most volumes traded at IDA1.

Traded volumes in MWh (top) and percentage shares of the three IDAs (bottom) in June – November 2024***

However, their total volume is still minor, as compared to continuous trading, about the 4th of the total volume traded in intraday.

SIDC traded volume in ID continuous and IDAs in January – October 2024 ***

?Now, as a market actor, why would I trade on IDAs instead of ID continuous? Two key words would be liquidity and more robust price signals. By aggregating orders by specific GCTs liquidity is pooled around those times as compared to continuous matching. Compared to the first-come, first-served approach of continuous trading, cross-zonal capacity is allocated more efficiently and the IDA outcome is based on an optimization. This also ensures that prices reflect the value of the available capacity at specific times, which can lead to more accurate pricing signals. It can also save some money to smaller market actors on algo trading software ;)

So while continuous trading offers flexibility and real-time responsiveness, IDAs provide structured, more transparent, and likely more cost-effective outcomes, especially for cross-border trades.

In the next installment of this post series, I’ll take a look at the developments in other markets, such as balancing and capacity markets.

*Kudos to Chitransh Lot for the DA market graphs in this post! Source of data: ENTSO-E Transparency platform.

**Source: energy-charts

*** Source: MCSC

Federico Rossi

Studente presso Università degli Studi di Milano

2 周

Ksenia Tolstrup What direct or indirect effects, in terms of costs that can also be indirectly traced back to the bill for end users, can be seen in the increase in volumes on XBID?

回复
Anibal Sanjab

Senior Researcher at VITO/EnergyVille

2 周

Very nice summary! Thanks for sharing :-)

Andreas Clark

CEO, Fr?lunda Tech Holding AB

1 个月

Academic topic for discussion: PUMPED STORAGE RESERVOIRS (NORWAY): - Skjerstad fjord (25 TWh) - Mj?sa (17,6 TWh) - Sandsfjorden (8,1 TWh) - Tingvollfjorden (4,1 TWh) - Lysefjorden (3,6 TWh) - Beitstadsfjorden (3,2 TWh) - Tinnsj? (2,8 TWh) - Hornindalsvatnet (2,6 TWh) - F?rdefjorden (2,4 TWh) - Troms? (2,5 TWh) - Tyrifjorden (1,9 TWh) - Salsvatnet (1,6 TWh) - Fyresvatn (1,5 TWh) - Suldalsvatnet (1,5 TWh) - Lundevatnet (1 TWh) - Less-Than-1-TWh-Reservoirs (3,8 TWh) Grand Total: 83,2 TWh . Downloadable (more detailed) table: https://lnkd.in/dEhEdniu . BACKGROUND: - Large quantities of electric energy can be stored in norwegian fjords (since they are very deep): - Wikipedia(Pumped Storage): https://lnkd.in/gTHZrjj . (FJORDS) Time-dependent wind/sun --> --> Water pumping (fjord --> sea) --> Height diff (fjord<-->sea) --> Hydropower (Or) (sea --> fjord) (Or) (upstream non-saline water --> fjord) --> "On-demand wind/solar electricity" . (LAKES) Time-dependent wind/sun --> --> Water pumping (lake --> downstream river) --> Height diff (upstream river <--> lake) --> Hydropower (upstream river --> lake) --> "On-demand wind/solar electricity" . Nuclear power (off-peak hours)--> (Same as above) "Nuclear power" (peak hours)(indirectly)

回复

Thank you for sharing these market insights.

要查看或添加评论,请登录

Ksenia Tolstrup的更多文章

社区洞察

其他会员也浏览了