Expert Iacopo Bertelli is on top of European electricity market developments. He deep-dived into the topic of counterintuitive flows: when electricity is traded from a bidding zone with a high clearing price towards a bidding zone with a lower price. Iacopo sheds his light on this phenomenon in this first article of a series.
Counterintuitive Flows in 2023 – Part 1 of 3
Recently, counterintuitive flows (or adverse, or non-intuitive) have gained attention. Several instances of counterintuitive flows were observed on the Italy North bidding zone borders (link1, link2). The Core day-ahead flow-based market coupling regime (and the Nordic day-ahead flow-based market coupling regime which is currently being tested) deliver counterintuitive results every now and then. Furthermore, for Viking Link (link) and, more broadly, interconnectors between Great Britain and the EU (and Norway), counterintuitive flows are relevant as they signal inefficient allocation of cross-zonal capacities (link). The same applies to the Swiss bidding zone borders.
Hence, it felt the right time to deep-dive into the topic of counterintuitive flows and share some analysis on it.
In the context of European electricity market integration, a counterintuitive flow is a (net) commercial exchange (or schedule) of electricity from a bidding zone with a high clearing price towards a bidding zone with a lower price. Such a flow is contrary to intuition (“counterintuitive”), as electricity should always be exported from low- to high-price bidding zones (also referred to as the “right economic direction”).
Counterintuitive flows can be challenging to understand for many stakeholders. This article aims to shed some light on this phenomenon.
As there is a lot of ground to cover on this topic, I decided to split it into three parts, which I will publish over the next month:
- Introduce why counterintuitive flows result from market coupling
- Compare intuitive flows to counterintuitive ones
- Analyse counterintuitive flows based on 2023 data, including identification of trends and patterns by (type of) region
Today we start with part 1. All data (day-ahead prices and scheduled commercial flows) were retrieved from the ENTSO-E transparency platform (link).
Before delving into explanations, let’s visualize what all the abbreviations for capacity calculation regions (CCRs) and bidding zones (BZs) in this article mean.
Left: Capacity Calculation Regions as of January 2024. Right: Bidding Zones as of January 2024. Note that SEM is expected to join the Core CCR following a decision from ACER expected for March 2024 (however, the Celtic Interconnector will not be operational until 2026/27). GB, CH, RS are not part of SDAC but are mentioned in this article.
History of counterintuitive flows
When the market-coupling algorithm EUPHEMIA (link) was introduced in 2014, it was set to automatically reject counterintuitive flows. Following ACER decision 04/2020 (link), counterintuitive flows started being allowed with a patch in 2020, as they can result in greater overall socio-economic welfare. This change was triggered by studies (link) that showed how allowing counterintuitive flows would have generated, on average, an extra 18000 €/day of social welfare over the four years of the study (which comprised the CWE region: FR-BE-NL-DE-AT).
What is the origin of counterintuitive flows?
There are several reasons why counterintuitive flows can occur:
Explicit allocation
When cross-zonal capacity on a bidding zone border is allocated via an explicit auction, traders bid for cross-zonal transmission capacity to allow the energy they (expect to) trade to be exchanged across the bidding zone border. Essentially, they “bet” on the electricity price difference between the two bidding zones based on forecasts (of load, (renewable) generation, and ultimately day-ahead prices). The resulting (netted) schedule, after each transmission right holder has nominated the cross-zonal capacity it intends to utilize, might point in the wrong economic direction, i.e. from high-price to low-price areas. The bidding zones borders where capacity is allocated via explicit auctions are those involving bidding zones not included in SDAC (e.g. CH, GB, RS). Let’s take a look at the case of Switzerland.
Counterintuitive flows on the Swiss borders on 01/01/2023 at 06:00 CET. DA prices (€/MWh) are in yellow. Intuitive flows (MW) are indicated by blue arrows and labels, counterintuitive flows in red. Flows not involving CH are shaded (red and blue).
In the figure above, CH has the lowest day-ahead (DA) price (-14 €/MWh) compared to its neighbouring bidding zones, but is (net) importing from FR, DE, AT. This is not a “transit flow”, as CH is importing far more than it is exporting towards IT NORD. Note: IT NORD applies allocation constraints, but in this situation they are not triggered (see next point to understand why this is relevant).
Allocation Constraints
Allocation constraints are limits to the amount of electricity that can be imported/exported on a certain (set of) bidding zone border(s) or an entire bidding zone (technically, the sum of imports and exports is called net position).
Counterintuitive flows on the Polish borders on 20/06/2023 at 14:00 CET. DA prices (€/MWh) are indicated in yellow. Intuitive (counterintuitive) flows (MW) are indicated by blue (red) arrows and labels. Flows not involving PL directly are shaded.
During the hour in the figure above, PL applied an allocation constraint of net zero imports (i.e., it could not be a net importer) on its SDAC bidding zone borders. As PL is importing 593 MW (more precisely, 593 MWh for one hour) from SE4, PL must export the same amount (252+231+69+41=593). This results in counterintuitive flows between PL and each of its Core neighbours DE, CZ and SK as well as LT, which shares an NTC bidding zone border with PL.
Allocation constraints are used by PL and Italy North (NORD in the figure below). In addition, SE3, NO1, NO2, DK1 use lineset constraints. While the used constraints have different names and purposes, they all affect the maximum imports/exports on a set of bidding zone borders (see table and figure).
Use of external constraints in SDAC.
Flow-based market coupling
An exchange from a high-priced bidding zone to a low-price one can occur if the generated welfare loss is more than compensated by the welfare gain caused by an exchange on a different bidding-zone border (often a congested one), that becomes possible only thanks to the counterintuitive one. This “congestion-relieving” effect makes sense if we keep in mind that the goal of the pan-European market coupling algorithm is to maximize the sum of consumer surplus, producer surplus, and congestion income in the whole European market.
If this sounds complicated and quite theoretical, that’s because it is.
It is difficult to pinpoint exactly why a counterintuitive flow occurs in any situation with more than 4-5 bidding zones - except for the case of explicit auctions, where the counterintuitive flows are a result of inefficient allocation of cross-zonal capacities. Additionally, it can be difficult to disentangle the effect of allocation constraints and flow-based market coupling. One good example of counterintuitive flows (including calculations) with only three bidding zones can be found in Section 4.6.2 of the Princess Elisabeth Zone public consultation (link, page 137).
Counterintuitive flows in Core CCR on 09/07/2023 at 11:00 CET. DA prices (€/MWh) are indicated in blue. Intuitive (counterintuitive) flows are indicated by blue (red) arrows. Counterintuitive flows in this situation derive from a combination of flow-based, explicit allocations, and allocation constraints.
Looking at the figure above, it is complicated to pinpoint exactly the root cause of these counterintuitive flows. One high-level explanation could be that, for the Eastern part, the welfare gain of the (intuitive) flow from RO to HU outweighs the losses of the following (counterintuitive) flows out of HU to the neighbouring bidding zones. On the Central-Western side, a similar argument can be made for the (intuitive) flow from FR to IT NORD, supported by the (counterintuitive) flows into FR from BE, DE.
I will analyze the MTU in the figure above in more detail in part 2 of this series on counterintuitive flows, coming next week.
Take-aways & outlook
In this first article on counterintuitive flows, I laid the basics for understanding the data analysis to follow in the coming weeks. Key takeaways:
Counterintuitive flows are (net) scheduled commercial exchanges from bidding zones with high prices to bidding zones with lower prices. They can take place for several reasons:
- Explicit allocation of cross-zonal capacity
- Allocation constraints to imports/exports
- Flow-based market coupling
In the second part, I will show what intuitive and counterintuitive market results look like in SDAC, and in the third, I will analyse data on counterintuitive flows from 2023.
Industry Topics
Market Integration
Services
Regulatory & Energy Analytics, Energy Modelling & Strategies