Signal Update: Autumn 2022

Welcome to the second release of Signal - the latest update following the launch of our battery energy storage revenue projection back in August. This new version of the forecast covers the period October 2022 - September 2025. In this article, we look at:

• What our latest update suggests for this winter (and beyond).
• A closer look at how duration impacts expected revenues.
• What’s changed in the market since our last update - and what this means for battery energy storage.

In case you missed it, we recently published our forecasting performance for last month, which you can check out here.

Key takeaways

Figure 1 (below) shows the latest p50 scenario from Signal, showing average battery energy storage fleet revenues from October 2022-September 2025 split by market. Here we assume site duration in line with that of the current fleet: 1.1h.

• Battery energy storage system (BESS) revenues have been steadily falling since reaching their all-time high level in June 2022, with value in Dynamic Containment tailing off due to market saturation. In the short term (to the end of 2022), we expect revenues to pick up. This is driven by both high gas prices and system scarcity over the winter months.
• High gas prices, and therefore high power prices, continue along the forward curve resulting in substantial wholesale opportunities for BESS. Projected annualized revenues for 2022 total £176k/MW/a, up 43% on last year's realized earnings. Falling revenues are projected across the following year, but maintain a healthy £147k/MW/a across 2023.
• DC remains a key component of the BESS revenue stack, making up ~45% of revenues over the next three years.

What’s changed?

Signal’s battery revenue projection is updated every quarter, to reflect the latest market developments, policy decisions, and market sentiment. In the interest of transparency, we want to show you how our projections have changed - and outline the key drivers.

Figure 2 (below) shows how the latest update for Signal compares to our August release. Here, we show our p10, p50, and p90 revenues for the latest forecast vs. our previous p50 revenue case.

• Our latest update sees increases across the whole forecast horizon, with the most notable differences across 2023.
• Increasing prices along the gas curve (see below) are a key driver of revenue increases. Gas prices impact energy storage revenues as gas-fired generation often sets the price. Combined Cycle Gas Turbine (CCGT) costs drive off-peak pricing and gas-peaking technology often set the peak prices.
• Moreover, we have brought forward the phase-out timeline for FFR. We expect to see increased procurement volumes in DR and DM before the end of 2023 - there’s more information below.
• Other changes include a refreshed outlook on both demand and margin based on the latest available data.

Signal inputs deep dive

Here, we delve into the updates to the various inputs to Signal:

• Gas prices.
• Battery buildout.
• The requirements for frequency response.

Gas prices

Figure 3 (below) shows the UK National Balancing Point (NBP) calendar month forward curve, used as our base assumption for a forward view of gas prices. Forward gas prices are used to calculate the short-run marginal cost of operation for gas-fired generation - a feature used to determine the minimum and maximum daily prices in the wholesale power price spread model.

• Continued pressure on European gas supplies sees high prices along the forward gas curve. December, January, and February are all currently trading in excess of 500p/therm (as of the 26 September forward curve).
• While the latest curve is significantly higher than July’s, it is actually down from the highs seen on the forward curve in August.
• High gas prices have had a knock-on effect on power prices. December base is currently trading at £646/MWh, and peak trading at £873/MWh (as of 26 September 2022).
• Elevated power prices with substantial increases to peak pricing are key drivers of BESS revenues. This is reflected in the latest update to the forecast.

Buildout

Figure 3 (below) shows our assumptions on GB BESS deployment over the forecast horizon, comparing our previous and updated views on buildout.

• Q3 saw the deployment of two new BESS sites totaling 70 MW. Sembcorp’s 50 MW/50 MWh Cowstead site, in addition to Tag Energy’s 20 MW/40 MWh Hawkers Hill site.
• With continued disruptions to supply chains, the buildout of BESS has been slower than anticipated over the last year. It is lagging behind Capacity Market expectations by ~800MW (as of September 2022).
• Our forward view of BESS buildout assumes a deployment rate consistent with energization rates over the past 12 months. This has been revised upwards from our previous forecast in light of Q3 deployment.

Response holding

Figure 4 (below) shows indicative frequency response holding over the forecast horizon, with high and low requirements shown on the negative and positive y-axis, respectively.

• Following new information from the ESO regarding the FFR phase-out, we have updated our view on response requirements.
• Increased requirements for DR and DM are assumed before the end of the year, to 150 MW each for both the high and low services. This is in line with the ESO’s latest communication on the matter: “[we] expect to be able to start offsetting legacy services and increase the volume of Dynamic Regulation by the end of the year.”
• We also assume a shorter phase-out of dynamic FFR volumes relative to our previous forecast. FFR starts phasing out in January 2023, with retirement expected by the start of Q3 2023. This reflects Modo’s updated internal view of the market and is consistent with the ESO’s derogation request for the continued procurement of dynamic FFR beyond May 2023.
• Volumes for DC high and low are assumed constant on 2022 indicative requirements. More is needed in the summer months, accounting for the higher risk of consequential rate of change of frequency (RoCoF) losses from embedded solar generation.
• We expect to hear further information from the ESO on 2023 DC high and low requirements shortly, which will be included in a future release of Signal.

Signal results deep dive

Here we go into detail on some of the results of Signal. Namely:

• Looking at the impact of battery duration on the revenue stack.
• How the value in each of the markets for battery storage evolves over the duration of the forecast.
• What that means for revenues this winter.

How duration impacts revenues

Since our first release of Signal, we’ve received lots of feedback from users. If you’d like to have your say, head over to our public roadmap! The most requested improvement was the addition of longer-duration assets, which we released at the start of September.

To estimate the wholesale market value, we forecast the single highest and lowest hourly price each day in the forecast horizon. To accommodate longer duration storage, we predict the 2nd (and 3rd, and so on) highest and lowest prices during the day based on historical daily price profiles. This is used to calculate wholesale revenues, the resulting response market opportunity costs, and, ultimately, BESS revenues.

Figure 5 (below) shows the p50 fleet revenue scenarios for two durations of assets, 1.1h (corresponding with the current fleet duration) vs. a 2h variation.

• Based on historical daily price profiles, the price in the second-highest hour is ~93% of the daily maximum. The second-lowest price is ~107% of the daily minimum.
• As such, across the forecast horizon, revenues for the 2h duration scenario are on average ~1.75x that of the 1.1h duration scenario.
• Figure 6 (below) shows the relative market share within the revenue stack for both durations. For 2h systems, wholesale trading makes up 73% of revenues across the forecast horizon, compared to 47% in the 1.1h case.

Changes to the revenue stack

Figure 7 (below) shows how revenues are distributed between markets, over the forecast horizon, for the 1.1h scenario.

• Revenues move out of FFR in line with our revised assumptions around the ESO’s phase-out plan - namely, falling away to zero by September 2023.
• DC revenues begin to fall away in line due to the saturation of response markets. Wholesale revenues are expected to increase from October onwards.
• Since our last update, prices in Dynamic Regulation High have frequently cleared at £0/MW/h due to the emergence of a new strategy for BESS; namely charging via the imports associated with providing DRH and discharging into the wholesale market (for more info see here). As such, the opportunity cost in DRH is effectively negative. Therefore, due to the absence of negative pricing in ancillary services, the value associated with this market is £0. This is reflected in both the market split and the absolute value of our revenue forecast.

Revenues rebound as we head into winter

Bringing it all together, Figure 8 (below) shows the p10, p50 and p90 revenue projections for Modo Signal in the case of the 1.1h fleet scenario.

• In the short term, we see BESS revenues rebounding following three successive months of falling revenues. This is driven by increasing wholesale market opportunities as colder weather approaches.
• Looking ahead to winter, high gas prices and a high likelihood of low-margin events drive revenues upwards. Our p10 scenario suggests annualized Q4 revenues of £264k/MW.
• Backwardation in the forward gas and power curves sees a high wholesale opportunity for BESS on the front of the curve, with revenue expectations falling over the forecast horizon.

Psst... we’d love your feedback! 🙏

For this iteration of Modo Signal, we’re super excited to hear your feedback. Please get in touch via Intercom if you’d like to discuss anything with the Modo team, and we’ll be happy to schedule a call!

Also, if you hadn’t noticed, you can now leave comments on articles. So, if you've got a question you'd like answered, an opinion to share, a concern, a comment, or a critique - you can let us know. We can’t wait to hear from you!