Beyond liquidity: why medium-term power forecasts need fundamentals

Since the energy crisis in 2022, European power markets have recovered strongly. Trading activity has increased, churn rates have moved higher and Germany in particular shows a strong rise in the future market activity. 

This looks like a simple recovery story. But there is more to find in the details. An interesting question is not whether liquidity has increased, but how is liquidity is actually distributed over trading timeframes? 

This distinction is important, because many decisions in power markets are taken over many years. Utilities, large industrial consumers and PPA originators need to make decisions more than three years ahead, for example. Over that time horizon, the quality of the market signal can change drastically. 

One useful indicator to assess this development is the churn rate. In power markets, the churn rate describes the ratio between the total traded volume on the futures market and the volume that is physically delivered. A higher churn rate generally indicates a more liquid and mature market, where contracts are traded multiple times before delivery and the traded futures volume is much higher than the finally delivered one. This is important, because liquid futures markets provide stronger, more reliable price signal and better hedging opportunities for market participants.  

Churn Rate for all annual future contracts (sum)

As can be seen in the graphic, there are big differences between the markets for Germany, France and Austria. After the crisis-year in 2022, market activity has recovered in all three markets. Germany remains by far the deepest market of those three and within Europe overall. France also improveds, but from a much lower level. Austria remains thinner in terms of liquidity and should be interpreted more cautiously as a standalone benchmark market. 

The different liquidity patterns are important, as they show that European power market liquidity is not in a uniform structure. Even neighbouring and interconnected power markets can differ a lot in their depth.  

So, we can say Germany is by far the strongest reference market. France is relevant, but less deep and Austria is thinner still. That does not make France or Austria irrelevant, but it does mean that prices in these markets should not always be treated with the same confidence as prices in the most liquid market. As a result, many market participants across Europe increasingly focus on the German futures market for hedging, pricing and risk management purposes, since its higher liquidity provides more robust price signals and allows larger positions to be traded more efficiently. 

The uneven structure of future market liquidity 

The recovery in churn rates since 2022 shows that trading activity has returned to the markets. However, higher aggregate liquidity does not necessarily mean that liquidity is evenly distributed across all traded contracts. In reality, trading activity remains heavily concentrated in the first delivery years. 

When looking at the distribution of trading volume along the future trading years for the German market, we see that a very large share of volume sits in the first delivery years. Especially for the next (Y+1) and second (Y+2) delivery year, we see high liquidity in the market. Beyond that time horizon, the liquidity drops sharply, even in the most liquid German market.  

So even if historical data suggests that a market has high liquidity and strong churn rates overall, this liquidity is mainly concentrated in the short-term maturities. In the medium-term horizon, trading activity and market depth decline significantly, resulting in a much thinner liquidity environment after just 3 years. 

For future products, exchange-traded prices are usually used as a strong benchmark signal. They reflect trading activity, hedging interest and changing expectations based on the latest available information, but in medium- and long-term maturities, the quality of visible market prices weakens. Fewer trades, lower depth and more estimated pricing also mean that the price signal is less robust and the market price carries less standalone information or confidence. 

When market liquidity fades, fundamentals become more important 

For many companies, the one-to-six-year horizon, (the medium-term), is where many important decisions are made:  

• Large consumers need to prepare procurement strategies and decide about their budget.  

• Utilities and portfolio managers need to decide on hedging size.  

• Producers need to evaluate their potential margins in the upcoming years. 

• PPA pricing decisions require an assessment of whether offered market prices are reasonable. 

• Risk managers need to understand how exposed their portfolios is to changing fundamentals (capacities, demand, policies, commodity prices etc). 

In addition to a price tag, those decisions require reasoning and explainability. The thinner the market signal becomes, the more valuable that explainability gets. For this time-horizon, a fundamental-based approach becomes relevant to interpret market prices in a less liquid environment. This is the market segment at which Montel’s Medium-Term scenarios is focused on. The goal is to provide a fair value power price based on the underlying power-system dependencies. 

The model used, Power2Sim, is based on fundamental input parameters and calculates expected day-ahead auction prices for the future. These hourly price expectations are then aggregated into the relevant delivery products for base and peak prices, Power2Sim to builds the future view from the bottom up (weekly, monthly, quarterly and yearly). 

The model builds on the basic logic of the power system: demand, renewable generation, availability of thermal units, fuel prices, CO₂ costs, hydro, imports and exports. From there, the model simulates the hourly price formation according to the merit-order and aggregates the results into the power contracts the decision makers need for the medium-term time horizon. 

The price curve does not represent a classic arbitrage-free PFC, but rather determine a fair-value price for power delivery, excluding psychological factors and risk premiums that are typically embedded in exchange-traded products. 

This is an important distinction because market prices include more than fundamentals. They also reflect the potential risks, liquidity (or the lack of), uncertainty of weather and politics, technical analysis and overall market psychology. These factors are real and relevant, but they can move prices away from the underlying supply-and-demand system value, particularly in thinner parts of the curve. 

A fair-value approach therefore should not be seen as a replacement for the future market price, but as an additional analytical layer. It helps users assess how strongly market prices are supported by underlying system fundamentals and where additional premiums, discounts or uncertainty may influence pricing. As shown earlier, liquidity and trading activity decline significantly in the medium- and longer-term maturities, which also reduces the informational strength of observable market prices. In these less liquid market horizons, a fair-value approach can provide additional support and interpretation for medium-term pricing decisions. 

The Medium-Term solution does this by combining several analytical perspectives and compares real settlement with modelled fair value prices. The following analytical setups are part of the Medium-Term solution: 

Forecast price in relation to marginal costs and clean dark/ spark spreads

Distribution of hours in which the system/ market is under pressure

Lead time: comparison of settlement and forecast price over time

The distribution views add a risk perspective. They show how often certain system states really occur. For example, the residual-load distribution indicates how frequently the system is likely to operate in loose or tight conditions. This helps users to understand not only the expected price level, but also the frequency of those critical market situations. 

The lead-time adds the market-behaviour over time angle. The lead-time compares how market prices evolve against the model’s fundamental forecast as settlement and delivery is coming closer. This helps identify when premiums build up, sentiments shift or periods where market psychology may be moving prices away from fundamental fair-value. 

There are more analytical diagrams not shown here, like the power balance view, which explains the system behind the price. It shows consumption, residual load, wind and solar production, thermal generation, hydro output and net imports or exports. This is helpful because power prices are not built by average supply, but by the marginal hours in which the system becomes tight or stressed. 

For later-dated power products, the more relevant question is no longer whether a price exists, but how much depth, structure and explanatory power sits behind it. That is where a fundamental fair-value approach adds value. It does not replace the market, but rather it helps to interpret the market, especially for medium-term decisions.