Convergence Point

Introduction: The Plant-as-Option Framework

Here’s a mind-bending idea: A power plant is not just a piece of machinery. It’s a financial call option with a strike price equal to its variable cost.

Every hour, the plant “owner” (or operator) can choose:

• Exercise the option: Run the plant, sell power at market price
• Let it expire: Keep it offline if market price < variable cost

This optionality has value—even if the plant loses money on average. Understanding this concept is fundamental to generation asset management and energy trading strategy.

Spark Spreads Explained: The Basic Profitability Metric

The spark spread is the fundamental profitability measure for thermal power plants:

Spark Spread = Electricity Price – Fuel Cost

Example: Gas-Fired Plant

• Electricity price: €80/MWh
• Natural gas price: €30/MWh (thermal)
• Plant efficiency: 50% (needs 2 MWh gas to produce 1 MWh electricity)
• Fuel cost: €30 × 2 = €60/MWh
• Spark spread: €80 – €60 = €20/MWh

If spark spread > 0: Run the plant (make money)
If spark spread < 0: Keep it offline (lose money if you run it)

For Coal Plants:

• Spark spread still used (technically “dark spread” but often called spark spread)
• Coal price + variable O&M costs determine fuel cost

For Nuclear:

• Extremely low fuel cost (~€5/MWh)
• Always profitable to run (baseload plants)

Clean Spark Spreads: Adding CO₂ Costs

In Europe, power plants must buy CO₂ allowances for every ton emitted. This creates the clean spark spread:

Clean Spark Spread = Electricity Price – Fuel Cost – CO₂ Cost

Example: Gas Plant in 2025

• Electricity price: €85/MWh
• Gas price: €35/MWh (thermal)
• Efficiency: 50% (2 MWh gas per 1 MWh electricity)
• CO₂ emissions: 0.4 tons CO₂ per MWh electricity
• CO₂ allowance price: €80/ton

Calculation:

• Fuel cost: €35 × 2 = €70/MWh
• CO₂ cost: 0.4 tons × €80 = €32/MWh
• Total variable cost: €70 + €32 = €102/MWh
• Clean spark spread: €85 – €102 = -€17/MWh (LOSS)

Decision: Keep plant offline this hour.

Why This Matters:

• CO₂ prices went from €5/ton (2017) to €100+/ton (2023)
• This destroyed the economics of coal plants
• Shifted merit order: Gas plants overtook coal as cheaper marginal plants
• Accelerated coal phase-outs across Europe

Real Example: German Gas Plants Lost Money for 12 Years (2008-2020)

This is one of the most striking realities of European power markets:

The Setup:

• Germany built significant gas capacity in 2000s
• Expected gas plants to be mid-merit (running regularly)
• What actually happened: Renewables + cheap coal pushed gas far back in merit order

The Numbers (Simplified Average 2008-2020):

Operating Costs (Gas Plant):

• Gas fuel: €40/MWh equivalent
• Variable O&M: €3/MWh
• CO₂ cost: €15/MWh (average over period)
• Total variable cost: €58/MWh

Market Prices:

• Average baseload price: €45/MWh
• Average peak price: €60/MWh

Spark Spread:

• Baseload: €45 – €58 = -€13/MWh (loss)
• Peak: €60 – €58 = +€2/MWh (tiny profit)

Annual Plant Performance:

• Capacity: 500 MW
• Potential output: 500 MW × 8,760 hours = 4,380 GWh/year
• Actual output: ~800 GWh/year (18% utilization—ran only during very high prices)
• Fixed costs: ~€40M/year (capital, staffing, maintenance)
• Revenue: €48M (800 GWh × €60 average when running)
• Variable costs: €46.4M (800 GWh × €58)
• Contribution margin: €1.6M
• Net result: €1.6M – €40M fixed costs = -€38.4M annual loss

Why Didn’t They Shut Down?

1. Optionality value: Kept plant ready for price spikes
2. Capacity payments: Some markets paid to keep capacity available
3. Regulatory requirements: Required for grid stability
4. Hope: Waiting for coal plants to retire (eventually happened)

What Changed After 2020:

• Coal retirements accelerated
• Russian gas supply disruptions (2022)
• Gas plants became profitable again (moved up merit order)

Lesson: Average economics don’t matter—optionality does. The plant’s value wasn’t average profitability but the option to produce during scarcity.

Power Plants as Call Options: The Deep Insight

Let’s formalize why a power plant is like a call option:

Financial Call Option:

• Right (not obligation) to buy asset at strike price
• Exercise if market price > strike price
• Value depends on: probability of market price > strike, volatility

Power Plant:

• Right (not obligation) to produce at variable cost
• Exercise (run) if electricity price > variable cost
• Value depends on: probability of price > variable cost, price volatility

Mathematical Parallel:

Call Option Value:

V = P(S > K) × E[S - K | S > K]

Where:

• S = market price
• K = strike price
• P(S > K) = probability market exceeds strike

Power Plant Value:

V = P(Price > VC) × E[Price - VC | Price > VC] × Capacity

Where:

• Price = electricity market price
• VC = variable cost (fuel + CO₂ + O&M)
• Capacity = MW rating

Key Insight: Even if price > VC only 10% of the time, the plant has value because:

1. Those 10% hours could have very high spreads
2. You have no downside (don’t run when unprofitable)
3. This asymmetric payoff is classic option value

Optionality = Value: Why Flexibility Matters

Example: Comparing Two Assets

Asset A: Wind Farm

• Capacity: 100 MW
• Capacity factor: 25% (runs 2,190 hours/year at full power)
• Variable cost: €0/MWh (no fuel)
• Must produce when wind blows (no choice)
• Revenue: 2,190 hours × 100 MW × average price = €219M (if avg price = €100)

Asset B: Gas Peaker Plant

• Capacity: 100 MW
• Can run up to 8,760 hours/year
• Variable cost: €100/MWh
• Chooses when to run (flexibility)
• Revenue: Runs only when price > €100, sells at avg €150 during those hours
• Runs 500 hours/year × 100 MW × €150 = €7.5M revenue
• Variable costs: 500 hours × 100 MW × €100 = €5M
• Contribution: €2.5M

Which is more valuable?

• Wind farm: Higher output, higher revenue
• Gas plant: Lower output, but controls when

The Optionality Premium:

• Wind farm is long-only (can’t choose not to produce)
• Gas plant can time production to high-price hours
• In volatile markets, this timing ability has significant value

Real Market Evidence:

• Peaker plants earn “missing money” from scarcity pricing
• A plant running 50 hours/year (0.5% utilization) can still be economically viable
• Why? Those 50 hours might average €500/MWh vs. €100 variable cost

Make-or-Buy Decisions With Hedging

Generation assets create a natural question: Should we sell the physical output or hedge financially?

Scenario: Combined-Cycle Gas Turbine (CCGT)

• Capacity: 400 MW
• Expected output: 3,000 GWh/year (85% availability × 8,760 hours × 40% utilization)
• Variable cost: €65/MWh
• Forward price Year+1: €90/MWh

Option 1: Sell Physical Output (Day-Ahead Market)

• Produce when spark spread > 0
• Sell each hour at day-ahead price
• Retain optionality (don’t run if unprofitable)
• Revenue: Variable (depends on hourly prices)
• Risk: Price volatility, volume uncertainty

Option 2: Full Hedge (Sell Forwards)

• Sell 3,000 GWh Year+1 forwards at €90/MWh
• Lock in revenue: 3,000 GWh × €90 = €270M
• Problem: Lost optionality (must deliver even if spark spread negative)
• Result: Revenue certain, but forced to run at losses some hours

Option 3: Partial Hedge (Structured)

• Sell 2,000 GWh forwards at €90/MWh (baseload portion)
• Keep 1,000 GWh unhedged (peak/option hours)
• Benefits: Revenue certainty on base + optionality on flexible portion
• Tradeoff: Some price risk remains

Decision Framework:

1. Risk tolerance: High → keep unhedged, low → full hedge
2. Price view: Bullish → hedge less, bearish → hedge more
3. Credit needs: Lenders prefer hedged revenue (bankability)
4. Operational flexibility: Physical constraints limit optionality

Real Example: Wind Farm Hedging

• Wind farm produces 500 GWh/year (variable by weather)
• Forward market: €85/MWh
• Problem: Can’t guarantee exact volume (wind dependent)
• Solution: Sell baseload volume (400 GWh) in forwards, leave 100 GWh upside unhedged
• Result: Hedged revenue on 80%, retain 20% merchant exposure

Multi-Market Optimization: Beyond Day-Ahead

Sophisticated generation operators don’t just sell into day-ahead markets. They optimize across multiple revenue streams:

Revenue Stack for Flexible Gas Plant:

1. Day-Ahead Energy Market

• Sell electricity production at spot prices
• Typical: 50-70% of revenue

2. Intraday Energy Market

• Adjust positions closer to delivery
• Capture price spikes from forecast errors
• Typical: 5-10% of revenue

3. Frequency Containment Reserve (FCR)

• Keep plant ready to increase/decrease output instantly (±X MW)
• Paid capacity fee just for being available
• Typical: 10-15% of revenue

4. Automatic Frequency Restoration Reserve (aFRR)

• Automatically adjust output based on TSO signals
• Paid capacity + energy when activated
• Typical: 10-20% of revenue

5. Manual Frequency Restoration Reserve (mFRR)

• Manually adjust output when TSO calls
• Paid capacity + energy when dispatched
• Typical: 5-10% of revenue

Example Annual Revenue:

• Day-ahead energy: €7M
• Intraday: €800k
• FCR: €1.5M
• aFRR: €2M
• mFRR: €600k
• Total: €11.9M

vs. Day-Ahead Only: €7M

Lift from optimization: 70% revenue increase

The Complexity:

• Must bid into 5 different markets simultaneously
• Constraints: Can’t commit same capacity to multiple markets
• Optimization: Allocate capacity to highest-value market each hour
• Software required: Manual optimization impossible

Why Complexity = Opportunity

The German gas plant example (2008-2020 losses) reveals a deeper truth:

Simple Strategy: Run plant whenever spark spread > 0 → Lost money
Complex Strategy: Optimize across energy + reserves + forward hedging → Survived

Value Creation Areas:

1. Market selection: Sell into balancing markets during scarcity (higher prices)
2. Forward hedging: Lock in favorable prices during market contangos
3. Option retention: Keep flexibility for extreme price events
4. Startup optimization: Minimize expensive starts/stops
5. Fuel procurement: Hedge gas prices separately from electricity sales

The Winners:

• Operators with sophisticated optimization software
• Traders embedded in asset operations
• Integration of fundamental analysis + algorithmic bidding

The Losers:

• “Set and forget” operators
• Pure merchant plants without hedging strategies
• Plants treating day-ahead as only market

Key Takeaways

✓ Spark spread = electricity price – fuel cost (clean spark adds CO₂)
✓ German gas plants lost money 2008-2020 due to renewables and cheap coal
✓ Power plants are call options: Value comes from optionality, not average economics
✓ Flexibility has value: Ability to choose when to run beats forced production
✓ Make-or-buy hedging: Balance revenue certainty vs. optionality retention
✓ Multi-market optimization: Energy + reserves stacking can increase revenue 70%
✓ Complexity = opportunity: Sophisticated operators capture value simple strategies miss

Next in Series: Post 7: Balancing Markets & Optimization: Where the Real Money Hides