PPAs: How to Avoid the Most Common Mistakes in Market Valuation as a Derivative

Long-term Power Purchase Agreements (PPAs) have become a key tool in the energy transition. They allow companies to secure prices, finance renewable assets, and meet ESG targets through origin certificates.

Although they may seem simple at first glance, PPAs hide considerable technical complexity. Valuing a PPA requires understanding how the market behaves, how energy is generated, and how both factors interact.

From an accounting perspective, many PPAs meet the criteria to be classified as financial derivatives under IFRS 9, which requires fair value measurement in accordance with IFRS 13. The practical question is how to estimate a credible fair value for contracts with horizons of 10, 15, or even 20 years especially when market liquidity disappears beyond the first years.

Everything depends on two pillars: the robustness of the model and the reliability of the data feeding it. Below, we review the most frequent mistakes that distort PPA valuations and how to avoid them.

1. Using baseload prices instead of capture prices

One of the most common errors when valuing PPAs is using “baseload” or market average prices instead of capture prices specific to each plant or technology.

Capture prices reflect the weighted average price based on the actual hourly production of a facility. A solar plant does not produce at the same pace as the daily average market price: it generates during daylight hours, precisely when the abundance of renewable supply tends to depress prices.

For example, a solar plant (which only generates during the day) will have a different capture price than a wind farm, which can operate at night. Even between two solar plants, geographic location can significantly alter the price profile.
Using average prices can overestimate the contract’s value and lead to material differences. Hourly modeling, adjusted to the reality of each facility, is highly recommended to correctly estimate the economic value of the PPA.

2. Ignoring hourly variability and degradation

Many models simplify PPA cash flows to an annual basis, as if production were uniform. In reality, both prices and renewable generation fluctuate constantly: a cloudy day, a wind surge, or a heatwave can completely change forecasts.

Moreover, every solar or wind plant loses efficiency over time (typically between 0.3% and 0.5% annually). Ignoring this degradation can cause significant overvaluation in long-term contracts.

Best practice is to model annual production as a function of expected irradiance, adjusted for degradation and variability, and to cross-check these results with historical data from the operator or grid. Typically, valuations use the P50 metric, but this figure should be reviewed periodically.

3. Failing to adjust curves to local market reality

Another common challenge is the lack of liquidity in power markets. In most European countries, forward price curves are only liquid for the first three to five years. Beyond that, prices must be extrapolated or derived from proxies, fundamental models, or comparable PPAs.

A frequent mistake is extending the price curve without verifying whether assumptions align with observable data. IFRS 13 requires prioritizing quoted inputs and properly documenting those that are not.

Therefore, curve analysis is as relevant as the fair value calculation itself: using unvalidated proxies can significantly distort the final valuation.

Another factor is that under IFRS 13, market credit risk (CVA/DVA) must also be calculated.

4. Not validating model assumptions against real data

The quality of a valuation depends directly on the quality of its input data. In many cases, PPA models are built on assumptions without empirical support.

Performing backtesting (comparing projections with historical results) is essential to measure model reliability. It’s not about predicting the future but ensuring consistency between simulation and actual market behavior.

When real production data becomes available, recalibrating the model is mandatory, as it is at contract inception. If discrepancies are systematic, the error may lie in irradiance simulation, efficiency, or hourly profile.

A solid model should combine historical series, contractual data, and market benchmarks to ensure traceability and transparency.

5. Relying on standard or generic models

Using universal models is tempting but rarely works in practice. Each PPA has its own price structure, volume clauses, indexations, and flexibility conditions that can significantly affect fair value.

“Plug and play” models do not capture these nuances. The key is to combine automation with professional judgment:

• Automation to ensure consistency and efficiency.
• Professional judgment to adapt assumptions to the economic reality of the contract.
  The result should be a traceable, auditable, and transparent model where every input is justified and documented.

Conclusion: The value lies in the data, not just the model

Valuing a PPA goes far beyond filling out a spreadsheet. It requires understanding how data, the market, and the physics of generation interact.

The most common mistakes (using average prices, ignoring variability, extrapolating without analysis, or failing to validate assumptions) can distort strategic decisions, audits, and even financing.

The right approach combines:

• Hourly granularity and realistic capture prices.
• Models calibrated with historical and observable data.
• Traceability and transparent documentation.
• Professional judgment to interpret results.

At EY, we combine accounting expertise, market knowledge, and advanced quantitative modeling to deliver robust, defensible valuations aligned with IFRS 9 and IFRS 13 requirements.