Tariff design and LCOE

Why CfDs are more efficient than PPAs

Written with Erkan Uysal, with credit to Simon Matthiessen and Kevin Feldman for the model work

This is the short version of an article that I started writing 4 years ago and never completely finished, but which is still relevant. The goal was to show the real-world consequences of certain market design choices.

The table below presents the key assumptions and results for the financing of an offshore wind farm under different tariff regimes, from various CfDs, to pure merchant and structures using corporate PPAs. The numbers were realistic in 2021, and obviously are outdated today, as costs, interest rates and power prices have all moved significantly. Since the most interesting result is the relative numbers rather than the absolute ones, we have not updated the table with more recent assumptions, as the conclusions would be identical.

The project’s cost structure does not change between the different scenarios, but its financing does, driven only by the evaluation by debt and equity providers of the power price risk, all other factors being equal. The tariff regimes influence how much debt lenders are willing to provide, and under what terms, and the cost of equity that the owners require for the corresponding residual risk.

Debt amounts are constrained in two different ways:

1. the cost side: the “gearing”, or “leverage”, as banks will require a minimum fraction of equity to fund the project that depends on the perceived risk (the higher the risk, the lower the fraction of debt that will be allowed); and

2. the revenue side: the “DSCR” or debt service cover ratio, which looks at how much revenue is expected in each period, and how much of that can be allocated to repaying debt (the more certain the revenues, the higher that fraction can be and the lower the DSCR will be)

The above table shows the numbers that could be realistically achieved in 2021 - they have not changed much today, as far as the ratios are concerned - for each risk profile. “Merchant” pricing (selling directly on the spot market) is the riskiest strategy, and thus allows the lowest amount of debt and the least favorable terms. PPAs are seen as less risky from a price perspective but bring in counterparty risk (that the buyers will go bankrupt or terminate the contract). CfDs are seen as the least risky. Longer durations are seen as riskier but also offer a larger revenue volume to repay debt so can lead to increased debt amounts.

Equity remuneration requirements move in tandem with debt perceptions (the riskier profiles lead to higher IRR expectations).

The first key result is that a CfD allows a project to bid a significantly lower price than if it has to rely on a corporate PPA: project owners need a PPA price of 66 EUR/MWh over 15 years and a CfD at 50.5 EUR/MWh over 20 years (or 59 EUR/MWh over 15 years) to receive risk-appropriate returns on the equity they invest.

This was proven in real life with the HKZ project, where Vattenfall gave away the equivalent of 10% of project revenues to BASF to secure a PPA for the electricity generated by the project, which did not get a proper CfD.

The second result is that merchant projects are even more difficult to finance, which means that investors relying on spot prices will effectively need to rely on high (and likely very volatile) prices to get the returns they expect. The utilities’ and oil&gas majors’ bets on “zero bids” and other merchant strategies for offshore wind (as have happened in the Netherlands, Denmark and Germany) are either deeply misguided (they will never get decent returns, knowing that their long term market price outlooks are usually quite conservative) or extremely cynical (betting on sustained high and/or volatile prices). The continuing link between natural gas prices and power prices suggests that parties that are experienced gas traders could see paths to long term profitability by benefitting the most from price spikes like those that occurred during the 2022-23 crisis.

From the regulator’s perspective, merchant projects do not provide any price protection for consumers even though their cost base is fixed - renewables projects will make “super profits” during price spikes. But they appear to be “subsidy-free”.

And if they lead to PPA-backed structures, the benefits of the fixed price will go to the buyer - which these days is most likely to be one of the GAFAs (Google, Microsoft, Amazon). Thus relying on PPAs rather than CfDs is akin to indirectly giving subsidies to some of the richest corporates on earth…

There are other elements that matter in the design of a tariff auction, and I’ll get to these in a separate post, but the core element here is that a CfD (like a PPA) is a swap of short term prices for long term prices, and having a regulated counterparty has a value that can translate into (i) lower cost of electricity for consumers, and (ii) increased stability of prices for them.