LCOE has reached its limits
We regularly read about new record lows for the Levelized Cost of Energy (LCOE); such as US$ 1.79 ct/kWh for solar photovoltaic (PV) projects in the Kingdom of Saudi Arabia or US$ 1.97 ct/kWh in India. Those who are more familiar with the subject know that these numbers sometimes represent the levelized cost of energy, and at other times they indicate the anticipated levelized revenues for the energy sold under the terms of an auction. If you dig deeper, you start to understand that this number is based on multiple assumptions: not only about Capex and Opex but also debt and equity levels, the cost and term of financing, the life time of the asset, availability, wind speed and solar irradiation, the period and method of depreciation, residual value, and the exchange rate. Where the number represents levelized revenues, you also need to factor in assumptions about how the offtake agreement is adjusted over time and what the market price for energy will be after this agreement expires.
With all these considerations in mind, you may wonder what the value is of an LCOE? Can you rely on the figure; or should you only trust LCOEs when they come from one source, because then at least you can compare one LCOE with another? Or is the LCOE about as insightful as the typical statement in a press release where the new owners proudly announce that their project will supply so and so many households with electricity?
Let’s take a step back and ask ourselves if cost per kWh is the right number to look at? Is the underlying assumption that every kWh has the same value correct? Can I use the LCOE index in a world where storage is becoming more and more prevalent? Batteries typically increase the LCOE, but perhaps these higher costs increase the value of the system at the same time? And if so, by how much?
We all like a method that allows us to compare different solutions and technologies in a simple way. As long as LCOEs are calculated using the right assumptions, then they provide a straightforward comparison. But we should be cautious. There is an old saying, “what gets measured gets done”. But is a low LCOE what we want to get done, or would we rather demand and supply are matched at the lowest possible cost? Do we want to have security of supply that is sustainable environmentally? If the answers to these three questions are ‘yes’, or even if it is ‘yes’ to just the first question, then we need to go beyond a simple LCOE number.
In future, we need to work with a “function” rather than with a single “index”. Our new Cost of Energy Function (COEF) needs to start by capturing the system cost of power generation, and then progressively show how this cost changes, typically increasing, if the generation (supply) profile is adjusted to a demand profile. The resulting curve may well end before full adaptation has been reached as this amount of flexibility may not be possible technically. The demand profile will be a new assumption that is factored into our equations. We may also need to accept that two demand profiles are necessary: one for summer daily demand, another for winter. We will also need to include whether we are considering a base load or a peak load generation source.
In addition to generation curves, we can construct demand curves that start with demand today. By creating two curves, we can show the cost of energy efficiency (how much investment would be needed to reduce energy consumption) and the cost of flexibility (how much it would cost to move the point in time when energy is consumed by 30 minutes or one hour, for example).
Moving from a LCOE index to a function is not an easy undertaking, and it may well progress in stages. The car industry currently measures gas consumption using different driving profiles, depending on whether it’s in the city, on country roads, or on the highway. Regardless of the question, if we use a function or a group of values that are profile-based, we need to ensure that we all mean the same thing. Here, an independent body or workgroup could play an important role in developing and codifying a workable framework: a framework that captures the complexity of demand and supply, that captures the value of storage, and that helps regulators to set policies because it measures what needs to be done.