5 minute read time

Understanding power markets: The levelised cost of energy

The levelised cost of energy provides a very useful way of valuing new renewable energy assets and determining the energy price a project needs to be viable.

What does it cost to generate a megawatt/hour of electricity? This simple question throws up a lot of complicated answers. For a solar farm, powered by sunlight, the answer could – arguably – be close to ‘zero’.

A natural gas plant, conversely, would have to factor in the price of the fuel, the cost of carbon emissions permits and its other operating costs. On top of this, the cost of the capital needed to build the plant needs to be taken into account. That cost might be zero for a 25-year-old asset, which has had its original capex amortised down to zero. Or it could represent the lion’s share of a new wind farm’s cost.

As we have seen, the price of electricity in most wholesale markets is based solely on the operating costs of the various generators that bid to supply power. This is expressed as the marginal price – the price required to incentivise sufficient capacity to meet demand at a given point.

However, because these operating costs do not include the costs of repaying debt and generating a return for equity investors, they only tell part of the story. To compare the costs of different types of generating capacity, analysts use the levelised cost of energy (LCOE).

The cost of energy on the level

Put simply, a power plant’s LCOE is a measure of its lifetime costs divided by the volume of energy it produces over that lifetime.

The calculation incorporates the costs of building, financing and operating the plant (including fuel costs, staffing, maintenance and emissions allowances, if applicable). It also includes a discount rate to depreciate the cash flows to account for the returns expected by the investor, and to factor in risks, etc.

It should be noted that there is no universally agreed methodology to calculate an LCOE, and different analysts could generate different LCOEs for the same asset, depending on the assumptions they make and the granularity of the data used. However, the US National Renewable Energy Laboratory provides a useful calculator.

Once derived, the LCOE can be compared with the expected revenues a project can earn from selling electricity (and, in some markets, from selling other attributes, such as frequency control, reactive power and avoided emissions). If the expected revenues are greater than the LCOE, then the project should be profitable and, all things being equal, a developer will develop the project. Conversely, where the LCOE is higher than expected revenues, the project won’t be developed. This disconnect from the wholesale market means that it is the LOCE, not the wholesale power price, that is the driver of the price of power from new-to-earth generation.

The LCOE is not without its drawbacks. It can oversimplify the complexities around project risks and the cost of capital. But it is a useful tool for comparing different technologies and project types. It’s also very useful in establishing the price at which a project needs to sell its output under a long-term contract, and was used by the UK’s Department of Business, Energy and Industrial Strategy in developing pricing for its Contracts for Difference.

Plummeting green energy LCOEs

Over the last decade or so, the LCOE of renewable energy has fallen spectacularly. Bloomberg New Energy Finance (BNEF) produces benchmarks that track the global LCOE of various power generating technologies. Between 2009 and mid-2022, the average LCOE for a fixed-axis solar photovoltaic plants fell from $304/MWh to $45/MWh. Onshore wind fell from $93/MWh to $46/MWh, while offshore wind has fallen from a peak of around $220/MWh in 2012 to $81/MWh.

Meanwhile, the LCOE of coal-fired power has ranged between $60 and $85/MWh and that of gas plants from $45 to $81/MWh. These changes have, to a very large extent, been driven by changes in fuel costs.

The main drivers pushing down clean energy LCOEs have been technological innovation and economies of scale. More efficient solar cells and wind turbines produce more power for each unit of cost. Mass production has helped components and manufacturing become cheaper.

As the chart shows, onshore wind and solar electricity are now cheaper, on a lifetime basis, than fossil fuel-fired plants.

The LCOE of wind and solar technologies are particularly sensitive to costs of capital. According to financial advisory firm Lazard, which publishes closely-watched LCOE analysis, capital costs accounted for $26 of the $30/MWh LCOE of utility-scale crystalline solar PV plants. For wind, the figure is $20 of the $26/MWh LCOE. Since the financial crisis in 2008, central banks have pursued ultra-loose monetary policy; the resulting ‘cheap money’ has helped keep clean energy costs down.

However, the various cost curves do not all bend in the same direction. In June, BNEF reported cost rises pushing up prices of wind energy by 7% year on year, and solar by 14%. It pegged those rises to increases in the cost of materials, freight, fuel and labour.

Despite this inflation, the current energy crisis that has driven up gas prices has made renewables more attractive from an LCOE perspective. While higher fuel costs have increased the LCOE of gas plants, the LCOE of renewable energy has been largely unaffected. This has led to higher profits for renewable generators while many fossil fuel generators have struggled as they have had to cover higher fuel costs.

This gap between the two types of generation, fossil fuel generation (low capital need, expensive to run) and renewables (high capital need, cheap to run), is driving many market observers to consider how the market could be split.

Looking beyond LCOE – value-adjusted LCOE or VALCOE

There is a further dimension to valuing electricity generating assets. As is noted above, every electricity generating asset has other attributes, not all of which are financially remunerated. The ability of an asset to provide capacity on demand, or its flexibility to provide system services such as helping to stabilise the frequency of the grid, may not generate revenue, but they have value to the system operator. Equally, assets lacking those capabilities can impose costs on a system.

To capture the value or costs of these attributes, a value-adjusted LCOE can be calculated. This adjusts the LCOE by comparing an asset’s performance on three metrics – energy, capacity and flexibility – against the grid average. Energy is its ability to capture wholesale power prices, capacity its contribution to system adequacy, and flexibility is its ability to provide system services such as frequency regulation or reserve power.

So, a peaking natural gas plant might have a high LCOE, but its dispatchability would increase its VALCOE. Conversely, a solar plant without energy storage attached would mean its VALCOE would be lower to the grid operator than its LCOE.

As long as these values are not compensated for, VALCOE as a measure is of most interest to policymakers and market operators. But, as renewables penetration increases, it is likely that more of these costs and benefits will factor into market payments and the investment case for new assets.