Understanding the riddle of who produces the cheapest electricity

Over the last months, the approval of the Hinkley nuclear power station has stirred a heated debate, in the United Kingdom just as much as in Brussels.

The discussion put the spotlight on one key question: which technology is able to generate electricity the cheapest?

The answer will have a significant impact on the way we will produce our electricity in the next decades, with consequences ranging from the cost for consumers to the impact on climate. At the same time, the continued low prices of gas and other fuels and the falling costs of technologies such as solar are bringing new elements to the debate.

To answer the question, economists typically calculate a levelised cost for each technology with one-off expenditures such as site construction and equipment being spread over the expected lifetime production of the plant they are analysing. To this, they add variable costs (fuel, personnel and maintenance, for example) to get a total quote – in euros per megawatt/hour. This number is in theory comparable across all the technologies being studied.

However, reality is more complex. Apart from the fairly common discussion about the value of the key assumptions used, two other points are key.

Firstly, being forced to do these calculations is a sign of market failure (in some cases, reinforced by sizeable support schemes for renewables). If the power generation market was fully competitive, undistorted by subsidy and encompassed all the facets of power supply (for example location, intermittency and carbon footprint) then policy makers would need not worry about developing levelised costs. They would not need to pick winners out of the available technologies as the market would do this for them.

But because there are market failures – and this is the second point – we need to be sure we compare apples with apples, when we assess levelized costs. For example, power generated by solar panels right where it is used (let’s say in a house) has very different characteristics to the power generated by a centralised coal fired plant.  Simply basing the calculations on the capital and operating costs of the plant ignores issues of location, intermittency and carbon footprint that have to be taken into account.

These parameters can have a significant impact in calculating levelized costs.

Here’s an example:

-To ensure intermittent power supply can have the same characteristics as a dispatchable plant (one that produces electricity continuously), it needs some form of back up.
-If back up was provided by battery storage, the current cost would be at least €1 million per megawatt.  For a source needing back-up for 60% of the time, this  would lead to an increase in the levelised costs of production by around €25 per megawatt/hour. But for a source needing back-up for 90% of the time, the additional cost would be as high as €98 per megawatt/hour.

The location of power production can also have a significant impact on costs. In Germany, last year industry paid, for the use of the power network, between €18 and €64 per megawatt/hour depending on the size and pattern of its consumption. These costs could be reduced significantly or removed all together if power production was located close to demand centres. The recently announced increase in cost of the Sued Link, the “Wind Power Line”, to €10 billion only reinforces the locational value.

Finally, even though currently it is low, the cost of carbon needs to be taken into account. (At the moment, the cost of carbon only adds a few euros per megawatt/hour to the cost of power produced from coal-fired generators).

Ultimately, the advent of a fully competitive power generation market in Europe will render the calculation of levelised cost obsolete. But in the meantime, decision makers use them to make policy choices: to make sure they make the best choices, we need to be sure we are comparing apples with apples.

Malcolm Rice-Jones is the Chair of GasNaturally's Power Generation Task Force