Comparing performance in the economics of energy production: seeing through theory

 In the economics of energy production, the eventual value of a unit of electricity – the figure most interesting to consumers and governments – is influenced by a complex array of interconnected factors, ranging from the generating unit’s ability to produce at peak demand to how much of its capacity it can effectively use. To take all these factors into account, and thus make the economic performance of different energy sources comparable, economists prevalently use the measurement of “levelised costs”. This quantity is synonymous with the net present value of all capital and operating costs of a generating unit over its lifetime, divided by the amount of electricity (in megawatt-hours) it is expected to produce. This intends to provide an indication of the economic effort required to produce a unit of electricity. Levelised costs have thus become the dominant parameter in comparing economic efficiencies, influencing energy policy.

Yet already in a 2011 paper, economist Paul Joskow of MIT noticed that this method of standardization inaccurately represents the value of electricity. It seems levelised costs are less useful for ranking and comparing alternative technologies than previously thought. And this especially applies to sources of renewable energy.

To re-evaluate the standard economic performance of renewable technologies, Joskow analyzes their interaction with the standard electricity. The economic side-effects of this interaction, so his argument, are not adequately represented by levelised costs.

For instance, most non-carbon energy units (take solar and wind technologies) can produce at only a comparatively small fraction of their capacity and have highly variable performance. The fluctuations in their generating capabilities may not concur with daily variations in electricity demand. Especially from countries increasingly emphasizing these technologies in the standard electricity system, this demands an awkward compromise. For stable grid output to be maintained, conventional electricity from fuel-based generation must be injected into the system as a supplement to the renewable energies. Since the fluctuations in non-carbon performance are unpredictable, this supplement must often occur through a process of redundancy production from the traditional energy infrastructure – which implies that conventional power plants are not only kept on stand-by, but are effectively kept running to be ready to supply injections. The costs associated with balancing the electricity system in this way when renewables go offline (what Joskow calls the cost of intermittency), among other factors, is not taken into account by standard levelised costs calculations. Therefore, levelised costs tend to understate the cost of electricity derived from renewable energy sources.

In a paper published in May 2014, Charles Frank of the Brookings Institution presents a more appropriate approach to ranking alternative technologies. He extends the spectrum of phenomena and side-effects taken into account by basing his parameter on a cost-benefit analysis. As formulated by Frank: “rather than using levelised costs to compare alternative technologies, one should compute the annual costs and benefits of each project and then rank those projects by net benefits delivered per megawatt (MW) of new electrical capacity”. Therein, “the benefits of a new electricity project are its avoided carbon dioxide emissions, avoided energy costs and avoided capacity costs [or the value of the fuel that would have been used if a fuel-based plant had produced the same amount of energy]”. The costs include, among others, the unit’s “own carbon dioxide emissions, its own energy cost, and its own capacity cost” as well as the cost of intermittency (which itself encompasses the costs associated with operating the supplement generating units).

It must also be noted that in itself the connection of renewable technology to the system is often an elaborate and expensive exercise, incurring broader costs for the grid. The most suitable sites for large-scale harvesting are often remote from regions of highest demand (urban areas, etc.). 

The Economist published a telling chart on the issue:


(from “Sun, wind and drain” by The Economist, Free Exchange column, July 26th 2014 issue)

For instance, comparing the costs and benefits of different non-carbon sources, it must be assumed that renewable technologies do not avoid carbon emissions or capacity costs (intermittency costs) when they are not running. The magnitude of their benefits in this respect consequently depends on their ability to run at a large percentage of their capacity (basically run for the longest). Therefore, within this parameter, nuclear power plants – which on average run at 90% of capacity – show the best economic performance of the zero-carbon technologies (avoiding almost six times as many carbon emissions per unit capacity as solar power plants). Taking intermittency cost into account, furthermore, a 1 MW solar power plant running at 16% of capacity could replace only roughly 0.15 MW of a coal-fired plant running at 90%. A nuclear power plant running at 90% of capacity could replace effectively the entire capacity of fuel-based energy.

Nonetheless, with respect to other cost and benefits, nuclear power plants have comparatively high capital and operating costs (taking into account nuclear waste handling and other associated hazard-management), as well as being uninsurable. Yet due to their high use of capacity, capital and operating costs are only 75% greater per MW for nuclear power plants than for solar plants.

Thus, solar and wind power appear very uncompetitive when compared to nuclear energy and conventional fuel-based production methods. And it must be noted that the avoided energy/capacity costs in Mr Frank’s analysis assume a carbon price of 50 USD per tonne. The economic inefficiency and general expensiveness of wind and solar energy, already shown to be worse than previously assumed by Mr Frank, would be even more pronounced if actual carbon prices (below 10 USD per tonne in Europe) were incorporated in the calculations. Carbon prices would have to surpass 185 USD per tonne for solar energy to show a net benefit with its current rate of emission avoidance.

According to Mr Frank’s analysis, on balance the economically most efficient non-carbon source is nuclear power – the least efficient sources are wind and solar power. This not only implies that the cost of solar and wind generation for the economy is larger than previously assumed, but also that these types of generation constitute the most expensive ways of reducing carbon emissions. Yet governments are spilling billions in subsidies onto solar and wind industries with the justification of helping battle carbon emissions.

The implications of Mr Juskow’s and Mr Frank’s insights are diverse. Artificially building renewable energies that are both economically inefficient and as of now highly variable in their performance is the most expensive and least effective method of reducing carbon emissions. The subsidization and promotion of cost-inefficiency within the energy mixes of developed countries may continue to raise electricity prices. And, if the reliable connection of already highly expensive renewable generating units requires conventional carbon-based plants to be kept running “just-in-case”, then it is in any way questionable whether significant increases in energy spending (and electricity prices) are worth that marginal reduction in carbon emissions they may effect. As The Economist summarizes: “governments should target emissions reductions from any source rather than focus on boosting certain kinds of renewable energy.”

Yet the implications of the above described insights are not satisfying to the solar enthusiast, or the solar industry in general. In a way, they appear to destroy any firm incentive to continue investment in solar technologies. But when departing from the theoretical world of economics and political criticism, the above implied pessimism is more illusory than practically appropriate. The derivation of recommendations from the economists’ statistical game has hitherto ignored public opinion with respect to the promotion of renewable energy, as well the innovation potential of certain technologies.

Furthermore, all the data above shows is that, in political settings, solar technologies are ineffectively instrumentalized in an artificial and overly expensive battle against carbon emissions.

Criticism towards this current mistreatment of renewable energy does not mean a serious, longer-term belief in solar’s contribution to solving the world’s energy problems is in any way unjustified. Moreover, we maintain that a solution to the above described problems requires investment in solar.

Needless to say, the incentive to invest in solar remains as strong as ever. Firstly, nuclear power remains too unpopular (and too hazardous on a large scale) to constitute an ultimate solution. To avoid any potential hazards, after all, the renewable energy industry has developed towards the pursuit of replacing not only hydrocarbon-based power, but nuclear power as well.

The aim remains to achieve stable grid output based only on renewable energy sources. In order for solar to become competitive in this respect, and thus make a powerful contribution to solving the problems described above, it must be able to neutralize the fluctuations in its performance. This will benefit the technology’s economic efficiency by eliminating intermittency costs and, in consequence, its ability to avoid significant quantities of carbon emissions. The development of storing technologies and pv-thermal combinations currently shows the greatest potential for maximizing solar’s capacity utilization. At the heart of this endeavor is a tight cooperation between material research and product development on one hand, and proper entrepreneurial deployment on the other in order to get economic efficiencies under control.


The Net Benefits of Low and No-carbon Electricity Technologies, by Charles Frank, Brookings Institution, May 2014

Comparing the Costs of Intermittent and Dispatchable Electricity-Generating Technologies, by Paul Joskow,  Massachusetts Institute of Technology, September 2011

“Sun, wind, and drain”, by The Economist newspaper, Free Exchange column, July 26th 2014 issue.

“Why is renewable energy so expensive?”, The Economist newspaper, The Economist Explains column, January 5th 2014,


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s