Cheap, widely available electricity from solar PV in the coming decades looks increasingly likely. It could transform the world’s energy system, and prospects for reducing carbon dioxide emissions.
Only a decade ago solar was a niche source of energy, accounting for a tiny fraction of world electricity consumption. It was also very expensive.
Since then its status has been transformed. The installed base grew by a factor of about 17 between 2010 and 2018, with consecutive doublings of cumulative installed capacity every two years or so. By 2018 output was over 2% of world electricity demand, and continuing to grow rapidly.
Over the same period prices came down by a factor of four. Analysis by the International Renewable Energy Agency (Irena) shows average prices to have fallen from $240/MWh in 2010 to around $60/MWh in 2018 (see chart, which also shows that prices for wind power have also fallen, but by a much smaller proportion).
Chart 1: Global average prices resulting from auctions for solar and wind, 2010-18
Some recent solar PV contracts are already well below the level found by Irena. Prices are below $20/MWh in some cases (see Table 1).
Table 1: Some contract prices for solar PV in 2019
Source: press reports[ii]
The reasons for the differences between these costs and the higher costs shown by Irena appear to include:
- Comparing global averages with world’s lowest cost. For example, the Portuguese contract is an outlier, with most contracts in Europe remaining above $50/MWh. Similarly, the Dubai project may benefit from cheap land, and perhaps low-cost capital.
- The inclusion of incentives reducing some of the contract prices. For example, the California projects’ prices would likely be closer to $30/MWh without production tax credits available to the projects.
- Continued fall in costs since 2018
So how far could solar PV costs go in the long term?
Projecting cost trends following such large changes is inherently uncertain. However there seems no clear reason why changes as large as those that have already taken place should not happen again. Costs fell by 75% between 2010 and 2019 in response to capacity increasing by a factor of 17, following a typical experience curve. There is scope for a similar increase in capacity – a factor of about 17 would take solar to around 40% total electricity production, plausible over the next few decades. This may lead to a similar reduction in costs of around 75%.
Indeed cost reductions may occur much more quickly than that. The transformation to date has happened very rapidly, and there may be as yet more unrealised gains as more fundamental R&D, which is longer lead time, comes through.
There may also be potential for currently higher cost locations where the industry is relatively undeveloped to catch up with the best. This may reduce average costs, even if the best projects do not reduce costs as rapidly.
Conversely, some efficiency gains, such as those in construction costs of large scale installations may be one-offs, and similar gains may not be possible in future. These limits have not been binding so far, but may be more so in future.
Taking these considerations together, a very rough and ready estimate would suggest widespread availability of unsubsidised solar power at prices at around $10/MWh or less within the next 2 or 3 decades or so, and possibly much more quickly than that.
This would make solar the among the cheapest form of large scale, high value energy the world has ever seen. By comparison, annual average crude oil prices dropped to $7/MWh (in $2019) only briefly at the end of the 1960s before the oil crises of the 1970s took effect and prices increased[iii]. Oil is currently around $25/MWh ($60/bbl), and is a less valuable and versatile form of energy than electricity. Coal may in some cases be cheaper per MWh but is expensive and inefficient to convert to electricity, so total fuel costs may exceed even some current solar costs. Coal is also, of course, highly polluting, with expensive CCS required to reduce its emissions.
What are the implications of this?
To bring low carbon energy prices to close to the levels prevailing for high carbon energy in the era of cheap oil would be potentially transformative for emissions reduction. Effects would be widespread and may include the following.
- The generally expected route of decarbonising the power sector and electrifying end use becomes much easier with plenty of cheap low carbon electricity. This is supported by falls in the costs of batteries, which will help solar based systems cope with large daily variations in output. Land availability for solar is not a constraint globally, but maybe in certain places where population density is high and there are other uses for land.
- It will also support electrification of transport, helped by electric cars being much more efficient than internal combustion engines anyway.
- Hydrogen is more likely to be produced by electrolysis than by natural gas reforming with CCS, especially as large amounts of surplus solar may be available at times of peak production. And with cheap electricity, hydrogen use is likely to be focussed on applications where its specific qualities are needed, for example some industrial processes requiring high flame temperatures.
- CCS is also likely to be focussed on a few special classes of emissions, especially those from industrial processes.
Some things seem likely to remain true even with cheap solar …
- Wind power is likely to continue to play an important role for reasons of system security, diversity, and different locations of resource.
- Improving energy efficiency, especially in buildings, will continue to be worthwhile, as it makes the transition to lower emissions easier in many respects.
- The emissions reduction challenge remains daunting, not least because of the scale of the transformation required is formidable, and existing assets are long lived.
- Long-haul aviation looks likely to continue to be a challenge, perhaps requiring synthetic fuels or biofuels.
- Land use emissions from deforestation and agriculture remain a large problem.
So far modelling of the energy system seems to have largely ignored the possibility of very cheap solar being widely available. This is an area which needs much more consideration. Nothing gives greater hope for solving the climate change problem than the prospect of abundant cheap low carbon electricity.
Adam Whitmore – 18th November 2019
[i] See here for contract prices quoted in the table and commentaries.
[iii] Based on $12/bbl (source: BP statistical review of world energy) and 1.7MWh per bbl of crude oil