IEA renewables report: Only solar PV exceed expectations
IEA estimates on costs of solar PV are getting more realistic, but they still overplay investment costs.
The main message in the new Medium-Term Renewable Energy Market Report from IEA is that sustained growth in renewable energy is at risk. Governments, therefore, need to strengthen efforts to facilitate growth of renewables in all energy sectors. In the power sector, both hydropower, wind and bioenergy fall short of the deployment speed consistent with IEAs 2 Degree Scenario. Solar PV, says IEA, is "the only source expected to exceed global 2DS targets by 2020, boosted by cost declines and an increasingly rapid scale-up in non-OECD markets."
This is troublesome news for renewables in general, but encouraging news for solar energy. And having studied the underlying assumptions in the report, I am tempted to add: This is just the beginning of the solar awakening of IEA; be ready for better news next year.
Deployment of renewables under different projections (source: IEA). Click on image for larger version.
There are three main markets driving the deployment to 2020, according to IEA. First, it’s China where renewables will account for nearly 45% of all new power generation capacity, clearly ahead of even coal. Secondly, it’s the more mature OECD markets, where renewable generation is expected to account for nearly 80% of new power generation between 2013 and 2020. And thirdly, it’s the emerging markets of Latin-America, Asia and Africa, starting from a lower level but with a higher growth rate.
In the report's baseline-scenario total installed renewables grow from 1700 to 2500 Gigawatt from 2013 to 2020. IEA projects that solar PV will reach 405 GW in 2020, with an average addition of 40GW per year (2013; 37 GW), i.e. more or less a flattening out after a 25 % growth the last three years. The new numbers are certainly more realistic than the "negative growth" forecasted by IEA in its World Energy Outlook last year, but still conservative in my view.
Fortunately, the report contains an enhanced case in order to show the potential impacts of a more renewable-friendly policy in a number of countries. Under the enhanced case, the most dynamic renewable electricity sources (i.e. solar and wind) grow faster, with solar PV reaching a cumulative 465-515 GW in 2020, starting from 139 GW in 2013.
Although of less consequence for the energy-mix in short term, the new "realism" from IEA is important because it reveals the significant role PV can play in the longer term. But IEA still does not seem to grasp the dynamics of the market, and therefore still exaggerates the costs of PV and underestimates its growth potential. There are two reasons for this; one is that they still do not get the solar costs right, the other is that they don’t get the cost of alternatives right either. This is unfortunate because it leads to the impression that solar PV generally is more expensive than the alternatives.
Let us take the solar costs first, where IEA seems to have been through a sort of awakening. Last year, in the Outlook, it projected large-scale solar PV would cost 2,2 dollar per watt in 2020, down from the 2.5-3 dollar per watt in 2013. This year IEA presents some very different numbers. The costs in OECD are rather in the range of 2 dollar per watt, they say, and will decline moderately to 1,7 dollar in 2020. For comparison: Fraunhofer Institute, for example, estimates systemcosts to be in the range 1.35-2.2 dollar per watt, a figure more similar to what the industry encounters in the competitive marketplace. Looking to 2020, Fraunhofer Institute,NREL (Laboratory of the US Department of Energy) and commercial research bodies like Citi Researchall expect investment costs of large-scale PV to approach or even drop below 1 dollar per watt, 40 percent beneath IEA's assumption.
On the way down: IEA projections of solar PV cost.
Regarding the cost of PV electricity, IEA this year comes out with a more realistic estimate. The Levelised Cost of Electricity (LCOE) for Large-Scale photovoltaics in 2013 ranges from 100 to 220 $/MWh, says IEA, reflecting the variations in solar resources and costs of financing. But as IEA also this year overplays the investment costs, they still overestimate the cost of power from PV. In regions with a solar radiation at the level of Southern Europe, says Fraunhofer, the LCOE of large-scale PV has already dropped to 0.08-0.10 $/kWh. These numbers are confirmed by tariffs currently offered by developers in sunny markets like India, US, South Africa and Chile, among others, even when corrected for tax and other incentives.
The other reason IEA still underestimates solar PV, is that they don’t get the alternatives right. One gets the impression that solar power plants compete with the wholesale tariff, but that is generally not the case. Large-scale solar PV competes with the cost of adding new generation capacity to cover the daytime consumption, i.e. building new gas, oil or diesel power plant including the exposure to rising price fuel and CO₂ costs during the project’s lifetime. In theory, thermal power still looks cheaper than PV in most parts of the world. But experience tells us that large thermal projects most often suffer from delays and cost overruns. E&Y, for example, found thatof 365 Oil and Gas megaprojects costing over $1 billion, 64% face cost overruns and 73% face delays. According to Citi Research, a large portion of the coal and gas power projects planned towards 2020 will reach a cost level where they at no or very limited cost can be substituted by wind or solar.
Investors are increasingly wary of these risks, and seem to put a higher hurdle rate on fossil fuel investments compared to wind and solar. Similarly, utilities are increasingly prepared to offer a premium to generation alternatives that mitigate fuel and CO₂ price risks, even if the tariff initially is higher than expected market prices the next decades. And, last but not least, governments struggling with growing budget and trade deficits as well as unemployment, are gradually beginning to see the added national economic value of harvesting domestic resources like wind and solar.
Nevertheless, solar PV is not baseload, and will therefore always be part of a broader energy-mix. Much has been written about the grid constraints limiting the magnitude and speed of the solar and wind revolution. But as highlighted by IEA, studies indicate that the power systems should normally be able to cover 25 % of annual generation from intermittent sources. With more flexibility in terms of demand management, interconnection and some possibility of curtailment, the share should be able to raise to 50 percent. Developments in battery technologies and other storage technologies, are expected to cut in half the cost of storing solar and wind power by 2020, radically increasing the potential for wind and solar to cover a larger share of power consumption than hitherto foreseen.
There are numerous signs, from states and cities in US, India, China, Brazil, Jordan, West-Africa among others, of the unforeseen fuel substitution described here actually taking place and driving the deployment of solar PV beyond current estimates. Take India as an example, where a recent report concludes utility-scalesolar PV is cheaper than coal, and that the country’s solar sector is set to boost the country’s PV capacity by more than 140 GW over the next decade, ten times more than predicted by IEA in 2020. In one initiative, the government aims at setting up so-called ultra-mega solar projects in every state in the country, with several states already announcing plans for single-site projects exceeding 1 GW.
As highlighted by IEA, abrupt policy changes, sustained grid constraints, macroeconomic and currency instability are still factors that can slow the deployment of PV. But although such set-backs are likely to occur in individual markets like we see in Australia now for example, it is hard to see what global risk factors could credibly threaten the double-digit yearly global growth in solar the coming decade.
