Photo Credit: COP28

The International Energy Agency (IEA) published its “Renewables 2023” report on Jan. 11. The report, rich with data as usual, highlighted the potential of solar photovoltaics (PV) in our energy landscape. Anticipation was high, given the importance of PV in the quest for sustainable energy. Considering the COP28 pledge made by more than 200 nations to triple renewable energy capacity by 2030, there is clearly a huge volume of renewables that must be deployed, and fast. And the world’s intergovernmental agency for energy sees solar leading the way.

“Solar PV manufacturing is ready to deliver much accelerated growth,” said the report’s lead author, IEA senior energy analyst Heymi Bahar during the press conference to launch Renewables 2023. “We are now back in a low-cost solar PV market and a lot of capacity ready to be deployed.”

The IEA noted that solar module costs had halved over the past two years. This follows a relatively short period, 2020 and 2021, in which price declines had ceased largely because of Covid-19 induced supply chain disruptions and strong end-market demand. Manufacturing capacity additions in 2023, occurring at a record-setting pace, have also accelerated cost declines.

At the Renewables 2023 report launch Bahar pointed to a doubling of solar manufacturing capacity in 2023 alone, across all manufacturing segments. Overwhelmingly, that capacity was added in China, some 97% according to the IEA, although there has been some increase in geographic diversity in the solar supply chain. Alongside these supply side dynamics, the IEA noted that solar demand is being bolstered by rooftop installations.

“There is increasing adoption of small-scale rooftop systems, because these affordable systems are helping people to save money on their electricity bills, which have increased since 2021 drastically in some countries,” said Bahar.

3x Expansion Within Reach
Given solar’s strength, the lofty COP28 renewables goals agreed to in Dubai look eminently achievable. The headline figure of the Renewables 2023 report was that the IEA expects a 2.5x expansion of renewable capacity by 2030, given current policies and deployment trajectory. “Not miles away from that [the COP28] target,” noted IEA Executive Director Fatih Birol.

In terms of generation capacity numbers, the published IEA modeling only extends through 2028, by which time the agency expects renewable capacity to grow to 7.3 terawatts (TW). Of that total, solar and wind are expected to account for 95% of the capacity.

Increased Ambition
It is gratifying to see solar being recognized as a renewable-energy powerhouse by the IEA. That it is acknowledging both the achievements on the supply side, in terms of solar’s scale and cost, along with the technology advantages, is particularly satisfying.

However, if you’ve been in the renewable energy sector or solar industry for some time, you’d know that it hasn’t always been that way

Since 1998, the IEA’s flagship publication has been the “World Energy Outlook.” And for many years solar’s potential was largely overlooked.

Last decade, while solar was in the grips of a truly transformational period of expansion, the IEA consistently underestimated what it was capable of. The IEA would recognize another year of expansion in its installation data, and then anticipate flat growth from that point onward. While there was year-after-year solar market expansion, when it came to its World Energy Outlook, the IEA always expected growth from that point onward to end.

Underestimating PV
It's a tricky thing to describe in words, so best to let a chart do the talking. Thankfully a researcher from the Netherlands compiled the data – his January 2018 version of his chart is below. His name is Auke Hoekstra and with a professional background in ICT (Information and Communication Technology – namely, the internet) he could spot a disruptive technology when he saw it.

Hoekstra’s chart tells the tale. The black line shows what solar was actually doing – growing almost each and every year, at a rapid pace. By contrast, the different scenarios envisaged in the World Energy Outlook would have solar growth flatlining, somewhat inexplicably. The chart has since been updated by others, but the trend remains the same.

The founder of what is now BloombergNEF, Michael Liebreich, never one to shy away from an argument, was another voice in the chorus of criticism – although he was praiseworthy of the IEA’s efforts more generally and acknowledged the difficulty in making forecasts.

Liebreich published a blog in September 2017, based on his keynote at the BloombergNEF Finance Summit in London that year. In it he wrote:

“You can immediately see the problem if you take the first derivative of those IEA solar capacity forecasts: each time, the assumption is that installations have reached some steady state and will no longer grow. The 2016 IEA World Energy Outlook New Policies Scenario shows PV capacity additions flat at around 50 GW per year out to 2040. In 2017, just one year later, the world’s PV industry will install around 90 GW.”

Spooling forward to today, BloombergNEF charts around 400 GW of PV installations in 2023 and expects that to rise to up to 570 GW in 2025.

Seeing the forecasting failure by the IEA, some more conspiratorially-minded renewables advocates suggested that the Paris-based agency’s analysts were being willfully blind – propping up energy incumbents in the face of evidence to the contrary. Speaking to me at pv magazine at the time, Hoekstra was kinder:

“What I do know, from my experience with the internet, is that those experts who got it wrong were not doing it on purpose. They were simply unable to think that what was true for them all these years, all of a sudden was not true anymore.

“What has been happening is that there are a lot of non-linear positive feedback loops that are radically changing the energy system and experts of the old, as I call them, can really not accept or rationally give a place to the fact that the energy system is going to be radically different.”

Both Hoekstra and Liebreich pointed to the same phenomenon: the IEA, steeped in generations of energy-sector knowledge and assumptions, simply failed to see the scaling of the solar industry, and wind, batteries, and EVs alongside it. As a result, the Paris-based agency continued to underestimate solar’s ability to continue growing, reducing costs, and expanding within a clean technology ecosystem.

As Hoekstra then summed it up: “Solar panels are getting cheaper, and when they get cheaper people buy them more, and when people buy them more, they get cheaper. Wind is the same thing. Batteries for electric vehicles are also the same thing: The batteries get cheaper, and that means for an increasing number of people electric vehicles are the least-cost option.”

Cost Counts
The IEA has defended its projections for solar installations by arguing that they were based on different scenarios, when taking into account existing, announced, and policy measures that may be deployed by national governments in the pursuit of ambitious climate goals. The other key input, of course, was the expected cost of solar. But here, too, the IEA appears to have been far too pessimistic.

Cleantech investor, author, and solar advocate Ramez Naam weighed into the discussion in 2020, charting how IEA cost forecasts for electricity from utility-scale solar arrays were woefully inadequate. Naam’s background is also in ICT, having worked on early versions of Microsoft Outlook, Internet Explorer, and the Bing search engine.

“The decline of solar costs over the past decade has been faster than almost any credible forecast,” wrote Naam. “Actual solar prices are well below even the most optimistic projections put forth a decade ago. Adding forecasts from the IEA 2014 and mine from 2015 as well, we see that solar prices are decades ahead of what these forecasts expected.”

The IEA itself explicitly pointed to solar's rapid cost declines as having informed its more optimistic outlook for PV deployment. And the outsized role China has played in achieving these declines was singled out for mention. 

“China by deploying a lot of clean energy technologies brings the cost of these technologies down and makes it affordable for many people around the world,” said the IEA analyst Bahar at the Renewables 2023 report press conference.

Speaking with me in a subsequent interview, Bahar also emphasized China’s role and the influence it has had in terms of solar’s growth and why it has made the IEA’s forecasting task particularly difficult.

“The Chinese solar PV market has been extremely unpredictable,” said Bahar. “Basically, in terms of the information available we have from China versus how we can analyze this information it has been an important challenge [to making reliable forecasts].

“We saw over the last few years the exponential growth of solar PV demand. If you separate China and non-China countries, you will see the trend that one is going exponentially, and the other is growing but not in the same way.” In fact, the Renewables 2023 report finds that China installed more PV in 2023 than the entire world did in 2022.

And herein lies a more-or-less reasonable explanation as to why the IEA has failed to anticipate solar’s expansion. China has an outsized influence on solar deployment and production, its policy landscape is complex and often opaque, and when the IEA draws on the policy information that is available to it for its forecasts, it misses the full extent of the Chinese market.

In its Renewables 2023 report, the IEA had to set its forecast for China to one side, given its enormous scale. It found that more than 2 TW of renewable energy capacity is to be added to China’s grids between 2023 and 2028, in the IEA “main case,” so not its most optimistic scenario.

“We are trying really our best to follow the market trends,” Bahar said, as a concluding remark. And it is entirely true that the Chinese PV market has been difficult to anticipate at the best of times. Given its vast size and dominance of the solar manufacturing landscape, small oversights by the IEA can entirely overshadow a clear understanding and accurate forecast.

It is a plausible explanation and there is likely some truth to it. There is no doubt that the IEA faces an enormous challenge in tracking the energy transition and anticipating the emergence of a new paradigm. But equally, Hoesktra, Liebreich and Naam have likely described well the shortcomings of agencies like the IEA, which are steeped in incumbent energy system thinking, and struggle to grapple with disruptive new technologies like PV.

While it is illuminating to place the IEA’s expectations for solar in light of its previous missteps, there are also troubling implications for forecasting failures. Back in 2017, Liebreich hit the nail on the head when he wrote,

“Bad forecasts from authoritative agencies have consequences. They are self-reinforcing. They also lead to bad decisions, which destroy wealth. Never mind the stranded assets of the future, just look at the stranded assets of the recent past.”

So what of IEA’s most-recent forecast? The agency’s rhetoric is certainly encouraging. But the forecast itself? I’ve isolated the latest forecast from the IEA – although in previous years they’d demur at the term. Too low? About right? Or getting it wrong again, like the bad old days?