Why the low-carbon transition can be much cheaper than models predict
To achieve net zero emissions by the middle of the century, global energy systems must en bloc shift to low-carbon, energy-efficient technologies.
However, many models used to map this transition imply that there are benefits to delaying investments in these technologies, rather waiting for R&D to reduce costs over time.
Such models overlook the results of a vast literature showing that the costs of these technologies decrease due to their use.
Our new paper in Environmental research letters draws on evidence from over 200 journal articles and concludes that policies promoting such âinduced innovationâ have been a clear factor behind the remarkable success of low-carbon technologies.
This suggests that not only is urgent action essential to ease the transition to clean energy, but the transition itself may also turn out to be much cheaper than expected.
Models vs reality
Government decarbonization strategies often rely heavily on integrated assessment models (IAM) which attempt to represent the energy system and the economy.
These models select combinations of energy supply and use technologies to achieve decarbonization with the lowest total cost over time.
Typically, the costs of low-carbon technologies in these models are assumed to decrease gradually as they become more mature, mainly over time, but also through policies such as support for R&D.
This means that even though the costs of mature existing technologies are generally assumed to remain largely stable, low carbon technologies remain more expensive in the future.
As such, switching to low carbon technologies to achieve decarbonization usually comes at a considerable cost.
While technological cost assumptions may be adequate for mature technologies with well-known characteristics, they may be broadly valid for immature technologies.
This is evident for solar power, which already hit costs well below the 2050 forecast just a few years ago.
In a remarkable turnaround, the International Energy Agency (AIE) said solar offer now “Cheapest electricity in history”, leading to a “new normal“With exceptional levels of deployment underway, well above its previous forecasts. This can be seen in the Carbon Brief table below.
Gigawatts of solar capacity added annually worldwide (red line) and IEA 2021 Renewable Energy Market update (red triangles), as well as IEA Global Energy Outlook released between 2009-2020. Source: Carbon Brief analysis of IEA reports. Carbon Brief chart using Highcharts.
Models that do not take into account the dynamics that lead to such developments are likely to greatly underuse key technologies and overestimate costs.
In general, the importance of “technology-drivenâPolicies, such as R&D funding, are well represented in models – or at least supported by projected cost reductions. But the role played by âdemand pullâ policies – those that create, expand or enhance the market for new technologies – tend to be overlooked.
A growing literature argues that factoring these policies into business models can produce results that involve earlier, larger-scale investment in low-carbon energy at a lower long-term cost.
Efforts to address this problem have been challenged by the disparate, ill-characterized and often contested evidence regarding demand-driven innovation.
In our new article, we attempt to fill this gap by synthesizing 228 academic journal articles from a wide range of research disciplines, sectors, technologies, time frames and geographies.
We note unequivocally that since the Oil price shocks of the 1970s, innovation in energy technologies has responded to energy prices, large-scale deployment and strategic low-carbon policies.
The drivers of innovation that drive demand can take a variety of forms, ranging from ‘market-wide’ phenomena such as rising energy and carbon prices, to ‘targeted’ policies such as subsidies. or technological regulations.
Different drivers seem to drive innovation in different ways.
Rising energy and carbon prices encourage the development of lower-cost or low-carbon technologies, but this innovation has so far been fairly gradual.
It was driven by incumbents that made marginal improvements to maintain a competitive advantage, such as automotive companies investing in more efficient fossil fuel vehicles, rather than electric or hydrogen technologies.
We find that more radical innovation – necessary to deliver newer and initially expensive renewable energy technologies – was much more responsive to targeted policy approaches.
Purchase rates (FiTs) are by far the most widely adopted instrument for deploying renewable electricity. They provide a fixed subsidy per unit of output, typically for 10 to 20 years.
Evidence suggests that the security provided by FiTs has provided a strong incentive for innovation in low-carbon technologies, particularly for solar power.
Some studies provide insight into how the combined policy instruments influence innovation. Overall, we found that:
- Radical innovation in low-carbon technologies is triggered by targeted policies that build market confidence and overcome barriers to entry.
- Firms are much more likely to innovate if they have already done so, so encouraging innovation can have positive and reinforcing accumulative effects.
- The broader policy environment is important in determining the impacts of innovation, in which factors such as overall policy coherence and coherence, future policy expectations and the institutional context can have a strong influence on the propensity to innovate .
Patent applications are by far the most common measure of innovation, in large part because the data is widely available, quantifiable and detailed. However, they only provide a partial picture.
Patents are a good measure of innovative âactivityâ, but not of âoutcomesâ such as reducing the cost of technology.
Examining the full link between demand pull factors and outcomes is extremely complex and, therefore, relatively few studies do so directly.
In our review, we try to partially fill this gap by including studies deriving from “curve experiment“And associate”learning rate– the percentage of cost reduction for each doubling of cumulative deployment – for low carbon technologies deployed as a result of demand attraction policies.
Solar and wind power were by far the most studied technologies of those we considered, the former having seen an average learning rate of around 20% for much of the past 40 years.
The solar learning rates from the literature we reviewed are shown in the figure below in terms of cost reduction per unit of installed capacity. The length of the colored line indicates the period over which the average rate was calculated.
Learning rate,%, for the cost / price of installed solar photovoltaic (PV) capacity. The position of the colored lines on the y-axis indicates the average percentage cost reduction for each doubling of the cumulative deployment, while the length of the lines indicates the period of time over which the average was calculated. Source: Grubb et al. (2021).
Much of this cost reduction is attributed to learning by doing and the economies of scale that arise as a new technology is deployed and the supply chains and markets around it expand.
Some studies that suppress the role of other influences, such as R&D funding and the price of energy and materials, generally find lower learning rates – but clearly positive.
Falling technology costs also induce additional deployment, which can confuse the issue when examining the factors behind the experience curves. Although few studies attempt to explicitly test the direction of causation, those that find an important role in deployment-induced learning and economies of scale.
Evidence clearly shows that expanding deployment correlates positively with cost reductions, across a wide range of technologies.
If we take better account of the role of induced innovation, our energy and economic models can paint a very different picture of the best path to decarbonization and its cost.
A recent article, Modeling myths, directed by Professor Michael Grubb – also lead author of our new systematic review – explores the impact of including induced innovation, as well as inertia and path dependence, in Nordhaus’s DICE model, with striking results.
He finds that the âoptimalâ initial investment in low-carbon technologies is multiplied by several, but that the overall system costs for the overall transition are reduced.
This implies that early investment leads to a low carbon energy system that may be less expensive than the fossil-based system it replaces, which stands in stark contrast to the high cost of decarbonization involved when such dynamics are not taken into account.
We recognize that incorporating the impacts of innovation into models introduces significant complexity which in many cases is not desirable and may not be appropriate to try.
However, it is essential that policymakers and those who advise them are aware of these complexities and recognize the opportunities that early large-scale investments can offer.
Grubb, M. et al. (2021) Induced Innovation in Energy Technologies and Systems: A Review of the Evidence and Potential Implications for CO2 Mitigation, doi.org/10.1088/1748-9326/abde07
Grubb, M., Wieners, C. & Yang, P. (2021) Modeling myths: on DICE and dynamic realism in integrated assessment models for climate change mitigation, doi.org/10.1002/wcc.698
Teaser photo credit: by Erik Wilde from Berkeley, CA, USA – crop of the wind, CC BY-SA 2.0, https://commons.wikimedia.org/w/index.php?curid=51105579