This section is based on work that was done for the October 2024 Markets & Society conference. See our paper 1 for a more detailed treatment of the subject.
The cost of a product tends to fall as more is produced. The effect, identified by Wright 2 in the context of aircraft, can be quantified and modeled. Cost reduction through production is an essential element of bringing a clean energy technology to commercial maturity. Following are observed learning rates for select technologies. The learning rate is the percent cost reduction that is observed for doubling cumulative production.
Not all of the cost decline observed for a growing technology is necessarily a result of learning curves. Some of the decline may be the result of other technological improvements that would occur regardless of how much the technology is deployed 8.
Measuring learning rates may be difficult, and for this reason, estimate of learning rates for similar technologies may vary widely 9. A major challenge is that authors account for exogeneous factors, such as the effect of subsidies or commodity prices, in different ways. The published literature on learning rates shows a survivorship bias, meaning that higher values are more likely to be published than lower values, skewing up observed learning rates.
When a relationship between cumulative production and unit price is found, the causality of that relationship is difficult to prove. Price reduction may be caused by exogeneous price declines or economies of scale, rather than by cumulative production 10. It may also be that cost declines are driving increased deployment, rather than the reverse 8. Some learning models, known as two-factor models, incorporate cumulative research in addition to cumulative production. These models tend to find lower learning rates for production and also that cumulative research shows a greater effect on price than cumulative production 11.
The Wright learning model, that the cost of some production falls by a fixed percentage with every doubling of production, is a simple and widely used model, but there is evidence that for many products, learning rates fall as production increases 12.
Learning curves are especially challenging from a policy perspective, since some learning is a purely private gain to the firm producing, in which case a subsidy is not justified, and some learning is a knowledge spillover that may merit subsidy 13.
Goff, M. "Use and Abuse of Learning Curves for Energy Subsidies". October 2024. ↩
Wright, T. "Factors Affecting the Cost of Airplanes". Journal of the Aeronautical Sciences 3(4), pp. 122-128. February 1936. ↩ ↩2
Chen, X., Kotlyarevsky, A., Kumiega, A., Terry, J., Wu, B., Goldberg, S., Hoffman, E. "Small Modular Nuclear Reactors: Parametric Modeling of Integrated Reactor Vessel Manufacturing Within A Factory Environment Volume 2, Detailed Analysis". Department of Energy, Office of Nuclear Energy. August 2013. ↩
Goldie-Scot, L. "A Behind the Scenes Take on Lithium-ion Battery Prices". BloombergNEF. March 2019. ↩
Hax, A., Majluf, N. "Competitive Cost Dynamics: The Experience Curve". INFORMS Journal on Applied Analytics 12(5), pp. 50-61. October 1982. ↩
Reeves, M., Stalk, G., Scognamiglio, F. "BCG Classics Revisited: The Experience Curve". Boston Consulting Group. May 2013. ↩
Samadi, S. "The experience curve theory and its application in the field of electricity generation technologies - A literature review". Renewable and Sustainable Energy Reviews 82(3), pp. 2346-2364. February 2018. ↩
Nordhaus, W. D. "The Perils of the Learning Model for Modeling Endogenous Technological Change". The Energy Journal 35(1). January 2014. ↩ ↩2
Söderholm, P., Sundqvist, T.. "Empirical challenges in the use of learning curves for assessing the economic prospects of renewable energy technologies". Renewable energy 32(15), pp. 2559–2578. December 2007. ↩
Lafond, F., Bailey, A.G., Bakker, J.D., Rebois, D., Zadourian, R., McSharry, P., Farmer, J.D. "How well do experience curves predict technological progress? A method for making distributional forecasts". Technological Forecasting and Social Change 128, pp. 1-4-117. March 2018. ↩
Jamasb, T. "Technical change theory and learning curves: patterns of progress in electricity generation technologies". The Energy Journal 28(3), pp. 51-72. July 2007. ↩
Anzanello, M.J., Fogliatto, F.S. "Learning curve models and applications: Literature review and research directions". International Journal of Industrial Ergonomics 41(5), pp. 573-583. September 2011. ↩
Bläsi, A., Requate, T. "Subsidies for Wind Power: Surfing down the Learning Curve?". Economics Working Paper; 2007. ↩