The meaning of price fluctations
For lithium-ion technology, overall raw material cost soared by 300% to 700% in the 12 months from March 2021 to March 2022 (Figure 1). This is outside of the uncertainty range of usual raw material price fluctuations and driven by a combination of extraordinary circumstances, mainly:
The supply chain disruptions and delayed expansion of mining and raw material processing capacity due to the Covid pandemic that started in 2020
The firm increase in lithium-ion battery production due to a strong uptake in electric vehicles and stationary lithium-ion storage in 2020 and 2021
The Russian invasion of Ukraine in February 2022 and the resulting war that created broad uncertainty on resulting economic implications
The foreseeable increase in interest rates by central banks for 2022 to counter high levels of inflation
As a result, the cost of nearly all raw materials used in lithium-ion batteries increased significantly. In particular, the cost of lithium carbonate (used in LFP) increased, which also affected the cost of lithium hydroxide as derivative (used in NMC). The raw material cost increase of 300% and 700% would increase battery pack prices by 100% for NMC and 150% for LFP respectively.
Figure 1 – Lithium-ion battery raw material price index from January 2020 to March 2022.
However, battery manufacturers are unlikely to be directly subject to these elevated raw material costs, because raw materials are largely purchased via long-term contracts. These elevated price levels are unlikely to sustain long enough to be reflected in future long-term contracts. Figure 2 illustrates this with the example of high-purity polysilicon used to manufacture solar PV modules. Three important insights can be derived from this chart.
First, the continuous alternation between shortage and oversupply of polysilicon is a great illustration of the Cobweb theorem, which also applies for lithium-ion raw materials and production capacities. The impact on polysilicon contract price (see Figure 2a) can be seen in the product price for solar PV modules (see Figure 2b). However, it does not materially distort the overall product price development from the underlying experience curve.
Second, there was an extraordinary polysilicon price increase between 2005 to 2008. This was due to the combination of an increase in feed-in-tariffs for electricity from solar PV in Germany, which quadrupled solar PV module demand in Germany in 2004, and a high growth rate for semiconductors in 2004 (23%), which also use high-purity polysilicon. However, this raw material price increase was time-limited because it triggered innovation (the amount of polysilicon in PV modules dropped from 13g to 10g per Watt) and additional polysilicon processing capacity. As a result, prices collapsed in 2008 exacerbated by the financial crisis and scrapping of lucrative feed-in-tariffs in Spain. It thereby shows that the extraordinary raw material cost increases seen in 2021/22 are unlikely to sustain for more than 2-3 years at maximum.
Lastly, while the spot price for polysilicon increased by 1,500% from ~30 USD/kg (2005) to ~460 USD/kg (2008), contract prices only increased by 100% from ~35 USD/kg to 70 USD/kg. This shows that extraordinary increases in sport prices do not directly feed through to the long-term raw material contracts that battery manufacturers are likely to have with raw material producers.
Figure 2 – Impact of high-purity polysilicon availability on a) polysilicon contract and spot market prices and b) product prices of solar PV modules. Red circles indicate times of polysilicon shortage or oversupply. In chart a), solid black line refers to polysilicon contract prices (left axis) and grey line refers to polysilicon spot prices (right axis).
Overall, these observations show that experience-curve based price projections for electricity storage technologies are unlikely to be significantly distorted by raw material, component or production capacity supply-demand imbalances. Any increase in product prices due to a shortage is likely followed by a return to the long-term cost-reduction trend once supply meets or overshoots demand. This can be the result of additional mining and processing capacities, product innovation and/or a reduction in the rate of demand increase.
But, this discussion also shows that experience curves are ill-suited for short-term projections of product prices. They do not capture commodity price fluctuations and market developments. Their purpose is rather to uncover underlying cost development trends and make long-term product price projections.
Schmidt, O., & Staffell, I. Monetizing Energy Storage - A toolkit to assess future cost and value. Oxford University Press. Forthcoming.