FACT CHECK:
"massive oversupply at present which will take 2/3 years to obsorb"I'm calling BS on this statement. Our "friends" over at GS recently released a new supply demand estimate. While others have strong growth assumptions, lets try the harder task of using the estimates from someone with a negative view on the supply demand balance.
So GS's view of the 2023 supply/demand situation is that new supply is 967kt with an additional 33kt from recycling (1,000kt total). Demand was only 971kt so there's an estimated oversupply of 29kt. That over-supply is 0.03% of annual demand. Multiplying by 365 the over supply is 11 days worth of demand. 11 days worth of demand taking 2/3 years to absorb? Reusing a classic line by Julia Roberts in Pretty Woman: "You work on commission right [yes]... Big mistake, Big, Huge".
GS estimated demand in 2022 was 827kt of LCE. In 2023 it is estimated at 967kt, the 2024 estimate is 1,173kt and the 2025 estimate is 1,571kt. That's annual increases of +144 (+17.4%), +202 (+20.8%) and +398 (+33.9%). Taking an average 23/24, demand is increasing at 14.4kt/mth. Across two months demand increases by around 43kt. That "massive oversupply" that is suggested to take 2/3 years to absorb, is under 2 months growth (if it even exists to clear).
GS's estimate was that the market was over supplied by 32kt in 2019 and then by another 38kt in 2020. By the end of 2020 there was a cumulative 70kt of surplus inventory and demand that year was 341kt (28kt/mth). The inventory surplus was about 2.5 months worth of supply or 75 days. 75 days is a oversupply, particularly when EV demand is just starting to build and large parts of the world are in lockdown and unable to use cars, let alone buy new one's.
The market was then under supplied by 56kt in 2021. The 2021 undersupply would have reduced the net 3-year position to +14kt. In 2022 the market was under supplied by a further 89kt while the estimate for 2023 is +29kt of over supply. Across this whole period there's a net deficit of 46kt. GS's prediction is that by the end of 2023 there is less inventory in the system than there was in 2019 and over that time the market is much larger.
Ok so demand is growing and growing quickly. The projected 2025 increase of +398kt is larger than total world supply in 2020 of 375kt. To meet the GS demand projection in 2025, every single operational mine/Brine in 2020 needs to be replicated as new capacity. All existing production capacity in 2023 needs to be maintained. The 2026 demand increase estimate is +307kt so that same process almost needs to repeat again. As context LTR's KV is under 100kt so to meet the 2026 demand estimate, you need over 3 LTR's coming online that year alone.
GS's supply modelGS's supply model is where things get a bit funny (in a sad way). Spod/Brine can be low cost ways of suppling LCE. Lepidolite is the high cost way of supplying LCE. GS have observed that lepidolite has had a huge increase in supply when prices are high, much the same way that expensive sources of oil come onstream when prices are high. GS then forecast that the 2024 production of lepidolite will increase from 140kt in 2023 (14% of world LCE supply) to 248kt in 2024 (21% of world supply). By all accounts lepidolite conversion to carbonate is loss making at current prices. Normal supply and demand economics would have it decreasing in supply. GS have chosen to ignore decades of supply/demand evidence and model that Lepidolite production will grow strongly despite being loss making at the suggested prices. If Lepidolite were to decrease back to 2020 volumes (13kt) then that would remove 235kt from GS's projected 2024 supply. Their projected surplus of 202kt would flip to a deficit of 33kt.
What GS's supply/demand model really means is that Spod/Carbonate/Hydroxide need to find the price were only a small amount of Lepidolite is profitable. That point would cause supply and demand to balance at a clearing price. That price point is probably above current prices.
(20min delay)
