EDITOR: | April 15th, 2016 | 9 Comments

Cheap Lithium-ion Batteries for EVs vs. The Cobalt Cliff

| April 15, 2016 | 9 Comments

Since early March I’ve written five articles that focus on supply and demand dynamics in the cobalt mining sector and explain why I believe the lithium-ion battery industry is facing a raw materials shortage of epic proportions. Today I’ll drill down into market dynamics within the lithium-ion battery industry and explain why I believe cheap lithium-ion batteries for electric vehicles (EVs) will be the first casualties of the Cobalt Cliff.

I want to begin with an explanation that I’ve derived most of the numbers in this article by digitizing graphs from Avicenne Energy’s presentation at AABC 2016 in Mainz, Germany, a process that’s inherently imprecise. While I believe my estimates are close enough to offer a good overview, digitization is dependent on the visual acuity of the human being running the software and like most humans I’m imperfect on my best days.

Based on a careful analysis of Avicenne’s graphs I’ve estimated that in 2015 the lithium-ion battery industry manufactured cells with 61,500 MWh of capacity and generated $17.2 billion in revenue, which works out to an industry-wide average revenue of $280 per kWh at the cell level. Total battery sales to automakers were roughly 15,000 MWh and total revenue from those sales was roughly $5.4 billion, which works out to an average revenue of $360 per kWh. That average will seem high to readers who are accustomed to an endless stream of EV hucksters’ happy-talk about $150 per kWh batteries, but it’s pretty accurate when you factor in the more costly cells used in PHEVs and HEVs.

In 2014, total lithium-ion cell production was 50,000 MWh and sales to automakers were 10,000 MWh. So while the lithium-ion battery industry grew at an overall rate of 23% in 2015, automaker sales grew by 50%. Without the Cobalt Cliff, I’d expect the 2014 growth trends to continue, if not accelerate.

While Avicenne’s AABC presentation didn’t offer a detailed breakdown by end-use sector for 2015, it did include this graph for 2014.

Battery Market by Sector

In my view the key takeaway from the graph is that four different end-use sectors; smartphones, portable computers, other portable electronics and automobiles, had roughly equivalent battery capacity requirements. As cobalt supplies decline, I expect each of these end-use sectors to become an aggressive and powerful competitor for the available supply of high-energy lithium-ion batteries.

In my first article on graphite, lithium and cobalt supplies, I explained that all high-energy lithium-ion batteries use cobalt as an essential raw material in their cathode formulations. I also estimated that cells with LCO chemistry need about 1.44 kg of cobalt per kWh of battery capacity while cells based on NCM and NCA chemistries only need 0.36 kg, and 0.22 kg, respectively. That leads to an easy conclusion that a significant cobalt shortage and the attendant cobalt price increases will make LCO less desirable over time while making NCM and NCA more desirable. In my mind, the key unanswerable questions are “When will the inevitable transition away from LCO begin in earnest?” and “How long will that transition take?” Since the answers to those questions will ultimately be driven by end-users that buy large volumes of lithium-ion batteries for integration into their high-value products, I won’t even venture a guess.

At the outset, it’s important to understand that even large swings in cobalt prices will only have a modest impact on lithium-ion battery prices. At the current price of $10.21 for a pound for cobalt, LCO batteries need $32.40 of cobalt per kWh of capacity while NCM and NCA need $8.10 and $4.95 of cobalt, respectively. So while cobalt represents 12% of the iImplied cost of LCO cells, it only represents 1% to 2% of the implied cost of NCM and NCA cells. With a return to the $50 a pound cobalt prices we saw in late 2007 and early 2008, the implied cost of LCO cells would increase by about 46%. However the implied cost of NCM and NCA cells would only increase by 6% to 9%. Flow through raw material cost increases of that magnitude simply aren’t big enough to materially impact consumer product prices or severely dampen end-user demand.

While I don’t expect the Cobalt Cliff to have a substantial direct impact on lithium-ion battery prices, I believe a cobalt shortage will give cell manufacturers who have secure cobalt supply chains a critical business advantage they haven’t enjoyed for the better part of a decade – pricing power!

The most troubling number in Avicenne’s AABC presentation is that lithium-ion battery manufacturers struggle to survive with average gross margins of less than 10%. For many, the gross margins on lithium-ion cells are negative, but they’re less negative than the cost of idle factories. This is not a sustainable dynamic when lithium-ion battery users routinely target gross margins of 25% or more. Frankly, unless everybody in a value chain earns a reasonable margin, the value chain itself is inherently unstable.

The biggest reason for today’s horrible battery industry margins is that the industry over-built capacity during the last decade and the ensuing capacity glut gave battery buyers immense negotiating power. As a result, most big battery buyers brought brass knuckles to negotiating sessions and cell manufacturers acquiesced to unreasonable demands out of fear that they might lose key customers to competitors.

As cobalt supplies tighten, cell manufacturing capacity will become less relevant and supply chain security will take center stage. Almost overnight, the negotiating dynamic of the last decade will reverse itself and cell manufacturers that have secure supply chains will find themselves holding the aces. Instead of meekly accepting threadbare profit margins to keep their factories working, cell manufacturers with secure supply chains will be able to say, “We can’t make more than XXX MWh of cells per year because our supply chains won’t support larger volumes. If you want to be a priority customer for our cells, you’ll have to pay a price that gives us a reasonable profit margin.” Heck, we may even see some cell manufacturers pull out their own brass knuckles as they try to recover some of the operating losses and asset write-downs they’ve absorbed over the last few years.

I can’t predict what will happen when the battery industry regains it’s pricing power, but I will invite you to engage in a simple thought experiment. Imagine for a moment that battery manufacturers find themselves with unlimited pricing power and decide to charge $1 per watt-hour (wh) for cells instead of the current industry average of $0.28 per wh. Outfits like Apple and Microsoft that need 10 wh of battery capacity for a smartphone or 50 wh of capacity for a portable computer or tablet should be able to absorb the cost increases or pass them along to consumers without losing a step. In the EV space where 24,000 wh packs are “so yesterday” and 60,000 to 90,000 wh packs are the new normal, the impact would be devastating. There’s no way EV manufacturers can absorb the cost increment and doubling EV sticker prices will be a non-starter with consumers.

The example is extreme bordering on the absurd, but it highlights the inescapable reality that manufacturers of high-value products like smartphones, portable computers and other portable electronics who only need a little battery capacity will be better positioned to survive the Cobalt Cliff than manufacturers of highly price-sensitive products like EVs that need massive amounts of battery capacity.

It’s going to be fun watching this battle of industrial titans unfold.



InvestorIntel is a trusted source of reliable information at the forefront of emerging markets that brings investment opportunities to discerning investors.

Copyright © 2016 InvestorIntel Corp. All rights reserved. More & Disclaimer »


  • Peter

    Conico Ltd. in WA is sitting on the worlds 4th largest cobalt deposit, if the cobalt prices rise it would certainly be a low-risk low-cost opportunity for every manufacturer needing to secure a producer in their supply chain, especially the chinese, korean and japanese ones due to positioning.

    April 16, 2016 - 5:06 AM

  • Hugh Sharman

    Informed, incisive and disturbing to a fault, as usual! Forget the forthy guff about $100/kWh which is still doing the rounds in the “green” press as of this week.
    Thank you and congratulations, John!

    April 16, 2016 - 11:38 AM

  • Seb

    Conico’s project economics is also dependant on the nickel price however.

    April 16, 2016 - 2:31 PM

  • Peter

    Thank you for the insight John, very much appreciated.

    Looking at any other virtual producers however, none gives me a better assurance (as vague as CNJ’s may be):

    Formation Metals have a PEA and a purchase of some heavy machinery with a construction halt since 2013, but I am unable to find a more comprehensive study about their deposit which would give further insight in the feasibility of their project which makes me wonder why were they starting the construction in the first place (but offers a probable answer why it was halted three years ago).

    Broken Hill Prospecting doesn’t have a JORC compliant study yet, even in a PEA stage, other than trending news there is no current information necessary for further assessment. The orebody at Thackaringa presented on their website is vertical, thus the mining costs will be a factor when compared to others.

    Conico Ltd. (CNJ) is I agree not better positioned than the competitors in terms of project progress, all of them own a site (with some even making roads on it), however Conico’s surveys are at least near the PFS stage giving additional insight. The project ore is totally oxidised, negating the need for drilling and blasting. The shallow ore body is amenable to low cost, simple, conventional open pit mining. According to the study they presented their production potential will allow it to command from start-up, at least 3-4% of the global cobalt production. The study is of course not a DFS, but in comparison to others it is still something to go by.

    I agree that it is improbable that any of the virtual producers will have a producing mine in the time necessary to have an impact on the cobalt shortage, however the cobalt shortage will have an impact on the cobalt pricing which will have an impact on them, as we currently see with the market caps of virtual lithium producers (holding nothing but fields) rallying in the last couple of months due to the increase in the lithium demand and pricing. My assumption is, the cobalt developers will experience the same.

    April 16, 2016 - 5:58 PM

  • Weekly Update: April 18th, 2016 | Power Trip

    […] An interesting piece about the raw materials that go into making batteries, which need cobalt to do their thing. The author believes that we are on the brink of a “cobalt cliff,” and are heading for huge shortages. And as cobalt supplies dwindle, battery makers will be hit hard, and the cheap lithium-ion batteries that are necessary to make EV’s marketable may be in great danger. http://investorintel.wpengine.com/technology-metals-intel/cheap-lithium-ion-batteries-for-evs-vs-the-cobalt-c… […]

    April 18, 2016 - 4:27 PM

  • Henk Mol

    Hi John nice article. Long term effects however can be expected from this one http://www.nature.com/articles/nenergy201639

    BR Henk

    April 19, 2016 - 8:06 AM

  • Henk Mol

    Hi John – I agree to that. It is long term. The Li-Ion technology has also developed for more than 25 years before we have full industrialisation reached. Moore’s Law of batteries has an astonishing 55 years clock frequency to double their energy and power density. I don’t expect to see 3 or 5 years. We run into the cobalt cliff while we knew this already 10 years ago because then already it appeared that Co was a scarce metal mined in politically unstable areas and as byproduct of other mining. Jack Lifton also published a list of metal volumes in a brief table somewhere in 2009 and Co was pretty low on the list – another warning. Basic research into the Zn / MgO2 system has not received much attention, much for the reasons you mention (low power density coupled with reasonable to high energy density). Yet, like the hybridisation of ICE cars today, it is the combination of energy sources with power rating optimised and other energy sources with capacity optimised. I.e. a “hybrid” energy storage has economic potential. Some Li-Ion storage, sure, to give you good road agility, but needs not be 60 kWh, as another but much “slower” storage bin has the other say 2/3rd of the energy. Dont’t get me wrong, I’m very skeptical about the reality of electric cars for many reasons (economic and environmental). It is a much too expensive medicine that only changes NET a very small amount of environmental loading – unless combined with a nearly perverse size of investments on electrical infrastructure, generation, and storage – with a smart grid that potentially exerts an undisired, “hydraulic despotic” some would argue, power over the individual consumer. But, when being with your back against the long term and short term walls of the energy and other personal needs that grow along with the siz of the demographic wave (we experience other quite dramatic effects of that also in Europe) we have to diversify and “hybridise”. Change must and will come like it or not. Best regards, Henk.

    April 20, 2016 - 2:13 AM

Leave a Reply

Your email address will not be published. Required fields are marked *