Published on May 15th, 2016 | by Zachary Shahan


EV Battery Prices: Looking Back A Few Years, & Forward Yet Again

May 15th, 2016 by  

I spent a couple of hours last night updating our main electric car page (and also the one on EV Obsession that includes European cars but not battery & environmental discussions). While in there, I was reminded of EV battery prices, price changes, and price projections from several years ago, which jumped out at me as an awesome article topic!

From that page, here’s the updated text:

The upfront price tag of EVs and PHEVs is higher than that of similarly sized and equipped gasoline-powered cars, mostly because batteries are expensive. How expensive? That’s hard to know, because car manufacturers generally won’t say what they are paying for their batteries, or what they expect to pay in 1 year, 2 years, 3 years, etc. Here are some of the best answers we’ve got for now regarding EV battery prices for specific models:

→ Tesla’s battery packs were estimated to cost $240/kWh in 2014, while the rest of the industry was projected to be no lower than $400/kWh (that seems dubious). But the latest figure from a Tesla representative pegs its battery pack cost at under $190/kWh. (Note that CEO and Chairman Elon Musk stated in February 2012 that the cost of EV batteries would drop below $200 per kWh in the “not-too-distant future.”)

→ GM has a contract with LG Chem to get battery cells for $145/kWh, which probably translates into a battery pack cost around $190/kWh as well.

Bloomberg New Energy Finance Battery Price Estimates

For some historical background, though, here’s some info from a 2012 BNEF report that found that the average price of batteries used in electric vehicles dropped 14% from Q1 2011 to Q1 2012, and 30% from 2009 to 2012 (I didn’t even realize/remember that I have been writing about EV battery prices for this long!):

“Electric vehicles such as the Mitsubishi Motor iMiEV, Nissan Leaf or Tesla Model S require between 16 and 85kWh of storage, with a total cost of $11,200 and $34,000, or around 25% of the total cost of the vehicle,” BNEF notes. “Battery pack prices for plug-in hybrid vehicles such as GM’s Volt are on average 67% higher in terms of $/kWh, than those for electric-only vehicles like Nissan’s Leaf. This higher price is mainly due to the greater power-to-energy performance required for plug-in hybrid vehicles.”

A more recent BNEF study found that EV battery prices fell 35% in 2015. It stated that prices fell 65% since 2010, but it estimated battery pack prices at $350/kWh, which is considerably higher than the Tesla/Panasonic & GM/LG Chem estimates.

US Department of Energy Aims & Estimates

For another broad view, here’s a statement from US Secretary of Energy Steven Chu, from back in January 2012, on battery costs (emphasis mine):

“Overall, the Department of Energy is partnering with industry to reduce the manufacturing cost of advanced batteries. While a typical battery for a plug-in hybrid electric vehicle with a 40-mile electric range cost $12,000 in 2008, we’re on track to demonstrate technology by 2015 that would reduce the cost to $3,600. And last year, we set a goal of demonstrating technology by 2020 that would further reduce the cost to $1,500 – an accomplishment that could help spur the mass-market adoption of electric vehicles.”

It’s 2016, and it seems EV battery prices have fallen faster than projected. The DOE at that time was targeting $300 per kWh in 2015 (the $3,600 packs) and $125 per kWh by 2022.

Battery Price Projections Consistently Too High

Lastly, a 2014 study found that EV battery prices were falling much faster than most forecasts anticipated. Here’s a chart from that report:


Cost estimates and future projections for electric vehicle battery packs, measured in $US per kilowatt hour of capacity. Each mark on the chart represents a documented estimate reviewed by the study. Source: Nykvist et al. (2015).

Looking at that chart, it seems that Tesla/Panasonic and GM/LG Chem battery costs are already (in 2016) down to the lowest projections for 2020. Will we achieve $100/kWh by 2020?

Overall, we have been seeing something I’ve presented about in Mumbai, India; Vancouver, Canada; Cocoa, Florida, USA; and  Berlin, Germany: once a technology is ripe, it takes over the market quicker than anticipated and costs come down faster than most people anticipated. Check out these three presentations for more on that (if you haven’t already done so):

Let me know if I’m missing anything big here or you have more thoughts/data on this topic.

Obsessed with Tesla? Join our Tesla Obsessed Facebook group.

Complete our 2017 CleanTechnica Reader Survey — have your opinions, preferences, and deepest wishes heard.

Check out our 93-page EV report, based on over 2,000 surveys collected from EV drivers in 49 of 50 US states, 26 European countries, and 9 Canadian provinces.

Tags: , , , , , , ,

About the Author

is tryin' to help society help itself (and other species) with the power of the word. He spends most of his time here on CleanTechnica as its director and chief editor, but he's also the president of Important Media and the director/founder of EV Obsession and Solar Love. Zach is recognized globally as a solar energy, electric car, and energy storage expert. He has presented about cleantech at conferences in India, the UAE, Ukraine, Poland, Germany, the Netherlands, the USA, and Canada. Zach has long-term investments in TSLA, FSLR, SPWR, SEDG, & ABB — after years of covering solar and EVs, he simply has a lot of faith in these particular companies and feels like they are good cleantech companies to invest in. But he offers no professional investment advice and would rather not be responsible for you losing money, so don't jump to conclusions.

  • maxholland

    Thanks for this update article Zach. All we keen observers know that battery cost is key to overturning ICE vehicles (and thus ending the fossil fuel era), so we are all interested in this topic. I was also interested to run some numbers to see at what $/kwh various EVs reach sticker price parity with ICEs in an equivalent market segment (assuming for simplicity that the non-drivetrain features/quality/costs/price are identical) and without subsidies. I emphasize that given the much lower running costs and many other benefits of EVs, they don’t have to reach sticker price parity to become more attractive to the average buyer than ICEs, but nevertheless to understand the pattern might be interesting. There are lots of variables, so I have assumed that a 200 mile range 50kwh battery is seen as ‘minimum comfortable’ in mid-market segments ($25k to $50K), as Telsa and GM figure and many commentators suggest. I am focussing in on mainly these (arbitrary) price points, but they do straddle the ‘average’ US new car sticker price (around $33k). More on my assumptions below.
    At 200 mile range, an EV is at ‘sticker price parity’ with an ICE at the following $/kwh:

    At anything > $75k price point, parity is already reached (at $200/kwh)
    At the $50k price point, parity is at $150/kwh
    At the $33k price point, parity is at $100/kwh
    At the $25k price point, parity is at $75/kwh
    (At the $20k price point, parity is at $60/kwh)

    Again – this is only sticker price parity – the mass-market value proposition of course comes in well before parity is reached due to big savings on running costs and many other EV benefits! But even for a Luddite who thinks that the sticker price of an EV should be ‘no higher’ than its ICE equivalent, and is satisfied by a 200 mile range, these are the numbers.

    My assumptions are based around a 2014 academic study (Kochhan et al. ‘An Overview of Costs for Vehicle Components’) suggesting that the non-battery cost of electric vehicles are around 90-92% of an equivalent ICE vehicle. They figured that the 76-78% of non-drive-train costs are similar, but the EV drive train (excluding battery) might make up just 14% of the remaining cost (compared to the ICE drive train costing 22-24% of an ICE vehicle). If this is correct (or was correct in 2013-2014 when the research was done) it leaves 8-10% of the remaining vehicle cost for the battery. In the longer run I assume this 8-10% will be conservative because in EVs, inverters, motors and other electrical components can get significantly cheaper with volume and technology learning, relative to 2013-14 costs. Non-drivetrain costs could also turn out to be cheaper relative to ICE vehicles, e.g. vibration tolerances can be lower, engine noise insulation and heat insulation lowered, electrical auxiliary systems relatively less costly (amongst other things, please suggest ideas). So to give a baseline to the variables, I assume that as volume increases, total non-battery vehicle costs might get down to 85% of the equivalent ICE vehicle costs, which leaves us 15% margin to absorb the battery cost. If anyone wants to suggest an alternative figure, that would be welcome. The above parity figures already assume this 85% figure.

    What is interesting for me is that the figures suggest that at $100/kwh, everyone happy with 200 miles range who is buying a new vehicle around the $33k price point (or above) will automatically choose an EV because it’s sticker is going to be cheaper. Musk is on to something! 😉 Given the running cost savings and other benefits, likely this will hold at lower price points also (likely anything above $20k-$25k).

    For folks who are happy with an urban-only vehicle with 100 mile range (and rent something else for occasional longer trips) the sticker parity comes earlier/at lower price points (I assume a 25kwh battery):

    At anything > $33,333 price parity is already reached (at $200/kwh)
    At the $25k price point, parity is at $150/kwh
    At the $20k price point, parity is at $120/kwh
    At the $15k price point, parity is at $90/kwh

    As I’ve mentioned – I am not claiming price parity is necessary for an average buyer to choose an EV over an ICE. There are lots of other savings and benefits to EVs, so this is just a baseline to work from. I’d appreciate others suggesting how these savings might add up to make the e.g. the 5 year ownership proposition at 10k or 15k miles a year an equivalent (or better) overall proposition relative to ICEs in different segments. Regulatory factors may well overtake these considerations anyway, as Netherlands and Norway already plan to outlaw new ICE sales for passenger vehicles after 2025. We will likely be at $100/kwh (or better) by then, so the market above $33,000 (or likely 20-25k given other savings & benefits) will effectively already have ‘banned’ ICEs. So it is within reason. The value proposition for EVs is higher in Europe anyway due to significant fuel duty, and generally less long-distance road trips etc. Carbon taxes will change the boundaries for ICEs also in favour of EVs. And of course fully autonomous vehicles will change everything completely, since most city dwellers likely won’t bother with car ownership at all.

    I welcome any feedback, cheers.

    • Bob_Wallace

      I don’t see any problems with your numbers but would like to suggest that purchase price parity is not as important as monthly out of pocket costs during payoff.

      Let’s assume a six year loan at 4.5%, 13,000 miles per year, $0.12/kWh electricity and $3/gallon fuel. Here’s a really rough monthly cost to own and operate….

      A $35k Tesla Mod3 with no subsidy = a $28k ICEV

      A $35k Mod3 with $7,500 subsidy = a $22k ICEV

      • maxholland

        Thank you Bob for filling in some numbers for the rough 6 year cost around this price point, making the EV proposition even more compelling. For a $33k segment EV to be the same longer term value as a $28k segment ICEV even without subsidy is great for consumers; there’s definitely going to be a word-of-mouth (and media) effect to spread this ‘good value’ message once the Model 3 and Bolt start getting into circulation – especially since they will both have sports car performance (more-so the Telsa). One other conclusion from my looking at the numbers, and the upper and middle price brackets favouring EVs first, is that ICEVs will increasingly be associated with lower price brackets and have a negative connotation, further boosting the image/status of EVs. Roll on the EV revolution! Cheers

        • Bob_Wallace

          ” ICEVs will increasingly be associated with lower price brackets and have a negative connotation”

          That’s a factor that I don’t know how to monetize, but ICEVs are likely to be seen as ‘old tech’ and will lose cache.

          “What? You’re still using a crt monitor gasmobile? You stuck in the past?”

  • Bob_Wallace

    I’m not buying the $215 number. The rule of thumb is 20% to 30% extra to convert cells into packs. We know GM is paying $145 for cells and 30% would take this to just under $190.

    Then there’s this –

    ” the raw materials cost for today’s lithium-ion cell chemistries is $160 per kwh if you assume a conventional 40-percent supplier markup”

    Someone has done a materials cost for lithium ion batteries and come in at just under $70/kWh using market prices for all the materials. Market prices include the supplier markup.

    How could the materials cost by $170/kWh when LG Chem is selling cells to GM for $145/kWh?

    I’d suggest not believing anything Bereisa said. Remember, he stated that the Mod 3 would have an all aluminum body when Tesla had already said that it would not.

  • vensonata .

    See the lecture on youtube by Argonne Labs battery project director George Crabtree. The aim by 2019 for car batteries is 400wh/kg and cost $100 kwh. That would make the Tesla model battery weigh 250 lbs at 45 kwh and cost $4500. 3 years to go in the Argonne deadline with the DOE.

  • Thanks. Yes, got it.

  • eveee

    This subject has been dealt with elsewhere. Both LG and Tesla, the two largest battery suppliers, have secured sources of materials anticipating the need in advance. The losers will be the third parties trying to do so now, after the demand has already increased.
    Graphite is a particular case.
    A discussion of flake graphite.
    Teslas response is to use local materials to r Duce costs and pollution.

    “It turns out Tesla is going to stay at home and source everything from North America, and the reason isn’t cost (as it is quite likely more expensive to acquire resources abroad), but rather the environment.

    “It will enable us to establish a supply chain that is local and focused on minimizing environmental impact while significantly reducing battery cost,” Tesla spokesperson Liz Jarvis-Shean said in a statement to Bloomberg.

    The topic “du jour” of late has been graphite mining around the world, a process which, if not done properly, puts out massive amount of pollution; so much so that countries like China are now actively shuttering many of their mines.

    Tesla’s answer at this moment in acquiring graphite for their lithium batteries is to source the material in its synthetic variant from Japan and Europe.

    Bloomberg talked to Sam Jaffe, an analyst at Navigant Research, and he predicts that the graphite for Tesla’s Gigafactory will most likely come from planned operations in Canada. If more is required than Canada can supply, resources in Idaho and Minnesota could also be tapped.”

    • Graphite Gus

      evee, a couple of key points from the link you included are: – synthetic graphite is $20,000/ton vs. natural @ $10,000 a ton.
      – a model S has 54 kg. of graphite for a 70 kwh battery. So Elon may be using synthetic for the first run of 500K model 3s, but he will have to go to natural very soon. He will be imposing a huge cost differential on a product – model 3 – that needs to compete on price. I repeat my point that there are no functioning mines, (the one in quebec suspended operations) and no upgrade facilities in north america. And they will take years to come on stream

      • Bob_Wallace

        907 kg in a ton. $11 per kg @ $10,000 a ton. $22 per kg @ $20,000 a ton.

        54 kg @ $11 = $594. At $22 per kg around $1,200.

        $594 / 70 kWh = $8/kWh. $16/kWh for synthetic.

        Tesla was paying Panasonic $180/kWh for cells in 2014. 4% to 8% of the cell cost was graphite.

        Someone did a materials analysis and tabulated about $70/kWh material cost. That would make graphite 11%/22% of material cost. ($70 may have assumed the lower priced graphite.)

        That look right?

      • eveee

        Tesla has announced its intentions. Do you think they would publicize their sources? I think not. The fact that you do not know them is not evidence that they do not exist.
        Then there is the possibility that new methods of creating synthetic graphite might be discovered lowering the cost differential between synthetic and flake graphite.

        But on the other hand, the cost differential of $430/vehicle in a Model S with 85kwhr pack is not that great.

        “Using today’s prices for synthetic (~$20,000/t) and coated spheroidal natural graphite (~$10,000/t), all other things being equal, a switch from all-synthetic to all-natural-graphite anodes for those 500,000 EVs/year would save $216M in material costs, which translates to over $6/kWh, or over $430 per vehicle. Not a bad start.”

        On a vehicle like the Model 3, a pack of 48kwhr could do the job. The cost differential would be $243, a reason to change, but not a completely compelling one when transportation, logistics, and pollution are considered, and hardly a show stopper for a 35k automobile.

  • Bob_Wallace

    China is not the only source of lithium. Tesla is accessing lithium from Mexico and Nevada. There are many sources of lithium around the world.

    Graphite is found in almost every country. Plentiful reserves are found in Sri Lanka, Madagascar, Mexico, Germany, North & South Korea.

    • Graphite Gus

      You are right – graphite is found in many countries. But there is only one functioning mine in North America (quebec), and there are no facilities for upgrading the output to battery grade spherical graphite. The facility is a $200M investment with a 2-3 year timeline. So it looks like Tesla will have to join the queue in China – unless they can jump it by paying a premium over the massive Chinese demand.
      That link I gave is a good website with all the ins and outs of battery precursors.

  • NRG4All

    As we all know, the delivery of Model 3s will include the batteries coming from the Gigafactory. Perhaps someone out there has a good idea as to the cost per kWh of the pack which could make them more affordable for long distance travel. My question is, what might the limit be (probably volume limited) for the amount of kWh that a Model 3 could hold? With about a 4 mi./kWh rate of consumption, could we see a 100 kWh battery give a range of 400 miles for something less than $90,000?

    • Bob_Wallace

      Why would anyone need a 400 mile range?

      But let’s assume someone does. My working number for Panasonic cells (Tesla’s price) is $130/kWh when the 3 starts shipping. That’s based on a statement by Sam Jaffe of Navigant Research in October, 2014 that Tesla was paying Panasonic $180/kWh then. And since then I’ve repeatedly seen an estimation that cell prices will drop 30% with the Gigafactory running. So somewhere below $130. At 20% to convert cells into packs, so $155.

      Assuming a 50 kWh ‘entry level’ pack the other 50 kWh would cost $7,750.

      Add $7,750 to a $37,500 EV and you’ve got $45,250. That doesn’t include profit on the extra batteries and any bells and whistles one might have to purchase along with the range boost.

      Actually, at some point I think we might have 500 mile range EVs. I think Elon has talked about a 400 mile range Mod S by 2020. And the Destination chargers are sized to pump in a bit over 500 miles in a ten hour “overnight” charge.

      Some people might pay for that. But for most people a ‘solid 200 miles’ is plenty.

      • NRG4All

        I like your math. To think we could get a 400 mi. range in a vehicle costing $45,250 is encouraging. The only reason I mention 400 miles is that some people will resist EVs until they can match ICE vehicles. I have a Mini Cooper with a 13 gallon tank that will go 390 miles on a tank full. It is 10 years old this year and has all the goodies on it (convertible, John Cooper Works treatment, etc.) Ten years ago it cost within $1,000 of your estimate. Of course unless you’re a car nut like me, $45K is out of most peoples willingness to spend on a car so the price needs to come down further. When it can get 400 miles and be priced at about the average for ICE cars (somewhere around $32K-$33K) the transformation may turn into a stampede, if the 200 mile Model 3 is any indication.

        • Bob_Wallace

          I’m sure there were people who were reluctant to move to cars because cars didn’t give them poop to fertilize their gardens…. ;o)

          With your Mini Cooper you seldom get up in the morning to find that your tank has filled itself up as you slept. Once people understand how getting up to find they have a 200 mile range waiting for them every morning the desire for a 400 mile range will drop. Then drop further when they understand how really fast charge stations add in the miles.

          Tesla Superchargers now provide 125 kW and some 150 kWh. Tesla is working on 200 kW chargers. Up to 170 miles in 30 minutes with a 125 kW charger. About 40% faster (< 20 minutes) with the larger charger. Just enough time to get to the loo and back.

          A $35k 200 mile EV costs the same to own and operate over a six year payoff period as a $28k ICEV.

          At $30k for a 200 mile EV the market should break wide open. A $30k 200 mile EV would cost the same as a $25,500 ICEV to own and operate over the first six years. (And cheaper after the loan is paid off.)

          • NRG4All

            I perhaps am a really special case. We live about 200 miles from the largest town (Phoenix). We are up in the mountains with about 5,500 ft. higher elevation. I’m sure a 200 mile EV could get us down to Phoenix, but the problem is in coming back up. I’ve put in a request to Tesla to install a Supercharger in Payson which is about the half way point. A rep even said that they would contact me, but so far nothing. If there were a Supercharger there, a 200 mile EV may be okay because it is about a 3-1/2 to 4 hour drive and we usually stop in Payson going either way. I hate using the Mini and so it sits in the garage on a battery tender for 90% of the mileage we do. I don’t know which will get here first, the charging infrastructure or the EV range.

  • OneHundredbyFifty

    This post has a log plot on it for those who are asking. Note, the new Tesla $190 figure has not yet been added.

  • Freddy D

    Would love to see similar coverage of utility scale batteries, such as flow batteries, thermal storage, or other.

    • OneHundredbyFifty

      Li-Ion is moving too fast. I think flow will go away just as GaAs did for semiconductor. Even though it was better for some things, the resources thrown at silicon kept it so far ahead on its learning curve that GaAs became irrelevant.

      • Mike Dill

        GaAs is still there for specific applications, as it costs more. I feel flow and ‘super-caps’ will retain a place, but agree that they will continue to be beat by Li-Ion due to scaling factors unless some other technology comes along to shake system.

        • Bob_Wallace

          Flows have the advantage of being able to store large amounts of energy in inexpensive tanks in the form of (possibly) inexpensive chemicals.

          They can serve for both short term and long term storage. Sort term, frequent cycling can pay for the hardware and staff, helping to lower the cost of long term storage. Same as pump-up hydro.

      • Kenneth Ferland

        I think GaAS was also fatally flawed by the limited production capacity of thouse rare materials, they would never be able to scale up to the production volumes that would have then warranted the research volume to make them competitive. Now Silicon is on the verge of being manufactured in thin and flexible forms so the one advantage of GaAS is disappearing.

  • Freddy D

    These charts would be much more useful with a logarithmic axis on the left. It’s very difficult to project and see what’s going on as prices go below $100.

    • OneHundredbyFifty
      • Bob_Wallace

        Right click the image to open it in a new tab.

        Copy the page address and paste it into your comment.

        Bob’s your uncle tutor….

        • Freddy D

          Nice – thanks Bob. With the Tesla announcement of sub-$180 or so for packs now, we’re well below the green band today. What’s a little strange is that BNEF seemed to show that it would take 1.5 decades to sell a cumulative million EVs? If I’m reading this right. According to CleanTechnica last september 1 million EVs had been sold

          • Bob_Wallace

            GM has already contracted for $145/kWh battery packs from LG Chem with delivery starting this year. My envelope math says that Tesla’s pack price will be $130/kWh or lower by this time next year when they loose the 3 on the world.

            What we’re seeing with the folks who predict stuff is that the world is changing faster than they predicted.

            The transition away from fossil fuels looks like it’s going to be faster than a scalded cat….

          • Freddy D

            The longer these vehicles command a tiny 1% market share, the more the cost structure advantage winds up like an ever more tightly coiled spring. Having said that, inportant to remember that 99% of the worlds auto drivers and 100% of the truck, bus, boat, and airplane drivers will burn oil today just like the last 50 years.

            When that coiled spring releases its cost advantage then there will be severe scale up challenges not unlike silicon for solar a few years back. That will work through.

          • Bobby

            Actually, that’s a misunderstanding. $145/kWh would be GM’s cell cost, not pack cost. Their pack cost is estimated to be $215/kWh, while Tesla is presumably under $190/kWh.

          • Bob_Wallace

            Sorry. I wrote “pack” when I should have used “cell”.

            Do you have a link for GM paying $215/kWh for packs? $145 + 30% pack cost = $189. $215 seems a bit high even allowing LG Chem some profit on manufacturing the pack.

            Tesla was paying $180 for cells a year and a half ago. There is the expectation that cell costs will drop 30% with the Gigafactory, $125/kWh. Pack assembly should be very efficient at the GF so let’s assume 20% and Tesla will make the packs, so their finished pack cost should be more in the range of $150/kWh.

        • neroden

          Someone misread that graph when they added the annotation. The green band is $130-$150/kwh.

  • Graphite Gus

    I saw a panel discussion between 3 battery manufacturers LG US, Sakti3 and someone else, and a key point stuck with me. There is a high BOM to COGS ratio for manufacture. i.e. the cost of the inputs are a big proportion of the cost of the output. (which is what you would expect in a product on the commodity path) So at some point we will be influenced by the cost of the raw materials.

    • I remember that discussion. Great one. Here it is:

    • Bob_Wallace

      I remember a cost of cells discussion where the cost of materials for a lithium-ion cell was found to be just under $70/kWh. The claim was that those cells could be marketed for $100/kWh and give the manufacturer a decent profit. (30% for manufacturing costs and profit.)

      The cost of materials is like to drop under $70 as material production increases. And an efficient factory like the Panasonic/Tesla Gigafactory is likely to take pack assembly below 20% of cell price. Battery packs in the $120/kWh range may be a possibility by 2020.

      Battery packs lower than $130/kWh pretty much mean that ICEVs are dead men walking….

      • Graphite Gus

        (this is all supposition) but I think as Li cells gain popularity, it will put pressure on suppliers. None of these materials – lithium, graphite, cobalt -are in short supply in the ground, but mines and refining capabilities take time – several years – to come on stream, and the demand ramp in the next couple of years looks quite steep, so I can see price spikes and shortages. We saw this with other commodities over the past decade as China demand exploded for copper, iron ore, etc.

        • Bob_Wallace

          Sure, there could easily be price noise during the ramp up to 90 million battery packs per year. But for this first round Panasonic/Tesla, LG Chem and BYD have given very public notice that they are going to up their needs and they did that a few years before they will start to consume at higher levels.

          It would have been very poor management if all those manufacturers, and any others that are ramping up, didn’t contract for supplies as soon as they pulled the trigger to start building more manufacturing capacity.

      • Kenneth Ferland

        Current Li-ion tech depends on metals like Cobalt and Nickel which are already produced in large quantities for steel production. Increased demand is likely to raise prices as this is not an mature mining industry with a learning curve. Also the mining of these minerals is a rather nasty process which we should be eager to end.

        Rather it is the transition to new cathode-anode techs which will be based on cheaper materials like carbon, sulfur and sodium which will drop the price of the batteries input materials. Lithium will remain in the mix but is already a minor part of the total cost.

        • Bob_Wallace

          not an immature mining industry?

          Any idea what the current scale of cobalt and nickel production is in relation to how much might be needed for battery manufacturing? How much demand might be freed up as China slows down?

          • Kenneth Ferland

            Oh yes, sorry, not immature.

            Cobalt production is ~110,000 metric tons per year, with the Democratic Republic of the Congo far and away the dominant producer with over half of all world production. Most Cobalt is produced as a byproduct of other mining such as Copper.

            It looks like production has been growing rapidly after about 1995 and Battery usage is already over half of world consumption, It appears that Cobalt prices peaked in 2007 (along with everything else) at $55/lb but is now down to $10/lb and still looks to be on a slow downtrend despite accelerated consumption.


            I suspect this is because of the byproduct nature of the metal, it’s relatively easy to add extraction of the metal to an existing mine which previously discarded the metal in slag, but we might face a ceiling once all such sources are being tapped. A slow down in world consumption of metals like copper would reduce Cobalt production as mines are shutdown.

    • eveee

      We don’t care much about the price per battery. Rather, we care more about the cost/kWhr. We can even increase the cost per battery if the net effect is lower cost/ kWhr.
      That’s precisely what has happened as EVs shifted from lead acid to Nickel metal hydride to lithium batteries.

  • JamesWimberley

    BNEF’s assertion that batteries for hybrids are inherently more expensive per kWh thatnthose for BEVs was news to me. It’s an important part of the case for a rapid transition to full electric. The ratio will presumably hold even as costs of all types if battery fall. A hybrid is also a more complex and expensive piece of machinery to make, and imposes design constraints that BEVs escape. A skateboard platform is as close as you can get to a blank piece of paper for a car body designer.

    • Yes, news to me too, even though I apparently inserted that quotes years ago. 😛

      • Eric Lukac-Kuruc

        By design, a hybrid is loading/depleting its battery – even if partly – many times during a single trip while a BEV does it only once. These different needs might explain the difference in cost per KWh.

        • Yes, it makes a lot of sense now that I think about it. Charging & discharging so much is quickly going to wear out the battery if it isn’t designed to be more robust (resulting in a higher $, presumably).

          • Mike Shurtleff

            Right, but hybrid battery can address much of this as costs for EVs and EREVs become more competitive. Volt 2.0 for example has 50 mile all-electric range, but would need only small portion of that battery for HEV mode over longer distance. Think Prius size battery for HEV mode. EREV (PHEV if you like) battery is much smaller than EV battery, so increase in battery cost $/kWh is not as big an impact on vehicle cost.

        • Perttu Lehtinen

          BEV’s still do regen.

          • Eric Lukac-Kuruc

            True. There’s still a difference in the percentage of the overall battery capacity affected by these charge/discharge cycles. The bigger the battery pack the less aggressive is the regen operation.

        • eveee

          Bingo. Read my comment above for explanation.

    • Bob_Wallace

      Let me bring out one of my favorite charts once again. The one that shows falling battery pack prices (horizontal axis) kills off hybrids and PHEVs fairly early in the process….


      • Looks like we need an image sharing tutor here 😀

        But I’m guessing this is the image that I heavily use in all my EV presentations thanks to you 😀

        • Bob_Wallace

          I started writing one but am suffering from massive overload at the moment. Need to get off the keyboard now and out to mow the orchard and make sure the drip system is working. We’re entering the dry season.

      • neroden

        I’d like to see their source data for that, because I think the McKinsey/EIA chart there is way too pessimistic. We have $190/kwh batteries; this would predict that gas would need to be over $2.75 to make electric vehicles competitive, but it’s clearly wrong; BEVs are clearly competitive at lower gas prices than that.

        • Bob_Wallace

          Those are battery pack prices, not cell prices. Just in case that wasn’t clear.

    • Mike Shurtleff

      EREV (extended-range electric vehicle) can be built far less expensively than other PHEVs. Use internal combustion engine only for generator and control power electrically without complex transmission.
      Wonder if we’ll every see these with battery tech improving so fast? Straight to EV only domination?

      • Bob_Wallace

        I suspect we’ll see PHEV cars die out fairly soon. We may see PHEV pickups appear and last longer due to higher energy needs when fully loaded/towing.

        Batteries shouldn’t have to even double in capacity to kill off ICEs and even jet engines. (265 x 2 = 530 W/kg. “According to Tesla’s Elon Musk, the concept of battery-powered transcontinental airplanes becomes “compelling” …”)

        Historically batteries have been gaining capacity at about 7% per year. If continued that would mean a doubling in about a decade.

      • eveee

        Straight to EV domination. With the model 3, the price differential between a Volt has been largely eroded. As battery costs drop, EVs benefit more. That’s the flip side of PHEVs not depending on battery costs. PHEV greater complexity costs cannot be reduced as much.

    • eveee

      This is slightly misleading and needs discussion. HYbrids require high cycle life than BEVs, because they cycle much more. The Volt achieved it by reduced Depth of Discharge only using half its capacity. It also uses a bigger capacity battery than the Prius which uses a high power/energy ratio battery with lower energy capacity. That’s the same as high C rate, batteries that are more expensive per unit energy stored. Both those issues contribute to higher per kWhr costs for hybrids.
      In general, the trade off is lower C for higher energy capacity. Depth of Discharge is traded for longer cycle life.

  • nakedChimp

    Can you do a logarithmic chart (base 10, price axis) with those prices?
    I just did from the above chart for 3 data points that I could read from the graph easily (2006:$1,300; 2010:$750; 2014:$400) and it’s a straight line obviously.
    But having all the other values on a graph like that would be pretty telling, no?

    Anyhow.. taking those numbers and extrapolating for $100/kWh for a pack one reaches 2022..2024.
    (2018:$220; 2022:$120)

    • If someone here has access to the paper, maybe they could drop the numbers here for us to do so?

      Otherwise, we can just purchase it here for $32 I guess … and hope the numbers are in there.

      • nakedChimp

        Sorry, the data is not in there 🙂
        It’s a 4 page pdf and I can make you a high res screenshot of that graph (which isn’t needed seeing yours), but that’s it.
        There is no further link to anything relevant I think.. just a more direct email for Bjorn Nykvist vs whats available on the overview page for the letter. So one probably has to contact him to get anything there.

      • voracity

        Since my previous post probably went missing due to it containing a link (why does that keep happening?), you can search for nclimate2564-s1.xlsx for the source data on Google. (It’s on the Nature site.)

Back to Top ↑