Study: Big Battery Factories “Won’t Do Much To Cut Cost”


Tesla Gigafactory

Tesla Gigafactory

Carnegie Mellon University joins the group of institutions/experts that question the meaningfulness of large battery factories as a way to reduce battery costs.

Four researchers – Professor of Engineering and Public Policy and Mechanical Engineering Jeremy Michalek, Associate Professor of Engineering and Public Policy and Materials Science Jay Whitacre, and Associate Professor of Engineering and Public Policy Erica Fuchs, together with Apurba Sakti, a postdoctoral research associate at MIT, analyzed the design and production of vehicle batteries in a study appearing in the Journal of Power Sources.

The results of their work in general stand in opposition to Tesla Motors CEO, who assures us that larger scale will bring costs down.

The press release title is striking “Big Factories Won’t Solve High Cost of Electric Vehicles“.

“Carnegie Mellon University researchers have found that the cost savings associated with manufacturing a high volume of batteries for electric vehicles may be nearly exhausted. Mass production lowers cost, say the researchers โ€” but only up to a point.”

This point is level of 200-300 MWh of annual production, which is equivalent to just 8,000+ to 12,500 Nissan LEAFs a year! This would mean that all major manufacturers like Nissan/NEC, LG Chem, Panasonic exceeded this mark long ago.

“Economies of scale are the cost savings that arise from spreading the cost of expensive equipment, facilities and other investments over a large number of units produced, reducing the cost for each unit.”

Professor of Engineering and Public Policy and Mechanical Engineering Jeremy Michalek stated:

“Electric vehicle batteries are expensive. Federal and state governments have been subsidizing and mandating electric vehicle sales for years with the idea that increasing production volume will reduce costs and make these vehicles viable for mainstream consumers.”

“But we found that battery economies of scale are exhausted quickly, at around 200-300 MWh of annual production. That’s comparable to the amount of batteries produced for the Nissan Leaf or the Chevy Volt last year. Past this point, higher volume alone won’t do much to cut cost.”

Associate Professor of Engineering and Public Policy and Materials Science Jay Whitacre commented:

“Our results raise questions about whether increasing vehicle sales is the best way to continue to spend limited resources โ€” as opposed to, say, more research on battery technology. For example, we estimate that finding a way to make batteries with thicker electrodes could lower the cost of long-range electric vehicle batteries by as much as 8 percent, while increasing production beyond current levels may only cut costs by less than 3 percent.”

Well, this does not sound too good for Tesla Motors’ Gigafactory; and if high volume production will not bring costs down (or perhaps just a little bit), then to achieve one third lower costs, Tesla and Panasonic must have something new to offer on the cell chemistry level or rather a combination of several different factors.

Michalek adds:

“At the end of the day, economics will determine the degree to which electric vehicles are adopted by mainstream consumers. Battery cost is the single largest economic barrier for mainstream adoption of electric vehicles, and large factories alone aren’t likely to solve the battery cost problem.”

Source: Carnegie Mellon

Category: General

Tags: , ,

82 responses to "Study: Big Battery Factories “Won’t Do Much To Cut Cost”"
  1. Bryan says:

    This from an organization that has never built a thing and stands to benefit from more research.

    1. DaveMart says:

      You don’t like researchers because they have not built anything and don’t agree with Elon, and at a guess you don’t like other car manufacturers who have built millions of cars more than Elon and don’t agree with him.

      It seems relatively safe to say that you ain’t too keen on anybody disagreeing with Elon whatever their experience or lack of it! ๐Ÿ˜‰

      1. Rick Danger says:

        More arrogant presumption on your part. Quelle suprise.

        1. See Through says:

          Agreed. The ‘economy of scale’ BS from Tesla was only to raise $2 B. They burnt $300M of that just in Q3. So what’s left?

          You guys talk like Elon has been building battery factories his entire career at paypal and Ebay. At least, these are 4 professors from a reputed university, showing their results with some analysis to back it up. What do you guys have to prove Elon’s thesis? 5 slides and a snap of Nevada desert?

          1. Mike says:

            Why don’t you read this more carefully.
            1) They are not saying the gigafactory isn’t needed, they’re just saying it won’t bring down cost. Tesla has to have a large factory to build the number of batteries they need, don’t matter if Size ALONE doesn’t bring down cost. They need the Quantity they need.
            2) Tesla & Panasonic never said they’re FREEZING their battery tech to 2014 levels, and Stopping. They’re R&D efforts are still ongoing.
            3) Panasonic has delivered a track record of 20% capacity increases a year, with 20% price decreases.

            Must be a slow news day.

          2. LuStuccc says:

            The only one I see here talking BS day after day is you.

      2. LuStuccc says:

        Elon Musk, nov. 5 : “We felt comfortable with at least a 30 percent reduction in cost just based on the location and economies of scale. Thatโ€™s without taking any technology improvement into account, and weโ€™ll certainly do technology improvement. If we canโ€™t get to 30 percent without technology improvements, someone should shoot us, because that would be in complete defiance of economies of scale and obvious cost savings. “

          1. LuStuccc says:

            And CTO Straubel said earlier this summer that 50% will be more likely.

            1. reguest says:

              Can I have some links to those talks please?

              1. DaveMart says:

                Got to the Tesla com site, investors at the bottom right of the page, then events and presentations.
                It is probably somewhere in one of the talks there.
                Straubel tends to say all sorts of things though, and who knows, some of them may even pan out! ๐Ÿ˜‰

              2. Lustuccc says:


                Take some time to listen to what he has to say, he is way more credible than DaveSmart ๐Ÿ˜‰


                1. DaveMart says:

                  I’ve heard it, thanks.
                  I usually prefer to assess technologies using other sources than a company’s technical officer.

                  Don’t bother to thank me for digging out the link for you, as that takes manners.

                  1. Lustuccc says:

                    It is NOT on Tesla’s site. I usually do not bother to thank pedant liars.

                    1. Boris says:

                      Oh, get a room you too…:)

                    2. Boris says:

                      Two not too

      3. Bryan Whitton says:

        Actually I don’t own a Tesla, Wish I could afford one but I do lease a Honda Fit EV. I just recently sold my Dodge RAM 1500 and our ICE car is an Acura CL, about 12 years old.
        I just thought it would be funny. I don’t take anything too seriously these days. Everyone is so polarized you can’t really have a decent discussion anymore.

        1. Robster says:


          1. danpatgal says:

            Using a factorial (the “!”) didn’t really help your โ€œ1โ€ ;). I like something like 2^100. Which is probably an exaggeration of how much I agree with the sentiment that we need to have less polarization and more dialogue. The internet seems to allow, perhaps even encourage, this kind separation.

            Anyway, back on topic, I think that the economies of scale the Gigafactory will deliver is not necessarily for production costs (which seems to be the focus of the research), but for reducing transportation costs. I also wonder if there would not be some effort to build Tesla modules, perhaps even a different form of cell to replace putting 7000 tiny cells, and just make it 250 instead, to reduce the costs.

            So reducing transportation and packaging costs, at scale, is a great justification for the Gigafactory. When they build the Model III it will drive a more efficient (lighter, more aero, and better drive-trian) car to go 200 miles (in good conditions), and deliver comfort, 5 seats, and with probably a smaller pack (40kwh/60kwh), to keep costs down (30-40k). If they could do that they’d have a lot of buyers … and the Gigafactory insures they will be able to supply those cars. Without it, and the additional battery capacity it will deliver, it makes no sense to even build the Model III.

        2. alain says:

          your right ,everyone is polarized.the cost has come down 800$ to 250, it’s just time to scale up !this debate is already over

      4. Mint says:

        As much as you don’t like it, it’s true. Manufacturing cost is notoriously hard to predict.

        I met with an MIT professor who was at the bleeding edge of displays, and he told me that he always thought LCD was a crap technology (transmissive, complicated structure, lots of light absorption in polarizers and filters), but was astounded at how far industry refined it over the years and got cost down.

        DLP is a similar case: How could anyone predict a >1 m^2 micro-patterned LCD or plasma panel (originally $5000+) could compete with a 1-inch DLP chip in cost? Yet somehow it happened.

        Unless you are actually in the business of building batteries, you really don’t know much about manufacturing cost.

    2. DonC says:

      Exactly. Imagine thinking that a CEO whose entire wealth is dependent on his company’s stock price would ever pile on the BS! Obviously a bunch of academics are far more likely to be biased.

      1. See Through says:

        When that CEO needs $2B of cheap cash, and burns $300M a quarter, anything is possible.

        1. Whatever says:

          You seem upset with Elon Musk actually spending the money he raised. What did you think he was going to do with it?

      2. Mike says:

        Musks statement doesn’t contradict this Mellon study, it agrees with it. Why don’t you not jump to conclusions.

    3. Tech01x says:

      I think it is very important for people to read the paper to understand what it really says and what it doesn’t. Given the inputs into this paper, what it tries to address, and very importantly the limitations that the authors acknowledge, this paper has almost nothing to do with Tesla’s Gigafactory.

      There are a lot of inputs from other academic papers, many of which are old and not relevant to Tesla’s solutions. Next, there are assumptions that just don’t hold in Tesla’s case. Just like a slew of FUD research associated with greenhouse gas emissions from battery production, studies like this one are misunderstood and misused.

  2. Rob Stark says:

    Those that can’t do teach.

    Genius is always surrounded by a Confederacy of Dunces.

    There where a long line of PhDs, battery scientist and assorted big wigs that said a vehicle such as the Model S was not economically viable just before the Model S was introduced. Battery technology was at least 15 years away. You can see some of them on Youtube.

    Elon is putting his money where his mouth is.

    1. liberty says:

      Looky here, the lead author of this

      Is the same as the lead author of this

      Of course many thought that study had some pretty bad assumptions, and we do find out that the public policy for larger battery sizes have significantly dropped the cost of batteries, far lower than that study thought about.

      Could it simply be that this follow on, is trying to advocate against big batteries? Like the last one? I don’t understand why study the tesla plant, either it will drop battery costs or it won’t. Arguing that it won’t drop costs is a prediction, and one founded on lack of knowledge of what tesla is actually doing differently.

      1. Mike says:

        Ouch. So this is really a Wall Street Short Funded effort. Shock.

        1. ydnas7 says:

          Its an academic effort, whose department gets Toyota and Ford funding.

          LEV50 cell (ie Mitsubishi) is a manganese based cel, not a NMC based cell

          nobody is modelling using high Co content cells in EV except CMU (and insideEVs love of NMC)

          which cathode has the least cost reduction going forward – yep its NMC 333

          if it from CMU, expect it to be biased against EVs

          1. ydnas7 says:

            what CMU is modelling, is that using NMC, then there is a meaning cost floor that is scaled to quite readily, they are also modelling quite high power cells, due to arbitrary thickness limits on cathode/anode thickness.

            its not representative of
            Tesla, Nissan, Mitsubishi, GM
            (ie Panasonic, AESC, GS Yuasa, LG, BYD)

            but it is representative of
            BMW and Kia

        2. liberty says:

          No its not wall street funded, and carnegie mellon is a fine institution, but this is political science based on a number of assumptions, and many of them in this case appear to be bad as they were in the first case. Toyota was a major funder in the first case. I don’t know who funded this one, but the authors may have carried their biases from the first study.

          Certainly some of the 30% tesla and panasonic are expecting have to do with Simply a better location with lower cost labor, electricity, and less transport of raw materials and cells. Without knowing the costs of tesla/panasonic and new we have no idea how much has to do with gigafactory or simply having a unified nevada factory that looks like it can supply tesla’s needs.

  3. DaveMart says:

    Actually, there are substantial caveats in the study:

    ‘ The study only considered the popular NMC-G chemistry, which is used either solely or in combination with other active material chemistries in the Ford C-Max Energi, BMW ActiveE, BMW i3, BMW i8, Mitsubishi i-MiEV, Volvo C30 EV, Honda Fit EV and Honda Accord, according to the team.

    Nor did the study explore the 18650 cell format that Tesla uses, opting instead for the prismatic format that everyone else is using, Michalek said. Although manufacturing cylindrical cells involves a few different steps, Michalek said he would expect economies of scale for these cells to be comparable. Indeed, with manufacturing for that format having already been cost-minimized for decades, there is likely less room for improvement in that format, he suggested.

    Although high volume alone may not provide the cost savings Tesla is looking for from its Gigafactory, the company may get additional savings from other factors such as supply chain integration, he suggested. ‘

    If I were Elon I would be rather more concerned with the fall of the yen against the dollar, and since Japan is easing massively to the tune of hundreds of billions of dollars it is likely to stay fallen.

    That means that sourcing in the US instead of Japan may yield much less savings than they had reckoned on, or cost money instead.

    it pays to put a gigafactory in the right place.

    It is pretty tough if your all your production is based in an expensive location.

    1. Rob Stark says:

      Currencies fall and rise.

      Tesla will still buy 15 GW of Japanese batteries from Panasonic even after the GF is built. Currently the buy 4 GW and next year it will be ~ 7 GW.

      Nevada is not an expensive place to manufacture particularly when all the tax abatements are considered.

      Elon has stated than Panasonic,Samsung SDI, LG Chem et al use generic equipment in manufacturing cylindrical cells. But at the level of the GF you can customize the machines which will result in higher efficiency.

      Not only logistics but purchasing raw materials. Going directly to mining companies and purchasing on long term contracts.No 20 years but more like 5 years. Unlike selling on the spot market it allows the Mining companies to plan better, make more efficient decisions and pass part of the savings to the GF.

      1. Rick Danger says:

        Rob, no point in explaining it to DaveMart – he knows everything, and Elon is some kind of mutant aberration who will be erased from relevance in short order once Mighty Toyota unleashes the Fuel Cell Flood.

        Still laughing at YOU Daaaaaaaaaave.

        1. DaveMart says:

          Thank you for your contribution, Rick, as witty as always.
          I wonder how you find the time to research all the useful information in your posts.

      2. DaveMart says:

        A lot of the savings you suggest are covered by the ‘other savings’ such as supply chain logistics which the authors themselves highlight – they are not in fact saying that savings can’t be had at all.

        I dunno, myself, but it hardly seem fair as some are doing to try to dismiss the study simply on the grounds that they are academics.

        Elon has been known to be wrong before about costs, judging by the mooted prices of his cars and what the average sales prices turned out to be.

        That is not to say that there is not some fine engineering in the Tesla, but Elon is perhaps above all a promoter, and talking big and optimistically is part of that.

        Of course, he also had great achievements to back that up, but to me at least it hardly seems wise to take every word he says as gospel, as there are a heck of a lot of them and some of them work out and some of them don’t.

        1. SIvad says:

          Everyone can just keep debating this for a few more years until I go pick up my affordable Model 3 and drive off.

          15 years from now the detractors will be debating how even though EVs cost the same and have the same range as an ICE will never be viable because they take 6 1/2 minutes to charge rather than 5 minutes.

          1. DaveMart says:

            However true or untrue your assumptions of the BEV markets progress may be, they have precious little to do with the subject at hand, the effects of scaling on battery economics, as not only as I note above is the study not about the batteries used in the Tesla, but as they themselves say technological progress is far more important.

            1. SIvad says:

              You keep pointing out to everyone that this study isn’t specifically about Tesla but battery manufacturing in general. Well that is certainly the flaw then of this study.

              I think the example of the Nevada factory is not only relevant but the outcome of it’s level of success absolutely determines the accuracy of this study. Their study is based on what they know of current battery production and how it would scale. I doubt they had access to all the detailed information on the equipment, construction and operation of the Nevada plant.

              If the Nevada plan cuts costs even remotely closer to the ballpark of 30% like Musk claims rather than the the “less than 3%” that the study claims “increasing production beyond current levels” could achieve then you can guarantee the other battery producers will follow suit leaving this study irrelevant.

              Of course technology like what they say in the study of “finding a way to make batteries with thicker electrodes could lower the cost of long-range electric vehicle batteries by as much as 8 percent” would be welcome and necessary to move forward but that can be implemented into existing larger scale battery plants like the Nevada plant.

              In fact their entire conclusion to their study states that: “Our results raise questions about whether increasing vehicle sales is the best way to continue to spend limited resources โ€” as opposed to, say, more research on battery technology,” “For example, we estimate that finding a way to make batteries with thicker electrodes could lower the cost of long-range electric vehicle batteries by as much as 8 percent, while increasing production beyond current levels may only cut costs by less than 3 percent.”

              So there is their argument distilled down. Continued research supposedly will pay off more (e.g. 8%)than scaling production (less than 3%). Well again if the Nevada plant can produce current battery tech for <8%, especially if it is closer to 30% then that would become the standard for battery production and this study is garbage.

              Whether that is true or not we'll all have to wait and see but I think the flaw in their study was making huge assumptions that there are technological advancements only in battery architecture but not in battery production as well when you scale.

              For example, did they include the plant lifetime cost of energy used for the claimed efficiency limit of 200-300MWh of production vs. 50GWh of production if both are using on-site renewables? Scaling up an on-site solar and wind farm to cover the cost of producing energy for a 50GWh plant will surely save over building a smaller renewable farm to cover the 300MWh of production and even more so for plants that use only grid electricity.

        2. Koz says:

          Do you think the high average selling price is a point of success or failure for Tesla?

    2. Lensman says:

      Thank you, Dave, for– belatedly– providing some light to a discussion mostly characterized by uninformed heat. Sometimes you surprise us and write a thoughtful, insightful post.

      Too bad you don’t do that more often. And too bad you didn’t start your discussion here that way.

      1. DaveMart says:

        I was replying to a post which read:
        ‘This from an organization that has never built a thing and stands to benefit from more research.’

        which was hardly a very liberal judgement on the study, and ad hominem rather than addressing in any way the substance of the report.

        In fact I meant my reply to be fairly light hearted, but in print is is sometimes difficult to convey tone, and clearly I missed it.

        However I don’t think the post I replied to deserved much better, as I was hardly mocking a substantial critique.

    3. TomArt says:

      Yep. Locating relatively close to lithium sources with generous tax incentives and ready access to freight rail will probably do as much or more in cost savings (and carbon footprint reduction) than mass-production. And, those advantages would be compounded further by the level of vertical integration involved in the gigafactory. Raw materials and old packs go in, and new packs come out. After the initial capital investment, it stands to reason that costs would be lower compared to separate facilities, since almost everything with respect to the packs is being done in the same location.

  4. mike w says:

    I think Elon Musk is in a better position to talk about Tesla and its gigafactory than these goobers.

    1. DaveMart says:

      They did not research Tesla.
      They researched NMC batteries, and what they found had only coincidental applicability to Tesla.

      1. mike w says:

        The article calls out TESLA by name several times and questions Tesla’s ability to reduce battery prices. The author also show “pictures” of the proposed Gigafactory. If they were not talking about Tesla then who are they talking about?

        1. DaveMart says:

          Fair comment.
          The study itself was clearly on NMC batteries though, and the comment about Tesla was an addenda.

          In my view it seems likely that the author tried to deflect journalistic attempts to link them by making a saying a bit about it himself, but was unwise as that simply made it more ‘about’ Tesla.

          Any relevance to the Tesla factory remains limited however, and there are more than economies of scale involved.

    2. Bret says:

      +1 Mike

      I trust Tesla’s predictions much more than the academics. Failed companies like LTCM are a prime example of theory vs. reality.

      I have worked for two Japanese companies and it is a very expensive place to manufacture, not to mention ship product from. An obvious example of the potential cost savings is the Japanese LEAF vs. the LEAF built in Tennessee.

      In addition, as others have already mentioned, the domestic supply chain of raw materials and rail proximity to the Tesla factory should yield big savings.

      Finally, a new cell format, larger than the 18650, could further reduce the cost, by reducing the number of cells.

      A 30-50% cost savings seems pretty reasonable and the trend is 8% per year in cost reductions, even if they do nothing.

      1. Tech01x says:

        I agree… Tesla is working with Panasonic who already operates one of the largest battery plants in the world which is currently producing cylindrical NCA cells for the Model S in Osaka, Japan. Further, Tesla has battery experts on staff and has far more visibility into actual costs, including the costs of building that factory in Osaka.

        I think some people misunderstand the Gigafactory situation. Some of the cost reduction is economies of scale – maybe even the largest reduction. However, just not having to ship raw materials to Japan, then ship completed cells to the U.S. is a big win for the Gigafactory. Further, the current Panasonic factory is not fully vertically integrated – the Gigafactory is going to have anode, cathode, electrolyte, etc. production collocated with the cell production which is collocated with the battery pack production. The economies of scale are not just in the actual cell production, but in securing the supply chain including the raw materials directly from the miners. Tesla is able to command supply chain prices drops just due to consistent demand at a high volume, much like the way Apple secured NAND flash during the iPod growth heyday. Further, Tesla is making a bigger cylindrical cell and tweaking the cell chemistry, possibly moving to a silicon anode. Finally, the Gigafactory has to be sized big enough to max out the auto making capacity of the Fremont factory.

        1. DaveMart says:

          Just so.
          And the actual report does not claim much different.
          It is some of the commentary on the report which assumes that it is a definitive statement on the cost at the Nevada factory, which it is not and the study does not deal with that.

  5. If economics of scale for batteries has been reached already, then why do smartphone, tablets and other portable devices keep getting thinner, batteries last longer, all the while keeping the same price points?

    hint: The quantity per year (GWh) of batteries used for portable electronics devices is an order of magnitude larger than the capacity used by PEVs.

  6. BravelilToaster says:

    I seem to recall that at the volumes Tesla expects to sell the Model 3, there’s no way that LG could ever supply their batteries *unless* they built the gigafactory.

    So the existence of the gigafactory isn’t only to drive down the cost of batteries, but to simply supply batteries to Tesla at the rate they expect to consume them.

    Plus, they also benefit from the fact that those batteries don’t need to be shipped from Korea. Musk did mention something about location being one of the big cost savers there.

    1. DaveMart says:

      No doubt LG Chem would love to get the contract, but at the moment it is Panasonic supplying the cells, not LG, and from Japan, not Korea! ๐Ÿ˜‰

    2. pete g says:

      LG does have a Gigafactory. They just use modular approach and keep the last few stages of production close to the auto assembly plants where they are used.

    3. pete g says:

      The main question this study answers is how big can a factory get before its logistics become a cluster f…

      An other study done over 100 years ago stated no city can grow to have over 1 million residents because the people will be walking around in 3 feet of horse manure.

  7. Ryan says:

    Those who can’t do, teach… and seek funding from sources invested in a particular outcome.

    1. Bob A says:

      I don’t know if the analysts and authors have an agenda. The facts and math they use to support their conclusions deserves fair scrutiny as it would get in any refereed journal.

      But say that they are right. It is an interesting result. You could use it to decide that your minimum battery factory size should be no less than 300MWhr/yr, and if the local raw/refined material supplier cannot provide that, find a better location. You can use it to decide how far you want to ship completed batteries in the overall cost functional for your business optimization. I am sure there are others who could come up with better uses for the information.

      A lot of the battery research that is probably going to come up with the next delta in battery performance (pick your metric) is gonna be done by those who teach. When I went to college here were many outstanding professors. There were a few loser professors, too (i.e. lousy teachers), but those who can’t get their stuff published in refereed journals usually don’t last long.

      1. DaveMart says:

        This sort of study is done all the time, to determine the correlation of scaling to costs.
        It just does not usually attract such an emotional response.

        They likely studied NMC batteries as that is what most are building, and so they could offer a resource for the engineers trying to determine the optimum size to build.

        Its pretty much like a thousand other studies academics routinely perform and which provide useful source material, for instance on the strength of alloys under compression.

        1. Marshal G says:

          “Its pretty much like a thousand other studies academics routinely perform and which provide useful source material, for instance on the strength of alloys under compression”

          Only they’re not studying physics, they’re studying the publicly available information for a whole manufacturing sector. Nobody really knows what Tesla’s costs are but Tesla.

          1. DaveMart says:

            The study is not about Tesla, so I don’t understand your point.

            It is simply, rather too freely, being interpreted as a critique of Tesla.

            It ain’t.

            It is on a different subject, which has some very limited degree of applicability in considering the Tesla factory.

  8. Bob A says:

    Tesla will need the batteries for Model 3. Without them, they’ll never get to 500,00 cars. The economy of scale for vehicle production should be considered. Not just the batteries.

    Nevada has many dry lakes which apparently is a place you find lithium. So they are near the source. Good tax deals, good sources, (kinda sorta) close to the location of vehicle production.

    3% isn’t a lot to save for scaling the battery production up. But its better than not getting it.

    Big picture, it still looks like a good thing. Big picture, it may well be optimal.

  9. Spec9 says:

    It is not just mass manufacturing . . . it is the ability to strike deals directly with mining companies to get materials.

    If you are the car company building the batteries then going directly from the mining company and into the car cuts out a lot of middlemen.

    Vertical integration can work to cut costs. It is not high-tech, it is old 19th century business acumen.

    1. Koz says:


      Not just middlemen, there may not be much cost there but handling and shipping for sure. Shipping/handling of materials in and and packs out to NUMMI will be the only shipping/handling costs, probably a single digit percent of what those costs add to currently. Since Tesla and Panasonic are the ones bearing the costs in the $B’s and doing the negotiations with suppliers, I’ll trust their judgement here. Tesla is also the one on the hook to come through with the Model III at $35k.

  10. Lensman says:

    As others have already pointed out, in several ways, this article rather misses the point. DaveMart pointed out that the premises underlying the conclusions in the paper are rather different than the premises used to design the GigaFactory.

    And as several have pointed out, the main purpose of the GF isn’t to reduce cost; it’s to produce a dependable supply of batteries large enough for Tesla Motors to be able to sell BEVs in large numbers. This is an absolute necessity; -every- auto manufacturer which wants to make BEVs in large numbers, numbers approaching that of a best-selling gas guzzler, will have to do the same– either by building one giant battery factory or by building several large factories with an equal cumulative output.

    Is Tesla Motors right, in thinking the best way to save costs is to concentrate the manufacturing in one carefully chosen place, near a lithium mine? Or will it be better to have distributed manufacturing, spreading around the demand between various factories and multiple vendors?

    Only time will tell. A paper from three academics with no manufacturing experience certainly won’t settle the matter, no matter how well-educated they are.

  11. Jouni Valkonen says:

    What we have learned from solar panel manufacturing, conclusions are just opposite.

    Good thing is this study can be falsified in about three months, because the cost of batteries are getting down more than 3 % in just three months.

    1. Jouni Valkonen says:

      Also this study does not refer to Tesla Gigafactory in anyway and those researchers probably have not even heard Tesla’s plans for Gigafactory.

      It discusses more that they claim that artificially increased demand for electric cars does not help with the cost reduction.. This is somewhat correct because instead of innovating, traditional car companies just filled the markets with uninnovative compliance cars.

      Therefore this study does not apply to Tesla that is aiming for unsubsidized car markets and therefore is motivated to innovate.

      Also this is methologically broken, because this tries to predict future from the information of past.

      1. Thomas J. Thias says:



        Thomas J. Thias

  12. Unless you accept that we are at the end of the road on cost reduction this year, the premise of the study is flawed.

    Battery price reductions of ~8% per year have occurred for decades. Part of this in increased energy density from continuously improving chemistries and nanostructuring, part of it is manufacturing efficiencies, and part is materials cost. But add them all up and the price pressure has been relentlessly down. Who believes this will really stop within the next 5 years? Can I bet against you somewhere?

    1. DaveMart says:

      The authors note that technological improvements are far more important than economies of scale, so it seems that they agree with you, and have said so! ๐Ÿ˜‰

  13. Dave86 says:

    Hmmm… wonder if the Carnegie Mellon University researchers ever did a budget for an R&D department before.

    If the goal is to get a battery pack for a 350 mile EV under $5,000, then maybe 10% of that $5,000 goes towards R&D? And if you sell 10,000 cars? $500 * 10,000 = $5,000,000?

    $5,000,000 doesn’t buy a whole lot of R&D.

    Scaling up volume like Tesla is doing is important to so that overhead, like R&D, can be spread out over a large amount of product, resulting in small overhead cost per unit built.

  14. Omar Sultan says:


    I do not think its asking too much of you guys to do some research before writing an article for the site and not just re-posting someone’s talking points. You should not be depending on your readers to do due diligence and provide context.


  15. Thomas J. Thias says:

    My Turn,

    I hate to post so late in the thread cycle.

    Looking at the actual paper outtakes from Carnegie Mellon University, a stunning fact, in my opinion, emerges and that if our favorite Journalist, Mr. Kane had dug deep enough demands attention.

    Quoting Mr. Whitacker, Mr Kane said this:

    “Well, this does not sound too good for Tesla Motorsโ€™ Gigafactory; and if high volume production will not bring costs down (or perhaps just a little bit), then to achieve one third lower costs, Tesla and Panasonic must have something new to offer on the cell chemistry level or rather a combination of several different factors.”

    This was in responce to Mr. Whitacre quoted statement abrigded here-

    “~batteries with thicker electrodes could lower the cost of long-range electric vehicle batteries by as much as 8 percent~”

    8% reduction from how much a kWh?….”

    Carnegie Mellon Highlights.

    “Large-pack BEV applications use lower cost cells with thicker electrodes (base case $230 kWhโˆ’1).
    Increased electrode thickness capabilities could reduce BEV pack cost by an additional 8%.”

    Link Goes To Carnegi Mellon White Paper for sale page-

    So, building Traction Batteries at the GigaFactory projected to result in a cost per kWh of $230.00 a kWh?

    Then less 8% if Increased electrode thickness capabilities thus hitting a stunning cost of just $211.40 a kWh?

    Unheard of! This is actually the mother load of efficiency!

    Calling my broker at once! Thanks Carnegi!



    Thomas J. Thias

    $211.40 kWh? WHAT!

  16. Priusmaniac says:

    Could it be that that study was only focused on the battery production process side and completely forgot the other important aspect of large scale production which is buying power?

  17. Murrysville EV says:

    The Gigafactory is also supposed to realize its savings – in part – from recycling old batteries, not just economies of scale.

    Nobody seems to have mentioned that lately.

    1. Koz says:

      True and it also outputs completed packs, not just cells.

  18. martinwinlow says:

    A/ All 4 profs are engineers not economists.
    b/ The actual cost of the materials used in 1kWh-worth of Tesla-type cells is no more than $50 (by my reckoning) and the process of making the cells from these constituents is not rocket science.

    So, my money is on Musk – time will tell!

  19. JohnM says:

    mass production dont cut down cost???these people are crazy…

  20. Ocean Railroader says:

    In terms of Tesla’s Giga factory cutting costs I notice Tesla’s biggest handle cap right now is supply of batteries from Panasonic. In that Tesla keeps trying to ramp up production of their cars but they keep hitting the battery supply wall. At least with the Giga factory being built they will at least have the battery supply to build a 100,000 to 150,000 Model S and Model X’s a year even if the model 3 never comes into existence. Also Tesla’s friend solar city is going to need a really large amount of batteries if they have plans of building 20,000 back up battery banks that are ten kilowatts or bigger. So even the battery costs never come down there are simply still not enough batteries for existing demand.

    At the same time this going on I think the biggest barrier for electric cars is battery production. In that Kia has a new generation of energy dense cells for their Kia Soul that has double the power of the i-miev batteries and 40% to 30% more power then the Nissan leaf. The biggest thing holding this new car back is they simply don’t have the battery production.

  21. Blipblop says:

    Well, Panasonic was hesitating about joining in at first so it is not surprising to hear professors to question the gigafactory. But if the same professors had analyzed the possibility to build considerably cheaper rockets before SpaceX came around, I am pretty sure the conclusion would be that you can’t build them as cheap as SpaceX are building them today. Rolls of cupper and aluminum come in, rocket engines come out from the factory. SpaceX also use 3D printing to manufacture their new Super Draco engines for the Dragon V2 capsule. I am not sure professors take new untested manufacturing methods into account. Their paper is based on how things has been made in the past. Mr Musk has shown us that it is possible to challenge old fashion ideas about how things should be made. I am pretty certain the Gigafactory will be another success chapter in the book of Elon Musk.