UBS Analyst Dubs Model 3 Unprofitable, Tesla Responds With Battery Pricing

Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!

Full disclosure: I own TSLA stock. I bought it quite a while after starting to cover the industry and plan to keep it for the long haul. I’m not looking to game the system, nor do I believe that my financial interests add any bias beyond just my passion for Tesla and the work they are doing to break the world of its addiction to oil in transportation and the combined beast that is energy storage/production. Having said that, let’s jump in!

Battery Model 3 Tesla
Tesla Model 3 at March 31 unveiling | Photo by Kyle Field

With the Model 3 holding down center stage when it comes to Tesla stock forecasts, analysts around the world are putting it under the scope, building cost estimates, and searching for profit (or loss), with batteries getting the majority of the attention, as they represent the single largest chunk of cash going into the vehicle.

Analysts at UBS recently talked with Jon Bereisa (hat tip to StreetInsider), President and CEO of Auto Lectrification, who earned his stripes in the electrification of cars over at GM, having served as the chief engineer on the Chevy Volt program. UBS and AL chatted about the profitability of the Model 3, with UBS coming to the surprising conclusion that the Model 3 would need to be priced at $50,000 to be profitable … leaving a gap of $15,000 vs the current price.

It’s difficult to put any weight behind these comments from UBS, as they have consistently been critical of Tesla’s stock price with regular calls to sell … in February, setting a price target of $140 to $160 with calls to sell. Seeing as how stock is around $250 today, it seems that the broader market is bullish on news of Model 3, in stark contrast to the UBS position.

Interestingly, even with extensive experience at GM, which recently locked in battery cell pricing with LG Chem at $145/kWh, Mr. Bereisa’s forecast for best-case battery pack costs for the Model 3 comes in around $133–155/kWh by 2025. Given that Tesla is building batteries in-house in partnership with Panasonic at a much larger scale than anyone else, whereas GM is buying packs from LG Chem, it sure seems like Tesla would have a leg up on the competition … let’s dig into the details.

Battery Cells to Packs

That comparison bridges a wide gap, so let’s build out some structure to pull those two data points closer. First off — moving from battery cells (like the AA or D cell batteries in stores) to packs requires additional assembly and hardware costs. The cells must be arranged in order and wired up into groups with management and monitoring hardware to keep them all in line. This step in the process adds 20–30% to the cell cost.

This overhead brings GMs price from LG of $145/kWh for cells up to ~$188/kWh today for an LG-supplied GM pack on the high side … today. Mr. Bereisa shared his estimate of Tesla’s pack costs at $260/kWh, with GM coming in at $215/kWh based on the pricing estimates and glide path shared back in October.

GM vs Tesla

GM has shared its pricing in what seemed to be a push to be taken seriously as a legitimate high-volume player in the EV market. Without the same detail from Tesla, it is difficult to compare and Mr. Bereisa made some assumptions in order to estimate Tesla’s pack pricing … which is where things got interesting.

Adding yet another chapter to the direct PR model that Tesla seems to love, Jeff Evanson, VP of Tesla Investor Relations, actually dialed into the call during the Q&A segment to add some hard facts to the estimations Mr. Bereisa had made.

Specifically, he mentioned that the Model 3 would only be partially aluminum, not all aluminum as Mr. Bereisa had assumed in his financials. This was known from day 1 with subsequent details coming out that it would not be all steel but that it would be a hybrid (gasp!) of aluminum and steel (oh, ok).

Most importantly, Mr. Evanson also shared that Tesla’s all-in pack cost is already below $190/kWh, blowing away the estimates of $260/kWh. This sheds light on just how little data the analysts have to work with when compiling these estimates, and makes one question motives. What is to gain by putting out an extremely negative position on TSLA stock … especially if it’s wrong? This type of estimation makes UBS look amateurish and out of touch with current technology trends.

Finally, Mr. Evanson added that the battery pack size for Model 3 would be smaller than the 60 kWh that was used in the financial estimates. This is an interesting note and indicates that the Model 3 will have an efficiency greater than the 3.3 miles per kWh of the Model S, with at least 3.58 miles / kWh minimum (215 miles / 60 kWh).

Analyst Assumptions

The article goes on to talk about the pricing estimates that were used by UBS and reveals some bold assumptions, specifically regarding the large supplier margins:

“Jon is skeptical about the <$190/kWh cost, as the raw material floor on current chemistries is ~$160 assuming 40% supplier margins. He also sees 55 kWh min. necessary to get to just 200 mile range.”

Tesla’s entire strategy for the Gigafactory has been to flatten the entire supply chain by sourcing lithium and other raw materials locally to minimize non-value-add transportation and warehousing costs, which typically add significant overhead to battery production costs.

Beyond that, the world has never seen the type of partnership Tesla has with Panasonic, with both investing on the scale of several billion dollars each into the shared factory, which necessarily drives a change to supplier margins, essentially requiring old methods for estimating battery costs be thrown out the window. EVs are driving scale and integration in battery production and supply like the world has never seen … which this virtual collision of worlds demonstrates all too well.

Wrapping Up

In summary, this call was great for exposing actual pack pricing for Tesla and a few other exciting details leading up to the release of the Model 3, while at the same time showing how out of touch conventional auto and tech analysts are when it comes to this new hybrid industry that electric vehicles fall into.


Exclusive Tesla Model 3 Test Track Photos

Exclusive Tesla Model 3 Test Track Video

Exclusive Tesla Model 3 Review From EVANNEX

Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.

Our Latest EVObsession Video

I don't like paywalls. You don't like paywalls. Who likes paywalls? Here at CleanTechnica, we implemented a limited paywall for a while, but it always felt wrong — and it was always tough to decide what we should put behind there. In theory, your most exclusive and best content goes behind a paywall. But then fewer people read it!! So, we've decided to completely nix paywalls here at CleanTechnica. But...
Like other media companies, we need reader support! If you support us, please chip in a bit monthly to help our team write, edit, and publish 15 cleantech stories a day!
Thank you!

CleanTechnica uses affiliate links. See our policy here.

Kyle Field

I'm a tech geek passionately in search of actionable ways to reduce the negative impact my life has on the planet, save money and reduce stress. Live intentionally, make conscious decisions, love more, act responsibly, play. The more you know, the less you need. As an activist investor, Kyle owns long term holdings in Tesla, Lightning eMotors, Arcimoto, and SolarEdge.

Kyle Field has 1638 posts and counting. See all posts by Kyle Field

165 thoughts on “UBS Analyst Dubs Model 3 Unprofitable, Tesla Responds With Battery Pricing

  • Wow, that is an interesting story. Pretty shocked that Tesla revealed the battery pack price….

    • Thing is, Tesla is now on record with <$190/kWh pack price. That statement was made by a company representative.

      If it is not accurate and if Tesla doesn't walk it back and soon then they could open themselves up to a shareholder suit. Publicly owned corporations can't shovel bull in the way privately owned companies can.

      • If they are under $190/kWh now and expect a 30% decrease when the gigafactory is running that would take them to <$135.

        Take a look at what $135 battery pack prices does toward kicking hybrid, PHEV and ICEV butt.


          • Maybe they played a game with Elon:

            below $250?
            head nod
            below $240?
            head nod
            below $220?
            head nod

            Elon: “okay, i’m not playing this game any more.”

          • As we found out with the Navigant paper in 2014 (Tesla paying Panasonic $180/kWh for cells) there is backroom information that isn’t released for public consumption.

          • Maybe they guessed, like most of the other analysts…

            “And of course our assumptions (guesses) assuming (another guess) aligning with our previous assumptions (more guesses) in this case going forward (really are you sure?) dovetail with our initial (guesses – sorry jumped the gun on that one) assumptions (yes guesses!) appear to be accurate value predictors (guesses).

          • Any analist is requested to deal with the data availeble to him. so they didnt do something wrong but they just may be cold bad analysts.

          • True. Omar, would you mind changing the first reference to analyst, LOL!

          • Haha, like we do with a child

        • I remember my first week in college as a 17 year old and seeing a graph very similar to this from one of the professors. This professor was one of the early leaders making technology for hybrid electric cars and he was always fighting against and testifying against the all-electric “ZEV” mandates for what was eventually GM’s EV-1.

          He had similar graphs for Pb-acid, NiCad, and “future” lightweight ones. That was far longer ago than I care to admit – before NiMH ot Li Ion changed the slopes of the lines.

          • If Panasonic/Tesla can produce battery packs for under $150/kwh then production cost parity is reached and probably surpassed.

            P/T should be trying to get some of the lower ranked companies to get in bed with them. Let Tesla design their battery/motor systems. Bring them into the Supercharger system. Start work on a couple more Gigafactories to supply the battery packs.

            For example, Suzuki manufacturers close to 3 million cars per year. They’ve got enough capacity to produce more than Tesla could produce before 2025.

            If “Suzuki” could put a “Tesla Inside” label on their cars they would instantly jump the line and get ahead of the big five.

          • That is a great idea! Tesla could still keep their high end cars (S, X, and 3) to maintain brand image as best ever, but also partner with current auto makers to produce cheap EVs quickly, becoming the Intel of electric drive trains. Absolutely brilliant idea.

          • Suzuki cars are cheap. Would they also have free supercharging?

            Tesla need to think about that.

          • At one time unlimited Supercharger use was a $2k option for the Mod S. I think Tesla has said that it’s worked out to cost less than they expected.
            I’d guess the people buying the “Suzuki” would be the most likely to not have a place to plugin where they parked. People living in cities and parking on the street. Having access to Superchargers might be very important to them.

            Basically it seems that if it’s offered on all cars then the use per car is a lot less. Those who don’t use the system as much help pay for those who do. The cost averages out to be not that much per car.

            I suspect “unlimited no fee charging” has to be a huge selling point.

          • I live in Melbourne, Victoria and there is only 1 supercharging station.

            Yet there are 5-7 luxury cars parked outside each night on our block.

            ie, BMWs, Mercs, Lexus, etc.

            50kWh of electrons here are worth U$11.50.

            Still a lot cheaper than paying for 200 miles of petrol.

            Charging U$5/hour to park at a supercharging station would allow a lot more to be built.

          • Remember renting a Suzuki “Samurai”. A toy car, and the “Viagro” was worse.

          • Do you have a point or just an excess of bile you needed to unload?

          • Probably depends all on the incumbents there now.. the first to move will win, the others will be crushed.

          • I don’t remember having the ability to pay attention to a graph like that when I was 17!

        • Hmm, not clear if $190/kWh is now or when the Model 3 will be produced…

        • Oh! my…
          Ahh. 🙂
          You need to extend your chart.

          • How about this version?


          • Yaar!! Pirates be o’er yonder plunder all yer ICEs and PHEV’rs!! Yarrrrr….

          • Beware the crusher, matey. From it gasmobiles do not return….

        • It’s likely that the $190/kwh includes part of the 30% reduction — remember the batteries are already being produced at the Gigafactory.

          So don’t get too optimistic.

          However, I’m suspicious of that graph. It would imply that ICEs are more competitive than EVs at current $190/kwh battery prices and $2.17 gas. But they’re not; EVs are cleearly more attractive. I think this isn’t factoring in the fact that EVs are simply nicer to drive.

          • Batteries for EVs are not being produced there yet. Just Powerwall/Powerpacks.

          • Do we have confirmation of that? I had heard some time back that pack assembly (not actual cell assembly) had already moved out of Fremont.

          • I’ve only seen confirmation that these are being produced there…with no evidence to say that vehicle packs are being produced there. See the recent HBO video and leaked interior shots from the press tour.

          • I’m still waiting for one of those Powerwalls. Here in Edison’s California, if you produce a DMV registration for an EV, you can download unlimited electricity from midnight to sunup for 12 cents/kwh. This is California where we pay 2 1/2 times the National rate to begin with, so that is cheap for us. Fill the Wall, fill the car, it’s much less expensive than putting up PV panels.

          • I’m with you 🙂 Edison isn’t terrible but adding solar makes it that much nicer. I’m looking forward to trying to keep as much as my power in-house (hah) as possible. I’m thinking 2 powerwalls or similar to cover most of our production/consumption. No real savings other than a few bucks/month on taxes…but I’m eager to try residential storage out. I love to learn by getting my hands dirty 🙂

          • I really doubt that the Gigafactory is producing at “Gigafactory prices”. It’s not a fully functioning battery with a full scale workforce showing up every day and trainloads of materials at the loading docks.

            I don’t know if the $190/kWh is a “exactly right now” or “our current number on what it will cost when the plant is running”. That, I hope gets cleared up fairly soon. There should be a reporter pushing to get that answer.

            The graph plots only two variables. Cost of battery packs and cost of fuel. And I suspect there’s an assumption built into the graph that the EVs will be priced at normal GPMs. Tesla sells product at a premium because demand is high.

          • When will we see a Tesla dominate NASCAR? ICE’ers haven’t seen the light yet, maybe Monica needs an edge?

        • Those “hybrid” things are quite annoying, more so every day it seems. They only sustain our miserable situation with petro and worse, make it even harder to talk with sales people.

          • I can’t afford a Model S. I need to go over 120 miles in one shot frequently. So, I bought a used Prius two years ago to tide me over till a +200 mile range EV is available. I get up to 60 mpg. I get at least 43 mpg pulling a 400 lb trailer on hwy doing 70 mph.

            Give me an alternative.

          • Would a used Volt have worked for you? I see that someone sells a trailer hitch.

            If most of your driving days are within the Volt range or not much over you should be able to that mostly on electricity.

            (Not suggesting that it would make economic sense to switch over now.)

          • Main reason: trunk space is like 10 cubic feet – prius is huge plus being pretty much a hatchback – plenty of room for our vacations. I really liked volt and REALLY liked the 40 miles on electric.

        • Very informative graph but how did they do it ?

          • Took what was known about the price of internal combustion engines for ICEVs, hybrids, and PHEVs. Set up a cost model for all the other “car stuff” and ran it with various battery pack prices.

      • Is GM’s $145/kWh chart subject to shareholder scrutiny too? They did that several months ago.

        As battery prices dip below $175/kWh for light durable deep-cycle batteries, an odd thing happens. As I see it, the total social value for batteries probably goes up $20Billion (or perhaps $3B/year) for each dollar dropped in price below that as they work their way into more and more applications.

        I think few economists or politicians have really realized the profound world-changing implications of this. While they superficially acknowledge benefits for surface transport and electric power, I don’t think they really get what this really means. Both are multi-trillion dollar industries.

        I also think few realize the impact of getting to this battery price inflection point milestone on the R&D RoI on battery tech. Before this inflection point, improving battery tech wasn’t that big of a deal. But below this, it now means a technology that improves the batteries just 1% is worth about $5Billion a year.

        The industrial learning that Tesla will go through in Nevada will probably result in a major durable first-mover advantage that will be several percent better than others.

        • $145/kWh for cells. That would mean $175 to $190/kWh for battery packs. I’m not a corporate attorney but I suspect leaving a piece of erroneous information out in public opens the door to a damage suit if things go a bit sour.

          “I think few economists or politicians have really realized the profound world-changing implications of this.”

          True. I see little realization at the political level of what is starting to happen.

          And I agree, Tesla has provided itself with a permanent advantage by being able to push raw materials in one door and finished product out the other. Permanent at least until other companies respond which could take several years.

          • Yeah, troops protecting key rail lines and battery factories instead of the Seventh Fleet patrolling oil lanes!

          • Troops freed up from oil duty to help us fight wildfires as we climate change ourselves to a crisp.

          • Protecting Canadian cobalt and graphite mines, surely. 🙂

          • Canada has good cobalt resources? Do you know or do I need to check? It plays to another discussion I’m having elsewhere.

            If we can’t find a renewable energy way to run oceanic shipping, other than biofuel, which will be limited, what absolutely must be shipped from landmass to landmass.

            We can manufacture cars, solar panels, refers, etc. in the Americas or in Europe/Asia/Africa and move product around those two landmasses by electric rail. What would we absolutely have to move from/to the western landmass and eastern landmass?

          • “I think few economists or politicians have really realized the profound world-changing implications of this.”

            Justin Trudeau will get it. A vid went viral the other day where he correctly answered a question about quantum computing. I think Angela Merkel will get it too – she was a research scientist. Many pollies come from Law, which for whatever reason seems to kill off every maths/science brain cell.

        • You betcha and how. But there are ways to fudge.p, still. If the battery can’t charge quickly or has higher series resistance, capacity can be traded for charge rate to get a cheaper battery or more specific energy without any real breakthroughs.

      • Definitely! And why it’s not just “speculation.” I think this guy must be pretty certain about the # to state it publicly — he’s an investment guy.

    • Do you think they were fishing to smoke out Teslas costs? It might be interesting to consider if an ex GM guy got a call from one of his buddies asking a favor. Naaaaa. We know auto mfrs are clean as a whistle.

      • From what we’ve seen, Tesla is a straight shooter.

    • With this prices, I can not explain why are so expensive the PowerPacks.
      Are they reserving batteries for the cars?
      Will they thought at first that they will need to sell batteries to avoid GigaFactory stops but they has been overwhelmed by Model-3 reservations?

      • I think you guessed right. Originally they thought they would need stationary battery sales in case they didn’t sell enough Model 3s to use the full Gigafactory production.

        Now that they have enough Model 3s on order to require a second Gigafactory, they don’t really want to use the batteries for stationary storage. So they jacked the price up to reduce demand. They still seem to be selling like hotcakes at the higher price, so they’ll probably keep the price up to control sales, until Gigafactory 2 goes on line.

      • It seems to me that Tesla (and SolarCity) has a basic business approach of selling for what the market will bear.

        In the early stages of their businesses they have limited or no competition. There are people who can, and will, pay a premium price for something they want. The companies let them.

        As long as there is more demand than product to sell then sell at a higher price. That gives the company more money to offset start up costs and to build the company into something larger.

        Keep doing that until demand slows or competition arises and it’s necessary to compete for business.

        That seems like a smart move to me. Earn more, build faster, cut profits later when necessary.

    • The Tesla Powerpack, that price surprised me, too. It surprised a lot of us. Let’s hope that Tesla delivers can achieve the numbers for the price of the battery packs for cars.

  • It’s great to have new data on pack costs. Really helps with the curve fitting.

    And this is before Gigafactory!

    • YES! I’m so freakin’ thrilled. Use these charts in basically every EV presentation.

      • Oh 🙁

        I was waiting for the amazed looks on the faces of many powerful leaders, and the shockingly sudden changes of allegiance that would occur right around 2019. But like Armchair Hydrogeologist said anyway, it isn’t sinking in 🙂

  • What is the motive of UBS?

    • Could be nefarious. Could be like most people digging into this: Just want to know the cost details and potential.

      • Could be nefarious but simply put these so called experts/analysts are just people and people are often wrong. I should have been an analyst, or a meteorologist, there seems to be very little accountability in both fields…Its going to be sunny with some clouds and a chance of rain, the stock will go up and down…. cash cheque. =)

      • Could be ‘living in a bubble stupidity’. Groups of people tend to create their own reality and some may not keep an eye on what is happening outside the bubble.

        • Tesla, who’s CEO keeps trying to convince the world to embrace sustainable transport, is building a giant battery factory, is suddenly, according to my analysis, going to commit corporate suicide by selling 80% of their cars at a $15,000 loss, and this is where I would go “what am I missing? This can’t be right” but not this guy. The words they speweth.

    • UBS does not know costs for the M3 like Tesla does. It seems to be a ploy to get more information.

      • Agreed. Looks like they’re fishing for what they can get and testing Tesla.

    • They earn when you sell and when you buy.. they don’t care what the stock does as long as people sell and buy.
      The more that sell and the more that buy- repeatedly the more they make.
      They are interested in their own thick bottom line.

    • theyve probably been shorting the stock.

  • Unsubstantiated speculations refuted by other unsubstantiated speculations. Yellow pages are like that.

  • Even a $70/kWh difference in battery pack of 60 kWh only equates to a price difference of $4,200. There is a difference of $15,000 to cover as per biased argument of UBS.

    At $190/kWh, the battery component is $11,400. The rest of the car should be produced at $23,600 to break even. I think it would be profitable for Tesla if they can further reduce the battery pack.

    • All aluminum vs part aluminum was another piece as was smaller pack size as you noted.

    • Marginal costs of each unit, after breakeven, are said to be far under $10,000 for a ICE car.

    • actually the math is even better. Not only do you not need to not include the gas engine, lots of ancillary parts such as fan belts, spark plugs, catayltic converter, air filters etc not needed. However given that Tesla intends to be a premium model they could offer a lot more features.

    • Hey, nice post there Marion 😉

  • Model 3 is supposed to be 20% smaller than Model S, and is supposed to weigh about 20% less. It’s also supposed to have a *better* Cd, but let’s assume the Cd is the same. If Model S is getting 3.3 miles/kwh, Model 3 should get 4.125 miles/kwh. For 215 miles this would imply a 53 kwh battery pack (though I’m sure it’ll be called “55”).

    Using slightly more realistic and worse estimates for Model S efficiency, I get a 57 kwh battery pack, but you get the point: it’ll be somewhere between 50 and 60 kwh, definitely less than 60.

    • Try plugging in 4.9ft^2 for your drag area, that’s the frontal area of a Merc CLA x their target of .21Cd.

      • There is a aero showdown compare with all the figures, too. MB didn’t live up to its claims.

    • ” $24113 to produce the rest of the car”

      The Mod 3 should be just about the same size as the Toyota Camry. A car with which many of us are familiar. The Camry has a MSRP just over $23,000 and that includes the ICE and support systems.

      Perhaps someone has better data but I assume about 40% of an ICEV cost is engine/engine systems. Take away the 40% of $23k and there’s a lot of money to build a Camry sized car and make it a hell of a lot nicer.

      • A nickel a watt was what my rust-belt professor would always say for ICEv technology.

        From Rand:

        “In Cuenca, Gaines, and Vyas, 1999, manufacturing costs represent 50 percent of the manufacturer’s suggested retail price (MSRP). The engine, transmission, fuel, and exhaust system represent 22 percent of the manufacturing cost (or 11 percent of the MSRP) for a subcompact, and 20 percent (10 percent of MSRP) for a minivan. Based on these percentages and a review of current MSRPs for subcompacts and minivans, we put costs uniquely associated with ICEVs at $16 to $18 per kilowatt of rated engine power when nonmanufacturing costs are excluded and $32 to $36 once nonmanufacturing costs are included”.

        • Geez, when I run these cost numbers, it looks like Tesla has a chance at a 36% gross margin on a base Model 3. They won’t achieve that; they have warranty reserves and there will be unexpected costs and they need an allowance for changes in raw materials prices, and so on, but wow.

          • It feels like we’re watching something unfold that has not happened before in car manufacturing. A major cost component’s price is rapidly collapsing.

            We should keep a close watch on the GPM between now and when the 3 starts selling. Assume battery cells were over $300 when the S was introduced in 2012. By late 2014 the price had dropped under $200/kWh. Now the pack cost may be under $200/kWh, cells under $150/kWh. If that’s the case then we should start seeing a big bump up in GPM over the upcoming quarters.

      • Here’s a SWAG:
        – EV power train $0.02/Watt so 200HP is $3k
        – ICEv power train $0.05/Watt so 200HP is $7.5k(including transmission, exhaust)

        Costs unique to EV:
        – extra tire/suspension cost for heavier EV: +$1000.
        – So we’re at $4k + 60kwhr*$190packcost=$11.4K vs $7.5K.

        At 30% gross margin that inflates to a delta of $5K MSRP for the EV power train.

        If both last 150,000 miles,

        Model3 3.5mi/kWh at $0.12/kwh for 150k miles is $5.1K
        Camry 30mi/Gal at $2.50/gal for 150k miles is $12.5K

        So basically EV has crossed over without subsidy for a high-volume mid-sized family sedan.

        • How dare you to support Toyota over Tesla. Toyota selling tens of million of cars is dying and Tesla with its humble 100k is thriving.

        • I can not argue with you. I am just an old and sick man. Spare me, pls.

        • I don’t assume the 3 is extra weight. At least not compared rpto rivals in its acceleration class like BMW 3 series which starts at 3800 pounds. I doubt a 20% smaller 3 will weigh more. If an a S is 4500 pounds.. The rest falls into place.

          • 80% of 4500 is 3,600.

            A ‘20% smaller than S’ Mod3 with a >20% smaller battery pack could come in with a curb weight not that different than a BMW 3.

        • Tesla is engineering for 1,000,000 mile lifespan.

    • That’s today’s cost. Dare we estimate $150 for initial Gigafactory packs? That puts us at $8250 for the battery and $26750 for the rest of the car and margin…

      The Bolt is seriously overpriced.

      • October, 2014 cells at $190/kWh. 30% cost drop when the Gigafactory is running. Cells drop to $135.

        20% to 30% to turn cells into packs. Pack price becomes $160 to $175/kwh.

        44 to 50 kWh battery pack. $7,040 to $8,750. Average $7,900.

        $35,000 – $7,900 = $27,100. Yeah, your math looks reasonable to me.

        Toyota sells a fairly nice car for $23,000. The Camry is 20% smaller than the Mod S as the Mod 3 is suppose to be. Pull out the Camry engine and engine systems. There seems to be a lot of room to make a Camry sized car with nicer features and still end up with a higher GPM than Toyota enjoys.

        As we hit production cost parity for EVs and ICEVs over the next five years Tesla looks to be positioned to produce a Camry sized car with Camry features and sell it for less than Camry prices while making a decent profit.

        ICEVs die.

        (Production price parity, not purchase price parity. I suspect EVs will command a premium price because of high demand and short term low supply. Later, as more EVs are produced, we should see profit shrinkage.)

        • Guess what my favorite part of your post is? OK, I’ll tell you. It’s the paragraph that says:

          ICEVs die.

          • Oh, yeah!

    • Is the weight expected to be 20% less – I did not see where Tesla released that spec?

      • A drop from a 75 kWh pack to a 55 kWh pack would be a 25+% battery weight reduction.

        • All else remaining equal…

          • Well, if the body is 20% smaller (length x width) then there’s less metal in the hood, roof, frame, etc. Lots of stuff gets downsized.

          • Yes, hopefully the weight continues to be rapidly reduced iteratively. Though I remember one iteration of new batteries where Tesla received higher density cells – that also weighed more per cell?

          • Over time battery capacity increases. That means few kg per the same kWh.
            I haven’t seen anything about what may be happening with capacity. I think Musk has talked about a Mod S with a 400 mile range by 2020 and that came up in a discussion about capacity (weight).

          • Weight goes as the cube of length or width in a mechanical structure. We know both length and width. Not hard to figure out. Fudge factor is how much steel vs aluminum. But really, chassis weight doesn’t matter. It’s frontal area , Cd, efficiency. That sets battery size and weight. Everything goes from there. That’s if you purpose build an EV from scratch (and know what you’re doing). Weight doesn’t matter much to efficiency. Aero drag is much more important. 2/3 of loss is aero at highway speeds.

          • “It’s frontal area , Cd, efficiency.” Yes, critical factors.

          • I think it’s generally unappreciated how much the milage improves with a decrease in speed. On three occasions in the past two years I’ve been called upon to extend my planned journey in the Leaf. It hasn’t been at all stressful – I just slow down to a little under the speed limit, until it’s clear I’m going to make it home – then resume my usual ‘burn em off at the lights’ driving.

          • Yeah. I tend to go slow, and the result is that I get very good mileage in my Tesla. The very steep hill saround here don’t help though.

      • The *size* is supposed to be 20% less (length and width). I estimated from that that the weight would be 20% less.

      • They didn’t give it. I might have started the notion. It makes sense. For the curious, estimate it another way besides battery size and frontal area. Try this. Estimate outside dimension reduction and apply a cubic approximation on weight ratio from the S. More steel in 3 alters it, but it’s a rough cut approx.

  • “assuming 40% supplier margins.”
    Yep, this is where UBS screws up on their assumptions.

    By in-housing *everything*, Tesla is collecting most of those supplier margins. They control a larger portion of the value chain than any other carmaker.

    Interestingly, this strategy was used by Ford in the early 20th century: he owned everything from iron mines and coal mines where the raw materials for the steel came from, to the forests where the wood was taken from, to the shipping fleets which moved the materials.

    • Hmm, didn’t know that about Ford. But know Musk has read a lot about him (or the company’s history if we want to get borader) and took several lessons from that history.

      • Zac he also raised the wage for his work force much to the dismay of this rivals.
        His idea was more money in a workers pocket more chance he will be able to purchase the product, an aside a model T is running down the road here using ethanol as fuel.

        • I remember taking a school field trip to the Ford’s River Rouge plant as a child. I remember seeing huge front-loaders and watching as yellow-hot steel bars were forged.

          I recently took another tour of the facility, and it is nothing like it used to be.

        • Ford wanted the Model T to use multiple fuels. He was disappointed that his battery partnership with his old employer (Edison) would never result in a practical electric version for his cars.

    • Ford slowly shed those responsibilities to better allocate investment capital, to increase flexibility and to survive downturns with less disruptions.

      • Where they align is that neither could look around and see any viable support industries. They had to do it in house to get what they needed at the cost and performance required. Once auto industry grew, It was big enough to stimulate a supplier base. We are not quite there yet. Six years ago, if you wanted an EV purposed electrical disconnect from faruz shawmut or GE, you had to order it. Prices for those and high voltage DC fuses and DC GFCI were awful if you could find a supplier. The only easy source of good prismatic was Winston, then Thundersky. It cost too much to make an EV because each part was in low volume compared to the millions supplied to Detroit. That’s changing. Not there yet. But changing.

        • Exactly – my next comment to continue the thread; if you were busy with something else, LOL!

          • Pulled an all nighter. Carry on.

        • You are correct. He made them in-house because there was no other good option. The same thing is going on with Tesla — they’re using suppliers as much as possible to avoid investing their own capital, but “as much as possible” turns out not to be very much. The suppliers just aren’t able to deliver.

          Tesla bought a *tool and die* company to secure their supply. Think about that.

          • Yes. I read about that. Great move. Seems they use it to make their aluminum stamping forms.

    • Yeah.. and then the beancounters came in and siphoned off all those internal profits (= price reductions for the end product or higher wages for the workers) and moved them into the profit zone of subsidiaries and suppliers they had connections/stock/friends/family at where they can be turned into shareholder value etc.
      Smart, no? 😉

  • What is Tesla’s targeted margin in their business model? Is it 40%? If so then calc’ing out 40% margin over raw supplier costs would be reasonable. But I think historically it has been around 25% and most auto companies are around 15%

    Does Tesla’s pack figure of $190 properly account for everything including amortized cost of gigafactory and equipment? It is very tricky to account for internal cost correctly.

    If the prices are really that low I think the key to getting batteries cheaper long term is to get the ~$100 materials cost down by (1) finding ways to thin the materials on the cathode which drives the need for expensive cobalt and nickel and (2) finding ways increase the deep cycle durability of the anode which is a key driver of the energy density.

    • I can’t recall for sure, but think it’s 15%. (Probably shouldn’t comment when not checking/knowing for sure.) Quite positive it isn’t >25%.

    • I don’t think Tesla has a targeted net margin, other than “positive” (i.e. profitable). They have a targeted *gross* margin of 30%. Obviously they have to cover the immense overhead and capital expenditures with that, and hopefully make a decent profit after that.

      But the key here is that most of the capital expenditures are not proportional to production volume. Those capital costs that are proportional to production volume should be included in the per-kwh cost, but most of them are not. The entire point of the Gigafactory was to achieve massive economies of scale. It is absolutely *essential* for Tesla to sell very large numbers of cars — they will not break even unless they do so. They are *relying* on the economies of scale.

      I have a pretty simple model for valuing Tesla stock, which has a number of inputs. The key input, it turns out, is “how many cars can they manufacture and sell per year”. Period. Multiplying R&D and capital costs by factors of 5 doesn’t make that much difference to the valuation. Neither does the selling price for the cars or the gross margin on the cars. But varying the number of cars produced gives *huge* variances in the valuation of the stock under my model.

      So it really is all about how much they can scale up and get economies of scale from mass production.

      • How could Nickel and Cobolt raw costs be reduced by mining at a larger scale? Is there significant potential there?

        • Material costs don’t figure that prominently in pack costs yet. There was a lot of assembly that was just plain inefficient. And volumes have been miserable until Tesla changed all that permanently. IMO, 3 has changed everything.(and PowrWall PowerPack.

          • I saw a number on how many miles some of the materials in batteries traveled before the battery pack was delivered to the car plant. It was enormous.

            That’s not only shipping costs. It’s capital tied up with stock that’s not in the plant but out traveling. It’s more places for the supply chain to be disrupted. It’s more people syphoning off parts of the profit just to deal with the business of shipping.

            Ship materials once. To the factory. Ship finished battery packs from the factory, not cells that still need to go through one more factory.

        • A huge portion of the costs are manufacturing costs, not raw materials costs.

          A recent estimate pegged the materials cost for a lithium-ion battery below $125/kwh:

          This includes a certain amount of refining, which can be made cheaper by improved manufacturing processes.

          If you look at the pack costs of $190/kwh you see that there was a lot of savings to be made in manufacturing cost.

          — Elon Musk’s next target is materials transportation cost. By sourcing the lithium and cobalt and so forth from North America, he intends to eliminate a large portion of the shipping cost involved in the raw materials.

          — There have been multiple technological developments reducing the cost of lithium mining and refining.
          — Replacing synthetic graphite with natural graphite should cut the cost there.
          — I’m not aware of major sources of cost-cutting for the cobalt, unfortunately. But it’s only supposed to be $10 to $15 / kwh raw. So refining improvements could be all that’s really nearded here.

          • I’ve seen a materials breakdown for lithium-ion batteries that set the cost a about $70/kWh. That leaves the door open for $100/kWh cells.

            And higher demand for some of the materials could take their costs lower. More efficiency as mining/processing scales up.

    • Looks like Tesla wants as many profit sources as it can command. They left out the dealer profit bleed. They own the rapid charging system. They own a company that makes the machines that makes the cars.

    • I think 44 kwh getting 220 miles EPA is a bit optimistic

      • Model S60 had an EPA rating of 230 miles.

        20% smaller car, less frontal area to shove through the air. Every pound of battery you remove is one less pound you need to propel down the road. Perhaps over the years Tesla has figured out some ways to improve battery/motor/drive train efficiency a bit.

        44 is 27% less than 60. I’d only be guessing but I’d guess 44 to 50 and likely less than 50 to get to 215 miles. (That’s 6.5% less range than the S60.)

        • We have known for a few years that the Model 3 will not be all-aluminium like the Model S.

          The 3 will be made of steel.

          Thus maybe not 20% lighter.

          • So what is the cost driver from aluminum? I know little about cars but I have designed die casts with aluminum and mag alloys and did one hydroform. Okay I had a mechE do it for me but I was doing the high level stuff.

            Hydroform seems really cheap for larger objects. The expensive thing with aluminum was braising, which I was able to avoid. Otherwise, aluminum seemed like a good deal. Obviously Tesla knows it’s stuff but it seamed to me you could hydroform a crinkled tube and get great performance on the cheap.

          • Aluminium is hard to weld I think.

          • Very hard to weld. Weird properties when you crush it, too, so the pressing has to be done differently.

            The biggest cost driver from aluminum, however, is that aluminum is seriously expensive to refine. This is why recycling of aluminum is so important. Steel’s very cheap to refine from iron by comparison

          • In future a less expensive refining cost, as the cost of solar panels reduces.

          • Yeah, they still have to deal with the poisonous “red mud”, though. No such waste product from steel refining.

      • His first criticism – they didn’t show the $35k version, they showed a higher priced model. Did Musk claim they were showing the lowest priced car? No. Tesla has been clear that the glass roof is extra.

        Second – M3 will “Osborn” the MS. I can see the market for the MS shrinking some due to Tesla buyers being offered the MX and M3 options. Is that really a problem? Some people will still want a larger car, MS annual sales might cap out at 50k/whatever a year. The M3 is not a substitute for the MS. People who would otherwise buy a larger Lexus, Benz, BMW will buy a larger Tesla.

        Third – ” by the fact that a TSLA spokesman has already said that the construction of the whole Gigafactory won’t happen until 2020, whereas this was supposed to happen until October 2017″ That’s just a bullshit argument. Tesla said full production by 2020.

        Fourth – Supercharger system won’t grow fast enough. We’ll see. It will double by the end of 2017. It will be about twice as large when the first M3s start coming out the factory door. We don’t know how much the SC system will increase in size in 2018, 2019 and 2020 – the year Tesla intends to hit 500,000 sales per year.

        Fifth – M3 deliveries may not start as soon as Tesla has suggested they will. So? Neither Tesla nor any other manufacturer can fully control the behavior of contract suppliers. The major transmission company who was to build the drivetrain for the Roadster said they had engineered it correctly. They delivered hardware that was under engineered.

        I’m sorry. I can’t wade through this any longer. The guy is just putting as much negative spin as he can on how history will play out. He’s assuming Tesla doesn’t know what it’s doing, that Tesla hasn’t planned ahead.

        There’s no corned beef and Swiss in his Reuben. Only sauerkraut.

        • Thanks. I appreciate your efforts. I’ve bet big on Elon and get a little nervous sometimes. Historically I’m good at separating BS from truth but he hit on so many areas I didn’t have facts on.

          • Depending on how one takes Tesla’s proposed delivery dates it’s possible to legitimately criticize Tesla. If you view “will release in month X, year X” as a contracted date then Tesla has missed some big ones. If you view it as Tesla’s intentions then all that you can say is Tesla didn’t get there as fast as they hoped.

            Reality check.

            Tesla was late delivering the Roadster, but they beat every other car company to the market.

            Tesla was late delivering the S, but they beat every other car company to the market.

            Tesla may be late delivering the 3, but they’ll probably beat out every other car company to market with a 200+ mile EV priced in the mid-$30k range except GM. (And I think Tesla is going to deliver more for less.)

  • With the number of Model 3 orders the situation should be absolutely positive for Tesla i am sure they did not think they would be pushing 400 k orders in their figuring. I do not know this but stands to reason.

    • Mo Money Mo Problems
      – The Notorious B.I.G.

      They have admitted that they need to redo “their figuring.”

  • “This was known from day 1 with subsequent details coming out that it would not be all steel but that it would be a hybrid (gasp!) of aluminum and steel (oh, ok).”
    That made me lol

    Thanks for the explanation between cell/pack!

  • I took UBS advice (not) and got a bunch (for me) of TESLA on Feb 11th. I’m smiling, only wishing I had experience (and balls) enough do have done 3 month options instead. So the fact that they and all the other shorting TESLA are bad mouthing them is no a surprise.

    • LOL! Yes, apparently analyst advice can be somewhat counterintuitive.

    • I’m probably being way too conservative on my TSLA purchases but it is already my second-largest holding (after BRK which is a conglomerate) so…

  • I said the 3 pack would be about 50kwhr. Let’s see when they give specifics. Like shooting fish in a barrel. Tesla says 20% smaller. Guess what. It weighs less. Frontal area lower. And since I know a few secrets, length to width and frontal area ratios are similar to S. But heck, anyone with aero experience would know that.
    How come Detroit can’t figure it out even with a roadmap, instructions, and a call from mother.
    Try this, why would Tesla, who knows they can go 208 miles in an S with a 60 kwhr pack, elect to use the same size pack for a car 20% smaller that needs to cost less? Apparently, Tesla is the only one that could figure out that extremely difficult technical question.
    One can always imagine higher costs if one designs poorly. Guess what. That’s exactly what Detroit is doing. Time fore Detroit to get real and do some real design, not just posturing. Outsourcing your whole drivetrain from a Battery manufacturer isn’t very impressive technically. GM deserves to be called a marketing company. Ford is a day late and a dollar short, and Chrysler is bringing up the rear with Fiat boss making embarrassing comments, probably to make up the loss of Lutz from Detroit.

    • (hopefully I added something to your post…I know I am repeating quite a bit)

      Yeah the 60kwh assumption seems super dumb now that you bring it up. The car will be smaller and have a significantly better drag coefficient than the S. Even with some steel it has to also be lighter. It is pretty obvious they will reduce the pack size. Even during the M3 reveal musk said stuff like better aerodynamics will improve their range. They will get more with less!

      Market manipulation or a ploy to get info leaked earlier. One is illegal and the other makes UBS look dumb.

      This is ignoring improving margins from the gigafactory once it is fully operational…I don’t get it. Perhaps they are just incompetent? I am not saying I know the model 3 will be profitable or not but I reject the UBS analysis.

    • Detroit has been about marketing and style for decades. When Honda was busy inventing CVCC engine and variable valve timing, GM was tweaking chrome for the next Oldsmobile, that remarkably, got 140 horsepower out of a 4liter V8. GM might have set a record with that power to displacement ratio. Detroit was always tweaking styling in the 1970s, 1980s, and even 1990s and yet last to market with fuel injection, multi valve, variable valve timing, overhead cam, independent suspension, and so on. Their market share loss shows it too. They’re much better now, but this is a huge technology inflection and clever technology and business strategy are essential.

  • Pfft…. these analysts weren’t right about their Tesla predictions 5-10 years ago either.

    Remember when so many of these analysts were talking about $1000/kwh batteries… while Tesla was already well below that?

    They have every incentive to cast doubt about their competitors. Just because they can’t get their costs down, or have to take huge margins to survive, doesn’t mean Tesla has to follow the same rules.

  • Actual costs are plummeting well below analysts estimates. They cannot revise down fast enough. 1 1/2 years ago I wrote this piece and my predictions of $100 / kwhr by 2029 were considered very optimistic. Now it looks conservative. With Tesla at $190 / kwhr now it looks pretty certain that Musks $100 by 2025 will look conservative in the not too distant future. Volume ultimately drives cost through the experience curve. Since EVs appear to be catching on faster than previously thought, battery prices will drop faster than previously predicted due to the experience curve effect – In other words I am not too worried about Tesla not being profitable when they need to be.

  • Model 3 EPA range of 215 miles is 4.3 miles/kwh from a 50kWh battery. Model S 70kWh has an economy of 3.4 miles/kwh.

    These figures appear to me to match as the M3 has lower weight, lower acceleration which is factored into the EPA test, and lower aerodynamic drag.

    We should also assume 20700 size cells which should have 40% more energy than the 18650 size cells as the internal volume should be 40% larger. External volume is 33% larger. Thus, 40% less battery components will be required.

    Elon said recently that the thermodynamics of batteries is tough. I expect the shift to the larger format cells is as far as they can go at present keeping the same cell chemistry.

    • JB Straubel said in a talk about a year and a half ago that 18650 were slightly smaller than optimal at the time, so 20700 would seem to be in the direction they were headed for some time now. And there’s some percentage points of price reduction right there just by tuning the size to be optimal.

      • That is great info. As the Gigafactory is now in operation, I wonder when the larger format cells will appear in MS and MX. It will help with profitability, if nothing else.

    • The 20700 are as large as they want to deal with from a thermodynamic point of view. Good size.

  • $145/kWh, those must be some crazy margins on the Powerwall at 6.4kW. Throw in 1k for the pretty cover and electronics, the margins still look great.
    Off grid, even at $200 per kW for a “Superwall” you’re looking at $5,000 + $1,500 for the internal electronics. 8k for 25kW nameplate capacity, they would make a killing. California and Nevada could go off grid. 5kW solar array, charge controllers, inverters and a 25kW battery under 25k.

    • It looks like Powerwall has its own entire heating and cooling system — possibly an entire heat pump air conditioner. Powerpack certainly does. Until they get the economies of scale going, that’s a solid $3000 piece of equipment, at least. Which doesn’t actually leave much room to pay for the battery.

      I suspect Tesla is working on getting economies of scale and price reductions on the temperature maintenance equipment.

      • $3000 is retail price. Tesla must be selling it to the wholesaler at $2500. That is under $400 kwh with all bells and whistles…not too shabby.

  • I do not know about UBS. But regarding the track record of this web site in what concerns Tesla, one can not possibly take it as “independent” or “unbiased”. Just take the rise and fall of the Powerwall, for instance.

    • This site, and most of the commenters, give a lot of love to stuff that works.

      Most of us are very concerned about climate change and most of us realize that the best way to cut fossil fuel use is to find acceptable replacements for fossil fuel energy.

      That means we end up cheering for wind turbines, solar panel, and electric cars.

      Then we cheer louder for the companies that are driving those technologies fastest.

  • My position has been that the better the Model 3 seems at announcement, the later the Model 3 will be produced in volume. The Model 3 looks great, so I expect a slow roll out due to costs.

    Musk forward-priced the Powerwall, and probably the Model 3 too.

    At low production volumes Tesla can manipulate how overhead is applied to batteries and show decent margins on both Powerwall and Model 3.

  • If the spokesperson has GM experience I am reminded of:
    1) hiding the ignition switch issue
    2) colluding with the state of Indiana to prohibit sales of Teslas in the state
    3) anybody remember the movie Tucker?
    4) the statement before Congress years ago, “What’s good for GM is good for the nation”
    5) the “Unsafe at any speed” Corvair.

    Only time will tell, but Musk himself has said that he doesn’t expect bottom line profitability before 2020. Then we’ll know who is correct. I would find any outsider to Tesla’s views to be extremely tenuous. Maybe GM should hire a Manufacturing Cost Accountant from Tesla before guessing as to the viability of the Model 3.

  • Forbes:
    LG Chem May Be On The Verge Of Dominating EV Battery Industry
    Sam Abuelsamid , CONTRIBUTOR
    OCT 28, 2015 @ 09:03 AM 15,229 VIEWS

    GM executive vice-president Mark Reuss recently acknowledged that LG is on track to get cell costs down to $145 per kWh in 2016, ramping down to $100 by 2020. Combined with the reduced assembly costs, the Bolt could have a pack cost of ******* $300 per kWh or less. That’s well below the industry average of $424 per kWh projected by Navigant Research in its Advanced Energy Storage for Automotive report.

Comments are closed.