Demand

[AnxietyRanger]

I could be an untapped demand I'm not sure it makes sense to open a store/service center here until Tesla has more models available (3, y etc) though. I'm between the two stores/sc in Sweden, a 5 hour round trip to the nearest store. It doesn't make sense to take out several vacation days per year to get to the sc, get a loaner, return it etc.

There is untapped regional demand, for sure, in underserved or non-served areas (and equally a significant CapEx negative to serve them). That goes without saying. I'd say most are talking about natural demand levels in existing markets.

 

[AnxietyRanger]

I really do think that demand for Model S and X is limited basically by who has enough money to buy 'em (and who is willing to spend that much on a car). Tesla's approaching total control of this price bracket (segment) -- if you compare it to the other cars in the price bracket, Model S is heading to over 50% of the market IIRC -- and I don't see the number of people willing to pay that much for a car expanding much.

To on develop this thinking, let's use an Audi example:

The lack of availability of other long-range EVs has already pushed the Model S into segments of buyers that would normally never consider something in the Audi A7/A8 range, which is the Model S group. We have tons of people coming from a Prius to Model S on this forum, for example. That is not the story on a forum about Audi A7/A8, definitely not. This is why the likes of Model 3, I expect, will hit Model S harder than a new Audi A4 hits the sales of Audi A7/A8...

Beyond that, I would argue the split between Model S and Model X could be more nuanced. Let's look at Porsche:

My basic proposition is that Model X is inherently in a much larger volume segment of cars (high-end SUV) than Model S is (high-end sedan/hatch). In 2015 Porsche delivered (Porsche Geschäftsbericht) 17,207 units of Porsche Panamera, a high-end sedan/hatch like Model S. In the same year Porsche delivered over four times as many Porsche Cayennes at 73,119, a high-end SUV similar to Model X. (And again, Audi also selling tons of Audi Q7.)

Now, we can argue the practicalities. Porsche Panamera as a four-seater is a somewhat compromised hatch, whereas Cayenne is a fairly high-utility SUV (fully folding rear, roof rack expansion). Model S is a high-utility sedan/hatch with up to seven seats and massive storage thanks to large trunk and frunk, plus the roof rack. Model X, in reverse, is a somewhat compromised SUV with large area folding seats only recently and only in the fiver (no edge with seven seats) and the unconventional doors nixing the possibility of using roof racks and, I assume, alienating some more conservative buyers...

So, it is possible the relative lack of utility in Model X can limit its market appeal more, and likewise the relative abundance of utility in Model S can expand its market appeal more, but still at some point they both come down to what they are: an expensive sedan/hatch and an expensive SUV. Electric or not, those categories do have certain ceilings overall and relative to one another, and the expensive SUV category in all likelihood is larger than the expensive sedan/hatch category is...

Especially when things normalize competition and availability-wise, i.e. when there are more alternatives, I fully expect Model S to become the lowest selling Tesla instead of the biggest selling one of the S3XY lineup - unless the falcon wings prove to be a fundamentally problematic proposition on the market once production constraints on the Model X clear.

 

[bonaire]

I believe we have reached the natural demand plateau for Model S at the curent price levels -- 80K to 100K per year. (In the near term the demand will probably be depressed outside the US by the exchange rate.) Model X has had a rocky start but should be approaching its natural demand plateau within the next couple of quarters too, and it'll probably be about the same number. If the price levels are cut, obviously demand will rise. I suspect we'll actually see Tesla cut car prices across the board at some point soon, but they're currently aiming for that 30% gross margin so that Musk can collect his incentive stock options (I think he won't get that tranche).

With a combined production of (14730 10450) last quater, I don't see this equating to a natural plateau of 20-25k MS and 20-25k MX just yet. They need to sustain something like this 15k + 10k for a while to see if even this plateau is natural.

 

[bonaire]

I'm not bonaire, but overall I would personally like to - and actually assume bonaire does too - separate two things:

1) Whether or not Tesla is production or demand constrained
2) Whether or not this is a good or a bad thing or a neutral thing

I actually think the argument by far more has been about the first point. It has been established in the Tesla lore so to speak that they are a production constrained company that sells cars without any marketing, discounts or sales push.

I would argue a lot of people would say that - while once true - this no longer is true. Tesla is now a more normal company in the sense that is has to market, discount, push sales to keep up with their demand and demand goals.

That doesn't mean much regarding being a good or a bad thing necessarily. It just means the old fact is no longer true.

One thing we must consider. Repeat buyers and/or collection buyers. Those who have bought their third or fourth car since 2012 are naturally helping sales. Would they have bought 3-4 BMWs or 3-4 Audis in the same time period? Is this "natural" or "super-fan" solution sales? They are not everyone but there have been dozens of posts here of those who started RWD P85, then P85D, then P90DL then maybe even P100DL. In that vein, where did their older cars go? Used cars or CPOs which moves more products "down market". Without access to ownership databases, we can only assume that these bulk buyers are small in number, say 5-10% of the total names who have become new-buyers. But this then turns into more consumerific sale who order up an MS60 to "be like Mike" and get into the club. What is the natural plateau? I don't think anyone can know - but it will be affected by Model 3. If M3 is "good enough" then MS sales in the lower-end will tank and more will surely choose a cheaper and "it's newer" with the Model 3. If M3 offers a compelling Ludicrous mode acceleration, then MS P and L models may also suffer slightly.

Some replies here indicate that they think that entire markets will be overtaken by Tesla. Don't forget that many people are buying "multiple cars" themselves (for family, or personally). So, it's like a big rock band with a big following selling out stadiums in nearby cities all to the same people at each show. It is a sale - but it may not always be a unique sale but rather a form of "super-fan" action. And other people bring up "competition coming" in their critiques. I haven't yet seen anything compelling from the Germans to be equivalent. Now, if VW and their Audi team actually mimic drivetrain capability (hot motors and big batteries) they still have the issue of charging networks to think about. VW and all its subsidiaries - if they actually do something compelling in the EV space, can own it. But are they interested to invest the indicative $5-10 Billion USD to do it?

 

[Spidy]

I keep reading about Tesla building a "lot of inventory". Does anybody have any specifics to add to this? How much is a lot?

Usually there are like 10 reports or more for 1000 VINs, between 160xxx and 170xxx there were far less than usual. This is also the range of Inventory cars showing up, even some with a bit lower and some with a bit higher VINs. In addition you can see that many of these customers cars were build in a very short time frame (most of the ones build earlier were overseas orders) so my guess is that addition to European cars etc. Tesla was building Inventory from the end of September to the start of November.

 

[vgrinshpun]

Well, I have to conclude that the data about "a lot" of cars built for inventory is pretty soft. I think that drawing any conclusions based on this is premature to say the least.

Just a reminder, we had a lot of very concerned members insisting that Q3 margin will take a "hit" because of "discounts". These concerns turned out to be overblown. So I am going to make this large picture prediction: whatever is going on with "inventory" production is likely NOT an emergency maneuver because something is going terribly wrong - there is just no data to back up this narrative. As I was advocating during the heated debates about Q3 GM suffering due to "discounts", I believe that whatever is going on with the production of "inventory" cars is a planned activity which is accounted for in the Q4 guidance, including overall deliveries and GM. So bottom line is that I do not believe we have any data indicating Tesla is at risk of not meeting this guidance.

 

[vgrinshpun]

With a combined production of (14730 10450) last quater, I don't see this equating to a natural plateau of 20-25k MS and 20-25k MX just yet. They need to sustain something like this 15k + 10k for a while to see if even this plateau is natural.

Well, if we are discussing demand, delivery numbers are more relevant than production ones. Furthermore, the only delivery information that is meaningful at this time is for MS, as MX is still being ramped up.

In Q3 MS deliveries were 16,047, which is equivalent to an annual rate of 64,188 cars. Assuming Tesla meeting their guidance for Q4 and delivering about 16K MS, the total for 2016 will be 54,211 cars. This is conservative as it appears that due to the extreme batching of the production based on geographical area Tesla is likely to deliver more cars in Q4 than produced. Tesla delivered 50,372 cars in 2015.

So here is how all of the above stacking up:

2015 - 50,372 MS delivered
2016 - 54,221 MS projected to be delivered based on guidance
2017 - 64,188 MS projected based on current rate of deliveries

To my eye, word plateau falls short of describing above cadence.

Another interesting thing is that Tesla, which has more visibility on the issue of demand and, most importantly, their plans of pulling the various demand levers, are actually in the process of expanding production capacity for MS/MX. Their statement about this in Q3 shareholders letter appears to be overlooked:

For Model 3, we have completed production line layouts and will soon begin installation of new body welding and final assembly lines. We have established a world class team of suppliers for Model 3 production equipment and components and critical long lead time equipment and components have been sourced. We are now testing vehicle systems such as chassis, the high voltage drive system, and low voltage subsystems such as vehicle controllers, HVAC, infotainment and lighting. As refinement of the Model 3 continues, we remain on plan for our timing, volume, vehicle capability, pricing,and margin targets.

The Gigafactory remains on track to begin cell production later this year for use initially in our energy storage products and later to support volume production and deliveries of Model 3 in the second half of 2017. In addition, we continue to expand production capacity at our Fremont facility and are exploring additional production capacity in Asia and Europe.

 

[dennis]

Well, I have to conclude that the data about "a lot" of cars built for inventory is pretty soft. I think that drawing any conclusions based on this is premature to say the least.

Just a reminder, we had a lot of very concerned members insisting that Q3 margin will take a "hit" because of "discounts". These concerns turned out to be overblown. So I am going to make this large picture prediction: whatever is going on with "inventory" production is likely NOT an emergency maneuver because something is going terribly wrong - there is just no data to back up this narrative. As I was advocating during the heated debates about Q3 GM suffering due to "discounts", I believe that whatever is going on with the production of "inventory" cars is a planned activity which is accounted for in the Q4 guidance, including overall deliveries and GM. So bottom line is that I do not believe we have any data indicating Tesla is at risk of not meeting this guidance.

I agree with you that any changes in producing inventory cars is planned. Whether it is a one quarter phenomenon because of the manufacturing shutdown to switch to AP2 or is a change in mix between custom orders and inventory sales remains to be seen. I'm sticking with my theory that the incoming rate of custom orders is less than the increased production capacity so Tesla is capturing additional demand with cars that are immediately available for purchase.

I find it very interesting that there are no inventory 60's available. First off if Tesla was concerned about insufficient demand to meet the Q4 guidance wouldn't they be making the cheapest cars available to take advantage of price elasticity? And you certainly wouldn't raise the price of your least expensive model like they did last week.

It appears that Tesla is doing a similar thing with inventory cars to what they used to do when demand for custom orders exceeded production capacity: make the more expensive models available sooner in order to potentially capture additional gross margin. Let's hope it works!

 

[AnxietyRanger]

Well, I have to conclude that the data about "a lot" of cars built for inventory is pretty soft. I think that drawing any conclusions based on this is premature to say the least.

We know Tesla built a significant number of inventory cars during Q2/Q3 and seems to have repeated that in Q4, this is based on known inventory stock as marketed by Tesla (as well as the VIN conversation). We know plenty of inventory cars were sold in Q3. Tesla even built an inventory system to the design studio/configurator/online orders system. At this very moment there is, what, 1000+ factory-miled Teslas are available from Tesla and judging by the rate they keep going and keep being replenished, that is just a small window to a larger whole.

If Tesla were merely fulfilling custom orders with production to the max (being production constrained), there simply would be no need for such inventory building. Obviously Tesla wants to build significant-sized inventory for some reason. I can not see how anyone could really question this?

I'm sticking with my theory that the incoming rate of custom orders is less than the increased production capacity so Tesla is capturing additional demand with cars that are immediately available for purchase.

Indeed. It truly seems the days of supply constrained, build to order only are over for Model S and now other methods are used to keep the demand going in the required direction. Nothing wrong with that.

 

[vgrinshpun]

We know Tesla built a significant number of inventory cars during Q2/Q3 and seems to have repeated that in Q4, this is based on known inventory stock as marketed by Tesla (as well as the VIN conversation). We know plenty of inventory cars were sold in Q3. Tesla even built an inventory system to the design studio/configurator/online orders system. At this very moment there is, what, 1000+ factory-miled Teslas are available from Tesla and judging by the rate they keep going and keep being replenished, that is just a small window to a larger whole.

If Tesla were merely fulfilling custom orders with production to the max (being production constrained), there simply would be no need for such inventory building. Obviously Tesla wants to build significant-sized inventory for some reason. I can not see how anyone could really question this?

Do you acknowledge that Tesla has a fleet of demonstrator and floor model cars? When you are using term "inventory cars" do you think that it is the same as "demo and floor model cars"? If not, how do you distinguish the two?

 

[neroden]

Indeed. It truly seems the days of supply constrained, build to order only are over for Model S and now other methods are used to keep the demand going in the required direction. Nothing wrong with that.

This is just because the price is too high now. I priced out a model S as close as possible to my model from 2013 and it costs >$10,000 more. If Tesla can cut the price they can go back to being supply constrained.

 

[AnxietyRanger]

Do you acknowledge that Tesla has a fleet of demonstrator and floor model cars? When you are using term "inventory cars" do you think that it is the same as "demo and floor model cars"? If not, how do you distinguish the two?

Obviously such a thing as a demo/floor model fleet exists, I just don't agree it is the inventory we are talking about here - although of course there is some overlap. Tesla's inventory stock does include some demo-mileage units, but a significant amount of these are brand new, unused cars, not demo cars being offloaded after use.

I feel we have seen enough to conclude Tesla is basically building what amounts to something like the U.S. dealer lot stock for other brands. They are purposefully pushing demand up by creating cars for quick sales - instead of simply operating demo/floor model logistics only.

 

[AnxietyRanger]

This is just because the price is too high now. I priced out a model S as close as possible to my model from 2013 and it costs >$10,000 more. If Tesla can cut the price they can go back to being supply constrained.

Price is definitely a factor. If it is as much a factor as you suggest, hard to say. There is plenty of reason to think the large, high-end sedan/hatch form-factor to be limiting as well. Tesla is already well beyond a regular share of that limited market segment (thanks to the unique no-competition EV form) - it is not impossible they would reach a natural plateau at some point, after which more sales require marketing/selling, rather than just taking orders... I would say they have reached that already for Model S, though with effort they can of course increase the sales a lot still.

 

[vgrinshpun]

Obviously such a thing as a demo/floor model fleet exists, I just don't agree it is the inventory we are talking about here - although of course there is some overlap. Tesla's inventory stock does include some demo-mileage units, but a significant amount of these are brand new, unused cars, not demo cars being offloaded after use.

I feel we have seen enough to conclude Tesla is basically building what amounts to something like the U.S. dealer lot stock for other brands. They are purposefully pushing demand up by creating cars for quick sales - instead of simply operating demo/floor model logistics only.

Well, I do not buy this unless you have precise answer to my questions. I am agnostic on the above. But what I see is that everybody throwing around terminology and seemingly quantitative judgement with no real data to back it up.

IMO the proper way to approach this would be to accept that there is fleet of demo and floor model cars and get an estimate on how large this fleet and how many cars need to be sold to maintain this fleet fresh under normal circumstances. Than one need to figure out what major updates like elimination of nose cone, addition of AP, addition of AP2 do to the refresh rate. This would be a base to determine how many demo and floor model cars Tesla needs to sell on ongoing basis.

Next step would be to determine are there inventory cars that are NOT demo/floor model cars. If yes, how many of these are being sold and how this compares with sales of demo / floor model car.

Without any of the above, when a read "inventory" cars and "a lot" I just see that there are a lot of words being thrown around with little data to back them up, and this leads to what seemingly is accepted as forgone conclusion, but in reality it is far from that.

 

[vgrinshpun]

I agree with you that any changes in producing inventory cars is planned. Whether it is a one quarter phenomenon because of the manufacturing shutdown to switch to AP2 or is a change in mix between custom orders and inventory sales remains to be seen. I'm sticking with my theory that the incoming rate of custom orders is less than the increased production capacity so Tesla is capturing additional demand with cars that are immediately available for purchase.

I find it very interesting that there are no inventory 60's available. First off if Tesla was concerned about insufficient demand to meet the Q4 guidance wouldn't they be making the cheapest cars available to take advantage of price elasticity? And you certainly wouldn't raise the price of your least expensive model like they did last week.

It appears that Tesla is doing a similar thing with inventory cars to what they used to do when demand for custom orders exceeded production capacity: make the more expensive models available sooner in order to potentially capture additional gross margin. Let's hope it works!

What do you think the current production capacity for Model S is?

 

[AnxietyRanger]

Without any of the above, when a read "inventory" cars and "a lot" I just see that there are a lot of words being thrown around with little data to back them up, and this leads to what seemingly is accepted as forgone conclusion, but in reality it is far from that.

Without going into a tit-for-tat on the details, the same applies in reverse: dismissing the notion of a separate inventory for sales due to lack of hard-enough evidence (whatever the subjective understanding of that may be) also results in lots of words being thrown around. It is obvious to me that barring access to confidential records, our visibility into a company will always be fraught with uncertainties.

I would argue there more evidence to suggest Tesla is building inventory for sales (beyond demo fleet needs) than there is of the reverse. Already today can one buy factory-miles AP2 inventory from Tesla. No way has this been through any meaningful demo fleet/floor model use. Q3 of course was filled with similar examples pre-AP2.

 

[AnxietyRanger]

What do you think the current production capacity for Model S is?

My answer would be: The current production capacity for Model S is larger than Tesla's custom order demand is, in the current markets. This is evidenced by pulling numerous demand levers this year as well as a shorter than industry average production time for custom orders delivered near the factory (less than a month is not rare, for some even a week). I feel we can say this with certainty, even before going into any numbers.

It is only logical. Had Tesla been production constrained in Q3, they would have not needed a sales push (which I'm sure all agree there was in spades) to reach those numbers they got. Were they production constrained, new orders would take months and months, instead of weeks to destinations near the factory. (I'm excluding distant orders because overseas logistics adds an unrelated layer into the schedule.)

Tesla of course has a number of demand levers and marketing/sales options they can use to push that demand upwards, it just no longer only inherent demand, but demand that also needs to be driven - as is normal for their competitors.

 

[vgrinshpun]

Without going into a tit-for-tat on the details, the same applies in reverse: dismissing the notion of a separate inventory for sales due to lack of hard-enough evidence (whatever the subjective understanding of that may be) also results in lots of words being thrown around. It is obvious to me that barring access to confidential records, our visibility into a company will always be fraught with uncertainties.

I am not dismissing anything, and clearly stated that I am agnostic on this issue. IMO the fact that there is so much defensiveness in response to basic questions is an indication that the conclusion about a "lot of inventory cars" is not forgone by far, and more over there is no even an attempt to define what distinguishes "inventory car" from "demo / floor sample car".

I would argue there more evidence to suggest Tesla is building inventory for sales (beyond demo fleet needs) than there is of the reverse. Already today can one buy factory-miles AP2 inventory from Tesla. No way has this been through any meaningful demo fleet/floor model use. Q3 of course was filled with similar examples pre-AP2.

What I am arguing is that without knowing what is the demo fleet refresh rate , augmented to account for major updates like elimination of nose cone, addition of AP, addition of AP2, one can't seriously insist that Tesla building a lot of "inventory cars". I believe that data collection should precede forming conclusions, not the other way around.

 

[AnxietyRanger]

If your suggestion is, the recent inventory builds were motivated by fleet refreshes, that is certainly possible. It remains to be seen in the long run.

But again, were Tesla supply limited, those inventory builds and use of demand levers should have resulted in major delays in custom orders - but in Q2 and Q3 they did not, suggesting Tesla's Model S numbers are not limited by production, but in fact by their ability to generate demand.

 

[Spidy]

If Tesla was refreshing the fleet, why are they listing the new 50 miles cars just build a few weeks ago? Shouldn't they list the old cars that are getting replaced? Why refresh with AP1 cars at this time (600 in Europe/HK)? Out of more than 700 Inventory cars in Europe just around 50 have more than 50 miles/km.