I believe these estimates of global sustainable Model 3 demand by you and @Troy are still very conservative, due to: No leasing of Model 3 offered yet by Tesla. In the U.S. about 30% of all new car sales are leased cars - especially in the premium and business price segments. There's good reason to believe that the smaller Model 3 will be non-linearly more popular in Europe and in China than in the U.S. Based on competitor stats the Model 3 sedan form factor addressable market is twice the size of the U.S. market in Europe alone. Lower entry price increases the size of the addressable market non-linearly as well. If we read income/new-car-purchase histograms like this one: Then the addressable market of the $46k-$65k ASP Model 3 is at least 5 times the addressable market of the $90k-$150k Model S, and SR would cover about 60-70% of all sales. The Model 3 also seems to be particularly attractive to car buyers from much lower price segments - i.e. even the 70% estimate based on ICE consumer behavior might be low. This maps to sustainable global Model 3 demand of literally millions of units, per year. The market inaccessible to the Model 3 are the SUV buyers (50% globally) and <$35k buyers (another 50% of the rest). I.e. in the U.S. alone the upper bound for Model 3 sales at $35k and above are about 20-25% of all new light vehicle sales, which is about 3-4 million new units per year. I don't think we are going to see a permanent softening of Model 3 demand before 2025 or so.
New Tesla-internal email from Elon earlier today, about them expanding FSD beta testing: Notes: The first wave participants back in September were about 100-200 employees. They are now expanding that to at least 600 I think - maybe as much as 1,000 employees. They only needed 3 months to grow confident enough to initiate the final wave of FSD beta testing. Elon doubled down on his previous performance claims: "over 10 times more capable". Just 4 days left from Q4: this $8k incentive to Tesla employees could generate a few more inventory sales. Introduction of HW3 will roughly coincide with the introduction of the Standard Range option. Both should significantly boost demand. EAP+FSD option pricing is unchanged at $5k+$3k = $8k. My speculative guess: the first HW3 based FSD features will be released between March and July 2019. (My not so speculative guess: in the following months the probability of occasional FSD progress leaks will go up by an order of magnitude. ) Source: Reddit thread.
In Europe entry level luxury branded vehicles come standard with lower specifications and lower prices, almost to the mid price of cars such as Honda Accord( which is being discontinued in Europe) and Ford Fusion/Mondeo. In Europe a 2018 BMW 3 Series comes standard with a 3 cylinder engine, manual transmission, and cloth seating surfaces. In the US a BMW 3 Series comes standard with a 4 cylinder engine, automatic transmission, and vegan leather. If you eliminate these base powertrain models which compete more directly with mainstream sedans and compare Apples to Apples then sales are much closer to USA. Mainstream midsize sedan market is much bigger in the US than Europe. BTW 2019 BMW 3 Series is a bigger better equipped car with a significant bump in base price and projected ASP.
I'm wondering about this $8K cost for FSD that he mentions. Right now, it is available off-menu for $5K; does this mean a price increase is coming in the future?
They have a far inferior battery, no battery supplier (for volume), no ability to fast charge, and the vehicles are a mish mash of OEM buttontopia and Tesla we-don’t-need-no-sticking-buttons-knobs-or-dials such that the user experience is adversely affected - that’s what I’ve heard. So, I don’t highly recommend NIO shares. I’ve yet to decide if I’ll sell mine. Playing a bit of wait and see.
I went through current BMW pricing a couple of days ago: 'The Model 3 changes all that: LR AWD starts at around €58k, versus €108k for a 100D with similar range. Performance Model 3 at €68k versus €145k for a P100D. BMW is in trouble: the highest performance BMW M2, which is a smaller 2-door coupé, costs €74k+ in comparable configuration, the M3 and M4 cost €70-80k in Germany - and none comes close to the acceleration and handling of the PM3. The Model 3 also offers features unmatched by any BMW: Remote start/heating is illegal in most of Europe for ICE cars AutoPilot is unmatched, Interior is generous even for tall people and there's no drive shaft running in the middle like on AWD ICE cars, Remote cooling is unmatched and will be a hit as non-garage street parking is a lot more common in Europe, so hot cars in the summer are a real problem in all but Scandinavian countries. Frunk space. Frameless doors look "luxury American car" and are impressive AF. Only the highest end European sedans are frameless, and it's a rarity in general. Tesla back seats are foldable flat, offering camping sleeping space with air conditioning and dehumidification. "Camper Mode" under a Glass Roof is something most Europeans don't even know exists, but will value a lot, because the denser urban environments are more removed from nature than the more generous living spaces in most of the U.S. There's a few luxury and convenience features BMW offers that the Model 3 doesn't (such as surround camera view or high end luxury interior trims) - but none is major. I.e. the Model 3 expands the addressable market much more in Europe than it did in the U.S., even at the current $66k+ entry price for the LR AWD." Note that another big factor I didn't list in that post are significant fuel cost and maintenance savings, to which European car customers are much more sensitive than U.S. customers: $5-7 per gallon of gasoline in Europe and significantly lower free disposable income, versus $2-3 per gallon in the U.S. and higher income levels. I also posted about the hidden form factor advantages of the Model 3, which is not apparent from the raw dimensions alone. I expect the addressable market of the Model 3 to be surprisingly broad.
Elon "doubled down on his previous performance claims" is an English idiom meaning "he reiterated the previous 10x claim". So yes, nothing changed, which is my point: he didn't lower the previous estimates but reiterated them, giving the 10x claim more weight.
No, the current EAP+FSD pricing is $5k+$3k = $8k, so nothing changed there. $5k is the current post delivery FSD price.
That could be what it is. Since FSD for the public is off-menu, I only see the add-on cost as someone with a P3D with EAP only. So it might still be $3K at the time of order.
The only references in the filings to the accounting treatment of referrals (both automotive and solar) appear to relate to leases/installment sales (PPAs): "We capitalize initial direct costs from the origination of solar energy system leases or power purchase agreements (i.e. the incremental cost of contract administration, referral fees and sales commissions) as an element of solar energy systems, leased and to be leased, net, and subsequently amortize these costs over the term of the related lease or power purchase agreement... We capitalize shipping costs and initial direct costs such as the incremental cost of contract administration, referral fees and sales commissions from the origination of automotive lease agreements as an element of operating lease vehicles, net, and subsequently amortize these costs over the term of the related lease agreement. Automotive referral rewards could be treated similarly to warranties, i.e. set up a reserve at the time the obligations are incurred so the expense matches the time period when the revenue/income from the underlying product sales are recognized. The expense (reserve) could be charged to either SG&A as in Marketing etc. or (less likely) COGS --the latter choice lowers GM%. This approach would likely also show up with entries on the Balance Sheet for Accrued Liabilities/Long Term Liabilities. (There has been no mention of referrals in the MD&A explanations or Notes so this approach may not be what Tesla is using.) The other approach would be to treat referral rewards similarly to regulatory credits, but from a mirror image perspective, i.e. income vs expense and asset vs liability. This is essentially more of a cash rather than accrual approach. Nothing is recognized nor reported until the ultimate transaction occurs--for regulatory credits when the OEM buyer enters into a contract for the credits, and for referral rewards when the referrer receives the benefit in goods or services. The increase in referrals during 2018 is closer to a factor of four times rather than two times. With around 50 referrers earning enough for a free Roadster II, and more than a dozen of those earning enough for two Roadster IIs (and several with three or more), it's not inconsequential. The 10k should be interesting.
@Fact Checking Are those the cost or the MSRP? (As my understanding is that the MSRP gets large discounts in Germany, even larger than the US.)
Yes, but note that I set the 2x 2017 threshold clearly above expected non-referral marketing/advertising costs, but below the 4x cost which is not certain: As you noted it too, probably not all rewards are expensed immediately, for example Roadster 2 might be expensed in the future on a CoG basis once that cost is known. Some of the lower level and thus higher volume rewards were also reduced in 2018, notably the free supercharging reward to new customers. At $3k/unit reserve this was a significant cost - while 6 months of free supercharging should be much lower. I think free supercharging was the primary cost of the referral program. Considering all that 2x 2017 costs are a plausible estimate I think.