Monthly Archives: February 2018

You are browsing the site archives by month.

The case for Tesla deferring 200K until July

So I have long been arguing that Tesla would cross the 200K US vehicles barrier in 2Q18, despite optimistic estimates that they would cross it in the first quarter.  Some enthusiasts even thought they would cross it in the final few weeks of 2017!  With the latest news from Tesla, I don’t think anyone believes it will occur in the first quarter any more.

With the latest news from Tesla, there are now some people speculating that Tesla may be able to defer hitting 200K until the third quarter (July 1).  Initially I was definitely not in that camp.  In fact in my post two days ago, I specifically thought that Tesla would hit the 200K mark in late April and there was no way they would defer.

However, between new speculation on what it means that some Canadians are getting earlier delivery estimates for AWD models that even those of us in the US, combined with news that a key Ontario EV tax credit is expiring in June, and a fellow forum member who kindly reminded me that there is a lag between an increase in production at the Gigafactory and actual sales, I have revised my thinking and run the numbers.  The results indicate that it is at least plausible that Tesla may try to do this.

Here is what that would mean:

  1. As was the assumption before (prior to the extended production delays) it will be necessary to divert additional Model S and X overseas above and beyond what Tesla would normally do.  I am assuming a 20% figure.  This will be somewhat hard to swallow for US buyers due to the extremely low inventory of Model S/X already, so build times might climb past 8 weeks from order to delivery.
  2. I am assuming that a best case scenario for the new equipment to be installed and running at the Gigafactory is the end of March.  This will allow the battery module production rate to spike to 2,500 per week, but because the cars using those modules still need to be built, sent to a distribution center, prepped and actually delivered, I am assuming a four week lag before we start seeing the sales ramp hit 2,500 per week.  So I effectively moved my step-wise ramp to the right by 4 weeks.
  3. Finally the key aspect of this plan would be to divert a significant number of Model 3s to Canada.  For simplicity I assumed that up to 2,000 per week (subject to actual production rate) would be diverted to Canada, starting in April through the end of the second quarter (June 30).  This adds up to a total of 23,000 vehicles that otherwise would have been destined to the US, and effectively meaning there would be a 10 week pause in US deliveries before significant numbers were seen again.  During the second quarter, only 9,500 deliveries would be made in the US.  Of course Tesla wouldn’t have to follow this exact plan, but the effectively the numbers would have to work out roughly this way.

With those assumptions, I have Tesla selling a total of 200,246 vehicles as of June 30th, so with only a day or two of halted selling, they could defer selling #200,000 until July 1st.

This is what the ramp I’ve described looks like visually:

The drop in Model 3 sales in April and May represents the diversion to Canada.  The increase in late June represents the factory throughput kicking into high gear, not an end to the diversion to Canada.  After the end of the quarter, this rate would further increase to the full 5,000 per week (6,000 across the whole product range).

The cumulative sales look like this:

The slowdown in sales to the US is noticeable, but not tremendously so.  Of course the impact will surely be felt by those with April to June delivery estimates!

I’m certainly not at the point where I’d be willing to say this is how Tesla will play this, but there are certainly some indications that it may go this way.  As usual it will be very interested to see February sales estimates, any indication of the Model 3 ramp, and probably most importantly if and when Canadians start to get more specific delivery estimates and even invites to configure.  Certainly if we don’t see any of that happening by late March, this plan will not be put into play.  But until then, it’s something to think about.



How does the Tesla earnings report and subsequent delivery slip affect the tax credit?

Tesla released their 4Q17 earnings report yesterday and subsequently changed the delivery estimates for reservation holders.  How does this affect deliveries and Tesla’s progress towards the EV tax credit phaseout?

Let’s start by taking a look at the delivery estimate shift.  A few days ago, a few reservation holders that had Nov ’17-Jan ’18 delivery estimate windows noticed that their windows had slipped to Feb-Apr ’18.  In and of itself this was not surprising since January had already passed and they still did not have vehicles in their driveways!  In fact, the only surprising thing was that they hadn’t done this shift in early January.  With an approximately 4 week time from order to delivery, if they hadn’t configured by the first week of January, it wasn’t likely they would hit their window anyway.  The community on the various Tesla forums had already pretty much considered the delivery estimate windows to mean not delivery, but invitation to configure, with the invitation itself coming during the last few days of the window.  This caused many reservation holders to mentally add one month to the end of their window for the earliest possible delivery of their vehicle.

Also recall that delivery estimates were generally shifted by a month after the 3Q earnings call.  The single month slip was surprising because at the time they had announced a 3 month delay in their initial ramp estimates.    At the time I noted the discrepancy, but technically with a 3 month window, they could potentially absorb a 4 month delay.  When Tesla further announced another 3 month slip to the ramp, holding the one month delayed delivery windows seemed impossible, but hope was held out that Tesla was sandbagging on their ramp estimates and they would be able to bring production back into the windows.

The net is, given the fact that the delivery estimates were clearly not going to be hit, it really should not come as any big surprise that the windows were reset.  Taking into account the original one month slip and the current three month slip, we are now at a four month delay compared to original estimates.  And with a three month window, we are at least in the neighborhood of Tesla’s estimated six month slip in production ramp.  Hopefully this means we can go back to treating the delivery window as a delivery window and not an invite window.

So for example, my delivery estimates are shown above.  Before yesterday, the First Production estimate read Dec 2017 – Feb 2018, Standard Battery Mid 2018 and Dual Motor All-Wheel Drive had a specific Aug – Oct 2018.  It may have appeared that First Production moved out three months, but at this point the best I was hoping for was an invite on February 28th.  I knew my car would not arrive until late March at the earliest, and probably later than that based on my position in line.

It remains to be seen whether people in the Mar – May window start getting invites in March, but I think they will.  This would confirm that really nothing has changed, but rather that they just reset the delivery windows to more realistic values.  At least for First Production.  The SR and AWD options have apparently moved out more significantly, and this is troubling for those that were waiting for those options as it may mean it will not arrive in time for the full tax credit to be in effect.

So that brings us around to what Tesla said in their earnings report.  Basically they re-affirmed their production estimates of 2,500 vehicles per week by the end of March, and 5,000 vehicles per week by the end of June.  And they gave a specific timeline and explanation of the solution to the bottleneck at the Gigafactory.  The issue, they say, is in moving product between the Zone 1 and Zone 2 stations in the battery module assembly area.  They have built an automatic system at the Tesla Grohmann subsidiary in Germany that is up and running well.  It now has to be taken apart, shipped, re-assembled and brought up to speed at the Gigafactory.  Tesla offered a timeline of late March for this activity to conclude, and this will bring them to the 2,500 vehicle per week level.  Hopefully this is a conservative estimate on Tesla’s part, because that is a significant amount of work to do, and I’m sure there will be hurdles along the way.

After that point, they will start the climb to 5,000 vehicles per week.  The bottleneck then becomes the parts conveyance system at the Fremont factory–basically the machinery that brings all the parts to the assembly line to be added to the car.

This should produce a ramp that looks somewhat like the following:

I believe they have probably gotten to peak throughput using the semi-automatic method described at the Gigafactory, so we are likely to see a fixed limit of about 1,000 vehicles per week until the new equipment from Germany arrives.  Then the rate should shoot up to 2,500 and be more or less fixed until improvements to the conveyance system in Fremont are worked out, at which time we will see a few more steps on the way to 5,000 vehicles per week.

While the total number of vehicles built is somewhat lower than most estimates, it only makes a difference of about 2,000 vehicles (or less than half a week) by the end of June.  So I don’t think this is really the huge deal that the news is making it out to be.

What is significant, however, is Tesla’s apparent decision to prioritize AWD over the Standard Range battery.  I know that there are probably a lot of reservation holders out there that reserved based on the opportunity to get a $35,000 base model vehicle.  However, I think that the feedback Tesla has received so far from reservation holders that are deferring is that the AWD option is actually even more in demand.  So it would appear that they are moving availability of that option up at the expense of the SR option.


Post-January sales update

So it has been quite awhile since my last update.  Unfortunately I had some personal matters that came up and I was not able to post a year-end summary.

At any rate, 2017 went out with a bang.  Total US sales across all manufacturers totaled 26,107 which broke the record of 24,785 set last December.  I’m sure the threat of losing the EV tax credit didn’t hurt, although buyers who want an “instant return” on their EV tax credit would normally buy in December anyway.  The end of year is also traditionally a time for sales pushes as well.

Of note, the Chevy BoltEV sold 3227 copies, and the Model S and X had great a great month with 4,975 and 3,300 estimated sales respectively.

On the Model 3 ramp, the news could have been better.  Remember that Elon Musk initially predicted 5,000 Model 3s per week by year end.  Well, the reality turned out to be significantly less than that (which I don’t think surprised anyone).

Tesla did try their best to spin the disappointing news with marketing statements (that seemed like they had the desired effect, as many took the bait) such as their manufacturing lines reached a rate that extrapolated to over 1,000 units per week during the last few days of 2017.  The reality is that the actual production during the last week of 2017 was 793 units, not the 1,000 that most people quickly reading the statement assumed.

At the end of the month, however, only 1060 Model 3s were sold, with 860 “in transit”.  Considering there were also probably a few “in transit” at the beginning of the month, we can estimate that there were probably only 1600 or so Model 3s actually produced in December.  If 793 of those were produced in the last 7 days alone (and that was probably an all out push to generate good news), it doesn’t speak so much for the volume during the rest of the month.

Nonetheless, the news is at least positive that they were able to improve the production rate so dramatically.

This puts the total US sales for Tesla at the end of 2017 at about 161,571.

So how did the January numbers turn out?  As is the case every January, the big push at the end of the year, combined with lousy weather that is typical in January and post-Christmas bills, sales were down.  Way down.  Across the board.

Sales of the Chevy BoltEV plummeted to 1,177, just over a third of what they were the month prior.

Tesla Model S and X sales were estimated to be 800 and 700 respectively, slightly below what they were last January.  Model 3 deliveries were estimated to be 1,875.  Better than December for sure, but still nowhere near the milestone of 1,000 per week.  Rumors are that the factory was shut down for the first week in January, partly to give workers some well deserved R&R after the big December push, but also to perform routine maintenance on the assembly line machinery.  But still, even if you assume 3 weeks in January, and assume the same level of vehicles “in transit”, you’re still only talking an average of 625 vehicles per week.  Still a long way to go.

I actually feel that the number of vehicles “in transit” might be higher.  The reason is that Tesla is having to build out its delivery infrastructure as well.  The first Model 3s were delivered to relatively big delivery centers in California.  Now that vehicles are being distributed across the country, not only does the transit time itself increase, but you have to ship cars to many smaller stores and delivery centers that may not have the throughput that the ones in Fremont and LA have.  However, that excuse only works once.  Now that the pipe is full, we should expect to see more consistent deliveries.

So that puts us at around 165,000 vehicles from Tesla with only 8 weeks left in the quarter.  Even if Tesla were able to produce 4,375 vehicles per week today, they wouldn’t hit 200,000 before the end of the quarter.  I think we can safely put to rest any fear that they will hit 200K this quarter.

But do we need to consider the possibility of Tesla delaying all the way until July 1 to cross that barrier?

I don’t think so.  Assuming a ramp that gets them to 2.5K/week at the end of 1Q18 and 5K/week at the end of 2Q18, I predict they will hit #200,000 around April 25.  This is far too early in the quarter to talk about deferring until 3Q.  So I put the likelihood that they will hit it in 2Q extremely high.

As for other manufacturers, I think we really need to see where sales wind up after looking at February and March.  January’s sales were just too depressed to make any meaningful predictions at this time.  In particular, will Bolt sales end up picking up where they left off?  And what about the Nissan LEAF 2.0?  How well will that sell in the US market?

Stay tuned!