Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable

This site may earn commission on affiliate links.
I'm pretty sure that unlike Reddit the thumbs up "Count" just counts reactions... including the thumbs down icon. So that count doesn't really mean what you think it means.

I haven't seen anyone with a negative score and I'm pretty sure a few people around here would definitely be negative or at least permanently zero.

(Everyone should Thumbs down this post to see if it reduces me below my current total of 17,013...)

I added one step to your experiment. I thumbed you down and then refreshed your count and it was EXACTLY as you said! It's just a reactions counter, increments on ANY reaction.

Edit: Thanks to jw934, it seems that my test happened to coincide with someone else. disagrees did NOT add to the reaction score! SORRY!
 
Last edited:
How would replacing already purchased stamping dies and a fully installed and functional robotic body assembly cells with a new gigapress reduce the per vehicle cost 3-7k?
Casting lowers CapEx for a new plant, but for an existing line, I don't think the impact is that positive (if at all since you still carry the old equipment cost unless it gets reused).
Simple, don't throw away the old robots from the old body live that are no longer needed. Ship them to another new line/factory instead of buying new bots. Yes there will be new costs for the new stuff to support giga castings, but that improves margins in many ways going forward as already pointed out by others, plus you get to reduce costs elsewhere with the reused bots. It shouldn't take long to measurably improve things, at the rate they're going.
 
Not everyone in the UK supported Brexit, the true Brits will have a 'Stiff upper lip'. Ex-Pat Scot.
Sorry, in case any one missed the irony, please be in no doubt that I think Brexit is a disaster that has already damaged the UK and is going to get a lot worse for the people in the UK.

The relevance to Tesla is that Tesla was sufficiently sensible to eliminate UK as a candidate for a Tesla factory because of Brexit.

More discussion - or more to the point, monitoring of the unfolding slow motion trainwreck - is available on the TMC Brexit thread at Brexit
 
Why does France block these thing?



What is scary, the lack of manufacturing capacity, the huge order backlogs? Just trying to understand.
Re France : Quite simply because it is in France's narrow economic interests to block economic industrialisation of Spain. In French eyes, much of southern Frace is quite similar to much of Spain, and since they want investment/jobs/etc to flow into France it is a short term interest to block strategic developments that help Spain industrialise so as to become a more viable and attractive competitor. This is a very longstanding de facto policy, perhaps a century old. It has had a specific twist to it in the sphere of energy (protect French nuclear demand, so restrict oil/gas pipes and HV grid); rail/ports (protect French rail & port networks); auto (cripple rail-traffic); and of course general road. Additionally development in the Pyrenees border region of France gets all sorts of single-issue lobby groups fired up in opposition to it, and so it takes expenditure of scarce political capital to overcome. This is a more recent thing, primarily since the 1970s. Hence cross-border projects in this area have a habit of going nowhere. I happen to disagree with French policy as it prioritises short-term French interests over long term French interests. I've lived in the relevant areas of both France and Spain by the way, so I'm well aware of the local politics and the various schemes-to-nowhere. Plus I am in and out of enough factories in both France and Spain.

(a Tesla Gigafactory in Spain is a possibility imho, France less so)

Re HV grid : As an example one client (prob the largest manufacturer of HV grid widgets in the world) took me on a tour of their factory that was running at max capacity and quoting lead times of 5-years + . They explained that their clients wanted them to as a minimum triple capacity and were begging for even more than that. However the also know that $bn investments in a new factory and its associated workforce are long term projects that need at least 20-years of stability to be truly profitable (the requisite workforce with particular specialist skills doesn't just appear from nowhere; nor do the specialist production machinery). As a result they had deferred the decision on whether to build one new factory (on the same site) by a year or so, because they know how fickle the clients are. The clients of course are typically nation-state-owned HV grid companies, and they have a habit of truncating the budget for large scale infrastructure projects whenever there is a change of politicians. So that manufacturer would rather keep one factory fully occupied (and risk competitors entering the space) than risk $bn on doubling or trebling its production capacity and end up with stranded assets in a boom-bust big-ticket capital goods cycle. Knowing the status of pretty much all these factories around the world (because I ran a factory that was a critical supplier to all of them, and I was mulling my own corresponding capacity planning investments) I can say confidently that the only market that can be relied on over a multi-decadal timescale has been the Chinese market. At the senior levels of decision making in these organisations we are all very aware of the potential for batteries to restructure the grid plus many other technical/economic factors that impinge on the area. Anybody that deals with HV-stuff is well aware of the consequences of failure - in so many ways - and so decision making tends to be very conservative. (As a comparison making 1.2 million volt x GW-capacity grid widgets is the equivalent of making 5-nano-metre chips; and there are equally few fabs that can do it; and the science and engineering and costs are similarly large). The Chinese politicians are the only ones that get this and have set a corresponding long term industrial strategy that gives 'us' (globally, collectively) the necessary confidence to invest to deliver the necessary. One sees this in all the other large scale long term infrastructure plays in China as well, which are in many ways inter-related. I think it is only in the last few years that the Chinese politburo has included people who are not either engineers or econoimists, and it shows. Go and look at the state of your own country's grid (almost anywhere in the world), consider what I am saying, and you too will likely find this scary.

(This tends to play into the hands of Tesla Energy, if they can get their sh1t in one sock which so far has a poor track record)
 
Last edited:
Q3 exchange drop as of today are as follows.

Euro to USD: 7% drop
Yuan to USD: 6% drop
Pounds to USD: 13.6% drop

If Theaccountant can plug some of these in to see what kind of potential GM will be affected. Thanks!
The income statement is converted using Average Exchange rates for the Qtr (not the Qtr end rate).
For example, the Euro to USD:
Q2 Avg 1.066
Q3 Avg 1.047 (as of today)
That's only a 1.8% drop.

Since both Revenues and Costs (e.g. cost of sales from Berlin) are both translated to USD, the impact should not be material to the financial statements.
There is a larger impact with the Yuan (-3.3%) but again since revenue and costs are both impacted, should not be material.
By building factories in Europe and China, Tesla has created currency protection. If revenues decline because of exchange rates, then costs decline too.
 
The income statement is converted using Average Exchange rates for the Qtr (not the Qtr end rate).
For example, the Euro to USD:
Q2 Avg 1.066
Q3 Avg 1.047 (as of today)
That's only a 1.8% drop.

Since both Revenues and Costs (e.g. cost of sales from Berlin) are both translated to USD, the impact should not be material to the financial statements.
There is a larger impact with the Yuan (-3.3%) but again since revenue and costs are both impacted, should not be material.
By building factories in Europe and China, Tesla has created currency protection. If revenues decline because of exchange rates, then costs decline too.

My reply here generated a thought. Tesla gets significant revenue from Canada but there are very little Tesla expenses in Canada.
It may be one contributing factor to build a factory in Canada. A factory in Canada would result in expenses denominated in Canadian Dollars creating protection from currency swings.
 
Interesting article here on who reports to Elon.


Looks like Pete Bannon has 22. Anyone know much about him? We all heard from drew baglino but has Pete been on air or been on the spotlight yet?
 
Interesting article here on who reports to Elon.


Looks like Pete Bannon has 22. Anyone know much about him? We all heard from drew baglino but has Pete been on air or been on the spotlight yet?
Looks like Elon not creating enough controversy lately
Kordony is looking for the next level to begin to smell their dirty laundry as well ;)
 
Interesting article here on who reports to Elon.


Looks like Pete Bannon has 22. Anyone know much about him? We all heard from drew baglino but has Pete been on air or been on the spotlight yet?
Pete Bannon is very well known and respected having worked with Jim Keller for a long time at a number of stops. He helped develop the Apple A series chips for one.
 
Re France : Quite simply because it is in France's narrow economic interests to block economic industrialisation of Spain. In French eyes, much of southern Frace is quite similar to much of Spain, and since they want investment/jobs/etc to flow into France it is a short term interest to block strategic developments that help Spain industrialise so as to become a more viable and attractive competitor. This is a very longstanding de facto policy, perhaps a century old. It has had a specific twist to it in the sphere of energy (protect French nuclear demand, so restrict oil/gas pipes and HV grid); rail/ports (protect French rail & port networks); auto (cripple rail-traffic); and of course general road. Additionally development in the Pyrenees border region of France gets all sorts of single-issue lobby groups fired up in opposition to it, and so it takes expenditure of scarce political capital to overcome. This is a more recent thing, primarily since the 1970s. Hence cross-border projects in this area have a habit of going nowhere. I happen to disagree with French policy as it prioritises short-term French interests over long term French interests. I've lived in the relevant areas of both France and Spain by the way, so I'm well aware of the local politics and the various schemes-to-nowhere. Plus I am in and out of enough factories in both France and Spain.

(a Tesla Gigafactory in Spain is a possibility imho, France less so)

Re HV grid : As an example one client (prob the largest manufacturer of HV grid widgets in the world) took me on a tour of their factory that was running at max capacity and quoting lead times of 5-years + . They explained that their clients wanted them to as a minimum triple capacity and were begging for even more than that. However the also know that $bn investments in a new factory and its associated workforce are long term projects that need at least 20-years of stability to be truly profitable (the requisite workforce with particular specialist skills doesn't just appear from nowhere; nor do the specialist production machinery). As a result they had deferred the decision on whether to build one new factory (on the same site) by a year or so, because they know how fickle the clients are. The clients of course are typically nation-state-owned HV grid companies, and they have a habit of truncating the budget for large scale infrastructure projects whenever there is a change of politicians. So that manufacturer would rather keep one factory fully occupied (and risk competitors entering the space) than risk $bn on doubling or trebling its production capacity and end up with stranded assets in a boom-bust big-ticket capital goods cycle. Knowing the status of pretty much all these factories around the world (because I ran a factory that was a critical supplier to all of them, and I was mulling my own corresponding capacity planning investments) I can say confidently that the only market that can be relied on over a multi-decadal timescale has been the Chinese market. At the senior levels of decision making in these organisations we are all very aware of the potential for batteries to restructure the grid plus many other technical/economic factors that impinge on the area. Anybody that deals with HV-stuff is well aware of the consequences of failure - in so many ways - and so decision making tends to be very conservative. (As a comparison making 1.2 million volt x GW-capacity grid widgets is the equivalent of making 5-nano-metre chips; and there are equally few fabs that can do it; and the science and engineering and costs are similarly large). The Chinese politicians are the only ones that get this and have set a corresponding long term industrial strategy that gives 'us' (globally, collectively) the necessary confidence to invest to deliver the necessary. One sees this in all the other large scale long term infrastructure plays in China as well, which are in many ways inter-related. I think it is only in the last few years that the Chinese politburo has included people who are not either engineers or econoimists, and it shows. Go and look at the state of your own country's grid (almost anywhere in the world), consider what I am saying, and you too will likely find this scary.

(This tends to play into the hands of Tesla Energy, if they can get their sh1t in one sock which so far has a poor track record)

Wow, thanks for the complete answer. Very informative. I am guessing that China's lead is also extending in nuclear power plants? I had heard a while ago that they were building a lot of them. Other than the US, France and Canada, none of which are really building more of them, there doesn't seem to be a country that is has invested in nuclear.

On the other hand, nuclear is kinda yesterday's technology if you look at its need for massive transmission line facilities, which is in contrast with wind and solar which is mostly a local energy source (which, of course, need batteries to be considered a competitor to nuclear).

John Carmack of game programming fame, who recently started a general AI company, mentioned he had been exploring nano-scale nuclear. Meaning house or community level sized. The idea being that if designed properly, it need not be super efficient (so it isn't run on the knife edge of uncontrolled run-away fission), yet could just run with a completely sealed fuel pellet for a decade or so before being recycled. You'd use an isotope that would be useless for bomb making and other safety issues would need addressing, but I've always been interested in such a technology. Just like rooftop solar, removing the need for transmission cost removes half your costs right there.
 
Last edited: