OT: Sad how difficult it was to find this. The media avoids linking to primary sources. For example, Engadget's 3 paragraph article contains no less than 5 links back to engadget.com and not a single mention of this source.
This is a press release which was likely delivered to wsj et al through a wire service they pay to deliver them piles and piles of press releases. Nobody quotes those as sources because nobody but news organizations that pay for them has access to them. A press release is also a rather bad primary source unlike, say, a research paper or a blog post.
Also the WSJ article pretty clearly used the press release as a source, so mentioning that is a separate issue from whether it is a good or bad source.
(unless you think the WSJ called Tesla and verified specific facts like "4,820 Model S and X vehicles were in transit" vs "The company said 4,820 vehicles were in transit at the end of the quarter")
The newswire is the distributor of the press release. It is equivalent to Tesla emailing all major news outlets; just more convenient and reliable. In short, Tesla is the source.
It's a first-person source, coming from the company itself. Thus it's inherently biased toward the company and not the best source. This is as opposed to when the news article itself is a primary source, which is achieved by a reporter actually reporting what is happening in-person at the source of the event.
Of course, in this case it's a bit hard, since the only insight into Tesla's production goals is what they report externally, barring leaks and whatnot. A press release from the company is pretty much the best we can get. But just because it's the best doesn't mean it's necessarily good.
And as the very last sentence of my second paragraph says, "just because it's the best doesn't mean it's necessarily good". Your original post says "it is extremely odd to say they are not good sources".
You do realize that Tesla is a publicly traded company, and what you are saying (that they would misrepresent production numbers) would be securities fraud. They publicly stated their goals:
https://www.forbes.com/sites/alanohnsman/2017/08/02/tesla-st...)
They did not reach their goals. No need for paranoia.
Where did I say that they misrepresented their production numbers, or committed securities fraud?
In any case it's not paranoia. No matter what the law is, a press release is a first-party source. It is inherently biased toward the company. By no means should it be compared to an independent source, such as a report from an auditor or independent regulators.
You said that the information regarding either production goals or production numbers from Tesla's website was "inherently biased toward the company and not the best source". Meaning the published reservations, projections or production numbers would have been misrepresented in some way to benefit the company.
If they did so, that would be securities fraud, by definition. What about their projections, which were publicly stated and published in the past, or their actual production, which was publicly stated in the article (both of which are subject to SEC regulation review and legal recourse) are inherently biased toward the company?
Press releases are usually free. Companies use third party providers to ensure that every investor/subscriber receives information at exactly the same time. Paying for company publications is extremely uncommon. For a public company I'm not even sure if that wouldn't be illegal.
You can still pay for premium services (via Bloomberg, Thomson Reuters, etc) but subscribing to company news releases can be easily done by just visiting a company's investor relations website.
> Engadget's 3 paragraph article contains no less than 5 links back to engadget.com and not a single mention of this source.
That's Engadget's house style. They forbid external links the the body of an article, only allowing links within their own site. The best you can hope for is a single hard-to-find link at the end.
They've done it this way for years. It sucks. It's one of the reasons I never read anything on Engadget except by mistake.
I found it quite easily in a few clicks actually. I too wanted to see the original so I just googled "tesla ir", clicked first link, then at the bottom of the page I saw 'Recent Releases' and clicked 'Tesla Q3 2017 Vehicle Deliveries and Production'
News organizations work because the interests of the many finance a broad array of articles. Micropayments for articles would disrupt and distort the motives of unbiased news. There would emerge a hierarchy of articles chasing clicks and editors that favored certain writers with privileged access to click-generating traffic. underperforming articles would cease, and eventually, we'd get CNN. Joking aside, the newspaper model finances diversity of content. Per article payments finance creating lowest common denominator drivel.
Literally no entity with more than one line of business has each line of business growing at the same rate. This implies that the successful LoBs are "clickbait" to subsidize the less profitable LoBs.
Familiar (counter) example - loss leaders in supermarkets.
Maybe I'm not savvy enough but I don't really see what they offer over free publications like Reuters or Bloomberg. I've been tempted to subscribe to NYT and/or The Economist in the past but they have less meaningful competition in my opinion.
I’d much less rather live in a world of ads, because that’s exactly what created the perverse incentive to press all emotional buttons possible in order to get more eyeballs/ad impressions (see: fakenews epidemic).
People who produce good content expect to be paid for their labors. Do you have a better business model than ads (which incentivize twisting wording around to create emotional-button-pressing, and therefore mouse-clicking, content) or for-pay (which is incentivized by paying customers who are actually interested in good information)?
If you consider accurate journalism as a necessity of a well functioning economy and democracy then an argument can be made that it should be government funded and independent. Think BBC.
I don’t think the current model is working as well as it could.
Looking at PBS [1] and CPB [2] budgets from 2014, it appears that PBS gets more than 1% from CPB (if I am reading it right) but it is still a small percentage. Thanks!
Every government should fund journalism, and every citizen should read from several news sources (from different governments). The BBC is great, but you should not get your news about Britain from the BBC, get it from say Le Monde or NYT. And vice versa for France and the US.
This is one of those platitudes that are designed to make you feel smart about yourself despite being pretty silly when taken seriously.
Consider this: The funding of the judiciary comes from the executive and legislative branches of government. Yet, funding is never brought up as an argument in discussions of whether the judiciary is independent.
The reason is of course threefold: there is a large societal consensus that the judiciary should be independent; the independence of the judiciary is established in constitutions; and political norms, as eroded as they may be in some parts of the West, aren't sufficiently eroded that politicians feel like they can play hardball over funding with the judiciary.
All these factors could be established for an independent, government funded media as well.
There is a subtlety in underst and who feeds them. Public broadcasters in established western countries are beholden to a social class, not to the government of the day.
Based on his track record, Musk often sets unrealistic timeline goals (and unrealistic engineering goals, but he usually meets those). This has happened multiple times with both Tesla and SpaceX, not surprised this happened.
You understand that missing material estimates has legal implications if you do it too many times? Regulators can and do start thinking that you have something to hide.
Tesla isn't a startup working out of someone's garage -- it is a large player in a heavily regulated capital market. Estimates and guidance are for investors, not for people on the street: they are supposed to mean something.
Material in this sense means that it has relevance to the share price of the stock. Tesla's Model 3 ramp up schedule is material information. You are right they could have not volunteered it, but since they did they have a duty to ensure it was the most accurate estimate they had at the time of release.
If it turns out that they knew or should have known they wouldn't make the schedule then they are liable for misleading investors.
They'll be able to show the calculations that led to the assumptions. As long as those are valid, regulators won't act.
But many public companies have discovered that conservative forecasts with positive surprises are better for the share price than optimistic forecasts with profit/production warnings.
As it turns out, they knew full well that fully automated production wouldn't be possible in the upcoming quarter when the statement was made (i.e. they were manufacturing cars by hand, without an assembly line).
They lied to investors to keep the stock price pumped. Not cool.
Considering how outlandish and unrealistic his goals are usually, I feel like this is not even worth reporting on. I feel like it would be better to report on when he actually achieves the original deadline he set out.
The running joke that I've heard is that Elon operates on Mars years – if you multiply his projections by 1.88x, that's frequently closer to the mark. :)
This isn't news, it's gossip. ~200 instead of ~1500 means the bootstrapping is a little off schedule.
It's stories like this and the attitude that they're important which gets us very destructive behaviors like putting production quota ahead of quality. Production misses are news when they're for substantial cause. Like a 10% miss for an S or X might be news.
Yes, but it is news for investors. Missing productions goals as a startup is different than missing production numbers as a public company that is arguably quite overvalued. It matters because if it effects their share price, that can have important downstream effects on their eventual ability to execute (e.g. secure more loans).
Edit: I am not trying to say that Tesla should be punished, but am specifically addressing that it is, in fact, news.
It is especially not news for investors. It encourages terrible behavior when people focus on this sort of dumb number. Projecting problems and solutions at the beginning of production is more or less impossible. Valuing a company based on the accuracy of what is litte more than random noise makes you a bad investor, making it a headline news story makes you a bad editor. It's only news if it shows an _actual_ problem instead of showing what it actually shows ... which is random chance pushing around production scaling schedules.
It's news because Tesla projected to have delivered 1,500 Model 3 by end of 3rd quarter. They missed that milestone. Simple as that.
I wouldn't read too much into this myself. Scaling production is hard. Regardless, it's an important metric that speaks to Tesla's ability to deliver, at least short term.
Are you saying one should not value a company based on what they're saying? Then in Q2 when they wrote "Based on our preparedness at this time, we are confident we can produce just over 1,500 vehicles in Q3", then we should not believe them. So why do they bother saying this? If they got an uplift then, they get punished a bit now, that's fair, life goes on.
The only question is how this affects the ongoing production ramp-up. A relatively short delay on the delivery of a thousand cars is barely a rounding error to the underlying business case, and won't have any lasting consequences once they've ramped up production to full speed.
Significant portion of TSLA float is short https://ycharts.com/companies/TSLA/short_interest and that's before we get into the options activity, so any whiff of news usually impacts those side games in a significant way.
There's a strong incentive for either side to overblow a piece of news that fits their agenda.
Yield is a real mass production problem, not an imaginary one. You can verify it with your local manufacturer. For new products, new facilities, and new production methods scaling up the production often requires engineering breakthroughs.
How is being ~80% under quota not news? Such a large miss means that either Tesla is being unrealistically optimistic about their production capability, or they have serious production problems.
Because "~80% under quota" isn't a meaningful way to assess the schedule slip. They're still a few weeks/months off from mass production of the Model 3.
There are something like half a million Model 3 preorders. They just started production, 200 or 1500 – both are but a rounding error compared to the actual full production rate. (order of 10k/quarter)
It's something like criticizing someone for running 75 feet instead of 400 in the first 45 seconds of their third marathon because they had to re-tie their shoes.
Then what isn't an important number? Is everything worth a headline? How do you select what is and is not noteworthy or important for an investor or news consumer?
Why is Tesla missing it's Model 3 targets at the very beginning of the production run (hundreds of cars) more relevant than the Model S and X productions exceeding targets by "several thousand"?
Looking at the first two columns of WSJ tech page that lists that article ... articles are on (Apple, Tesla, Google, Facebook, Goldman Sachs/Bitcoin, Uber). Looking at that list of companies, would you guess that WSJ's editors are pushing the most relevant and important information, or the information their readers want the most? One is gossip, the other is news. People gravitating towards the news they want to see and news producing information tailored for their audience is the greatest force of evil in the 21st century. Popularity drives every news outlet to the bottom until you can't tell the difference between TMZ and WSJ.
Where do I go if I want to read actual news instead of the business gossip column?
> There are something like half a million Model 3 preorders.
Tesla is probably the only car company where the number of pre orders is deemed to be important. Car companies make money with sales and if they don't get production ramped up in time (before more alternatives become available), a lot of potential buyers will order somewhere else. Production capacity and consistent quality are the only important factors for Tesla, not how many people think about buying one at some point (in my opinion).
Only 220 Model 3 cars in an entire quarter? There's a big problem somewhere. That sounds like some key part of the production line doesn't work at all, and something is being hand-built to get some product out the door. Tesla isn't saying what's not working. Has anything leaked out?
When something like this this happens, the manufacturer has most of the costs of full production with the revenue of tiny production. Profitability just got further away.
As of August 2017, there have been about 63,000 cancellations of Model 3 preorders.[1]
It's extremely unlikely there's a big problem in the sense that you mean it. The Model 3 is easier to produce than the Model S and less complex. There's practically nothing new on the Model 3 that would plausibly be a big thing that's not working as to hold everything up in the scandalous manner you're implying (leaks, hand-building, not working, big problems, oh my).
More likely is that Musk set an overly ambitious target, for the hundredth time, and what they're doing for the quarter they probably should not have been doing at all. They produced something like 60-80 Model S vehicles in the first month. Check back in two quarters, if the Model 3 production is down 80%-90% from projections, that will be something meaningful.
I don't think OP's post is as hyperbolic as you characterize it.
The press release makes it clear that there was a major production bottleneck. Will it get fixed? Very likely. But as of today, they've got a serious problem.
Absent more data, I think the OP's guesses are plausible and not malicious.
Rather than a bottle neck, I'd say the production line wasn't up for the majority of the quarter. At 220 cars in ~120days, that is an 8-12 hour average takt time. Since nobody does that in production, it's more likely that they take the line down, fix things, and bring it up again days later. I'd be more interested in the change in delivery rate over the 3 months. If 218 cars were delivered on the last day, it would be 6k/mo or 70k/yr, great! More likely progress has been slow and hopefully steady (30=>60=>90 per month) for their ramp.
Workers at Fremont are complaining of overwork, time pressure, unsafe working conditions, and long hours. They're trying to join the UAW. It's not like they have production employees standing around doing nothing while the production engineers try to fix the line. "I’ve seen people pass out, hit the floor like a pancake and smash their face open. They just send us to work around him while he’s still laying on the floor."[1]
Comments on Glassdoor indicate that lower and middle management are disorganized.
It sounds like “kitchen chaos” startup mode while their “TPMS” workflow, tooling, fixtures and techinques get figured out for higher capacity, routine production. Unionizing is important because people shouldn’t be treated like animals and should make a living wage.
Regarding leaks, yes, there were rumors of battery pack replacements due to corrections on welds and a separate production issue with ground terminal bolts.[1]
These issues could be due to the different composition of the 3 compared to the S and X - more steel, less aluminum - or with them trying to build a denser automated production line, or just general gremlins setting up new production.
Is anyone inside the plant talking about what's wrong? That's what I meant by "leaks".
(Back when newspapers had real reporters, some reporter would go to a bar near the plant, buy some of the guys a few beers, and listen. The recent stories about Tesla are all from press releases or CEO interviews.)
As a car company with some years of experience you should be able to factor in uncertainties. They could've made the estimate less optimistic, giving some upside in case everything goes right. Using the best case scenario as your base case is not necessarily the smartest thing (but can be good for marketing).
Might very well be exactly what they did; it might be that 1500 was the tip of the bathtub curve so to speak and 260 was on the low end. We'll never know.
The 31% increase for the entire year is actually holding S+X production flat from the last 2 quarters of 2016. Which is apparently being done on purpose to put resources into the Model 3 launch, and, there are signs of softening S+X demand, like a lot of S+X cars being built for inventory. This softness hasn't made its way into reduced sales, but who knows...
106 comments
[ 3.6 ms ] story [ 175 ms ] threadOT: Sad how difficult it was to find this. The media avoids linking to primary sources. For example, Engadget's 3 paragraph article contains no less than 5 links back to engadget.com and not a single mention of this source.
Anyone can setup a real time account for free.
Also the WSJ article pretty clearly used the press release as a source, so mentioning that is a separate issue from whether it is a good or bad source.
(unless you think the WSJ called Tesla and verified specific facts like "4,820 Model S and X vehicles were in transit" vs "The company said 4,820 vehicles were in transit at the end of the quarter")
Of course, in this case it's a bit hard, since the only insight into Tesla's production goals is what they report externally, barring leaks and whatnot. A press release from the company is pretty much the best we can get. But just because it's the best doesn't mean it's necessarily good.
They did not reach their goals. No need for paranoia.
https://en.wikipedia.org/wiki/Securities_fraud
In any case it's not paranoia. No matter what the law is, a press release is a first-party source. It is inherently biased toward the company. By no means should it be compared to an independent source, such as a report from an auditor or independent regulators.
If they did so, that would be securities fraud, by definition. What about their projections, which were publicly stated and published in the past, or their actual production, which was publicly stated in the article (both of which are subject to SEC regulation review and legal recourse) are inherently biased toward the company?
You can still pay for premium services (via Bloomberg, Thomson Reuters, etc) but subscribing to company news releases can be easily done by just visiting a company's investor relations website.
That's Engadget's house style. They forbid external links the the body of an article, only allowing links within their own site. The best you can hope for is a single hard-to-find link at the end.
They've done it this way for years. It sucks. It's one of the reasons I never read anything on Engadget except by mistake.
The comment was not regarding the quality of Google's search engine results for given keywords.
Familiar (counter) example - loss leaders in supermarkets.
People who produce good content expect to be paid for their labors. Do you have a better business model than ads (which incentivize twisting wording around to create emotional-button-pressing, and therefore mouse-clicking, content) or for-pay (which is incentivized by paying customers who are actually interested in good information)?
I don’t think the current model is working as well as it could.
Corporation for Public Broadcasting gets $450 million/year.
PBS: “less than 1%” of its “operating budget comes in the form of grants from CPB and federal agencies and departments.”
1. http://www.pbs.org/about/about-pbs/financials/
2. https://en.wikipedia.org/wiki/Corporation_for_Public_Broadca...
Whether it's conscious, most people/groups/governments have an easier time rationally evaluating others vs themselves. Just like I'm doing right now.
Consider this: The funding of the judiciary comes from the executive and legislative branches of government. Yet, funding is never brought up as an argument in discussions of whether the judiciary is independent.
The reason is of course threefold: there is a large societal consensus that the judiciary should be independent; the independence of the judiciary is established in constitutions; and political norms, as eroded as they may be in some parts of the West, aren't sufficiently eroded that politicians feel like they can play hardball over funding with the judiciary.
All these factors could be established for an independent, government funded media as well.
You can be government funded or independent, but not both.
Pick one, please.
The German public broadcasters are well known for biting the hand that feeds them.
Tesla isn't a startup working out of someone's garage -- it is a large player in a heavily regulated capital market. Estimates and guidance are for investors, not for people on the street: they are supposed to mean something.
Are you sure about that? The only thing I start thinking is that the person making the estimates is being consistent.
Other manufacturers don't really do that, which is why you rarely see the headlines of Dodge Durango being three weeks behind schedule.
If it turns out that they knew or should have known they wouldn't make the schedule then they are liable for misleading investors.
But many public companies have discovered that conservative forecasts with positive surprises are better for the share price than optimistic forecasts with profit/production warnings.
They lied to investors to keep the stock price pumped. Not cool.
You don't read safe harbor statements?
It's stories like this and the attitude that they're important which gets us very destructive behaviors like putting production quota ahead of quality. Production misses are news when they're for substantial cause. Like a 10% miss for an S or X might be news.
Edit: I am not trying to say that Tesla should be punished, but am specifically addressing that it is, in fact, news.
I wouldn't read too much into this myself. Scaling production is hard. Regardless, it's an important metric that speaks to Tesla's ability to deliver, at least short term.
Some investors will value on random noise, and some other investors will figure this out and capture the mispricing, if there's any.
There's a strong incentive for either side to overblow a piece of news that fits their agenda.
It's something like criticizing someone for running 75 feet instead of 400 in the first 45 seconds of their third marathon because they had to re-tie their shoes.
Why is Tesla missing it's Model 3 targets at the very beginning of the production run (hundreds of cars) more relevant than the Model S and X productions exceeding targets by "several thousand"?
Looking at the first two columns of WSJ tech page that lists that article ... articles are on (Apple, Tesla, Google, Facebook, Goldman Sachs/Bitcoin, Uber). Looking at that list of companies, would you guess that WSJ's editors are pushing the most relevant and important information, or the information their readers want the most? One is gossip, the other is news. People gravitating towards the news they want to see and news producing information tailored for their audience is the greatest force of evil in the 21st century. Popularity drives every news outlet to the bottom until you can't tell the difference between TMZ and WSJ.
Where do I go if I want to read actual news instead of the business gossip column?
Tesla is probably the only car company where the number of pre orders is deemed to be important. Car companies make money with sales and if they don't get production ramped up in time (before more alternatives become available), a lot of potential buyers will order somewhere else. Production capacity and consistent quality are the only important factors for Tesla, not how many people think about buying one at some point (in my opinion).
1. https://www.wsj.com/news/author/8483
When something like this this happens, the manufacturer has most of the costs of full production with the revenue of tiny production. Profitability just got further away.
As of August 2017, there have been about 63,000 cancellations of Model 3 preorders.[1]
[1] https://www.recode.net/2017/8/2/16087432/tesla-model-3-elect...
More likely is that Musk set an overly ambitious target, for the hundredth time, and what they're doing for the quarter they probably should not have been doing at all. They produced something like 60-80 Model S vehicles in the first month. Check back in two quarters, if the Model 3 production is down 80%-90% from projections, that will be something meaningful.
The press release makes it clear that there was a major production bottleneck. Will it get fixed? Very likely. But as of today, they've got a serious problem.
Absent more data, I think the OP's guesses are plausible and not malicious.
Comments on Glassdoor indicate that lower and middle management are disorganized.
[1] http://www.businessinsider.com/tesla-factory-workers-detail-...
These issues could be due to the different composition of the 3 compared to the S and X - more steel, less aluminum - or with them trying to build a denser automated production line, or just general gremlins setting up new production.
[1] https://electrek.co/2017/09/29/tesla-model-3-production-ramp...
(Back when newspapers had real reporters, some reporter would go to a bar near the plant, buy some of the guys a few beers, and listen. The recent stories about Tesla are all from press releases or CEO interviews.)
> In total, we expect to deliver about 100,000 Model S and X vehicles in 2017, which would be a 31% increase over 2016.
31% increase from 2016, that's a huge bump in delivery and I wonder how/if this affected the model 3 production.
Looks like they have a long way to go before they will be a threat to the big car companies..