Editor's note: Seeking Alpha is proud to welcome The Battle Of Giants as a new contributor. It's easy to become a Seeking Alpha contributor and earn money for your best investment ideas. Active contributors also get free access to the SA PRO archive. Click here to find out more »
Tesla is struggling to produce enough Model 3s. The problem may stem from its production philosophy. A philosophy that's different from Toyota’s lean manufacturing system, which I believe tops on both quality and quantity.
Toyota developed “just in time” manufacturing to tackle three problems:
The manufacturing of vehicles involves thousands of parts. Each part has undergone its own unique production process and can cause problems to overall vehicle production
Car production involves many different models with numerous variations (options) and large fluctuations in demand for each variation
Vehicles are completely remodelled every few years
Tesla faces the same problems, but deals with them very differently.
Toyota Production System
Tesla Production System
What’s the goal?
To produce better quality goods having higher added value and an an even lower production cost than competitors.
To lower production costs by thinking about the factory as a product.
How do they achieve their goal?
Toyota attains low cost production through elimination of waste. Anything other than essential amount of equipment, materials, parts, and workers are considered waste.
One priority is to minimize dead time spent resetting and configuring machines necessary for short production runs of different car models. So it seldom uses sophisticated machines. Allowing workers to quickly tweak and improve processes. Toyota can in turn quickly change production schedule based on the latest sales trends.
One key characteristic of the process is Toyota’s emphasis on putting humans at center of manufacturing. Not robots. As Will James, president of manufacturing in Kentucky, shared: “Machines are good for repetitive things… but they can’t improve their own efficiency or the quality of their work.”
Toyota is effectively relying on continuous improvement to lower costs. Betting that building things right the first time, and constantly getting better at it, is cheaper than an automated production line that cranks out poor quality products they need to fix later.
Tesla wants to attain low cost production through automation. As discussed in their 2017 Q3 report, Tesla is focused “on highly automated manufacturing processes that we expect will ultimately result in higher volumes at significantly lower costs.”
Mr. Musk explains: “You really can’t have people in the production line itself. Otherwise you’ll automatically drop to people speed… There’s still a lot of people at the factory, but what they’re doing is maintaining the machines, upgrading them, dealing with anomalies. But in the production process itself there essentially would be no people.”
Mr. Musk’s words have been backed by heavy investments in manufacturing equipment. Assets listed under “Property, plant and equipment” grew from $3.4B in 2015 to $10B in 2017 as the Model 3 production line came online.
Mr. Musk is effectively relying on machines to achieve economies of scale and lower production cost. Betting that robots can crank things out faster than humans. This requires a production line that never stops. As result, Tesla’s manufacturing process tends to be inflexible and unresponsive to changes in the market. There’s little opportunity to improve.
Because Tesla’s automated production line can’t quickly adapt to changes in demand, pre-orders are a necessity. Not a nice-to-have.
Who’s in command?
It is not the conveyor that operates men. It’s men that operate the conveyor.
Any production line can be stopped by workers at any given moment to prevent making too many parts, control for defects, make improvements, or prevent accidents. Allowing for continuous improvement.
Nothing must break the continuity of production.
One line worker recalls a moment when a robot broke down and the supervisor came screaming: “That’s $18,000, $20,000, $30,000, $50,000 because you guys can’t get this done.” Another employee shared that safety is not a top priority.
Tesla’s reliance on automation to lower costs means it cannot afford to stop the machines. Even when improvements need to be made. A robot offline is a robot losing money.
Since we humans tend to hate losing money, as described by prospect theory, Tesla is likely to forgo small improvements to their production line they perceive as money losers, and only stop machines for major works. Such as the recent days long production pause in February and again in April. It’s clear they have difficulty tweaking on the go.
How do parts flow?
The need to minimize waste led to the creation of Toyota’s just-in-time system: All processes produce the necessary parts at the necessary time and have on hand only the minimum stock necessary to hold processes together.
This means all production lines need accurate knowledge of timing and quantity needed for their products. Information on priority of orders is gathered from the final assembly line and shared with foremen on subassembly lines. These front-line leaders then decide on job dispatching and overtime.
Control is effectively decentralized.
Keeping an automated production line running tends to promote surplus manufacturing of parts to lower risk of downtime. Tesla’s recent acquisition of a large warehouse indicates its production indeed has buffer stocks to help smooth out imbalances between flows of parts into main assembly.
This tends to be expensive not just in terms of storage costs, but also because parts may become obsolete should designs change.
It's also difficult for analysts and investors to use supplier order data as proxy to production performance. Tesla may simply be stocking the parts. Not using them.
Poor management adds fuel to the fire. One employee complained that the company is disorganized from the top down, with priorities changing daily. Another shared that nobody knows what needs to be done on a daily basis, leading to much more confusion and frustration.
These signs point to information bottlenecks and centralized command.
Mindset on quality?
Quality before quantity.
Setting up a new production line is a matter of building it up slowly, making piecemeal improvements in the process over years.
Toyota believes testing for quality is far more expensive than building it right the first time.
Build quickly. Fix it later.
Employee comments show that Tesla believes trial production runs are pointless. They jump right into production. Coupled with sky high goals, this leads to all hands on deck situations to meet targets.
This is consistent with Silicon Valley’s move fast break things mentality and agile software development philosophies. The difference in manufacturing however is that one can’t patch problems later with a software update. Customers must take time away from their lives and visit the shop. Not to mention increased warranty cost for Tesla.
It becomes clear that Toyota’s human-centered system is best suited to produce different products that change over time. It makes full use of humans’ ability to improve. To spot problems, identify their root cause, and creatively imagine better solutions.
By contrast, Tesla’s robot-centered system is best suited to produce one product that meets the needs of many. Andrew Sayer explains:
“Where advanced automation, such as flexible manufacturing systems, is introduced, its effects on productivity are greatest where it is applied to a production system that has already been rationally organised; otherwise it is likely to perpetuate the inefficiencies of the old technology and working practices, as has been found with many major new technologies, including computer-integrated manufacturing and office automation.”
In other words, Toyota’s system thrives in a changing world. Tesla’s system thrives when nothing changes.
Production costs will stay high as long as Tesla continues to develop new models or to improve designs of existing cars. It’s simply expensive to take robots offline for tuning. It also heightens the risk of running out of cash.
Tesla’s “build it first, fix it later” mentality is also causing high servicing costs. Service cost as percent of total cost of revenues grew from 7.2% in 2014 to 12.9% in 2017. For every $1 customers paid to get service, Tesla had to pay $1.23 to get the job done last year. A sign of high rates of in-warranty repairs. Mr. Musk’s cash problems will likely persist as long as he puts quantity over quality.
Interestingly, Tesla and Toyota were partners. They worked on an electric version of the Rav4 together. This ended in 2016, supposedly because of a culture clash. It wouldn’t be surprising to find they broke up because of opposing manufacturing philosophies.
The irony is that Mr. Musk instinctively believes in continuous improvement. He shared in an interview how:
“I always see what's wrong... When I see a car or a rocket or spacecraft, I only see what's wrong… It's not a recipe for happiness."
Mr. Musk should really think twice about investing more in automation. Robots don't allow him to quickly fix and improve designs. Likely leading to further displeasure.
More recently, Elon Musk appears to be awakening to the fact too many robots can slow down production. He tweeted in April 2018: “Yes, excessive automation at Tesla was a mistake… Humans are underrated.”
That’s a good start. Tesla however has a long road ahead in convincing its human workers it has their best interest at heart. That it’ll create a safer work environment. That it won’t again try to replace them with robots. Necessary for workers to trust their employer, and in turn, lead continuous improvement initiatives.
Unlike other car manufacturers, Tesla tends to both design and manufacture its parts in-house. When battery problems surfaced with a supplier, Mr. Musk cut the supplier and directed his engineers onto the issue, bringing manufacturing in-house. The same happened when problems arose with the Model X’s second row seat supplier.
This means Tesla only has one supplier for many of its parts: itself. It greatly increases the risk of production delays should any of these parts face manufacturing or design problems. Most manufacturers hedge that risk by using multiple suppliers for the same part. Some even use suppliers in different geographic regions to hedge political and natural disaster risks.
The tendency to build parts in-house may stem from Mr. Musk’s success in reducing cost of rockets at SpaceX through vertically integrated manufacturing. Over 80% of parts used in Falcon rockets are made in-house.
There is however a fundamental difference in rocket production and car production: Delays experienced in launching a rocket are expected and unlikely to affect revenue, whereas delays in car production directly leads to revenue loss. Threatening liquidity.
Tesla’s production is reminiscent of Ford’s 100 year-old just-in-case system used to produce Model Ts. Model T production also focused on lowering cost by using more machines and by making parts in-house at the Highland Park plant. The result was the first car every American could afford. Ford’s Model T only cost $370 in 1921 ($5,060 in 2017 dollars). Less than half the cost of the cheapest GM vehicle: A Chevrolet costing $795 ($10,900 in 2017 dollars).
Low cost was partially achieved by producing only one car model at Ford: The Model T. Ford believed that the Model T was all the car a person would, or could, ever need. He was wrong. The Model T fell out of fashion as GM both reduced cost of its vehicles and introduced different models suiting the needs of different customers. Production of Model Ts ended after 18 years in 1927.
Tesla is already producing three vehicle models by contrast. Each with dozens of options. Its robots therefore need to be much smarter and flexible than Ford’s, having to adapt to different specs. So unless Tesla’s robots are super easy to tweak, they’re bound to continue experiencing a lot of downtime.
Both Tesla’s Semi and Model 3 also face a greater challenge than the Model T. They’re competing with existing car manufacturers with far more experience mass producing vehicles. The Model T didn’t even have a competitor capable of mass production in the early days. They competed against horses. Heightened competition could effectively lower demand for Tesla vehicles.
Investors should be scared Tesla compared itself to Ford and the Model T.
Peter Hochholdinger, Tesla’s VP of Production, may not be of much help. The Volkswagen Group was already a manufacturing powerhouse producing over 3 million cars in 1994, when he joined them at Audi. Mr. Hochholdinger’s current challenge thus lies in boosting production without the support of an experienced manufacturing organization. Something he depended on from the very start of his career. During his time at Audi, production increased from 617,000 in 1998 to 1.9 million cars in 2016. An average growth of 6.6% per year. Tesla’s total production would grow from 103,184 in 2017 to around 172,000 in 2025 at such a rate, missing all production targets.
It's our opinion that Tesla will continue to face production challenges and high costs for the rest of 2018, considering:
None of this takes away the fact the Model 3 is disrupting personal transportation.
Editor's Note: This article discusses one or more securities that do not trade on a major U.S. exchange. Please be aware of the risks associated with these stocks.
This article was written by
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: This article is intended to stimulate awareness and discussion on the impact Tesla's production processes have on revenue and costs. Investors should view my work in this light, and seek other competent technical advice on the subject issues before making investment decisions.
Forward-looking statements found in this article are based on assumptions regarding the Tesla's business and performance, the economy and other future conditions and forecasts of future events, circumstances and results. As with any projection or forecast, forward-looking statements are inherently susceptible to uncertainty and changes in circumstances. Tesla's actual results may vary materially from those expressed or implied in its forward-looking statements.