Heijunka
One of the fundamental goals of lean manufacturing is to "level the line" whenever practical. The Japanese word heijunka is commonly used in industrial engineering to encapsulate this concept because of the Toyota Production System framework.
An ideal levelized process has continuous one-piece flow, in contrast to batch-and-queue processing in which batches of product are processed simultaneously in lots and then queued as inventory waiting for the next downstream step to accept the completed lot.
Heijunka helps optimize:
Unit costs
Inventory
Predictability
Working capital needs
Labor efficiency
Safety
Quality
Team morale
Lead time
Heijunka reduces waste of muri, which is the Japanese word that the Toyota Production System uses to describe the waste resulting from of overburdening people, machines, and systems. The basic idea is that smooth flow keeps the machine that makes the machine running efficiently without piling up work-in-progress in between production steps, without making people excessively tired, without drama, without variation, without excessive haste and mistakes that go along with rushing.
Bottleneck elimination is another benefit of heijunka. In a state of perfect leveling and consistent flow, every asset in the production is being used precisely to its maximum potential; no more, no less. There is neither excessive slack nor excessive strain anywhere. In contrast, when flow is uneven, output of the entire production system is constrained by the bottleneck, typically whichever particular machine or human team is the most overburdened at each moment in time.
Every capital asset (machines, buildings, land, etc.) represents upfront investment of money and ongoing costs for inspection and maintenance that are incurred irrespective of how much the asset is used. Return on investment is maximized when each asset is used to its maximum annual potential. Thus, heijunka’s goal of eliminating slack, or in other words maximizing full-speed uptime, goes hand-in-hand with ROI improvements.
Sometimes, continuous flow isn’t actually optimal. For example, this would be true for a process that has strong economies of scale as a function of batch size and relatively low costs of carrying work-in-progress inventory. However, generally leveling the flow is the best industrial design.
Wikipedia has good introductory material on this subject, but you can find plenty of other information on this subject elsewhere.
Production leveling - Wikipedia
Muri (Japanese term) - Wikipedia
The leveling principle extends beyond the factory walls to outbound logistics as well.
Unwinding the Wave
At long last, Tesla is ending its quarterly vehicle delivery wave. Historically, Tesla would batch deliveries regionally in an attempt to minimize the number of cars in transit at the end of each quarter.
The company has been talking about doing this for years and is finally following through, starting in Q3 2022 when, according to Tesla, hard limits on the sheer amount of available peak transportation capacity forced the issue.
Tesla has commented on this topic frequently in recent months:
Q3 Production & Deliveries report:
Historically, our delivery volumes have skewed towards the end of each quarter due to regional batch building of cars. As our production volumes continue to grow, it is becoming increasingly challenging to secure vehicle transportation capacity and at a reasonable cost during these peak logistics weeks. In Q3, we began transitioning to a more even regional mix of vehicle builds each week, which led to an increase in cars in transit at the end of the quarter. These cars have been ordered and will be delivered to customers upon arrival at their destination.
Q3 Shareholder Update:
…we are reaching such significant delivery volumes in the final weeks of each quarter that transportation capacity is becoming expensive and difficult to secure. As a result, we began transitioning to a smoother delivery pace, leading to more vehicles in transit at the end of the quarter. We expect that smoothing our outbound logistics throughout the quarter will improve cost per vehicle.
Zach Kirkhorn on Q3 earnings call:
Specifically on cars in transit, as noted in our press release on October 2, we've started to experience limits on outbound logistics capacity which we didn't anticipate. This issue is particularly present for ships from Shanghai to Europe and local trucking within certain parts of the U.S. and Europe. Our historical operating pattern of batch building by delivery region leads to extreme concentrations of outbound logistics needs in the final weeks of each quarter.
Just to put this in perspective, roughly two-thirds of our Q3 deliveries occurred in September and one-third in the final two weeks. As a result, we have begun to smooth the regional builds throughout the quarter to reduce our peak needs for outbound logistics. We expect this to simplify our operations, reduce costs, and improve the experience of our customers. As we look ahead, our plans show that we're on track for the 50% annual growth in production this year, although we are tracking supply chain risks which are beyond our control.
On the delivery side, we do expect to be just under 50% growth due to an increase in the cars in transit at the end of the year, as noted, just above. This means that, again, you should expect a gap between production and deliveries in Q4, and those cars in transit will be delivered shortly to their customers upon arrival to their destination in Q1.
Elon Musk on Q3 earnings call:
Actually, I want to caveat, I should say, growing production by 50% every year because deliveries -- we're trying to smooth out the deliveries and not have this crazy delivery wave at the end of every quarter. So, in fact, we're just fundamentally running out of -- there weren't enough boats, there weren't enough trains, there weren't enough car carriers to actually support the wave because it got too big. So, whether we like it or not, we actually have to smooth out the delivery of cars intra-quarter because there aren't just enough transportation objects to move them around.
Q4 Production & Deliveries report:
We continued to transition towards a more even regional mix of vehicle builds which again led to a further increase in cars in transit at the end of the quarter.
So, Tesla is finally beginning to mature their outbound logistics into a smooth, steady operation working under the principle of heijunka and reaping all the associated benefits.
Demand Issue?
Meanwhile, it seems that the stock market and many institutional analysts have focused on the large gaps between production and deliveries in Q3 and Q4 as concerning signals of plummeting demand, or at least a sign of difficulty maintaining pricing power as deliveries have grown to more than 400k per quarter. If so, this has negative implications for Tesla’s gross margins as annual sales volume continues to grow into the millions, and by extension Tesla’s expected net income growth.
Granted, the evidence does indicate that demand has dropped relative to early 2022. There has been plenty of data indicating a broader auto market decline since mid-2022 and Tesla has not been immune. Elon has said so himself. Prices for Tesla cars had been greatly elevated in 2022, and with rising financing costs for customers the true ownership cost had increased by thousands of dollars of interest expenses relative to the first half of 2022.
However, when Tesla was raising prices in 2021 and 2022, interest rates were at rock bottom, gas prices were sky-high, and customers in China still had access for government EV incentives. In retrospect, this period appears to have been an anomaly of high car prices in general and especially for EVs as consumers sought alternatives to gas and diesel cars. It’s also unclear if Tesla ever actually delivered much volume at the highest prices set in March 2022, given how long the order backlog was and how the global average selling price only rose by a few thousand dollars in total from early 2021 to the end of 2022.
Furthermore, when Tesla aggressively dropped prices in January 2022, orders came roaring back—even though the prices were still mostly higher than in the beginning of 2021 before the series of price increases, even after adjusting for inflation (Tesla Daily did a good breakdown in this video). In the US, for example, Google search interest for Model Y tripled after the price cuts and today remains about double the previous baseline. Spikes in Google search interest have strongly correlated with the timing of Tesla price hikes over 2021 and 2022.
On the Q4 2022 earnings call, Elon wasted no time addressing this point:
The most common question we've been getting from investors is about demand.
Thus far — so I want to put that concern to rest. Thus far in January, we've seen the strongest orders year-to-date than ever in our history. We currently are seeing orders at almost twice the rate of production. So I mean, that — it's hard to say whether that will continue twice the rate of production, but the orders are high.
And we've actually raised the Model Y price a little bit in response to that. So we think demand will be good despite probably a contraction in the automotive market as a whole. So basically, price really matters. I think there's just a vast number of people that want to buy a Tesla car but can't afford it.
And so, these price changes really make a difference for the average consumer.
So, basically Tesla had raised prices to outrageous levels to stem excessively hot demand in 2021 and 2022, and then at the end of 2022 made adjustments back to more reasonable levels, at the same time as their input costs should be declining from peaks around Q3 2022. Many other analysts have covered this topic thoroughly, so I’ll stop here for this essay. Suffice it to say that I think it’s safe to conclude that Tesla has more orders than they can handle at current prices, and there is no special reason for concern that the increase in inventory was caused by an inability to find customers willing to accept delivery.
What’s happening here is that Tesla is filling the pipeline with vehicles in transit.
Discussion
Ending the wave is beneficial, and I wish Tesla had done this sooner. As far as I can tell, the primary purpose of the wave was to make short-term quarterly delivery and financial numbers look more impressive by minimizing vehicle inventory at arbitrary calendar dates four times per year. This was not an optimal approach and was inconsistent with Tesla’s first-principles engineering ethos. The wave was optimal for maximizing reported revenue, profits, cash balance, and other key financial metrics, but only on the first days of January, April, July, and October. The rest of the time, this strategy was actually harming the business and making the long run results worse.
Cash, especially, was negatively affected by the wave. Working capital needs are determined primarily by peak needs, not average needs. As the sawtooth wave chart at the beginning of this article illustrates, Tesla had been stuffing the transit pipeline with a huge wave of vehicles in the middle of each quarter. Tesla doesn’t get paid until the customer takes delivery. Thus, Tesla was paying suppliers continuously but effectively delaying the bulk of customer payments until the end of each quarter. With smooth delivery patterns, there will be hardly any gap between peak and average numbers of vehicles in transit, and as a result no periods of waiting weeks for receiving cash for waves of vehicles. This will increase Tesla’s stores of working capital free for other purposes.
As Tesla has explained, this strategy also came with unnecessary operational complexity plus higher expedite fees and other costs, yet it resulted in a worse experience for customers and employees. It probably caused increased rates of mistakes, or in other words it was detrimental to the first-pass quality of the delivery process. I know of no method to estimate the exact magnitude of the benefits, but I am confident it’s substantial and well worth the negative impact on short-term results while Tesla completes this transition.
Inventory Remains 5x Lower than Industry Norm
Tesla's vehicle inventory, as measured by the numbers reported at end-of-quarter, is still far, far below the industry average, by a factor of 5! I think too many investors are missing this context.
The National Automobile Dealers Association published in their mid-year 2022 NADA Data report this graphic showing that in 2016 - 2019, before the impacts of COVID, supply chain disruption, and Russia’s invasion of Ukraine, in the US the average dealership had around 60-85 days worth of inventory on lots and in transit. In stark contrast, even with the Q3 and Q4 pipeline-filling caused by ending the wave, Tesla reported only having only 13 days of inventory at the end of Q4. This is still about half as much inventory (measured by days of sales) as Tesla had a few years ago prior to Giga Shanghai beginning to meaningfully contribute to overall deliveries, thus reducing the average number of days required to ship cars to customers.
Also, it’s important to realize that Tesla’s 5x advantage is actually a severely biased comparison in favor of the competitors, because cars in transit account for nearly all Tesla's inventory, with very few cars sitting in parking lots at any given time. The norm for the rest of the industry is to have cars sitting for an average of a couple of months at dealerships waiting for the right customer, whereas Tesla still doesn't even have standard same-day purchasing like other car companies or just about all consumer retail businesses in general. I have not been able to find data on the industry norm for days worth of sales of vehicles in transit to dealerships, so this apples-to-oranges comparison is the best I can do, but even this pessimistic analysis shows Tesla having a 5x advantage. Tesla’s actual fundamental business model is a true just-in-time, inventory-light design because they don’t park cars for months accumulating dirt, bird droppings, and costs from capital sunk into the inventory and less-obvious costs like insurance, security and property tax. The big picture for Tesla’s inventory is not the modest rise in the second half of 2022, but rather the online ordering and lack of traditional dealerships.
Localization
But wait, there’s more! Tesla's inventory will eventually come down again as more factories come online.
Tesla’s demonstrated profitability has been truly amazing for a company that’s still barely cracked the top ten in the industry production volume ranking. Tesla will sell approximately 2 million cars in 2023. Toyota and VW normally sell 10 million per year. Economies of scale are critical in manufacturing and especially the auto business. Logistics are one area of the business with serious scale efficiency that Tesla will increasingly enjoy as it scales to 10 million and beyond.
In the beginning, all Tesla vehicles were shipped from California. With only one factory, Tesla had no choice but to pay all the direct and indirect costs of long-distance shipping. Other major car companies like Toyota and Volkswagen have many more factories than Tesla and thus have a much shorter average shipping distance for each car they sell. Although California has an extreme concentration of Tesla customers, the in-state market still constitutes a minority of Tesla’s overall sales volume. All too often, Tesla has been paying $1,000 per car or more on freight expenses. As Elon succinctly put it in 2020 in a Third Row Podcast interview:
“The biggest problem we have to solve right now is having production on each continent, because it's insane to be making cars in California [and] shipping them to Europe and Asia"
This statement came when Gigafactory Shanghai was in low-rate initial production. Nowadays, Tesla is cranking out cars at an annualized rate of 1 million per year from Shanghai, which has massively reduced shipping and tariff costs for the Chinese market. Still, there’s a long way to go for localization because Giga Berlin and Giga Texas are only just now beginning to hit substantial production rates of several thousand cars per week.
Europe, in particular, has been a thorn in the side of Tesla’s logistics operation since the very first time Tesla delivered a car there. Europe has plenty of affluent consumers wanting to drive EVs, high fuel prices, and a regulatory environment favorable to EV sales. The problem has been actually getting the cars to the customers. Until the arrival of Gigafactory Berlin-Brandenburg in Germany, every vehicle Tesla ever sold in Europe was transported either from San Francisco or Shanghai, both of which are truly terrible locations for shipping by sea to Europe. It takes freight ships a whole month or more to travel from Shanghai to Belgium. Then after arrival Tesla must pay a 10% import tariff to sell in the EU. All in all, this has been contributing thousands of dollars of extra cost per car for every sale Tesla makes in Europe. I think this severe importation challenge is the main reason why Tesla has much higher share of the automotive market in the USA than in Europe, despite the much higher overall adoption rate of EVs by Europeans.
Giga Berlin changes everything for Tesla’s European logistics. Germany is an EU member state and is positioned in the center of Europe. Just as Giga Shanghai revolutionized the economics of selling Teslas in China, so too will Giga Berlin for the EU and the broader Europe, Middle East and North Africa (EMEA) region. No more import tariff for EU deliveries, which make up most of Tesla’s EMEA sales. No more multi-thousand-dollar boat ticket for a trip from China.
Giga Texas will have a similar, albeit smaller, impact for North American deliveries. About 80% of the population of North America lives in the eastern half of the continent. Shipping to these customers from Austin instead of Fremont cuts roughly 1,000 miles of land transportation from the average shipping distance. This will save a few days of transportation lead time and several hundred dollars per car, and thus add approximately 1% to gross margins for these cars, all else being equal, compared to shipping the same cars from California.
More factories in more locations are coming, and this pattern will continue. As with all economies of scale, this logistics efficiency effect has diminishing marginal returns, because as the average transit distances are shortened, each additional shortening matters less on an absolute basis. Still, as of today there remains plenty of room for substantial improvements. It’s still a major undertaking to ship multi-ton objects all over the planet from just four starting points. Ten would be much better, and cheaper. Will Tesla open a factory in Brazil to serve South America? Indonesia for the region around the Indian Ocean? Nashville? Detroit? One thing is for certain, and that is that more factories are coming.
Conclusion
While Tesla’s inventory buildup produced some concern and consternation amongst shareholders in the last two quarters, it’s actually a good sign that Tesla is finally moving on to a more rational logistics approach.
Serious cost savings of potentially a couple thousand dollars per car are coming as Tesla ramps production in Berlin, Austin, and future locations.
Disclaimer: I am about 99.9% all-in on Tesla via TSLA stock and call options. These are notes for my model that I’m sharing. I fully believe what I have written and I’ve put all of my money where my mouth is, but this is not advice regarding your personal investment or financial decisions. I'm not a certified professional investment or financial advisor and even if I were one, broad buying or selling advice would be inappropriate because there are so many variables to take into consideration such as your goals, risk tolerance, financial outlook, income, age, tax situation, dependents, and so on. Make your own choices. Thanks for reading and I hope you find this analysis helpful and comment on any mistakes you may find.