Think Model 3 cash burn is bad? Wait till Tesla sells Semis.

[This is an excerpt from our special report on Tesla’s Semi Strategy: comparing operating cost of the eletric Semi against diesel trucks, evaluating whether Tesla would make generate more revenue selling delivery trucks instead, and analyzing risks facing the truck program.]

Tesla investors are worried about the company’s cash flow. To alleviate the risk of running out of cash, Musk is working to speed up production and delivery of Model 3s. Hoping to sell more cars to stop the bleed. Will the same strategy work for Tesla’s next mass market vehicle: The Semi truck?

We explore below costs of building a Tesla Semi, whether selling delivery trucks would make more sense, and why Musk even bothers to sell a commercial vehicle.

Selling Semis may be a financial disaster

Tesla’s Semi reportedly consumes 2kWh/mile and has a range of 500 miles. We can thus expect a 1,000kWh battery pack. The Li-ion batteries used to power the Semi are predicted to cost around $180/kWh in 2020, when production trucks are planned to hit the road. This means the Semi’s battery pack alone may cost $180,000. That’s the price tag for the entire truck. It’s thus unlikely the the Semi gets build at a profit.

Musk also promised truck operators a charging rate of $0.07/kWh to compete aggressively with diesel fuel. That may be below Tesla’s true electricity cost. The average commercial rate for electricity in the U.S. is around $0.103/kWh. $0.03/kWh more than what Tesla charges. But electricity price also varies by the peak rate at which electricity is drawn from the grid, so Tesla’s megacharger stations are likely to pay a premium. One source estimates the true electricity cost of Tesla’s megacharger at $0.40kWh. $0.37/kWh more than what Tesla plans to charge.

How much will subsidizing electricity cost Tesla?

Let’s assume each Semi truck will charge at standard rates overnight, giving it a 500 miles in the morning. Let’s also assume they will use the megacharger once in the day to get another 400 miles range. Charging the Semi will therefore cost $0.103/kWh 55% of the time (standard charging overnight), and $0.40/kWh 45% of the time (megacharger in the day).

A long-haul truck travels around 85,805 miles per year. This is based off 2016 fleet performance of Knight-Swift, a publicly traded long-haul freight operator. Knight-Swift spent $129.7 million on fuel for its 4,286 trucks in 2016. That’s more than $30K per truck. The American Trucking Research Institute reports that median fuel efficiency of tractor trucks varies between 6.3MPG to 6.8MPG. Let’s assume for our analysis that tractor-trailers operate at 6.55MPG (average of median values found by ATRI). With diesel averaging $2.31/Gal in 2016 and trucks running at 6.55MPG, each of Knight-Swift’s trucks therefore averaged 85,805 miles.

A Tesla Semi traveling 85,805 miles a year would use 171,610kWh of electricity at 2kWh/mile. Charging at standard rates 55% of the time at $0.103/kWh, and at megacharger rates 45% of the time at $0.40/kWh, electricity will cost Tesla $40,612 per truck annually. After accounting for $0.07/kWh in revenue, Tesla stands to lose $28,599 per truck annually.

What may help Tesla lower charging costs includes batteries installed at charging sites that can draw energy from the grid at normal rates, and discharge into trucks without peaking the grid. Such technology is already in use by a fleet of UPS delivery trucks in London. Add some solar panels and it’s possible Tesla’s electricity will only cost around 10 cents per kWh. Normal market rate. All this would however necessitate a large capital investment.

It becomes clear Tesla is unlikely to make profits from selling Semi trucks. It may even lose money on an ongoing basis by subsidizing electricity costs.

More money selling delivery trucks?

One question Tesla hasn’t answered is why they aren’t building a delivery truck instead of a semi. Delivery trucks represent a much bigger opportunity.

For a start, there’s 8.46 million single unit delivery trucks versus 2.75 million semis on the road. They deliver goods from distribution centers to brick and mortar shops and end-users. Completing the last-mile. Despite being three times as numerous, they only traveled 109,597 million miles in 2015 compared to 169,830 million miles for semi trucks. These class 3 to 7 single unit trucks carry the bulk of local and regional shipments traveling less than 249 miles; representing 67% of total truck freight.

The rapid growth of e-commerce is also boosting demand for short-haul shipping. Online sales grew 16% in 2017. As result, courier and package shipments are growing much faster than the 1.6% expected growth for overall truck freight. UPS’s domestic packages segment experienced annual volume growth of more than 9% in the past two years, with trucking playing a vital role. Research indicates the development of next-day and 2nd-day courier services, often used by e-commerce stores such as Amazon, has made trucks critical to operations. They reduce airport congestion and allow greater schedule flexibility.

This unprecedented growth is even attracting long-haul freight operators into the courier business. XPO entered the U.S. home delivery business in 2013 and is actively seeking to grow their market share; specializing in last mile transportation for appliances, large electronics and heavy goods. Schneider National is following a similar growth strategy with their recent acquisitions of Watkins & Shepard and Lodeso. Both companies that handled last-mile deliveries of oversized items.

Is the demand for class 3–7 single unit trucks growing just as quickly?

Class 3 to 5 trucks, carrying out the bulk of urban deliveries, experienced average sales growth of 11.2% per year since 2010. From 204,000 vehicles sold in 2010 to 382,000 vehicles in 2016. Class 6 and 7 trucks, capable of delivering oversized items such as furniture, grew at a similar pace of 10.6% a year. From 67,000 vehicles sold in 2010 to 122,000 vehicles in 2016. Unlike semi trucks, sales have yet to peak.

Navistar International, one of the top single unit truck manufacturers, experienced a 16% increase in unit sales in Q1 of 2018 versus the same period last year. They also had a 43% boost in orders and 40% increase in backlog. Ford’s F-Series trucks (mainly class 2 to 7 vehicles) saw a 2.2% increase in sales in January 2018 versus the same period last year. Utilimaster, one of only two manufacturers of walk-in vans used by UPS and FedEx (the other being Morgan Olson), saw their order backlog increase by 198.9% in 2017 with the award of a $214 million UPS contract.

There’s a clear opportunity for electric trucks to disrupt the delivery truck market. Electric vehicles are highly efficient in the stop and go traffic delivery trucks find themselves in. A Dodge Ram Promaster diesel van achieves 13mpg in the city versus 18mpg on the highway. 28% less efficient in city driving. A Tesla Model X achieves 81mpg-e in city and 92mpg-e on the highway. Only 12% less efficient in city driving by comparison.

Let’s thus imagine Tesla designed a class 5 electric delivery truck. One that any operator driving class 3 to 5 trucks could use to haul their loads. How would it compare against diesel delivery trucks?

Diesel vs. Tesla delivery truck

Price

A delivery van can cost anywhere between $50,000 to $85,000 depending on specs. For our analysis, let’s assume we’re buying a brand new class 5 Freightliner MT55 walk-in van priced at $85,000.

Assuming Tesla’s lower operating cost can justify a 20% premium just like the semi truck, a class 5 delivery electric truck could therefore be priced at $105,000.

Fuel

The top-of-the-line, composite body, Utilimaster Reach reports 15MPG at the pump and claims to be 35% more fuel efficient than standard trucks. We can thus deduce that a standard diesel delivery truck to average 11MPG.

This translates into $0.26/mile at 2019 diesel prices of $2.86/gal.

Let’s look at an existing electric delivery truck for clues on energy consumption: The Workhorse E-Gen class 5 electric truck averages 1kWh/mile. Let’s assume our imaginary Tesla delivery truck to have a similar consumption rate.

This translates into $0.07/mile for fuel if Tesla sells electricity at 7 cents per kWh.

Environment

A diesel delivery truck emits 2.7 lbs CO2/mile when accounting for 11mpg and 29.6 lbs CO2/gal well-to-wheel emissions for diesel.

The electric truck is expected to emit 1.085 lbs CO2/mile when accounting for energy consumption of 1kWh/mile and 1.085 lbs CO2/kWh grid emissions. Less than half that of diesel.

So what?

What does this mean to a carrier like UPS? UPS operates a fleet of 101,863+ small package delivery vehicles worldwide. They travel anywhere from 60 miles to 100+ miles a day depending on whether they’re delivering downtown or in suburbia.

Let’s assume a UPS truck averages 60 miles a day. At $0.26/mile for 313 days a year (working 6 days a week), fuel cost amounts to $497 million. That’s compared to $134 million if they were to run an entire fleet of electric trucks sipping energy at $0.07/mile. Savings of over $363 million or 75%.

It’s no surprise UPS is actively investing in electric trucks, yet they remain an experiment. UPS operates 8,500+ alternative fuel vehicles, representing only 8% of their fleet. 84% of new vehicles purchased are still standard diesel trucks. As result, UPS still spent $2.69B on fuel in 2017 (for planes and trucks). A lack of charging infrastructure may very well be the biggest obstacle to the electric revolution.

Tesla could be forsaking $7.7 billion selling semis instead of delivery trucks

Is Tesla missing an opportunity by not selling an electric delivery truck? Let’s compare the potential revenue of selling a semi versus a class 5 delivery truck.

First, we must estimate the market share Tesla can realistically gain over time.

Elon Musk has set the bar at 100,000 semis to be sold annually by 2022. That’d result in a market share of over 50% two years after launch as 200,000 semis are sold each year. And wishful thinking at best.

Mr. Musk set a similarly outlandish goal for his Model 3 sedan in his Q1 2016 investor call. He envisioned 100,000+ Model 3s produced before end of 2017. He updated that figure to 20,000 in July of 2017. The reality? 1,550 Model 3s delivered in 2017. 1.5% of his original target.

Mr. Musk clearly suffers from chronic planning fallacy. He’s over-optimistic about Tesla’s ability to hit goals. Overlooking obstacles and uncertainties on the way. For example, manufacturing problems could push back the semi’s production schedule. Or freight companies could be slow to adopt electric trucks due to uncertainties around infrastructure. Perhaps planning fallacy is a prerequisite for big dreams.

For more realistic growth estimates, let’s look at the story of Hyundai Translead. The top semi truck trailer manufacturer in 2017 with 19% market share. One would never have guessed that they only started selling trailers to U.S. freight companies in 1994. They took advantage of lower wages south of the border by establishing a plant in Mexico: $1.30/hour in Mexico vs. $12.60/hour in the U.S. That allowed them to offer lower prices and added value. They claimed 3.1% of the market share in the first year. Hyundai Translead’s market share then grew on average 8.2% a year for the next 23 years. Nothing short of amazing.

Many similarities exist between Tesla’s Semi project and Hyundai Translead’s early days. Both started with no market share in the trucking business. Both promised lower operating costs and added value.

There is one important difference though: Hyundai was already a manufacturing powerhouse when they entered the U.S. trailer market in 1994. They manufactured over a million cars that year. Tesla produced less than 30,000 cars last year.

It’s safe to say that replicating Hyundai Translead’s growth would be a best-case scenario for Tesla. Let’s thus use Hyundai’s market share growth as proxy to Tesla’s future growth.

We now need to forecast how the overall market for trucks will grow. Two measures we found to predict truck sales include the industrial production index (IPI) and advanced retail sales (ARS). Assuming a linear relationship, we found annual sales of class 8 trucks to correlate meaningfully with the IPI, with a correlation coefficient of 0.654. We then found a strong relationship between annual sales of class 3 to 5 trucks and ARS, with a correlation coefficient of 0.94.

These relationships allow us to perform linear regressions to forecast future truck sales.

We can estimate future IPI and ARS figures based on historical growth rates from 2001 to 2017: 0.794% and 3.279% per year on average respectively. This in turn allows us to create trendlines to forecast future market size for truck sales.

The number of semi and delivery trucks Tesla could sell can finally be estimated based on Hyundai Translead’s market share over time: 3.1% of the market in 2020, their launch year, and all the way to 19% of the market 23 years on.

Potential annual revenue can also be estimated by assuming Tesla sells its semi truck and delivery truck at $180,000 and $105,000 per unit.

It becomes clear Tesla would make much more money selling delivery trucks. By 2030, Tesla could generate annual revenue of $5.2B by selling a class 5 truck compared to $3.65B by selling the semi. 42% more.

The cumulative revenue forfeited by selling a semi instead amounts to $7.7B+ over these first ten years. During which Tesla could make $17.3B selling delivery trucks versus $9.85B selling semi trucks. The loss in cumulative revenue grows to $49.5B by the 20 year mark in 2040. The delivery truck market is simply larger and growing faster than the semi market.

Mind these are rosy forecasts that assume retail sales will keep growing at historical rates. And that no disruptive technology comes in to replace trucks altogether.

Tesla delivery trucks could be cash cows

Let’s assume a Tesla delivery truck is designed to travel 200 miles. Double the average distance of UPS suburban trips. At 1kWh/mile, the vehicle only needs a 200kWh battery. Similar to those used in its passenger vehicles.

A 200kWh battery will cost $36,000 at $180/kWh. Much less than our imaginary Tesla delivery truck’s sticker price of $105,000. That leaves $69,000 to build the rest of the truck. Likely at a profit.

Most delivery trucks won’t travel more than 200 miles per day, so won’t need megachargers. They’ll simply recharge overnight at normal rates. This means a delivery truck traveling an average of 13,116 miles a year will cost Tesla $1,350.95 to charge at $0.103/kWh. After accounting for income of $0.07/kWh, Tesla will only lose $433 per truck per year. That’s $28,166 less than how much it stands to lose per Semi per year.

It becomes clear Tesla could turn a healthy profit by selling delivery trucks.

It’s climate change, stupid.

So why is Tesla building a tractor truck instead of a delivery truck? Certainly not to deliver above average returns to investors. Maybe to save the planet? Let’s compare the environmental effects of electric semis versus that of delivery trucks.

Tesla could have 57,418 semi trucks on the road by 2025 based on our growth model. Or they could have 118,979 delivery trucks. This assumes every truck produced from 2020 and on stays on the road.

The fleet of semis would travel a total of 4,754 million miles that year versus 1,560 million miles for the fleet of delivery trucks. This is based off the fact a long-haul Semi averages 85,805 miles a year as we previously estimated, while a delivery truck averages 13,116 miles a year.

How much CO2 will be saved by these electric trucks?

Burning one gallon of diesel emits 22.38 lbs of CO2. This means a diesel semi truck averaging 6.55MPG emits 3.36 lbs CO2/mile. This figure however doesn’t account for the emissions generated producing diesel fuel.

An investigation by the state of California reports the Well-to-Wheel (from extracting crude oil to burning diesel) emission of diesel to be 92g CO2 per MJ of fuel (0.203 lbs CO2 per MJ). This translates to 29.6 lbs CO2/gal as diesel has an energy density of 146MJ/gal. Or 4.52 lbs CO2/mile at 6.55MPG. This represents a more accurate emissions measure for diesel tractor trucks.

On the other hand, the department of energy estimates emissions from electricity production at 1,085 lbs CO2/MWh. For a Tesla truck consuming 2kWh/mile, this translates into emissions of 2.17 lbs CO2/mile.

A fleet of Tesla semis could therefore save 11,173 million lbs of CO2 from the atmosphere versus 2,536 million lbs of CO2 for a fleet of delivery truck. More than four times the amount of CO2.

Semi trucks in the U.S. simply travel far more miles, use much more fuel, and in turn emit a lot more pollution than delivery trucks.

Thanks, Elon.

A fleet of Tesla semi trucks could as result significantly help the U.S. meet the Paris Climate Change Accord. The agreement calls for a 26% reduction in carbon emissions in 2025 from 2005 levels. The U.S. emitted 7,313M metric tons of CO2 in 2005. The goal is thus to emit 5,412M metric tons of CO2 in 2025. Americans generated 6,587 metric tons of CO2 in 2015, so we still need to cut 1,175M metric tons or 2,590B lbs of CO2 per year. About 1.45% of that amount or 37.54B lbs CO2 comes from trucking. A fleet of 57,418 Tesla Semi trucks could single-handedly cut that by 30%.

Mr. Musk and Tesla are effectively building a semi truck to save the planet. Money comes second.

That leaves the delivery truck market for others to profit. Daimler is already experimenting with their Vision One and eCanter trucks for regional and local deliveries. Workhorse is working with UPS and others to build electric walk-in vans for courier service. We also find startups such as Chanje and Boulder Electric fighting for a piece of the pie. A pie that’s big enough for many players to thrive. Maybe even robots.

[This was an excerpt from our special report on Tesla’s Semi Strategy: comparing operating cost of the eletric Semi against diesel trucks, evaluating whether Tesla would make generate more revenue selling delivery trucks instead, and analyzing risks facing the truck program.]