Cost Accounting can be a bit of a dark art. It's pretty tricky to apportion part of the cost of a factory to every different widget made in it.

But in this case, it's likely just BS. They're probably just taking the yearly loss and dividing it by the number produced. Which is different from each marginal car produced costing them 30K.

edit to add: the article claims the main reason for them losing so much money is lack of throughput, which demonstrates that the claim is bullshit. If they made even more cars and "lost 30K" on each one they'd be worse off.

> edit to add: the article claims the main reason for them losing so much money is lack of throughput, which demonstrates that the claim is bullshit. If they made even more cars and "lost 30K" on each one they'd be worse off.

Hah! I knew I wasn't the only one who noticed that. It's like the old joke:

> Q: How do you expect to stay in business if you lose money on every sale?

> A: Easy. We'll make it up on volume!

More seriously, a production line has huge fixed costs associated with it, so you really need to get the most throughput out of it that you can, in order to maximize profits.

> the article claims the main reason for them losing so much money is lack of throughput, which demonstrates that the claim is bullshit. If they made even more cars and "lost 30K" on each one they'd be worse off.

Generally speaking, the cost per unit goes down as throughput goes up, especially in the car manufacturing space. Production rate is insanely important to car manufacturers making any money at all.

Virtually every single new mass produced vehicle ever made started off losing money on a per-unit basis until manufacturing scales.

Exactly right.

cost per unit = cost per unit + total non-unit costs/total units.

So they're loosing money overall, but the unit economics may still work.

Generally speaking, you can't generalize vehicle production with new companies, or even old ones. Vehicle production is extremely complicated and expensive. Minor quality errors lead to increasing delays and costs.

Famously, GM used to spend ridiculous amounts of time and money to fix cars coming out of some of its plants. It wasn't uncommon for some cars to be incapable of being driven out of the plant. The more cars they made, the more errors, and the more time and money spent on fixing them. This was a major factor in the development of the NUMMI plant.

What brings cost down is increased quality, reduction of waste, and increase of speed. If you don't have all three you aren't gonna save money.

Tesla famously was at one time spending almost $200K per car because it sucked at building so badly. It's completely plausible that a SV-inspired company with no experience manufacturing cars would lose a whole bundle on pretending it could do it cost effectively, while bilking investors for more cash to stay afloat

Automotive pricing is complicated. The cost probably was predicted to be profitable at some point, but costs change over time and changing cost targets is difficult. Here's some random speculation:

* The amount a vehicle costs to produce decreases dramatically depending on how many you're producing because of economies of scale, vendors charging less, and higher negotiating power in volume. They're currently not producing at the scale they expected, so their production costs are inherently higher.

* It's extremely difficult to know where you can cut costs in ways that won't affect customers significantly at design time. If you're a new startup in a competitive market, a bad reputation is a death sentence. It might be worth starting out conservatively and cutting those costs when you have better data.

* They went through production/design in one of the worst supplier markets in decades. Production costs might have increased due to last minute part changes, now they need to claw that back down again.

Etc.

I assume the plan is to become more efficient over time to the point of being profitable. This is a very common pattern even in software. Basically they are dumping product using VC money.

Not selling any vehicles would cost them even more. They spent a lot of money developing cars and if you only sell a few, you're going to make a loss.

I haven’t seen Rivian financial reports but I have on some other EV manufacturers like Ford. A lot of those have negative bottom lines because they are making large capitol expenditures on equipment and factories for batteries and production lines. You can’t just take that number and divide by the number of vehicles sold this year and say “look at the huge loss!”. That’s not how capitol investment works.