As you might have heard by now, Elon Musk, CEO of Tesla (TSLA 0.79%), took a view on his company's stock price last week that was...perhaps not perfectly aligned with the interests of Tesla investors.

But was he right? Of course he was. In my view at least, Tesla's stock is absurdly overvalued. I think it's because Tesla bulls insist on believing some things that just aren't true -- like that Tesla is anything but an automaker. 

Tesla is an automaker, not a software company

Tesla is in Silicon Valley, its products are sleek and futuristic, and run on advanced software. Clearly, Tesla is a software company that happens to make a bunch of things including cars, and so it's deserving of a software-company valuation. Right?

Nope. Nearly all of its revenue comes from making and selling cars. It's an automaker with some great software, but it's an automaker. That has big implications for its valuation. 

But what about self-driving?

What about it? Tesla isn't a leader in that space now. And despite what Elon might have said in the past, true Level 5 self-driving isn't happening for many years.

Despite the hype, Tesla has no advantage in self-driving versus the mainstream auto industry. (Meaning, that to the extent that self-driving eventually adds to Tesla's value, it'll add to the value of other automakers, too.)

A close-up of the Tesla logo on a Supercharge recharging unit.

Tesla's valuation has been supercharged in recent months. Is Elon right that it's now overcharged? Image source: Tesla.

Automakers, even Tesla, have very high fixed costs

Here's the thing about making cars: It's expensive in ways that tech investors just don't understand. Manufacturing cars requires insanely costly tooling that has to be paid for whether you sell 10 cars or 10,000 this month. Those costs, and the fact that Tesla can't generate much revenue without paying them, are why Tesla really isn't a software company.

Despite what Tesla fans on Reddit might tell you, the company doesn't get around these high tooling costs because its cars are electric. Example: The most expensive pieces of automotive tooling are the dies used to stamp sheet-metal body panels. Tesla has to pay just as much for those as every other automaker: hundreds of millions of dollars. 

Automakers don't generate software-company profit margins

That's just one example. The point is that the costs of making autos are huge, whether you're making electric cars or internal-combustion vehicles. Many of those costs have to be paid in advance, before the company can sell a single vehicle.

That's why automaking, unlike software, is a low-profit-margin business. Tesla doesn't get a magic pass on that just because its cars are electric, or because Elon Musk also runs a rocket company, or for any other reason. 

Companies like BMW (BMWYY -0.99%) or Mercedes-Benz parent Daimler (MBGY.Y -1.11%) typically have operating margins around 10% to 12% when everything's going well. Tesla has a terrific brand, but there's no realistic scenario where it's going to be able to do much better than that, even in the best of times.

Autos are a cyclical business, and Tesla isn't immune

Those high fixed costs are also why automakers' profits get squeezed (or go away entirely) when sales fall, as during recessions. That's why we say that autos are a cyclical industry: Profits rise and fall with consumer confidence.

We haven't seen it show up in Tesla's results yet, but it's a super-safe bet that its sales will also fall in recessions, including the one that we're probably in right now. It's all about consumer confidence: Consumers (and companies that operate fleets) tend to postpone purchases of vehicles -- even super-sleek electric vehicles -- when they're feeling uncertain about their income or cash flow.  

The realistic best case for Tesla: It could look like BMW -- in 10 years

We hear a lot about how Tesla is the leading electric-vehicle maker. It's true! But the 367,500 electric vehicles that Tesla sold in 2019 represented just 0.04% of the 91 million vehicles sold around the world last year.

Simply put, Tesla is a big fish, but the pond is still tiny. It'll grow, but assumptions that it will retain a big-fish share as it grows are absurd. 

Sure, its sales will probably rise as electric vehicles become more mainstream. But Tesla doesn't -- and won't -- have the scale to compete with global giants like Volkswagen, which is spending tens of billions of euros to ramp up to 3 million electric vehicles a year by 2025. (Even if demand exists, Tesla can't build factories and supply chains quickly enough to match that.)

That said, some bullish Wall Street analysts, including Morgan Stanley's Adam Jonas, think that Tesla could get to 3 million sales a year (about the size of BMW) by 2030. If we assume an average selling price of $45,000, that means that Tesla could be generating $135 billion in revenue in 10 years. At 10% operating margins, that would be operating income of $13.5 billion. 

That would be a nice business, but at that point, Tesla is almost certainly a mature automaker, with limited sales-growth potential (at least in autos). Assign a mature automaker's valuation of 8 to 10 times earnings, and we get a valuation between $108 billion and $135 billion.

In 10 years.

I think that's the realistic best case for Tesla, or at least the automotive portion of it. 

What's that worth now, given the (huge) risks between now and then? 

I can't say for sure, but I'm confident that it's a whole lot less than Tesla's current $142 billion market cap. Elon is right.