Hacker News Reader

Comment by roughly | Hacker News Reader

EVs are cheaper per mile than ICE, with the notable drawbacks that the range is lower, the refill time is longer, and the batteries mean the vehicle weighs more. None of those 3 matter for Amazon deliveries - they’re doing local deliveries (in stop & go traffic, where the efficiency benefits of EVs are even more pronounced), they sit in one place overnight, and for cargo vehicles the weight difference isn’t as much of an issue. Amazon’s got a history of investing in long-term logistics projects (there was an article here a while back about their investment in shipping paying off during COVID), so I’d imagine they’ve penciled this out and are doing it for financial reasons. The company is notoriously focused on financial efficiency, but not necessarily short-term financial efficiency.

View on HN

They may be cheaper per mile, but I doubt they are cheaper per delivered package when you take into account the purchase price of the vehicle at amazons 5.3% corporate bond rate (effectively the cost-of-money). The Rivian+EV charger probably costs ~$100k, whereas a regular mercedes would cost $60k.

Again, I’m sure that’s something that Amazon’s famously meticulous finance people missed in their calculations. What do you think the payoff horizon is for that kind of purchase for them?

They're doing it, so obviously they think it makes financial sense. And considering how insanely successful and good Amazon is in logistics and online retail, I'm inclined to believe they know what they're doing over some random HNer who probably doesn't know the first thing about the topic. What's your field exactly? Moreover, where are you consuming your opinions about EV from?

They're also custom made vehicles for delivery use case [1]. And they're also still a major shareholder in the company.

And long-long-term maybe some kind of integration with Zoox for an tsla/uber like delivery experience (with drones?).

[1] https://youtu.be/3CWCqJl0BEs

Even if all that is correct, 300 miles a day or so could eat up a $40k difference really fast (gas alone could be $60 a day, or $15k a year - so even baking in the cost of electricity you could save the difference in only a few years).

I would be quite surprised if the Amazon fleet average did 300 miles per day (which would be five hours at 60mph).

Over what timeframe?

Are they more expensive year one? Sure. New vehicles + chargers + infrastructure upgrades.

Year 5? Maybe it’s paid off. Energy costs, brake pads, no oil changes or other frequent maintenance.

When you refresh the EVs the infrastructure is still there, that’s one time. Gas prices go up? Your investment looks even better. Carbon tax appears? You’re golden.

Amazon must know it’s worth it. You don’t spend that kind of money for good PR alone.

And then, after putting solar on top of the warehouses and charging the batteries of the vans that aren't currently making deliveries, they get to have a closed system so they don't need to pay an external party.

Not sure about the US ones, but in Ireland they seem to use Mercedes eSprinters. These cost about 45k after grants (the new version, with a significantly bigger battery, is more like 55k, but I think they're on the old one). A normal Sprinter is more like 30k. It's not a huge difference, given the fuel cost diff.

looks like in the US, Mercedes sprinter vans start at $50k and e-sprinters at $72k.

Huh, that's a _huge_ gap. I'd wonder are those higher-spec/prosumer variants; I was using fleet prices.

EDIT: Hrm, no, a Ford Transit is also way more expensive in the US than in Europe. Wonder what's going on there. Is it just limited competition/more niche part of the market or something?

How could it be cheaper per mile and more expensive per package if the packages per mile is equivalent?

Reply on news.ycombinator.com