doyouwanttoknow?

Tesla shareholder since 2015, FSD supervisor

[doyouwanttoknow? @climateyupa]

@JuanFelipeRiano I know heatmap collaborated with MIT recently to develop a 5 year view of electricity prices on the very-local level.

https://t.co/QSTfoKiPKM

[doyouwanttoknow? @climateyupa]

At least on the delivery side, it is broadly true that data centers and other large loads spread fixed costs over more usage and lower electric bills, especially with the added protection layered on from many states adopting large load tariffs.

The supply side is where upward pressure may occur in deregulated markets, due to the fundamental market mechanism that consistently elevated prices must incentivize new generation.

[doyouwanttoknow? @climateyupa]

@prieurdp Backing in is considered a safer maneuver than backing out:

https://t.co/AOGq4qhnrs

https://t.co/ZuuTNrMcbj

https://t.co/iPxR6IIDLJ

[doyouwanttoknow? @climateyupa]

@BryceJohanneck Transmission rebuilds?

[doyouwanttoknow? @climateyupa]

@saeverley PA has some of the most stringent cost allocations for data centers and other large loads (!!!) after the recent order. They require large loads to pay for all system upgrades that weren’t already in upgrade plans entirely up front! This is a decidedly anti-development stance

[doyouwanttoknow? @climateyupa]

@duncancampbell @mitrajoy_ @jaeporeon These are wells, not public water

[doyouwanttoknow? @climateyupa]

@swingingcarrot @nerdalert Tesla must be valued at some multiple of SpaceX for me to vote yes on the merger. Merger of equals won’t cut it.

[doyouwanttoknow? @climateyupa]

@SlBrandin What does this mean in practice? How do they advocate for the consumer other than building as much new generation as possible

[doyouwanttoknow? @climateyupa]

“Unless the differential is directed towards the new customer” is doing a lot of the heavy lifting here.

New cost is always directed towards new customers to some extent, but especially so with power-intensive customers - the allocation of transmission cost to a rate class is based on their power usage. So when new customers drive usage higher, more of the total grid costs is allocated to their class.

The chart attached from that report is most instructive on how new load affects rates. And the most consequential question is ultimately does incremental revenue exceed incremental costs? (bottom right).

It turns out, as historically has been the case, the embedded cost of the system is large, incremental costs are relatively small, and generally new load spreads the cost over more usage. The cost pie gets bigger but the new load eats more of the cost pie, which on net creates downward pressure.

There is an incremental cost per unit of power at which that breaks, but there is no evidence we are at that point.

[doyouwanttoknow? @climateyupa]

https://t.co/cuI1JNQF5f

As has been the case historically, load growth generally has downward pressure on delivery rates.

Incremental cost being higher than embedded cost isn’t a problem. The problem is if incremental cost exceeds incremental revenue, as is well laid out in a recent ESIG/Brattle white paper: https://t.co/kkamU0KlKH

[doyouwanttoknow? @climateyupa]

Absolutely not fair. Tesla has historically traded 2-4x the value of SpaceX, and sometimes much much higher. The lowest ever was 1.3.

So 1:1 would be the lowest ever and the reason would be Elon having higher ownership of SpaceX and Elon canceling cheaper models because he thought robotaxi would scale faster.

[doyouwanttoknow? @climateyupa]

@Everman @robotaxi Historically, Tesla has traded 2-4x SpaceX, and a few times it was much, much higher. The lowest ever was 1.3x.

“A merger of equals” is not fair.

[doyouwanttoknow? @climateyupa]

@arne__ness Sounds like the problem isn’t the data center demand, but rather the market mechanisms and the barriers to bringing in new supply.

[doyouwanttoknow? @climateyupa]

There is really solid data and economic theory support that, at least on the delivery side, load growth lowers rates because it spreads fixed costs over more usage, and generally incremental cost to serve is relatively low compared to the large fixed costs.

https://t.co/cuI1JNQF5f

On the supply side though, prices can rise if the markets fail to procure incremental generation to match the new demand in a timely manner.

[doyouwanttoknow? @climateyupa]

https://t.co/2NUauYz1GB…

Exactly! At least on the delivery side, load growth has historically lead to downward pressure on rates.

That’s how Samuel Insull built the modern utility model - by growing electric load to spread high fixed infrastructure costs over more usage, lowering average prices and driving even greater adoption of electricity.

We forgot this because load hasn’t grown much in 50 years plus

[doyouwanttoknow? @climateyupa]

https://t.co/cuI1JNQF5f

At least on the delivery side, load growth has historically lead to downward pressure on rates.

That’s how Samuel Insull built the modern utility model - by growing electric load to spread high fixed infrastructure costs over more usage, lowering average prices and driving even greater adoption of electricity.

We forgot this because load hasn’t grown much in 50 years plus

[Ramez Naam @ramez]

AI data center growth is not correlated with electricity price changes.

[doyouwanttoknow? @climateyupa]

@BoBbyPleWniaK Historically, Tesla has traded at 2-4x SpaceX valuation, and never <1x.