Let's dissect some of the EV charging hype using $tsla Superchargers as an example

Spoiler: The hype is way overdone

$beem $blnk $chpt $clii $snpr

Backdrop:

EV charging pubcos are trading at crazy multiples (e.g. $blnk at 65x 2022 revs, $chpt at 40x 2025 EBITDA) on hype around the Biden plan to install 500k public charging "stations" by 2030, up from ~45k today

Big goals like 500k stations are seen as necessary to address range anxiety

Tesla's Supercharger network density/utilization may offer some clues about whether 500k are actually needed

Range anxiety not really stopping Tesla from selling cars, i.e. the network is sufficient

Tesla had to build the network to in order to remove range anxiety as an obstacle to selling cars

It's a mostly closed loop, i.e. it satisfies pretty much all public charging requirements for Tesla owners

It's DC fast charging (DCFC), i.e. you can get an 80% charge pretty quickly

Most non-Tesla public chargers are not DCFC, i.e. they take many hours to get an 80% charge

In order to be commercially viable, the future has to be DCFC. It's the only way to for the charging station pubcos to sell enough kWh in the limited about of time a customer is willing to park at a spot with a charger

People won't (and don't) hang out for 5 hours to charge

14 million EVs on the road in 2030 is an approximation of various sell-side estimates

There are typically <3 plugs/station nationally . That means 500k stations would have 1.5 million plugs

That's 9.3 EVs per plug. $chpt underwrites to a 7:1 ratio. Cowen assumes 10:1

Do we need that many plugs?

Tesla's Supercharger history indicates the answer is no

Never went lower than 22 EVs per plug. Currently ~60 EVs per plug

7-8-9-10 EVs per plug as implied by the Biden plan and assumed by charging pubcos would be a dramatic overshoot

Tesla has disclosed # of stations since 2013 but only started disclosing # of plugs in 2017

I assumed 7 plugs per station for pre-2017 years. Seems reasonable I guess.

The most dense the Supercharger network got was 22 EVs per plug in 2014.

If that number is the north star for the country, would imply that if we get to 14 million EVs by 2030, only 635k plugs will be needed

At 3 plugs/station, would imply 212k stations (vs 500k goal)

The least dense the Supercharger network got was 60 EVs per plug in 2020

If we get to 14 million EVs by 2030, the 60:1 figure would imply 235k plugs needed. At 3 plugs/station, would imply the country needs 78k stations

500k stations is way too many

EVgo $clii is trading at 7x 2027 EBITDA

They seem to assume their utilization will ramp from 0.8% in 2021 to 12% in 2027, enabling them to flip from a $58 million EBITDA loss to positive $500 million in 2027, as they build out their DCFC network

Is it realistic to assume they can reach 12% utilization as the country builds towards 500k stations and a ratio of just 9.3 EVs per plug?

Here is an ominous anecdote

In July 2019 Tesla disclosed the Supercharger network delivered 72 GWh

At the time they had installed 13,881 plugs and had sold almost 250k cars cumulatively (18:1 car to plug ratio - far less dense than the Biden plan at 9:1)

Was utilization high enough to validate the >10% rates in EVgo's SPAC forecasts?

Not even close

Utilization was just 5%, assuming 150 kW capacity per plug and the 13,881 plugs

Keep in mind that was before WFH, which reduced VMT

It's plausible that utilization is lower now

So, are we as a nation going to build to 2x as dense as the Supercharger network (9 EVs/plug vs 18 EVs/plug), but utilization for pubcos like EVgo is going be more than double that of Superchargers (>10% vs 5%)?

I don't think so

This dog don't hunt