Seems all normal though?
Warranty reserve increase should scale with sales.
Warranty payments should scale with fleet size and age.
Hertz hasn't purchased 100k Teslas yet, and the repair issues they had were likely non-warranty things like bodywork or other user induced damage. The data from that accelerated mileage rental fleet should provide Tesla guidance on the accuracy of their per-unit warranty reserve number.
"Seems all normal though?"
It did through 2022, but diverged in 2023. That is why reserve allocation per vehicle sold rose is 2023.
"Warranty reserve increase should scale with sales."
Yes, with reservations. The model mix is a material fact. Generally the least complex vehicles have much lower reserve allocation, so base model 3Y should be lower reserve. Clearly they are not.
Not really. It should typically scale with fleet under warranty, adjusted for time in service. Generally, most warranty claims happen in the first year of service, but claim severity increases with time in service. That means processing costs reduce over time, claim severity increases and claim number deciine. Sounds pedantic but is not.
As for Hertz, that is the only reply obvious departure from the norm, that I can find thus far anyway. There si not enough public information to judge its relevance, but claims history equivalents show the highest single correlation with warranty claims is vehicle care and condition. This assertion comes without data shown here, but rest assured it is true. Fleet use always has higher warranty claims than non-fleet use in every bit of data I have ever seen. It even works for aircraft as well as cars and trucks.
The proof that there is something wrong is found in that PWC note that i attached. Such a note NEVER happens without a significant concern. That the per vehicle charge to reserves rose could be innocuous, but not when that note was included. That this happened without a qualification meant there was some compromise and, probably, dissent. Undoubtably that was both within PWC and TSLA.
I do not want to blow this out of proportion, but it is significant. We'll soon see evidence when fleet Tesla buyers ahem more extensive support and end-user training begins to happen.
Past cases include some major ones, such as BMW 7 series iDrive in 2001, Cadillac V8-6-4an a litany of others. There is a distinction : The Cadillac was a technology mistake. The BMW actually had zero technical issues, but the first iDrive was so odd that they had to include instruction on starting the car.
Were I to guess I would think this Model 3 situation si comparable to that of the BMW. Their si nothing wrong with the car, it's just hard for 'civilian use. That ends out with a warranty problem, in bean-counter terms.