Tesla can easily squeeze *at least* 4,000 cycles out of an LFP cell with *today's* technology. A 250-mile car at 4,000 cycles is 1 million miles. We already know the motor and vehicle structure can last that long.
Right now an SR Model 3 with the CATL LFP cells from Giga Shanghai probably costs less than $30k to produce. With full casting, new paint shop etc it could easily get down to $25k or less, which over a million mile life is $0.025/mile.
Solar PV, wind and batteries are still setting new records every year for cheapest energy in history. If you look at RethinkX's projections (Tony Seba's team) let's say the price of electricity is going to fall 10x from today's typical price of about $0.15/kWh to perhaps $0.015/kWh by 2030. Model 3 even driven inefficiently gets at least 3 miles/kWh. Energy cost: $0.005/mile.
Then, Teslas will continue to get better at avoiding accidents with software upgrades. So insurance cost eventually also comes down 10x because the risk comes down 10x. Let's say that comes down from today's average vehicle insurance price of around $0.10/mile to $0.01/mi.
Reupholstering the interior and repainting the exterior 3 times over vehicle life is maybe $10k total, or $0.01/mile.
Tires, cleaning, wipers + fluid and other basic maintenance is perhaps $0.02/mile at most, same as it is today.
Road tax will need to be added, let's estimate $0.01/mile.
Total cost of Model 3 SR with LFP pack: $0.08/mi.
Let that sink in.
Currently driving costs at least $0.60/mi for an equivalent experience. This is a 7.5x cost reduction!! This is literally getting close to competing with the shoe replacement cost of walking!!
Not all of this will show up in margins but the value proposition is incredible. And no one else has most of what needs to be in place for this to work at this level, plus the packs very well might last longer than 1million miles. Additionally, this assumes no end of life scrap value of the battery and the rest of the car. So $0.05/mile is a plausible outcome by 2030.
4/4/21 addendum: This also doesn’t account for revenue from using the car as a grid battery nor the efficiency gains from using Boring company tunnels for a portion of miles. And in a boring tunnel the pavement will be nice and smooth, energy efficiency will be more like 4 or 5 miles/kWh, stopping and starting will be uncommon, and crashes will almost never happen. So, potentially drag down further the costs for electricity, tires, brakes, and insurance.
Finally, they also will not use only Model 3 SR. Tesla Network vehicles will have many 1-2 seat cars and larger 8-16 seat vans . So the $0.025 per seat lifetime average cost estimate might be 2-3x too high when considering the overall reduction in empty seats vs Uber’s current fleet.