If the cost of buying the land, installing the equipment, and maintaining it (thieves come and cut away the copper cables, for example) were all free, then this

If the cost of buying the land, installing the equipment, and maintaining it (thieves come and cut away the copper cables, for example) were all free, then this would be a valid argument that the charging stations turn a profit. In other words, you can't ignore fixed costs and cost of capital. If you take out a loan to buy $8 billion worth of land and deploy chargers there (about how much Tesla's network cost) - say 40,000 chargers, so a fully loaded cost of 200K per charger (land + equipment). At a 5% rate, that charger would need to generate a profit of $10K per year, but at a rate of 10% it would need to earn $20K per year. Average electricity costs are 14 center per kwh, but in California or the Northeast, it's closer to 20 cents. Let's say Tesla makes 10 cents per kwh profit by adding 10 cents to their local costs, so it would need to sell 100,000 kwh per charger per year to break even at a 5% cost of capital, and 200,000 kwh at a 10% rate. This assumes zero maintenance, labor, or operational costs. Assuming L3 chargers operate at 200 kw that means operating 500 hours each year at a 5% rate, or 1,000 hours each year at a 10% cost of capital. That's certainly doable, except you are not supposed to use L3 to regularly charge your car, as it kills the battery. This is why Tesla has been opening up the charging stations to other manufacturers -- it wants as many people to use their chargers as possible, even as it tells people to please not use L3 charging except in rare circumstances. In other words, the biggest factor limiting the viability of this system is whether they can get enough cars to regularly use what is supposed to be an emergency charging system. We can also do the reverse calculation, and let's say the superchargers are humming at being used 8 hours each day 365 days per year, so ~3000 hours. That's a gross margin of $60,000. Subtract out the cost of capital -- say $10-20,000 and you are left with $40-50K of operational costs. That's money spent on labor, replacing worn parts, paying local taxes, etc. If each car is supposed to be supercharged only 20 hours a year, (road trip or emergency use case), then you'd need a fleet of 120 million cars to use the existing network of 40,000 chargers sufficiently to generate that 8 hours per day usage. Again, a tough call. Obviously if you tweak these assumptions, you'll get different answers, but I did this to illustrate the types of trade offs needed beyond just looking at the energy spreads. My guess is that's it's too early to tell whether this pencils out for Tesla or not, but the key limiting factor is going to be the size of the fleet using the chargers rather than the electricity spreads. IMO this is a big bet on growth.