In the US, land is widely available at around $1000/acre. The current cost to cover that with PV is around $100,000/acre (depending on the fill factor).
So, land costs are not a substantial obstacle to reduction in the cost of PV energy.
Land is just an example, running new HV power lines from ultra cheap land to consumers is another cost that’s not dropping.
If hypothetically ~10% of costs are fixed the maximum theoretical cost reduction is 90%. Which the quoted 10% annual cost reductions in solar hit in 22 years. But more realistically the closer you approach theoretical maximum the more difficult it is to improve at the existing rate.
Or, you know, we move industry out to be closer to the PV fields. Globally, energy intensive industry probably moves to places like Chile with very high insolation. Too bad, Europe.
This is wrong. Steel manufacturing happens near ports.
We can't shift cities to electricity generation. We historically choose population centres in places with optimal conditions like having good water flow or natural protection with mountain ranges or good level land.
Most industry needs access to transport with shipping being the cheapest.
It helps out a bunch that there is a nuclear power plant outside of Phoenix and the Hoover dam in Vegas generating all kinds of carbon-free electricity to power the air conditioning.
1000 per acre, maybe in the Mojave desert with a parcel size over 1000 acres. Acreage near anything that isn't a ghost town is well into 10-20k per acre. Land has an inherit scarcity. Investment funds and even other countries are buying up US Land quickly. Land is always the problem. Proximity and transition are also a problem.