A new law in California will require that every new home built starting in 2020 must have solar panels. I have studied the housing price premium commanded by solar homes in this and this paper.
Who is this new law binding for? There are homes that would not have installed solar panels in the absence of the law. This is the "marginal set".
An energy economist will ask the following questions;
1. Where are these homes? Does the sun shine brightly on these areas? That the home owners would not have chosen to install solar suggest the answer.
2. What is the hourly price of electricity in the areas where the law is binding? Economics would predict that the marginal price is low and this is another reason why the buyers of new homes would not have wanted the panels in the first place.
So, either the law is not binding (because the panels would have been installed in the absence of the law).
Or the law is binding and it is forcing new home buyers to have panels that won't generate much power or that generate power but have a low marginal benefit (because the price of electricity is low).
This "essential heterogeneity" caused by the mandate to take the treatment (i.e install solar panels) suggests that the marginal benefits from forcing the marginal homes to adopt solar will be low.
Now this pessimistic logic is false if this demand mandate leads to learning by doing on the supply side by solar installers and solar panel builders. But, this needs to be tested.
There is also the Gruenspecht effect. The new law (requiring solar panels on new homes) is an implicit tax on new capital. Given that homes are long lived durables, some incumbent home owners will choose to keep their old home longer and renovate it versus tearing it down and building a new house. If new homes are more energy efficient than older homes, this new law could reduce the California residential sector's energy efficiency gains!
Substitution effects cannot be ignored.