Good points. I'm ambivalent about this requirement myself. From a technical perspective, I wonder what they're going to do in new subdivisions where every house is overproducing at certain times of the day/year. From a financial perspective I wonder how net metering rates are going to be pushed down when a lot more houses are participating. And I also agree that large-scale solar farms are going to be more cost-effective.
But I'm not totally against this change for a couple of reasons. One is that one of my ecologist friends is really alarmed at California's use of heretofore undeveloped desert land for solar farms when rooftop capacity is so underutilized. "Just because it's not full of trees doesn't mean it's a disposable ecosystem." Now personally I think the tradeoffs are worth it to get more solar built faster and cheaper, but it's not a universal perspective. A second related reason is that I've seen large solar farms in other regions prompt community pushback from people who dislike their appearance -- similar to, if less severe than, people objecting to visible wind farms. Those objections don't come up with distributed rooftop systems.
Right now the annual energy production per dollar of upfront cost is much lower from rooftop PV systems than from large-scale utility PV systems. If California installation costs can get down to German or Australian level, the cost effectiveness will be closer to (though still less than) large solar farms. There are reasons to believe that making rooftop systems mandatory will drive American costs down closer to German/Australian costs. For example, according to the presentation you linked, German rooftop PV installers spend $0.07/watt on customer acquisition. American installers spend $0.69/watt. If every new house is required to have PV, I expect acquisition costs to go down significantly; installers can court builders instead of trying to persuade one homeowner at a time. And investment into producing a design and a bid won't ultimately be rejected with "we decided not to add solar after all." (Though it could be rejected with "the other installer made a better offer.") Having every rooftop designed to support solar from the outset also makes installation less complicated, which should lower labor costs over retrofit-installs on 20th century housing stock.
But I'm not totally against this change for a couple of reasons. One is that one of my ecologist friends is really alarmed at California's use of heretofore undeveloped desert land for solar farms when rooftop capacity is so underutilized. "Just because it's not full of trees doesn't mean it's a disposable ecosystem." Now personally I think the tradeoffs are worth it to get more solar built faster and cheaper, but it's not a universal perspective. A second related reason is that I've seen large solar farms in other regions prompt community pushback from people who dislike their appearance -- similar to, if less severe than, people objecting to visible wind farms. Those objections don't come up with distributed rooftop systems.
Right now the annual energy production per dollar of upfront cost is much lower from rooftop PV systems than from large-scale utility PV systems. If California installation costs can get down to German or Australian level, the cost effectiveness will be closer to (though still less than) large solar farms. There are reasons to believe that making rooftop systems mandatory will drive American costs down closer to German/Australian costs. For example, according to the presentation you linked, German rooftop PV installers spend $0.07/watt on customer acquisition. American installers spend $0.69/watt. If every new house is required to have PV, I expect acquisition costs to go down significantly; installers can court builders instead of trying to persuade one homeowner at a time. And investment into producing a design and a bid won't ultimately be rejected with "we decided not to add solar after all." (Though it could be rejected with "the other installer made a better offer.") Having every rooftop designed to support solar from the outset also makes installation less complicated, which should lower labor costs over retrofit-installs on 20th century housing stock.