The panels are dirt cheap compared to the other costs (installation, permitting, inverters, etc...). I suspect that the installation costs and any permit modifi

The panels are dirt cheap compared to the other costs (installation, permitting, inverters, etc...). I suspect that the installation costs and any permit modifications required during the expansion would greatly extend the ROI timeline vs just including the panels in the initial install. If you were really trying to save costs, maybe you could significantly undersize the inverter, as those tend to only last 7-8 years, and you would only suffer losses during the peak hours in the summer months when you are likely to have way more energy than needed. More generally, overbuilding is not really going to be economical right now. We do not have the proper incentives in place for that. But from what I can tell it is our quickest/cheapest path to a sustainable grid, and that likely means we will need to adjust our consumption patterns to account for the absurd amount of excessive power we will have in peak solar months (usually late spring). This means we will need to start developing power intensive applications that can take advantage of free (and possibly negatively priced) power. So if you want to get ahead of the curve, put a massive solar system on your roof and start thinking up interesting ways to use that excess power.