This is all very well and good, but does it scale? Suppose every single house and business the grid served had enough capacity to go net zero. That is, they generate their entire day's budget of power, including nighttime use, using their panels. This means that on sunny days, the power has to get stored, because by definition everybody is generating more than the total they need during daylight hours. Or else it just goes to a resistor bank or something.
And then on cloudy days, and at night, the energy has to come either from grid-tied storage, or from non-solar generators (one would like for that to be wind, but it's not a perfect solution at the moment). And of course there is a very substantial cost to actually maintaining the grid. So in this scenario, the cost of the non-solar generation capacity and the grid has to be paid for by someone; if everybody who is connected to the grid is paying zero, or slightly less, then the money has to come from somewhere, and that's going to be the power company.
Of course, that's one extreme; the other is no site-generated power. You can draw a graph; on the left zero net-zero sites, on the right, 100% net zero sites. For some part of the left-hand side of the chart, solar is making the power company's life easier, because demand for power is higher during the day than at night. For some part past that, solar is neutral—it doesn't particularly benefit the company, but it's not a negative either—they are able to sell the power, assuming they are paying a fair price for it. And then at some point on the right, there is no longer sufficient revenue to pay for the grid and the non-solar generating capacity needed to run all those net-zero houses when the sun isn't out either because it's cloudy or nighttime. Now the money to pay for the grid has to come from people who are net zero.
And somewhere before that happens, to the right side of the graph, the money that pays for the non-solar generating capacity and the grid will all be coming from people who don't have solar, even though the people who have solar also benefit both from the grid and the non-solar generation.
So when you talk about "usefulness to the grid," it's important to keep this graph in mind. At some points on the graph, that term is meaningful. At other points on the graph, it isn't. And it is absolutely unfair for users who do not have solar to subsidize users who do—in general, users without solar will be poor, and users with will be affluent, so you have a reverse subsidy.
Andrea and I have solar on our roof, and we're happy to have it, and we get a subsidy, which apparently works out well for Green Mountain Power at the moment. I'm happy that's the case, because at the moment our excess generating capacity actually benefits other users of the grid. But when the point comes where the grid exists largely to spread out the power and provide night capacity, that will no longer be the case, and it would be shameful if I were to ask for the subsidy to continue at that point, paid for largely by people less fortunate than I am, or by running the power company that maintains the grid I depend on into the ground.