"Grid parity" is a term used by a lot of folks, and its definition is not solidly fixed. Here's one explanation. Here's another. And this. And here's a grid parity map plus a 2017-projection. Since much of Solar PV electricity is generated during the peak demand period of the day (when demand for electricity peaks), this "peak load cost" analysis is useful.
All of the "grid parity" analysis leads one to thinking about future Solar PV costs and thus the economic viability of Solar PV itself. Here's a $1/watt projection, by the way, plus a subsidy-free, grid-parity projection for India. Here's North Carolina's projection, too.
Here's a 2013 analysis showing unsubsidized solar PV growth projections in Germany.
Here's a September, 2013 analysis showing how close solar now is to parity. My grid-operator friend's take on it:
*7 to 10 cents/kWh for utility scale solar in the US currently.
*5-ish cents/kWh for utility scale solar in 2015
All these clowns fighting solar. This is why. The WSJ and Bloomberg are going to have to change their tune soon. I've got to think those greedy bastards are all going to start jumping ship.
Obama's move against new coal is odd considering that these numbers suggest that solar can EASILY beat new coal in a straight up fight. Obama's move is empty under these circumstances. Now, I know you're worried about the variability problem. For me it's not an issue because when you aggregate lots and lots of small plants that are individually variable you end up with a smooth predictable output. It's the same with load. If everybody turned on their hair dryers at the same time we'd be hooped but that just doesn't happen. In utility speak we call that "load diversity". Basic idea is that load is this million ups and downs that aggregate into a smooth demand. Someday we may talk about "solar diversity" in the same way.