This is because California has so much solar that it can’t use it all during the day:

https://www.gridstatus.io/charts/curtailment?iso=caiso

From here on out, batteries are where it’s at. Overprovisioning solar helps, but the demand isn’t there and they aren’t going to pay much for it. (More utility-scale batteries would increase demand for solar somewhat, for charging them.)

Meanwhile, California needs to maintain and improve the electrical grid to lower the risk of wildfires, but more solar in the wrong places doesn’t help much with that.

Batteries are just an extension of the thinking that PV or Wind are 'cheap'. There is slight U shape in the price where an initial fraction lowers the cost, but then it just additional cost overhead. Cost shifting some of the excess solar into the evening to reduce the peak load there is fine, but then your still paying for a pile of excess backup capacity to sit around idle for those days when the sun doesn't shine, and adding more batteries beyond a certain point is the same. They just sit around idle most of the time adding to the cost. I've posted here napkin math for how much a W of solar or reliable PV actually costs and been down voted but the math is easy when you stop believing that a W from PV/Wind is the same as a W from your local Gas/Coal/Hydro/Nuke plant.

All those gas plants and batteries sitting around idle soon start dominating the cost structure because the price of their produced watts starts to go exponential.