AC is obviously a main driver of peak demands, and why the US summer peaks are higher than the winter peaks. Refrigeration is not as peaky, but is primarily focused on the hours of the day that the door is being opened and closed ~ and as a heat source inside the house, increases the load on AC.

Energy efficiency might cut 8 percentage points out of that. If half of the balance can be shifted from on-demand to on-supply-available, shifting 10% of average demand into, say, 8 off-peak hours could easily represent a doubling of off-peak demand.

In both of those cases, there are substantial efficiency untapped gains available right now if there was Connie Mae style financing available with payments made as part of utility bills, and the first step is to reduce the size of the peak by reducing wasted energy during the peak. But for the refrigeration case especially, its the energy efficiency that requires an overhaul in what people expect to see when they buy a refrigerator ~ given a switch from intrinsically inefficient uprights to intrinsically efficient drawer refrigerators, the switch to the smart grid shifting of refrigerating load to periods when power is being offered cheap is so inexpensive to add that you'd not notice it.

Of course, you have to have the smart grid available to pick all the low hanging fruit that may be there to pick once the smart grid is in place.

I've been accused of being a Marxist, yet while Harpo's my favourite, it's Groucho I'm always quoting. Odd, that.