You did bring it up, perhaps your wording was sub-optimal. So, "clearly" if you have laws on the books that allow you to get credit for the power you feed back at retail / TOU rates, this benefit is not sustainable, and I would expect it to change. YMMV. Many places that had these rules, bootstrapped these subsidies so rich people could nail up some solar on their roof and not pay as much, have already lost them, and if they are still in play are likely going to lose them at some point.
Thus as it relates to TOU experience, I'd say if you could capitalize on it with some sort of guarantee of a time frame, I would factor all that in if you are trying to get an ROI on your solar investment. TOU/Peak shaving can be a significant benefit if your peak rates are high enough, but I wouldn't bet on any significant buyback sticking around. I know some folks here are still grandfathered into some generous agreements. If you want to grid tie here you must now run on a demand plan not TOU. I would expect to see more of that as well, because the biggest advantage to the power company is that demand shaving. Since utilities are heavily regulated, you can always use the pressure of the voters box to vote your desired subsidy in.
The math on the benefit of TOU/load shaving is fairly straightforward, I would just factor in a 20% loss, and discard any potential back-feed credit, then you should do better than your estimate. The maximal ROI is going to be with micro-inverters, the power company would prefer you had a small battery setup to smooth the dips on partly cloudy days ( if you want to be a good Samaritan ). If your peak TOU rate is not over ~$0.18 then you are likely not going to get any ROI.
Here, winter TOU is < .08/.11ish. It's obnoxious, 5-9A,5-9P. The PM setup it eay, the 5AM is tough without sufficient battery. You really need a good handle on usage if you don't get an excess of battery. My winter usage is around 45KWH/day, pretty consistent day/nite depending on the heat pump. With 60KWH of battery at leas 60%, I can make thru the nite. When that fails, (it's been crap the last 5+ days) I'll leverage TOU times and directly charge the cars from the grid, otherwise I just shut down the output on my inverters until I get enough battery buffer. It's not really worth the $$ to try and save 3 pennies a KWH, for at most 8 hours, figuring 2kwh/hour that yields a whopping 0.16*3 = 0.48/day at best. The summer is a totally different game, TOU is 1300-2000 pretty much the heart of solar output, with peak rates hitting 0.24 in July and August. In this case with a modicum of battery, you can get a huge benefit, as my electric usage hits 100+ KWH / day, and the battery stretches my output well past 8PM. Even so, the ROI is not all that great, TOU with rates much lower than what I'm paying simply would not pay pack.
Since wholesale (negotiated) non-demand rates run about 0.03+/kwh over most of North America, I would not expect to see any off peak rates much below 0.07/KWH. . . . Back to that margin thing. Demand rates OTOH can really spike, so depending on the capability of your power company to ramp up to meet high demand, they may have to purchase at obscene rates. Long term, if solar builds out significantly in places like here where there is a direct correlation: More sunshine = more demand, an abundance of solar could actually stabilize or even lower TOU peak pricing, since they can provide service without tapping into more expensive generation, or spot-market pricing. I think the general cost of generation is going to continue to rise, which is also likely to narrow the TOU benefit gap from a percentage standpoint, but could improve overall ROI for a solar investment.