3 ms·
AIUI (as someone who as on one of these plans), the primary provider, Griddy, went into bankruptcy (and was scapegoated for making a plan that had regulatory ap
by abathur 4y ago
AIUI (as someone who as on one of these plans), the primary provider, Griddy, went into bankruptcy (and was scapegoated for making a plan that had regulatory approval...).
A lesser provider, Octopus Energy (which is rooted in the UK) disabled new subscriptions for their market-rate plan and has probably phased out all users by now. They have been running a month-to-month plan, though they were giving customers $100 to leave the month-to-month service (in late May? Early June?).
I'm not 100% sure what this is about. They framed it as being about having to charge really high rates as the summer went on, but it could since it too is a pre-deposit plan, they might be fearing a similar circumstance (grid goes into crisis, prices once again get pegged at $8.99/kwh, and they get stuck with the choice of either turning people off and letting them potentially die in the heat, or being stuck with a massive shortfall).
FWIW, Griddy tried this as well (shedding customers), but the notice came late on a Friday and it wasn't clear who (if anyone) would be taking customers on a Saturday.
That aside, a cap would help, though it's not clear to me that the power infrastructure has controls that are fine-grained enough to actually enforce it? Unless maybe you had your own ~smart-breaker?
- xoa 4y ago>That aside, a cap would help, though it's not clear to me that the power infrastructure has controls that are fine-grained enough to actually enforce it? Unless maybe you had your own ~smart-breaker? Yeah I was thinking it'd be CPE, not on the provider-side since I have no idea what the Texas' grid capabilities are like but my default assumption for anything grid related based on experiences up north is that there is a lot of ancient stuff there that nobody is touching ever. But CPE should fine in this case (installation cost aside), the provider can offer pricing information in real time and each customer could then decide when to drop off the grid themselves. Which I think would also be politically helpful in terms of agency and thus avoiding that scapegoating situation. With customer set CPE nobody ever gets "cut off" they "decide the cost is no longer worth it until price goes back down" so it's still their choice, they get very cheap power much of the year. Could also offer opportunities for partnership with players like Tesla or other newer power storage providers, the Tesla PWs have some customization modes already (see Powerwall Modes [0]). Not hard to envision extending that interface to handle going off grid based on price supplied by the provider, which then has the added advantage of meaning the customer doesn't actually lose power immediately they have 13+ kWh of buffer first. That could further encourage users to drop sooner vs later. Sort of a shame Griddy died, I don't think the idea of that kind of provider or rapid market response is necessarily a bad one long term. But they didn't flesh things out enough in terms of the customer experience and ability to react to negative price signals not use the upside of positive ones. Maybe it was just too early, and some sort of energy storage as a practical matter would be a requirement for a market-rate plan to work for the general public. I think in the next 10-20 years that might become sort of a gimme, if only because BEVs (and 2-way power work seems to be rapidly developing too) will become common enough that lots of people will have a big chunk of storage "by default" when they're home. Perhaps then a market-rate plan can see a revival all over. ---- 0: https://www.tesla.com/support/energy/powerwall/mobile-app/powerwall-modes https://www.tesla.com/support/energy/powerwall/mobile-app/po...