> $50 saving per bulb- holy shintos!
... and that's with your cheap US electricity prices!
> Light bulbs are a small part of household electricity usage- water heater, heating/cooling, appliances, etc take more energy.
Well, that really depends on where you live, though. Here in Germany, heating mostly uses natural gas and oil, and AC is uncommon in homes, so lighting is a bigger percentage of electricity use. And potentially also of energy use, because heating with electricity is way less efficient than burning the fuel in your home in terms of primary energy use.
But also, you just have to take the efficiency gains that you can get, and that one was, overall, an easy one to get. If heating is 90% of your electricity use, but you have no way to make that more efficient, it's of no use that it's 90% of your electricity use. Lighting was overall easy to do, plus it didn't even cost anything, but rather saves everyone a not that insignificant amount of money now. Plus, for the migration to renewable energy, the efficiency of lighting is more important than the efficiency of other energy uses: You only need AC when the sun is shining and thus solar power is available, while you need lighting when the sun is down, and people generally want to be able to switch on light whenever they feel like it, while most bigger consumers of electric energy can deal relatively well with shifting their load profile: You can stop charging your car, or heating your house, or cooling your fridge, or washing your clothes for an hour of high demand/low availability without any impact on usability, but just shutting off all lights isn't really an option.
> changes in industrial and transportation would have a bigger impact
Well, maybe, but at what cost? It's trivial to have a "bigger impact" if you have infinite resources to spend. The interesting question is how much impact you can have per investment, and one that gives you a 50 bucks return on every dollar invested (or so) is pretty good based on that measure.
> As a general rule it doesn't make sense to target areas that will cause the most disruption
Except ... this didn't cause any disruption? You had good lighting options available all the way, the migration to LEDs was to be expected, and now everyone is saving money. Where is the disruption in that?
> I would be curious if electricity usage did not drop at all.
Chances are it didn't, but that's simply because people have more devices that use electricity now. I.e.: It would have risen if not for LEDs. But then, the switch to LEDs, while most visible, is not the only thing that's going on, the EU at least also has regulation on the efficiency and standby consumption of power supplies now, which also reduces energy consumption of new devices (or old ones, if you swap out the power supply).
> "oh it is an LED and is efficient so I will leave it on all day" I am sure there is a name for the effect/paradox.
I think "rebound effect" is the term, and I suspect there's probably some of that. But then, it's an about 8-fold increase in efficiency, so it's not that easy to use up again. I am pretty sure people don't use more than eight times the light now than before, and if they use twice as much, it's still clearly a win.
> As far as I know impact wasn't measured.
Well, it's probably difficult to do so, to any degree of accuracy, so possibly not. But it seems very unlikely that it was useless.