I think the key difference is whether the distortion is intentional or just incidental. All things being equal, we'd prefer fewer incidental distortions of smal

I think the key difference is whether the distortion is intentional or just incidental. All things being equal, we'd prefer fewer incidental distortions of smaller impact. As for intentional distortions, I think the idea goes that they're all harmful in some way (some more than others) but that the benefit of behavioural change can outweigh the cost. I don't think that would be a terribly controversial statement among economists.