Okay, the idea I was trying to set out is that one of the things I think we would like, and that anybody would like in an electoral system, is that there ought to be what we call a monotonic relationship between changes in the votes and changes in the legislative power. If a party gains votes, then it ought to gain legislative power. If it loses votes, and particularly if an incumbent government loses votes, then it ought to lose some power to continue to effect its policy agenda.
If we take that as a metric—and I'm open to hearing counter-arguments as to why that might not be a good metric—it's clear that majoritarian systems outperform proportional ones in the sense that if you take the votes from an incumbent government under a majoritarian system, then their power falls at a much faster rate than under a proportional system.
That can be tempered with a couple of remarks. It's still a positive relationship under PR systems. The worst performing system wasn't a proportional one, it was the Japanese electoral system, the old electoral system and the single non-transferable vote, which had the remarkable property that changes in votes were entirely uncorrelated to shifts in power. That's quite an accomplishment. It's good for incumbents, perhaps, but less so for voters.
We can take the New Zealand case as one, and there the responsiveness fell by half. These numbers are not for any given metric, but we can say that responsiveness declined in New Zealand. What did they get in return for that? They got more proportional representation.
It's up to how people made that choice. I don't know, I'm not a political philosopher, so I don't have guidelines for them on that.