relationship between unanimity and efficiency

your last interpretion is right. I'll say a few words on this in the next tutorial again, to make the point clearer. But so far:

if there's unanimous agreement on a move from situation A to B, you know that this is a Pareto-improvement. Whether B is a Pareto-Optimum is not yet clear: only if there's no alternative (e.g. C) to be reached by additional Pareto-improvements, you are sure to be at a Pareto-efficient point. Unanimity ensures that we don't run into Pareto-inferior situation w.r.t. status quo, but it is not a sufficient criterion for arriving at a Pareto-Optimum.

good question!

there's two main possibilities: checking if such rules could have been consented to under some form of veil --> hypothetical contract

or: we believe, like Hayek, in the survival of the best (efficient) norms by evolutionary approaches

additionally, one can at least argue if there's not way of reforming the actual constitution with unanimous change, than Buchanan's check for constitutional efficiency means that it is efficient. that's discussed in the lecture, too