What is changed in the "could have" scenario is how capital is allocated. For example, the public sector could allocate a larger fraction of the capital than th

What is changed in the "could have" scenario is how capital is allocated. For example, the public sector could allocate a larger fraction of the capital than they do currently, maybe by spending more to prevent homelessness and build infrastructure. This would require changing policy, and you are right that I think the policy of letting the market determine such a large fraction of our capital allocations is driving the inefficiency. It's a "know it when I see it" kind of utility measurement. Probably closer to a "happiness index" than GDP. It's fuzzy but I don't think that makes it meaningless. I think most people have a similar idea of "better utility" (not exactly the same, similar). For example, what's most utility: New boat for Bezos or dental care for 1000 kids? In any case, some kind of utility measure is necessary to determine when markets are efficient or not. If the measure excludes people just because they have limited means then few would call that a good utility measure. If the utility measure doesn't exclude people with limited means then markets can only be efficient when means are relatively well distributed.