These are great points, and I regret that I used an example that inadvertently framed the discussion around "women marrying for money"- In fact I believe that

These are great points, and I regret that I used an example that inadvertently framed the discussion around "women marrying for money"- In fact I believe that women have a surprisingly catholic breadth of mate choice criteria, and the data available on reproductive variability in modern societies seem to bear this out. As the preceding comments note, the existence of a Zipf-distributed hyper-elite is not necessarily a big problem for certain categories of goods. I am actually more interested in the effects of the collapse of alternative status hierarchies and how this affects different status 'tournaments'. Global competitions seem to converge on a single figure of merit, and I wonder whether jagged multimodal distributions on a single variable produce different 'competitions' than normal distributions or multivariate distributions. I would also note that an individual or group can affect a resource competition without directly consuming the resource itself.