I do agree that boiling it down to a single comparison is always gonna be reductive, but the median is IMO still the best you can do here, because it ignores outliers completely (on both sides).
From what I can tell the prosperity gap is also large enough that small errors don't really matter; I already looked at urban china vs average UK (systematically favoring China) and the numbers are still not close (something like $30k vs >$40k even after PP adjustment).