Absolutely this. We know that income is generally power-law distributed, so we also know that the majority of incomes will be below the mean. An analysis like

Absolutely this. We know that income is generally power-law distributed, so we also know that the majority of incomes will be below the mean. An analysis like this is bordering on nonsensical. It's too sensitive to outliers to be useful, in my opinion. A worked example: the median income per capita in Boston is $37,288[1] and there are 685,094 residents. The total income of people in Boston is then 685094*$37288 = $25,545,785,072. According to [2], Tim Cook's total compensation was $377 million in 2013. If Tim Cook moved to Boston, the mean per capita income would rise from $37,288 to ((25545785072+377000000)/(685094+1)) = $37,838. One person moving can make a 2% difference in mean income by moving to a top 25 city in the US, nevermind any of the smaller cities on that list. 1 = https://www.census.gov/quickfacts/fact/table/bostoncitymassachusetts/IPE120216 https://www.census.gov/quickfacts/fact/table/bostoncitymassa... 2 = https://www.therichest.com/rich-list/world/top-10-people-with-the-highest-salaries-in-the-world/ https://www.therichest.com/rich-list/world/top-10-people-wit...