“Income per natural” is a concept developed by development economists Michael Clemens and Lant Pritchett to overcome the measurement problems created by compositional effects. To summarize: compositional effects can result in the per capita income in two countries falling due to migration even though individual incomes rise. The problem is that when an individual moves from one country to another, the individual exits the aggregate statistics of the source country and enters the aggregate statistics of the target country.
“Income per natural” remedies this defect by tracking the mean annual income of persons born in a given country, regardless of where that person now resides. Their defining paper for the concept is at the page Income per Natural: Measuring Development as if People Mattered More Than Places – Working Paper 143. Quoting from the summary webpage:
It is easy to learn the average income of a resident of El Salvador or Albania. But there is no systematic source of information on the average income of a Salvadoran or Albanian. In this new working paper, research fellow Michael Clemens and non-resident fellow Lant Pritchett create a new statistic: income per natural — the mean annual income of persons born in a given country, regardless of where that person now resides. If income per capita has any interpretation as a welfare measure, exclusive focus on the nationally resident population can lead to substantial errors of the income of the natural population for countries where emigration is an important path to greater welfare. The estimates differ substantially from traditional measures of GDP or GNI per resident, and not just for a handful of tiny countries.