I guess that because assuming unchanging branch sizes over time (which must follow the patterns of real estate), the number of branches should correlate with po

I guess that because assuming unchanging branch sizes over time (which must follow the patterns of real estate), the number of branches should correlate with population size, therefore you take population out of the equation