>I read this book a few years ago called _The Maximum Wage_, which argued
>for a maximum wage equal to 10 times the minimum.
>
>In the chapter that argued for why the 10 times rule was decided upon (as
>opposed to a 100x rule or a 2x rule), the author presented a graph of
>income on the y-axis and %tile on the x-axis, with 0% at the origin.
>Basically it was one more way of looking at income inequality, but it was
>instructive for this purpose.
>
>The graph itself started out as a line which sloped gradually upwards as
>you moved along the x-axis, with a roughly constant slope. However, there
>was a drastic change around 98-99% on the x-axis, where the graph turned up
>virtually instantaneously and shot up vertically.
Check out <http://www.davidchandler.com/lcurve> for a graphic on this.
Doug