A problem with indicating the taper is that a single measurement doesn't tell you if there is angular runout. You can have zero runout at the taper but a test rod mounted could still have runout an inch outside the collet. You really need to measure runout at two places on the taper. Even then, because the distance between the measurements is so short, you could see barely noticeable runout at the two measurement points but still have significant runout on a test bar.
Additionally, the ER nut also contributes to the runout measurement and there is no practical way that I can think of to measure that contribution directly.
A solution to this would be a test bar made to fit the ER taper, as is done for checking spindle runout with a Morse taper test bar.