Reloading dies work off the outside surface of the brass. The outside surface geometry establishes one axis.
Neck turning is based on the mandrel fit to the inside diameter of the case neck. That diameter ,or cylinder,has its own axis. That axis may or may not be coaxial to the outside of the case,including the shoulder taper Bart refers to.
If the brass,as drawn,has non-uniform neck wall thickness,the case OD,which is what the die and the chamber reference,will not be co-axial to the neck ID,which is what the neck turn mandrel references.And the bullet.
So long as the neck thickness is not uniform,a resizing operation will only deflect the expander spindle . The OD and ID will not be coaxial.
Turning the neck will make the OD of the neck coaxial to the ID of the neck.
The OD of the neck is no longer coaxial to the outside geometry of the case.
If the brass is sized again after neck turning,this might be corrected.
But if you resize virgin brass,neck turn, then charge and seat bullets, the neck and bullet may present off center to the bore as the bolt is closed into battery.(Assuming there WAS unequal neck thickness,corrected by turning)
I can't say for sure how that will go as pressure builds,the bullet enters the bore,and the case fireforms.I assume the fireform will correct the problem.But the bullet may be offcenter as it enters he throat.
Without a resizing operation after neck turn,the first firing of neck turned ammo might see some variability.
I don't neck turn. I'm just thinking it over.