I created an account to post this, but in Cura 4.6.1 (the only one I have tested) the flow parameter has no effect on the E value.
I have verified this both experimentally, trying to modify a fixed thickness calibration cube --- and by examination of the E values generated in the g-code of some simple test shapes.
Based on the above discussion, I can only concur that this is a bug.
After printing a dozen calibration cubes at differing flow rates on a .6 nozzle (as little as 50% flow) and seeing zero impact this was maddening. This was verified by some very expensive measuring equipment.
An M221 value inserted into the startup g-code fixes this and creates the appropriate reduction in flow rate.
M221 S85 ; Set flowrate to percent
After doing this my ability to print PETG went from almost impossible to results similar to what I am able to get with other filaments.
One of the most maddening things about dealing with open source project maintainers is trying to convince them an obvious bug in fact exists.
If you can point me to the unit test that handles checking the flow calculations I'll see if I can fix this one myself, or at least track down why it isn't working.
Cheers!