How about put each current carrying conductor in it's own ferrous raceway? NEC of course doesn't allow this but would likely reduce the available fault current better than ferrite beads would.
If they were used as sleeves rather than raceways, I think the NEC would be silent on the issue. For example if the 'wire in pipe' assemblies were all together in a trough.
You would need to insure that you didn't have current loops that would act as shorted transformer turns.
The biggest problem, however, is that the ferrous conduit has no specification for magnetic qualities. You have no way (other than experiment) to determine how much pipe you would need for the required choke effect, how much heating to expect, etc.
What ferrite brings to the table is that it is magnetic while being an electrical insulator, along with magnetic specifications. But it is optimized for much higher frequency use, expensive, and has poorer permeability than common steel.
I'm guessing that some form of tape wound transformer steel tubed would be optimal for 'bead chokes' to reduce fault current.
Jon