Yeah, it’s not that the Romanian factory isn’t useful, it’s just that it takes too long to show a return on investment, even in long games. If you spend 30 ipcs in Romania on G1, and then max place infantry there on G2, G3, and G4 (unlikely, but I’ll give it to you anyway), then you have only just then broken even vs. buying mech. infantry in Berlin…except you have a stack of 30 infantry in southeast Europe, instead of a stack of 30 mech ing, so your stack is less able to penetrate deeper into Russia.
If you carry on max placing infantry in Romania on g5 and g6, you could theoretically make a profit of 20 ipcs…but at this point you really should have captured Leningrad and/or Kiev, and those factories are more useful. Plus, a 20 ipc advantage on turn 6 from a 30 ipc investment on turn 1 is pretty shabby. If you had instead spent those 30 ipcs on transports, tanks, bombers, etc., they probably would have helped you capture additional territories worth at least 5 ipcs per turn, and then over the course of five turns, those territories would pay you at least 5 x 5 = 25 ipcs, ie, more than the 20 ipcs you could possibly hope to save with your Romanian factory.
In general, I think factories make sense when you need either extra unit capacity (too much money and nowhere to drop units) or a way to get any land units into the refion, not just when you want the most efficient possible infantry drop. UK factory in Persia makes sense because Persia is seven turns from south Africa for foot soldiers; without that factory you need expensive planes or transports to contest the region. Mechs are just not expensive enough to justify building a factory to avoid purchasing them.
I am interested in minor factories in Romania to get German boats into the black sea, but that’s different.