How big of an issue was this?
If it takes 24 hours to run 11 T2 units, and recycling results in enough materials to make 11.5 units, agreed that materials are created out of thin air, but it would take 22 days or 1 player running 22 lines for 1 day to create enough material for a full run.
What's the end result here?
That a producer can create 11 units T2 units per day free, but they will need a second account building 23 lines of T1... extrapolated, that's roughly (4) maxed out accounts running everyday to create 11 'free' T4; that still require the creation and continual replacement of the T2,3,4P's...
For robots, a run of (11) mechs takes about 6 days.
Assuming about the same .5 return, that's 23 lines of bots running each week, to get 11 'free' bots. In the case of bots, your tying up a huge amount of material creating 23*11 bots per week per producer.
Lastling, the assumption of recovering the ".5" is probably even exaggerated.
That means, if you use 20k titanium per bot, that's 220k per run of 11 with a return of 10k which is something like 104.5% recycling effciency.
The insurance exploit was worth it because in volume they could get millions of NIC per bot back.
The duplication exploite was worth it because the 'time' factor was 0.
It would take months for an entire corp to set up this, creating CT's, gathering the initial raw resources to run these lines, so finally start producing what amounts to a 'few' free bots per WEEK!
Also, the whole system falls apart if you lose lvl 4 refining OR lvl 4 recycling.
Unless the % material 'gained' was well over 4.5% it simply an exploit that isn't worth the effort to exploit. The key here being that you had to be on beta with access to large amounts of Epi ANYWAY.
Tl-DR; Recycling nerf was probably over-kill and negatively effected many more players than it did to protect the integrity of the game. A better solution was noted, which was to reduce the % of the outposts instead of the individual skills.