I'm not convinced metal is inherently safer.
Agreed, there are a number of reasons why it is electrically less safe than an insulating material such as plastics.
I suspect the change to metal-only is one of those 'we need to change stuff to justify our existence' changes that do little but mean more work gets divvied out in the name of regulation..
Actually, it was those well known electrical experts the London Fire Brigade that pushed it through.
They saw a rise in fires where the CU was implicated as a source - though there are questions as to the stats in terms of "is there a case of 'sees melted CU case, assumes CU was the cause ?' thinking" involved. Therefore, their electrical expertise told them, the problem is that plastic CU boxes aren't sufficiently fire-resistant. The fact that over the last few years we've seen a massive rise in the number of meters being replaced, by inexperienced people effectively dragged off the street and given a few days training, and failing even basic precautions like checking that the cable terminations are all tight seems to have been ignored. As does the general lowering of standards for the equipment itself - the old Wylex boards had double screws in its terminals, everything's single screw these days.
We now see a demand for SPDs where the need is questionable*, and a move to AFDDs where the supporting evidence is even less so**.
* I strongly suspect that the rising risk of damage to electricals is down to lowering design standards rather than increasing risk.
** We suspect, being driven by the suppliers/manufacturers who smell another opportunity to sell more expensive boards.
Strictly speaking the regs don't demand metal only that it is of a non-combustible material, of which steel is given as an example.
Correct. Depite it being common knowledge, and the IET having been asked about it - they failed to take the opportunity to fix a complete and utter f-up. How that reg ever got past review beggars belief.
Bakelite has thermosetting properties and you could argue it is compliant to current requirements.
Actually, no you can't. It's not about being thermosetting or not, it's about its combustion resistance.
The reg that should have been strangled at first mention, or revised to make sense, does not define what "non-combustible" means - you point an oxygen lance at almost anything (including steel) and it will combust, therefore you cannot claim (almost) anything as being non-combustible. If it weren't for that note saying that ferrous metal (e.g. steel) is deemed to be non-combustible then it would not actually be possible to meet that reg at all - but because steel is "deemed to be" non-combustible (even though it isn't) then that's the only material you can use and argue that the reg has been met.
I suppose you could make an argument that "X is no more combustible than ferrous metals therefore it meets the reg", but then you have to also agree on what the measure of combustible is - there are different measures, and what might be good in one measure, might not be as good in another.
And all this hassle and debate because a) the LFB are idiots, and b) the JPEL64 committee are idiots who failed to add "when measured in accordance with standard blah" to the rag.
at the end of your post 
