For normal conductor that is no doubt. Some use the neutral for the return.
An unsafe installation is not permitted nor required. If the system can't clear the fault is a reasonable amount of time, they are not complying with the NESC.
Clearing a fault is not the issue, even though many of these street lights are being tapped to the transformers secondaries without any OCPD's, the main issue is the fact that they are using the grounded conductor as the fault clearing path, and because this conductor also carries normal load current you have the danger of electrifying the grounding of the metal pole if the grounded conductor looses connection or in the case of no OCPD is burned open when there is a fault, not only to add that the voltage drop on long strings of lights will also raise the voltage of anything that is bonded to this grounded conductor above the Earth, depending upon how long the circuit run is and the size of the conductors are, we have read in this thread that it is common for some POCO's to use a 12/2 cable, the string of lights I saw was hundreds of feet long, maybe over a thousand feet, the voltage drop at the end fixture on the neutral could have been as much as 10 volts with the fixtures being 250 watt HPS, this voltage can double when the fixtures start up as they draw twice the current when starting, also what if a person was leaning against the pole with bare feet at the time a line to neutral fault happened, the voltage could reach close to the circuit voltage if the 12/2 has a high enough resistance in this long run.
This is why the neutral should never be used as a ground fault path when the metal grounded parts of a pole or equipment can be contacted by a person who also can contact the Earth.
I think it is more a matter of getting them to understand what is required.
The argument about who should rule (NEC or NESC) is what I was addressing. For the utility, the NESC rules and a safe install can be made if they follow the NESC. If there is a failure to be safe, then obviously they need to change what they are doing.
But saying that following the NEC is the only way to go is not going to fly. One could look at the two ground-rod requirement (which is a complete farce) and say that the NEC is unsafe and we should all follow the NESC. The two codes are for two separate arenas and we should work within the codes to be safe, not just jump back and forth between codes because we don't like how someone is handling a particular install.
saying a safe install can be made is like saying that there is nothing wrong with using the neutral as a EGC, I hope you do not believe this? unless you are agreeing that a separate EGC needs to be run?
Only a separate non-current carrying conductor that does not have a path through a load to a conductor that has a voltage potential to Earth (hot/ungrounded conductor) this is what an EGC is and is why the NEC requires it.
We all know that the grounded conductor (neutral) will become the potential of the hot when the connection is lost at the source end, and anything bonded to it will also become hot in reference to earth.
I don't know if the NESC requires fusing at the source for these street lights, but I see many up on poles just tapped to what ever secondaries that might be located on the pole or several poles down, I know most pole mounted street lights that are up on the pole are done this way without any OCPD's, the problem is many linemen are carrying this method over to metal ground mounted poles that a person could be in contact with it and Earth, and because they use the grounded current carrying conductor as the fault bond to the metal pole it becomes even more of a hazard.
I talked to a lineman about this and asked him what would happen if the grounded conductor opened up, and his response was shocking literally, he didn't have a clue that the voltage of the ungrounded conductor would flow through the load and energize the grounded conductor and the pole, and when I got him to understand that part, he made the commit that the ground rod at the pole should keep it at earth potential, when I told him that the ground rod could not do this, and showed him that even at 25 ohms it would not pass more then 4.8 amps, and if the load on the circuit pulled more then this you would still have close to 120 volts on the pole, he then couldn't believe this was allowed, he said they never taught them anything like this in training, so its no wonder that many linemen don't have a clue on how dangerous this can be, they are being taught that all you have to do is install a ground rod and everything is safe, and we see this with them requiring a ground rod at services even though we have other electrodes that far exceed what a rod could ever do.
As I said before and Don has pointed out, light fixtures up on wooden poles or other insulated type oles, where a person would not have a way to make contact between the fixture and Earth I have no problem with using the grounded conductor as a fault path, but not when it is down close to Earth on a metal pole.