My opinion is that it depends on installation.

It's simply not feasible to distribute large amounts of power via underground cables when the cable calcs are performed on the basis that the ground thermal resistivity is 2.5 K.m/W. Cable sizes become borderline impossible to terminate and the cost per metre of the trench and cable sky rocket. Have a read of ENA TS 97-1 or search up Cement Bound Sand (CBS)... CBS has a thermal resistivity of <1.2 K.m/W.

More to the point if you are looking at more than a few £k of  cables, it does become worth putting in the effort to do a test cable heating measurement in the local ground,  or if needed to back fill the channels with material that is selected to improve the cooling rather than just lobbing back what was dug out. That however needs good control of who is doing it, and not perhaps the cheapest un-inspected contract labour - who may skimp on the expensive 'fancy sand' due to not realising why it is important.
Or stay at HV until nearer the load (!) nothing like having fewer amps to keep things small and neat. For high enough power levels, such things become economic, although the regulatory hurdles as to who can do that work, set that bar quite high in the UK.

Mike.

More to the point if you are looking at more than a few £k of  cables, it does become worth putting in the effort to do a test cable heating measurement in the local ground,  or if needed to back fill the channels with material that is selected to improve the cooling rather than just lobbing back what was dug out. That however needs good control of who is doing it, and not perhaps the cheapest un-inspected contract labour - who may skimp on the expensive 'fancy sand' due to not realising why it is important.
Or stay at HV until nearer the load (!) nothing like having fewer amps to keep things small and neat. For high enough power levels, such things become economic, although the regulatory hurdles as to who can do that work, set that bar quite high in the UK.

Mike.