After seeing this and realizing that no matter how you look at it everything is a fuse I tossed a 250 W SMPS into the freezer and set it to -20°C.
Deficiency before freezing: 8 %
Deficiency after freezing cycle: 6 %
More experiments will be required because the change in efficiency isn't in line with the resistance change measured on a simple fuse. The problem is that I'm not a metalurologist.