Nigel, that's always been the argument against using a regulated supply for an audio power amplifier. The supply voltage to an op amp or an output amp has no affect on the output voltage for a given input voltage. In only affects how high that input voltage can be before clipping the output. If you have to be able to handle the current levels of any signal using a regulated amp, the regulator has to be capable of regulating at those high current levels, meaning a supply that is at least four times larger than necessary for most program material.
Please don't insult me, as I'm not a golden-eared audiophool nor a 5th Avenue marketing man. I measure my amp power in RMS watts and also specify the distortion at that level. To tell you the truth, I've NEVER seen an audio amp with a regulated supply, whether Sansui, Pioneer, Marantz or whatever. Not to say they don't exist, but that they're plain not necessary. With an unregulated supply, an amp can supply the rated RMS power and still pop out the necessary power for transients that a regulated supply designed for the same AVERAGE current could not do unless it fed a big output capacitors -- in which case, what's the point of a good regulator?
Back in the 1970s, there were a lot of fabulous high-power amplifier designs hitting the market every year, anywhere from 50W to 500W. Here in the U.S., Southwest Technical Products was a big provider of such kits. They all used unregulated supplies and specified their power in RMS watts into 8 ohms.
So, yes, they don't do regulated supplies because of the expense, not only of the extra parts count, but because of the extra size they'd have to have because of the extra current requirements.
Dean