Of course the thinking may be that enough people don't have a Korngold concerto yet, but would be put off by a recording of two unknown concertos (in which case the performance and recording of the Korngold had better be really competitive with better-known versions by better-known artists, then...) - a larger group than the admittedly rather small market of repertoire-driven collectors. No idea, though. It would be interesting to see if they've ever actually tested this. Instead of another Korngold concerto or another lesser-known concerto, another work by the same composer might be preferable still.