I understand the points about distributing resources within a system and certainly agree that if other areas are lacking, money could likely be better spent but not necessarily so. Cables always fall to the bottom of most peoples list and then we blame the source, amplification etc if we are searching for "more".I would argue you can't really evaluate those variables until you tie them in some fashion with good cables. I think your dollar ratio system is flawed for one reason in this example and many others: The Callistos cover 90% of the audioband in world-class fashion which is 95% of most music. You would be hard pressed to out-class them with anything you could put in front of them regardless of pricetag. I could care less what their retail value is, I only care what their potential is. By this reasoning only 10K speakers (likely full range floorstanders with multiple drivers) would justify $1500 speaker cables.
Now that I think of it, I have bedroom system where the total cost of all the cable is close to the total of the amp/speaker/source.(used values) This was not intentional, just how it ended sounding the best. the amp is the Music Reference RM10 which can be had used for around $700. Speakers are Meadowlark Kestrel I HR, also cheap. I would not hesitate to pair it with a $1500 (if necessary) speaker cable that can show off the glorious midrange of a well designed dead quiet EL84 amp such as the RM10 paired with an uncanny speaker such as the Kestrel I. Link them with some mediocre cable based on some "formula" and you could easily miss the magic.
One last idea since we are kind of indirectly on it anyway. Average cables smear the signal with phase distortion in the same way average or overly complex crossover networks do in many expensive speakers. Speakers like the Callisto that do not do this really show cables that do.
regards, Paul