IBMs Rental Life Assumption
I suggest the statement, "the four-year life assumed by IBM," is a gross over simplification of the rental assumptions used by IBM. As I understand it, there was a RESP committee (Rental Equivalent Sales Price) that determined the expected rental life of each product so that IBM could comply with the 1956 consent decree and that the results of such calculations were generally longer than four years. My understanding is that the basis of the leasing companies lower rental price for IBM hardware was much more related to a lower imputed interest rate than a longer life assumption, but I am more of an engineer than an economist.
PCM hardware rental prices benefited from both lower price to the leasing company as well as lower imputed interest rate.
This of course was the subject of much discussion in the various anti-trust matters and you might want to look for material in some of the works coming out of that matter, e.g. Big Blue, or perhaps you can contact the US government's economist, Alan McAdams for his input.
BTW, I strongly support the S/360 a a STARS event.