Indeed, this is part of accepted UNIX lore. However, to me, "productivity" in this context is a measure of how much time it takes to implement a specified objective. UNIX did not have a specified objective. From "The UNIX Time Sharing System," (Ritchie & Thompson):

"Perhaps paradoxically, the success of UNIX is largely due to the fact that it was not designed to meet any predefined objectives."

So, does this productivity advantage really mean anything? It's comparing a research project to an industrial development.

The UNIX development methodology would seem to be this: Get a very small number of top people together with a common vision, ideally fewer than three, and see what happens. However many mistakes were made on the OS/360 project (very completely documented by Brooks), the need for an OS for the emerging 360 series would probably not have been met by the "see what happens" approach.

Marc