That argument seems a bit circular, and sounds too much like a rationalization for building ISS. If space manufacturing were the only reason to build ISS, then it says that we don't know for sure that manufacturing's a good reason, so we need to build ISS to find out if the expense of building it is really justified...

My reaction when somebody tries to pitch me on space manufacturing is uncannily Libertarian: If Intel needs a manufacturing facility in space to build perfect wafers for the upcoming Septium, let 'em build it. Odds are they won't, because a certain number of defects--deviated Septiums, as it were--are economically acceptable. They won't be able to justify increasing the yield from 90% to 99% if it raises the cost by an order of magnitude.

A more logically-defensible argument is that ISS is the only place we can learn how to do long-term space missions. While true on the facts, it's only of academic interest if there isn't a program of long-term exploration that needs the capability. If we keep postponing deepspace exploration then we can also keep postponing prerequisite programs like space stations.

I still think that if we put the horse before the cart we have a stronger argument. If ISS is supporting a deepspace exploration program, then it has plenty of justification, first as a laboratory, later as a shipyard and transfer point to the Moon and beyond. Putting it that way casts it as an argument that almost anybody can understand by commonplace analogies: Build the ships, build the port, send 'em out. We don't need to educate anybody about crystallography and the value of perfectly-round ballbearings and try to whip up enthusiasm for what are basically dismally geeky arguments.