Glyn Moody is quite stubborn on matters of language (grammar, euphemisms, poetry, etc.) and knowing that counterfeiting is very different from theft he had to tell them apart and he therefore had to respond to Canonical's COO Matt Asay (who almost went to work for Microsoft) after he had said that: "I don't mean to diminish the wrong nature of stealing software. Theft is theft and should be punished."
It is very hard to steal software: unless you creep into a computer store and steal the boxes (do they still exist?). As you know, what really happens is that somebody makes a copy of software: that is not theft, of course, that is copyright infringement. If I make a copy of a piece of software, the original still exists, but there is now a copy that I have. I have stolen nothing - I've actually created something - but I *have* infringed on copyright.
But wait, you will say, when you make that perfect copy you are *effectively* stealing the money that you would have paid for a legal copy. Except that you yourself write: "$1 in "lost" licensed revenue would not magically become $1 in proprietary software sales if piracy were reduced. It's very likely that users would elect to spend their money elsewhere." Exactly: couldn't have put it better myself. You can't start talking about that money that wasn't spent as if it were real and concrete: it's not, it's notional.
Of course, it is probably true that some fraction of the people with pirated copies *would* have bought genuine ones had they not made the copy: so that is truly lost revenue. But it also probably true that pirated copies act as marketing samplers and encourage other people to buy legitimate copies that they wouldn't otherwise have bought – to get support, updates etc. (Indeed, in the world of music, there are half a dozen studies that suggest this is the case.) After all, giving away software for free is the basis of many businesses based around open source.
Calling copyright infringement "theft" really plays into the hands of organisations like the BSA that put out these deliberately misleading studies. "Theft" is an emotive word that biases the reader against the people involved. If you call it "copyright infringement", and note that copyright is a time-limited, state-granted *monopoly* - and I think everyone accepts that monopolies are generally bad things - then copyright infringement simply means infringing on a monopoly. That's a rather different emotional bundle, I think, and a better one to place in opposition to the BSA's manipulations.