That is not clear... Addresses with zero balance could be created for many other reasons, such as sprinkling and collecting dirty coins to confuse detectives, stress-testing of wallet software, gambling, pay-out by miner pools, ...
It's a while ago that I read his methodology, but what you seem to have in mind is actually /creating/ an address, one that is empty though. Correct?
If so, that address would not appear on the blockchain. What gbianchi means (again, iirc) is the /existence/ of empty addresses which necessarily means they a) received an input (thus appearing in the blockchain) and b) were used in an ouput (the "spending" part).
Hence: an empty address on the blockchain implies a transaction took place, and more specifically, one that was not just for "storage" reasons, but also spending the input again (even if, in principle, that spending could be to another storage location)