If the presser is correct (and I am extremely skeptical), then the $7.5B cost would be trivial - even if it had to be done annually.