If I can answer that last question first—there isn’t any consequence for the CPI. They actually measure motor vehicles differently in the PPI and the CPI. In the PPI, when they get to October, they start measuring the prices of just the 2007 model years, whereas in the CPI they are trying to phase in new models as people are actually buying them. The problem with the PPI that we discovered was that General Motors has two very popular truck lines and apparently did not actually introduce a 2007 model for these trucks but labeled them “2007 Classics.” [Laughter] They discounted those heavily because they were anticipating that they would, in fact, introduce the 2007 model shortly. The statisticians at the BLS saw those 2007 Classics models, saw them as 2007 models, and said, “Gee, they’ve introduced a new model, and the price is dropping.” So, on a seasonally adjusted basis, the prices really fell a lot. That just isn’t right, and fortunately it doesn’t contaminate anything beyond the PPI and its consequences.