Quote from Martinghoul:
1) Firstly, what data might that be? 'Cause I think you're mistaken in your simplistic interpretation of the relationship. For example, what are we to make of the particularly "predictive" 2008 period of high commodity prices? Secondly, TIPS have been arnd since 1997. How much more data do you need? Do you think data from the 1700s (to give an exaggerated example) would be relevant? As to QE (POMOs are QE; there are no other Fed bond-buying activities), you can look at the ultra long part of the curve, where the Fed doesn't go. TIPS breakevens out there are not pricing any crazy inflation either.
1) 1970s data are certainly helpful when we're talking inflation and commodities. Data since 1997 may or may not even be statistically significant depending on how it's sliced and diced.
2) As for Fed activities, it can't be quantified precisely, but many think they're doing a lot more off the books than official QE/POMO $ numbers suggest.