<p>4th floor</p>
<p>Absolutely. That's a complicated issue and leads some to argue that you should never correct for cost-of-living in the extreme. A reasonable position would be it's somewhere in the middle. Figuring out exactly which part is consumption and which cost of living is left to complicated econometrics and to what assumptions you want to make about heterogeneity of individual preferences (i.e. which models of human behavior and choice you want to invoke). I am with Prof. Deirdre McCloskey on this -- all these econometric exercises are mostly qualitative calibrations at best. They're not the equivalent of lab experiments.</p>
<p>Take them with a grain of salt, but do not be foolish and ignore them.</p>