Such a strange comment. Yes, like most macroeconomists he uses a low pass filtered number (so-called “core inflation”) to look for underlying trends but is quite aware that headline inflation (includes the high frequency fluctuations) is what actual humans see at the store.

And he discussed the used car numbers — such an interesting outlier — as an indicator of supply-side shock.

You can agree or disagree with his Keynesian positions but he’s pretty clear with his thinking and calls out when he was wrong. To the degree macroeconomics can be considered a science, that’s what you want from a scientist, isn’t it?