> However, because CPI is done at the national level while people typically live in one area, it is possible that HN readers in SF or wherever experience higher inflation than the CPI calculates.

It could even be that most people experience above-average inflation, because there are places with relatively low population but very low, relative, inflation dragging the averages down.

If that is the case, it might be possible to alter how geographic aggregation works to calculate a CPI-for-median-consumer that better captures the common experience. This might be better for some policy purposes, as well.