Inflation of What?

Yglesias: 

Whenever a stat about flat or declining real median wages or income comes out, an immediate fight always breaks out as to what that statistic “really” means. These conversations often suffer from a lack of awareness of the extent to which the different price categories diverge over time. Here, for example, is an illustration of the main expenditure categories that have grown faster than the overall price level:

What this is telling you is that someone whose “real” income was the same in 2010 as in 1978 is going to have suffered a steadily eroding ability to afford health care services and school tuition. He’ll also have experienced a long period of increased gasoline affordability, followed more recently by a deterioration sufficiently severe as to have left him worse off on net terms. Housing, which many people on Twitter think has gotten more expensive, has actually kept pace with overall inflation (more on this later). Food is the same as housing in that regard. Most everything else has gotten cheaper relative to other goods and services. That’s computers, cars, etc., and to an extent the falling real cost of cars lets you offset the gasoline problem by getting more fuel efficient vehicles.
This means that the extent to which your flat real income represents a “really real” decline in living standards depends entirely on what stuff you buy. College tuition isn’t something everyone buys the way food or even something “everyone” buys the way cars are. Many people don’t go to college, and many of the people who do go to college went to college in the past (or their kids did) and are not impacted by ongoing developments in these prices. Some people are much sicker than others or are differently situated vis-a-vis ailing relatives. Technological progress in the health care sector isn’t uniform, so some diseases can be much better treated today while others are simply more expensively treated. Talking about averages is often important, but it obscures a great deal of variation.
Now onto housing. Have housing costs really increased at the pace of overall inflation since 1978? ...Housing tracks overall inflation for two reasons. One is that a house is made up of lots of different kinds of stuff and lots of different kinds of labor, so the fluctuations tend to even out. The other is that the differential quality of land isn’t factored into this. So the fact that specific parcels of high-quality land have grown more expensive isn’t considered in the analysis. The United States doesn’t suffer from an overall shortage of land, so it’s still perfectly possible to get a cheap house somewhere or other, but specific places have become more expensive. This concentrated increase in land prices is experienced as higher costs by some (the people who’ve moved to Brooklyn over the past 10 years) but as increased wealth by others (the parents of the kids I went to Grace Church School with in the ’80s). So here, again, the average masks a lot of differences.