Kevin Erdmann has written a detailed and thoughtful response to my post from last week on housing spending as a percent of income. My goal in that post was to look at consumer spending as a percent of income for a variety of different sub-groups (my primary interest was by age group, but I tried to get into more detail for other sub-groups).

As Erdmann emphasizes in his post, I left out one set of sub-groups that the CEX data allows us to use: renters vs. homeowners. And these are very important groups to look at, since for homeowners (as he points out) many of the costs are implicit (such as the opportunity cost of those that don’t have a mortgage). Lumping all of these households together may obscure some of the different trends.

Be sure to read Erdmann’s post in full (he says many smart and correct things), but the key result is in his Figure 2 (reproduced here). Renters have seen the share of their income spent on shelter rise from 19% in 1984 to 24% in 2021. This is not a trivial increase. Owners, by contrast, have seen their share of spending fall, which is how it all gets washed out in the average.

I will concede that Erdmann is probably right on many of his many points. Still, I wanted to see this at a much finer level of detail, since national aggregates might be giving us confusing results. The micro-data in the CEX is probably not detailed enough to give us good breakdowns by MSA.

Continue reading →