I'm afraid the economy isn't improving for everyone

Alex Tabarrok has a post at Marginal Revolution linking to a post from Jeremy Horpedahl at Economist Writing Every Day about rising real incomes. Figure 1 is the topic.

They make the point that, in general, everyone is getting richer. So much so that one-third of American families now earn more than $150,000 and the portion of American families earning less than $50,000 (adjusted for inflation) is at the lowest level ever.

The middle class is shrinking because they are becoming richer.

Figure 1

Tabarrok and Horpedahl are both thoughtful and thorough economists. And, I agree with this point in general for most people for most of American history. I also agree that it is an underappreciated point and that “fixed pie” thinking leads many people to underestimate the generalized economic improvement that we all share by being a part of a dynamic growing economy.

Horpedahl responds to several of the standard complaints in a follow-up post. This apparent improvement isn’t because there are more wage earners or because workers are working more hours. It isn’t due to inflation. The numbers are adjusted for inflation.

But….

In a nutshell, this data is not fully adjusted for inflation. Or, rather, I should say, the adjustment is biased. The housing shortage creates rent inflation that is highly regressive, down to the neighborhood level. Go 5 miles in one direction to a neighborhood with residents earning more than $150,000 and go 5 miles in the other direction to a neighborhood with residents earning less than $50,000. It is likely that rents in the first neighborhood have risen along with general inflation and that rents in the second neighborhood have risen something like 40% more than general inflation since 2016.

Adequate new housing supply is that powerful and that progressive, and we now know how powerful and economically progressive it is by living 20 years in its absence. Figure 2 shows rent inflation since 2016 in ZIP codes covered by Zillow, arranged by the starting monthly rent level.

Figure 2

Rent in the most affordable ZIP codes has increased by about 50% more than rent in the most expensive ZIP codes. Think of it this way. Over the past decade, about one-quarter of the variance in rent has vanished. Another 3 decades like this, and every home in America would rent for $4,000 per month.

Note, for those who, like me, appreciate the broader point being made here, it is tempting to respond to this by saying something like, “Everyone wants a single family home in the suburbs on a quarter acre lot. Not everyone can have one, so the price just gets bid up.” This is a common reply.

It couldn’t be more wrong. The homes that are least favored in the worst locations are the homes with the most rent inflation. You can’t explain this with preferences. Unfortunately, economists primed us to have that wrong reaction when they explained rising costs in New York, Los Angeles, and San Francisco with agglomeration value. Agglomeration value is real. It just isn’t the cause of our housing cost problem. When rising costs were limited to those regions, costs were rising the highest in the least favored neighborhoods, and families were flooding out of those cities out into the rest of the country. Now those cost patterns are everywhere. Figure 2 has no controls for regional fixed effects. It doesn’t matter if it’s the Rust Belt or the Sun Belt or the coasts. Rents are rising the most in the least favored neighborhoods in every region. Every region didn’t suddenly gain agglomeration value. Every region suddenly started building homes at the same low rate that New York and Los Angeles had been.

Before 2008, that meant that families were regionally displaced. They had to move away from New York and Los Angeles. Now that this is the case everywhere, there is nowhere to move to, so now the margin that gives is household formation. Now, rents in the most affordable neighborhoods in every city have to rise until a household doesn’t form.

Figure 3

This is the sort of lifestyle shift that archeologists would notice, and they wouldn’t attribute it to economic success. There are some details hidden in our economic condition that Figure 1 isn’t getting at.

A uniform inflation adjustment understates the problem of rent inflation in 2 ways for the poorest Americans. First, it is a larger percentage of their consumption basket, and, second, their rents have risen much more than average. My guess is that the inflation adjustment used in Figure 1 would attribute a bit more than $1,000 in excess rent cost for the poorest families since 2016. That much would be accounted for in the figure. The inflation adjustment in Figure 1 does account for high rent inflation, but it just doesn’t distribute it accurately. (I could be wrong. If costs have been tracked at the household or ZIP code level, then this post is incorrect.)

Figure 4 shows how much annual rent expenses have increased above the general rate of inflation in each ZIP code. Families with the lowest incomes probably have incomes about $3,000 lower, on average, relative to 2016, than uniform inflation adjustments would show. That’s a lot for households with incomes under $50,000.

Figure 4

Note, the inflation adjustment in Figure 1 probably attributes more than $4,000 of additional rental costs to the richest families. But rent inflation has been moderate in the most expensive ZIP codes. This means that, distributing inflation properly probably means that there are even more households with more than $150,000 incomes today, relative to 2016.

Claims about the “rich getting richer and the poor getting poorer” are usually overstated or false. In this case, they are real - at least in comparison to the stated data. Some additional families shown to be in the middle class category are really in the low category or the high category.

And, keep in mind, this is nominal spending. These are the costs that are still being expended after accepting regional displacement, downsizing, delaying household formation, etc.

When they say the economy isn’t growing for them, they feel both the compromises and the higher expenses. They feel both the red line and the blue line in Figure 5.

Figure 5

And, those compromises also are regressively salient. Families with incomes over $150,000 can give up the media room and the third bathroom. Families with incomes under $50,000 are more likely to be regionally displaced or to have to accept living in a dangerous neighborhood.

I hate to say it, because it’s usually wrong, but the economy is now rigged. Real incomes after fully accounting for rents are down, for the average household with less than $40,000 income. The number of families in the “under $50,000” category is probably about the same as it was in 2016, and many of those families are truly not moving in the right economic direction.

Figure 6

I say it is rigged because as long as new home construction is less than adequate, something like the right panel in Figure 5 is necessary - inevitable. As long as housing supply is bad enough to obstruct household formation, rents in every city must rise until the number of delayed households increases enough to match households to homes. Some of that works out still through regional displacement, but in the aggregate, somewhere, there are thousands of households not forming each year that would have formed under the economic quality of life that was normal in the 20th century.

Maybe those families aren’t worse off in absolute terms each year compared to the year before. But they are worse off enough to give up on a standard step in the life cycle of a normal modern human being. It is understandable that they think the economy isn’t working. It is understandable that they think it’s rigged. And telling them otherwise will only serve to damage our own credibility.

It pains me to say that because those who are prone to “fixed pie” thinking don’t think growing the pie is an option that will improve their economic lives. Those who normally think the economy isn’t improving for everyone won’t become allies on the solutions needed to unrig the system. They will call that “trickle down” thinking. So, it serves less of a political purpose to acknowledge this problem than it should. But, we must first and foremost strive for seeing things truthfully, regardless of the politics.