Erdmann Housing Tracker · Housing & Cities
TIER 4 Tue, 9 Jun 2026 14:02:19 +0000
I'm going to build on the discussion from the earlier post about the ways in which it is really easy to talk past each other on the topic of housing costs. ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ | | ---|---|--- | | | Forwarded this email? Subscribe here for more --- --- # Facets of the Housing Crisis, A Series: Part 1 | | Kevin Erdmann --- | Jun 9 --- | --- --- | | | --- | | --- | | --- | | --- | | READ IN APP --- I'm going to build on the discussion from the earlier post about the ways in which it is really easy to talk past each other on the topic of housing costs. Upgrade to paid I'm going to start with a small point. One of the common explanations that both academics and laypeople have about high housing costs is that people are willing to pay a lot to escape crime and the underclass, to live in good school districts, etc. This seems like a reasonable point. Nice places are, in fact, more expensive than less nice places. What makes this an especially telling assertion is that, in our era of elevated housing costs, the difference between the prices of nice places versus not nice places has shrunk. It's more expensive to live with the underclass now. That's what makes the housing crisis era different than previous times. It costs more to be poor, and it costs less, on the margin, to move up. Using Zillow ZORI rent estimates at the ZIP code level, across the country, since January 2015, rents have increased proportionately by about 35%, on average. In addition to that, every ZIP code has a $464 premium on it compared to January 2015. The rental value of the median home is $2,264. So, about one-fifth of the difference between the median home and the least expensive homes has been erased. (A ZIP code that used to have rent 50% of the median now is 60%.) But, if I show you this on an arithmetic scale, it sure doesn't look like it. In the left panel of Figure 1, the rents in expensive ZIP codes increased a lot more than the rents in the cheap ZIP codes. But, that's just a visual issue. The regression line intercepts the y-axis at $464. The only way to show the regressive change, visually, is by using a log-scale axis. In the right panel, it is clear that rent inflation has been higher in cheap ZIP codes than in expensive ones. And, if the shift was repeated 4 more times, the trendline would be flat. There wouldn't be any difference between expensive and cheap ZIP codes. | | ---|---|--- Figure 1 This creates a problem, because we are arithmetic creatures. We live our lives in arithmetic scales. Aziz Sunderji has been sharing screen shots on Twitter of different home price patterns, using his new visualization tools, and he showed a map of Los Angeles, where Manhattan Beach has an average home price of $3.3 million, while Compton, just over 6 miles away, has an average home price of $641,000. That's a wild gradient. And, when I saw his visualization, I immediately thought of this problem. Because, as wild as the gradient is, it is smaller than it used to be. But, it is completely understandable why the arithmetic creatures that live in Los Angeles would loudly disagree with that. After all, in January 2000, the average home in Compton was $123,000 while the average home in Manhattan Beach was $719,000. It used to be only a $596,000 difference, and now it's a _**$2.6 million**_**** difference. $2.6 million is a heck of a lot more than $596,000. Figure 2 compares the nominal home price in Compton and Manhattan Beach. The red line goes up a lot more than the black line does. | | ---|---|--- Figure 2 Figure 3 compares their percentage changes since January 2000. I have adjusted for compositional changes and inflation by adjusting by average US income growth, to reduce the upward drift that mostly comes from inflation. And, that's a big part of the perception issue. Inflation acts on every price proportionately. And, in Figure 2, if they had both retained the same real average home value and just increased with inflation, it would still seem like Manhattan Beach got a lot more expensive than Compton did. | | ---|---|--- Figure 3 Los Angeles and the Closed Access markets aren't great examples of this. (One reason LA isn't a great example of the problem is that the displacement has been playing out there for so long that even small increases in rent now motivate a lot of residents from places like Compton to move to Phoenix.) Figure 4 compares prices (after adjusting for income growth) of the cheapest and most expensive ZIP codes in the Phoenix area. The mortgage crackdown pushed the low end prices way down after 2007, but rent inflation that followed the construction collapse more than made up for it. Unfortunately, I think the recent convergence has more to do with additional mortgage access tightening and temporary migration shifts than it has to do with increased supply, though supply may finally be high enough to arrest the sharp upward trend of 2012-2019. | | ---|---|--- Figure 4 By the way, long-time readers probably don't need me to point this out, but new readers might notice how different the pattern was from 2004 to 2008 in Phoenix versus Los Angeles. Los Angeles during that time had mass outmigration and many of those families moved to Phoenix. That's what caused the boom in Phoenix, and that's why the patterns look so different. Inelastic supply, not high demand, causes low tier homes to be more expensive. High demand causes all home prices in a region to inflate. Note, that there is room for a "Superstar City" element to this story. From 2000 to 2003, home prices inflated in both Compton and Manhattan Beach. For a brief period before the main "housing bubble" period, Closed Access home prices really did have a pattern associated with high demand. And, those increases in Closed Access home prices have persisted. On the other hand, from 2000 to 2003, Phoenix prices were flat. In "We aren't as wealthy as we thought we were." I tried to quantify what portion of high housing costs has been related to aspirational changes and what portion is related to inertia against economic displacement. The blue bars in Figure 13 represent symmetrically rising housing costs, like those in Los Angeles before 2004. I sometimes joke that it has been a cosmic coincidence that a handful of cities all became the slowest growing cities and became spectacularly popular and productive at exactly the same time. That 2000-2003 period could represent an actual coincidence. Some of it could be related to declining interest rates. I usually am skeptical of a strong interest rate effect, but to the extent that there is one, elevated land value - either because of scarcity or locational value - would be where it would mostly show up. All the "superstar" cities are filter-up cities. The fact that there was also some increase in the "base value" in metro areas where homes filter down, suggests that not all of that increase was related to being "superstars", but it would be reasonable to believe that some was. Something caused home prices to rise in cities that have binding growth controls from 2000 to 2003 in a way that doesn't give the normal signal of supply inelasticity. Since then, the additional year of income that it takes to buy the average home is all from supply shortages and inertia against displacement. | | ---|---|--- The difference between LA and Phoenix is one of the foundational errors in the existing academic literature on the housing bubble. Except for a few exceptions, they don't distinguish between the high inflows regions and the high outflows regions. They toss all their numbers into their data blender, and control for metro area fixed effects (which basically only leaves the difference between Compton and Manhattan Beach as the factor creating statistical output before 2008). Then, they conclude that poor neighborhoods inflated more than rich neighborhoods, which coincided with a lending boom. And, they think that must explain why Phoenix became inflated so much in an unsustainable way, even though the difference they're measuring didn't happen in Phoenix. The entire literature is confused because the academy simply doesn't know what it doesn't know. Many people step into this conversation, with their arithmetic brains, and sincerely but mistakenly perceive that expensive places keep getting more and more expensive, and that the things that make expensive places more valuable to people are obviously the cause of it. And, that's backward. I think the two factors here work in tandem. One: Nice places really are more expensive than less nice places, and living in nice places really is a motivation for paying more, or for opposing changes in your neighborhood. All of those things are clearly true and important - so important that within 7 miles, you can find some places that cost nearly 6 times more than other places. Two: The scale and monetary illusions aren't defensible, though they are sincerely felt, but their sense of soundness and validity sort of borrows some of the authority brought by the first factor. Then, academics have a third factor: They have existing models that do a pretty good job of explaining why nice places are more expensive than not nice places, like the Rosen-Roback model, and so in addition to having misplaced certainty because of misperception, they have additional certainty because of education. It's worse than knowing something that just ain't so. It's knowing something that really is so, and has been useful in analyzing contexts which you think are the same as the current context, because of your misperceptions, but are actually much different. I think, right off the bat, fighting sincere observation with this abstract mathematical point, I lose half the audience. There's an understandable sense of, "This joker is trying to tell me that being a nicer location with better amenities isn't causing home price inflation, when anyone can see that homes in Manhattan Beach are selling for $3.3 million." Or, economists will say, "This joker doesn't know Rosen-Roback." It's understandable. I have probably used similar reasoning to dismiss things that I have seen in other topics. We all have to try to resolve the conflict between the abstract and the experiential. And, when a topic gets complicated, you don't necessarily have the bandwidth to dig into the details to see if your perceptions could really be that off. Upgrade to paid I have noted that this mistake is so easy to make that it tricks people into admitting that they fundamentally don't understand some basic details of the housing crisis. Many people, from the guy at the end of the bar, to Nobel prize winning economists, understandably, assume that they can assert that the best locations have seen the most price inflation without providing evidence. It's a really useful tell because it is such an easy assumption to believe, so it is frequently expressed voluntarily. And, I think because of everything I noted above, it's kind of a natural IQ test. How capable is somebody of handling counterintuitive new information when they find out that assumption was wrong? Unfortunately, on this issue, an education in economics hasn't been kind to IQs. Not because economics isn't a useful field of study. Ironically, it's because it was useful before the world changed. PS. I'm laughing that I started out attempting to write a "small point", and if I tried to make an index of all the ways you can identify in this post how perceptions can get twisted and reflected into mistakes, the post would probably double in length. I don't know if that's because of how I think or write, or if it is because this topic is just fraught with so much embedded misdirection that any chart I use can illustrate 5 mistakes. I'm sorry if I am the reason these posts get so complicated. You're currently a free subscriber to Erdmann Housing Tracker. For the full experience, upgrade your subscription. Upgrade to paid --- | | | Like --- | | Comment --- | | Restack --- (C) 2026 Kevin Erdmann 548 Market Street PMB 72296, San Francisco, CA 94104 Unsubscribe