Erdmann Housing Tracker · Housing & Cities
TIER 5 Mon, 15 Jun 2026 17:31:34 +0000
Talking past each other ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ | | ---|---|--- | | | Forwarded this email? Subscribe here for more --- --- # Facets of the Housing Crisis, A Series: Part 5 ### Talking past each other | | Kevin Erdmann --- | Jun 15 --- | --- --- | | | --- | | --- | | --- | | --- | | READ IN APP --- In this series, I have been using figures like Figure 1 to highlight the peculiar patterns that characterize the American housing market. To recap, in the before times, nominal housing costs rose roughly in proportion to incomes. Over time, as cities grew and incomes rose, dots just kept moving up that upward sloping vector. In very large cities with some amenities associated with density, the velocity up that vector increased a bit, and small variations developed, like the gray dots compared to the red dots. In general, as cities got richer, incomes and homes both increased up those relatively stable trendlines. Sometimes, cities got bigger, and accelerated their movement up those trendlines. They climbed the agglomeration hill. Upgrade to paid The reason housing is a topic of conversation, and the focus of my accidental career change is that most housing markets don't look anything like that anymore, in the US. In order of occurrence, some cities cut off housing supply so sharply, they became the blue dots. Homes got uniformly more expensive, pricing out poorer residents and causing mass regional displacement. Then, the cities where those migrants landed became the orange dots because the local demand boom led to a classic bubble where all homes in the destination cities increased in price proportionally. Then, the mortgage crackdown meant to push the blue dots and the orange dots back down to the old gray and red dots caused a collapse of single-family home construction so thorough that there is now a shortage of homes across the country. Now, the cities that weren't blue before 2008 are becoming blue, and are now somewhere between the red dots and the yellow dots as the shortage leads to similar outcomes everywhere. All of this was set in motion because cities are always changing (the arrows in Figure 1). Dots are always being teased to jump around in line. An infill development here. A depreciating neighborhood lacking adequate reinvestment there. New neighbors with higher or lower incomes than the existing residents here. Newly appreciated amenities attracting novel demand for a location there. | | ---|---|--- Figure 1 Over the course of the 20th century, we gave locals the power to block some of those changes by forcing all physical change to go through an obstructionist permission process. Eventually, city-building became, effectively, illegal. And, it was off to the races toward blue-dot-istan. And an important key here is that the scale at which blocking new housing supply moves a city from gray to yellow to blue is _much larger_ than how far up the agglomeration hill the new housing would have pushed the same city. At least an order of magnitude larger. If there is demand for a city to grow 10%, and it does, incomes and prices might increase by 1% because of the economical advantages of large cities. If it doesn't grow, home prices might increase 50%. So, understanding that cities are yellow or blue dot cities and not red or gray dot cities is really important. These different patterns are distinct. When I have run the numbers on cities across the country, as to how far toward being yellow or blue they are, the estimates, using the coefficients I have suggested in this series, always suggest a shortage that looks reasonable relative to changes in permitting trends across time and across cities. Figure 2 shows a test with 5 metro areas that have been rising up from red to blue in Figure 1. The blue bars in Figure 2 show past building rates. The orange bars show recent building rates. The gray bars show my estimate of the permitting rate required to stop moving away from the red dots and toward the blue dots. They are generally in line with historical rates of building, or less. The city that is second from the right is like the Closed Access cities. Compared to current rates of building, it looks like a crazy ask, but that's not because the model calls for so much building. It's because they permit so little now. And the farthest right city is a relatively average-growing Midwest market. Even Midwest markets, at this point, are moving from red to yellow. | | ---|---|--- Figure 2 In short, Figure 2 shows the rate of new home permitting required for these cities to return to normal affordability. The quantities shown are based on the estimate that 10% growth in the housing stock will lower prices by around 50%. Based on current prices, if you think more homes doesn't have that strong of an effect on prices, then you have to expect the required rate of permitting to be higher. And home prices are some of the most closely tracked data we have today. The elevated real estate value isn't debatable. You can know of it or be ignorant of it, but it would be hard to debate it. So, either we need millions more homes across the country than even I assert that we do, or new permits have a strong effect on prices (even in the "superstar" cities like New York). Once these markets start permitting more homes than those gray bars call for, each additional 1% (that's about 4 permits per capita x 1,000) increase in their local stock of homes will lower home prices by about $15,000. Let me stress, this is not a marginal issue. There isn't some tightrope that every market is on, where if construction drops by one unit, there is suddenly a one unit shortage. The drift from red or gray to yellow and blue is historically peculiar. This never happened historically. American cities aren't just having a bit of a building downturn. The downturn has been so deep and so persistent that a tipping point has been reached where families couldn't make marginal changes to keep nominal housing expenditures normal, and started taking on historically unusual nominal housing expenditures rather than making further compromises on the housing they consume. (One of the compromises they do make is to delay household formation. And, that is why affordability measures with "per household" in the denominator are weak indicators of these problems.) Some academic literature, like some of what I discussed in the previous post, asserts that home prices still just follow incomes. They think we're still gray and red, even though we are objectively not. All the novel 21st century deviations in home prices have been very much not associated with income growth. When mortgage lenders took the blame for high home prices, the idea that home prices were unusually and unsustainably high was pretty universally held. So, it's weird to me that we can't even get broad agreement that housing costs are high. That seems like a claim that doesn't just lack the constructed framework of Figure 1, but denies the basic observable facts that we already had universal acknowledgement of. We created a very popular financial crisis as a reaction to the strong consensus that prices were well above a justifiable range. | | ---|---|--- Many other claims basically assume that we are still in the spatial equilibrium housing market that always made red and gray dot cities before the 20th century. This is also odd, because the pre-2008 boom was concentrated in the Closed Access markets (New York, LA, etc.) plus a few regions that experienced sharp migration spikes from those markets. Since then, the Closed Access cities have maintained inflated prices similar to the pre-2008 levels. They were about as blue in 2005 as they are today. The elevated price levels have spread much more broadly across the country. So, the "everyone wants to move to the superstar cities" explanation also is much weaker than it was when everyone agreed that prices were inflated. One other comment I see a lot is that young people expect too much. The average home today is nicer than the average home was in 1970, and they shouldn't be so entitled. This isn't really true. On a socio-economic basis, if a young household has a 50th percentile income, and their parents and grandparents also had 50th percentile incomes, the young household will have to live among families with lower income than than theirs in order to spend the same portion of their income on housing as their parents and grandparents did. But, also, just on a size and quality basis, this hasn't been true in many locations for some time. If you have a 25th percentile income, and you move to Los Angeles, you will have to pay more to live in a worse home than the 25th percentile income household would have 30 years ago. In the gray and red before times, you _could_ have asserted that people expected too much, by assumption. What other reason could there have been for spending too much on housing? There was a dependable range of homes across prices and incomes, and given where your income dot was, it was up to you to distribute yourself into the stock of homes where it was appropriate. "Spend 3 times your income on a house" was a rule of thumb _because of_ the state of the world. A lot of discussion about housing comes from assuming we are still red and gray. Since turning blue has a much more powerful effect on home prices than climbing the agglomeration hill does, those who think that cities like New York are still on the agglomeration hill greatly exaggerate the demand for living in New York City. I recently saw a twitter conversation that I will translate using Figure 1. Matt Yglesias tweeted: > I think a lot about how none of my friends from growing up in Greenwich Village in the 80s and 90s could possibly afford apartments in that neighborhood the size of the ones we grew up in. > Downward mobility among objectively successful people induced by housing scarcity. He's basically personally observing my point about a 25th percentile income in Los Angeles. Yglesias gets it. He knows that New York City is blue dots. Here are some of the replies: > Under the conditions in which they could afford apartments like that, would it still be Greenwich Village? It is true that New York City would have had to grow to maintain affordability. I would say this reply is about scale. First, if you think that New York City is still gray dots, then the high cost and high average incomes reflect big moves up the agglomeration hill, which leads to an overestimate of how many newcomers would like to move there. I think that is how blue dot cities become misinterpreted as superstar gray dot cities that are expensive because of their suddenly limitless potential. Some analysts will claim that New York City has practically unlimited growth potential. It could never build enough to keep up with potential demand. I think it comes from mixing up the causality here, and not accounting for the scale of these effects properly. So they look at a blue dot city and say, "If home prices weren't so high, it would have grown even more, and attained even more agglomeration value, and driven incomes and prices even higher." It is plausible that San Francisco would have grown at Austin rates, if it didn't limit its growth. Or maybe even a bit more than Austin. Maybe part of the idea that they could build forever and never become affordable is just a reflection of the scale of the hole they are in. If San Francisco did have the economic potential to grow like Austin, it would, at this point, seem like they could build forever and never be affordable. But, even for San Francisco, if it built enough to catch Austin's growth, even if its average income was still higher than the US average after that, it would be lower than it is now, and its housing would certainly be _much_ cheaper. | | ---|---|--- New York City is mostly expensive because it stopped building. Changing that prior greatly lowers the estimate of how much building it would have taken to remain affordable. I think a New York City of 25 million today, versus its current size of 19 million, would likely have been enough growth to remain on the agglomeration hill. Second, this reply is about the arrow problem - an inflated sense of how much of that internal change happens over time. To get to a population of 25 million over a period of 20 or 30 years, would have required, for instance, 5% of New York City lots to be changed from single-family to 6-plexes. It's possible that an affordable New York City would leave Greenwich Village unchanged. It's also possible that it would have changed drastically. But, from an approaching airplane, a 25 million strong New York City would be only slightly noticeably different than a 19 million strong New York City. The Austin metro has remained relatively affordable and doubled in size since 2000. It has permitted about 5 times as many new homes, per capita, as New York City. Certainly, downtown Austin has visibly changed, but there are many nice neighborhoods across Austin that are practically untouched by that growth. > Other people got even richer, so relative mobility means they can be richer but still outbid on any fixed quantity good (land). Would have to make the whole village 20 stories to meet demand? There are similar things going on here. This reply is very much in the gray dot world and, to them, it's a story of the dots moving around. And, there is the overstated scale, which, again, comes from inferences you would make about New York City if you thought it was still a gray dot city that had just moved way up the agglomeration hill. It could very well be the case, though, that under more lax supply conditions, Greenwich Village could remain similar to its current form while its residents increase in relative socio-economic status while the rest of the city densifies and becomes more affordable. The dot could change its relative place, and it may have already changed its relative place. And, that could happen as a gray city or a blue city. I think making the village 20 stories to meet demand is kind of mixing up levels. More housing anywhere in NYC would make the village more affordable. I think the inference about how much more demand that would create to live, specifically in the Village, is much lower if you model New York moving from blue to gray than if you think New York has, instead, been flying up the gray trendline as a super-agglomeration machine. So, I think this replier mistakenly thinks New York is a gray city. On the other hand, they may be spot on. Blanket upzoning tends to attract development in neighborhoods with the most locational value and the most potential for increased density. Maybe the structural form of Greenwich Village would be especially vulnerable in a counterfactual affordable New York City. > Or is it real estate speculation? This one thinks New York City is an orange dot city. They're going to be waiting a long time for that correction to come. > NYC is unaffordable because of mass immigration. The majority of people in the city are foreign migrants. You can't out-build mass immigration. If you build more, the cheap labor advocates will just import more foreigners. More people, all else equal, means high home prices, if you assume fixed supply. They aren't assuming fixed supply. They have just internalized that same mistake about scale. They think New York must be expensive because it's so popular, it just couldn't realistically meet demand. It has a bit of a factual error, in that what has changed in New York City is that there are fewer households with lower incomes than there used to be, foreign or domestic. Also, it's kind of funny to imagine that the reason New York City was affordable a century ago was that it didn't have so many foreigners. Also, this reminds me of a moment. I had a YouTube debate with Cameron Murray, and this is the one time in my life where I had a visceral feeling of being speechless. I was making the case that the "superstar city" label is overstated because New York City permits so few homes, and that it is possible that it would be affordable if it had just grown at the rate of Kansas City or Omaha. His reply was that New York City permits so few homes because it's a city only for rich people. That's where it is in its life cycle, which he suggested was natural and normal. There just aren't that many rich people, so demand to move there doesn't lead to much of an increase in the quantity of homes. New York City is a Bentley. Other cities are Toyotas. He seemed to be saying that every city becomes a blue dot city and perennial mass displacement of its previous residents is a normal part of being a mature city. Elsewhere in the video he argues that urban housing prices aren't really any different than they have ever been, citing a historical complaint about high prices in a much smaller city. Case-Shiller erasure. I am unable to translate the Murray position in terms of Figure 1. I'll leave it to the reader to try to make sense of this, if you choose to spend your time that way. But, I'd like to see the replier above and Murray argue it out. They seem to be in conflict about where the demand for living in New York City is coming from. Aziz Sunderji responded to Yglesias, and they had some back and forth. > S: So, to be clear, you think Greenwich Village is expensive because of zoning? Same thing for, say, Martha's Vineyard? Aspen? This is a common gray dot intuition. If New York City is expensive because it has been climbing the agglomeration hill, then it's the highest dots at the top that must be leading the way. Greenwich Village is expensive because it's special because New York City is expensive because it's special. It's all just the same story. So, it doesn't seem dissonant to Sunderji to bring in Martha's Vineyard or Aspen, which are places known for attracting wealthy households. As I have mentioned in previous posts, focusing on the most favored, high-tier neighborhoods in the blue cities is a common and helpful way that people voluntarily announce that they don't know the expensive cities are blue cities, where prices have appreciated the most, in percentage terms, in poorer neighborhoods. > Y: I think that if the zoning allowed more dwellings to be built there, then more people would live there. > Do you disagree? > S: I don't disagree that more people would live there, but that's not what you wrote in your original post--you're talking about affordability: I do not believe adding more housing to Martha's Vineyard would mean journalists could afford to live there. > Y: MV is a weird case because of so many non-residents. > But today, X people live in Greenwich Village. If tomorrow X + Y people lived there, then more people could "could afford to live there." It's a tautology. > S: The tautology is about the number of people who'd live there, not affordability. Do you think the residents who'd occupy those newly-built Greenwich Village homes would be journalists like you and me, or bankers? > Y: If one additional person moves to a currently expensive neighborhood, that will probably be a rich person. But as the supply of dwellings keeps growing so does the number of people who can afford to live there. I don't really understand what we're disagreeing about. > S: I think most of our disagreement is about *the degree* to which upzoning will make housing accessible to a group that's *lower down* the income totem pole. You think it's meaningful, I think it's marginal. Again, this seems to be drawing from the gray dot city on steroids intuition, like the economists I mentioned earlier in the the series. The blue dots are a consequence of the strong motivation of existing residents to resist displacement. The ones that remain are willing to pay for it. This creates a uniform premium on homes in those markets. The elevated prices get misinterpreted to mean that there is a lot of demand for moving in from high-income households. If prices dropped, even just a little, they think that more households would be highly motivated to move in, and keep prices from declining. And, since they think New York City is a super star agglomeration city, those added people will raise local incomes and make it even more expensive. To some extent, there may be some truth to the basic intuition about how much new construction would lower prices. The roughly 200,000 net leavers from New York City in any given year that would stay if more homes were built, probably do have very elastic demand for aspiring to remain in New York City. It is possible that 80,000 extra homes would induce those residents to stay, and until New York City starts permitting more homes than that, it could be true that rents will be relatively sticky. But, those typically aren't Greenwich Village residents. And, they aren't rich newcomers. They are the current leavers, whose cumulative absence makes local incomes appear to be buoyant. If those 80,000 units were built in neighborhoods like Greenwich Village, then, residents who would have lived in New York City, anyway, would now live in the Greenwich Village-like parts of New York City, and the marginal growth in the population of New York City, in total, would come mostly from the 200,000 poorer residents that stay instead of leaving. The growth would largely come from fewer leavers, not more newcomers. In support of Sunderji's point, if Greenwich Village has moved to the top of the New York market, then the correction of the uniform scarcity price premium would have much less of a proportional effect on Greenwich Village affordability than it would have on lower tier neighborhoods. If this has happened - if it is represented by an arrow that moved it higher within the distribution of New York neighborhoods - then it might be true that Greenwich Village has changed in a way that would make it hard for Yglesias' original complaint to be rectified. Note that Sunderji's comparison to Aspen and Martha's Vineyard echoes Murray's claim about New York being a location only meant for the rich. But while Murray used that as an explanation for very low construction activity, Sunderji uses it to explain why no amount of construction activity could significantly lower New York prices. Murray seems to be at odds with me, with the anti-immigrant replier above, with Sunderji, and with many varieties of the super-agglomeration story that claim New York homes will gain value the more it grows. > Y: I mean, yes, literally all change happens at the margin! > Shifting zoning in one neighborhood is not that big of a deal, but in the aggregate it's a huge deal for the national economy. Matt really pretty accurately is working with the Figure 1 framework here. The last reply brings it all together. Let the arrows work. The arrows might leave Greenwich Village alone, or they might densify it. If it gets left alone, it might actually move up to a higher part of the gray dot city range. It could also do that if it densified. Or it could move down and become more affordable. What is important is that it is allowed to do any of those things. And, if it was allowed to, and if all of New York City was allowed to, then a more dense Greenwich Village might make the Bronx more affordable, or maybe a more dense the Bronx would make Greenwich Village more affordable. And, in either case, millions of families would still be living in New York City, instead of moving to Florida. And, if it was total anarchy, it would probably still amount to much less stressful displacement than New York City creates as a blue city. Yglesias understands that New York is a blue dot city. It should be a gray dot city. It isn't a superstar gray dot city. It is failing at being gray. It is blue. Upgrade to paid 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