Aug 6, 2018

Kevin Erdmann on Housing Shortages and a New Understanding of the Great Recession

A shortage in the housing market, not a surplus, may have been one of the principal causes of the Great Recession.
David Beckworth Senior Research Fellow , Kevin Erdmann Visiting Fellow

Hosted by David Beckworth of the Mercatus Center, Macro Musings is a new podcast which pulls back the curtain on the important macroeconomic issues of the past, present, and future.

Kevin Erdmann is an independent researcher and blogger at Idiosyncratic Whisk . Kevin is also the author of a book titled ‘Locked Out: How the Shortage of Urban Housing is Wrecking Our Economy.' Kevin joins the Macro Musings podcast to discuss his book.

Read the full episode transcript:

Note: While transcripts are lightly edited, they are not rigorously proofed for accuracy. If you notice an error, please reach out to macromusings@mercatus.gmu.edu.

David Beckworth: Kevin, welcome to the show.

Kevin Erdmann: Hi, David, it's great to be here.

Beckworth: All right, well fun to have you on. You've got a very provocative new book that's going to reshape the narrative, or at least that's your hope and your dream.

Erdmann: Yes.

Beckworth: You're going to change our views, and you're here to convince myself and our listeners today that our views about the housing recession are wrong in many ways. You're going to fix them. But before we get into the book, tell me, how did you actually get interested in this topic and what led you to write the book?

Erdmann: Yeah, this whole project has been sort of an accident. It's been really a three year process at this point, and my first sort of baby steps into it was I was just sort of managing personal investments and whatnot, and doing independent research, and had some ideas about the home builder market, maybe some tactical investments you could make there back in 2014, 2015. And the original germ of the idea was that there's been this big decline in home ownership, there's probably sort of a regime shift in the market from a homeowner price point to a landlord price point. As that reverses, maybe there were some tactical gains to be made there.

Erdmann: So my original sort of ways of looking at it was, okay, there's going to be some bounce back, but since we had all this excess in 2004 and '05, I need to sort of look at the data and figure out how much of that pendulum swing was just going back to neutral, how much has been an overswing, just for personal investment strategies. So I would go look at, say, Survey of Consumer Finance, or any sort of data like that, census data on home building.

Erdmann: So like, Survey of Consumer Finance, I go look and there's nothing, there was no pendulum swing. There was never a shift downward in home buyer incomes or FICO scores, all those measures were pretty level throughout the boom. And so for, I don't know, six months or a year, well, I originally maybe planned on spending a month on this and then figuring out what I wanted to do in the market and go from there. But every time I'd look at a new piece of data, it would be nothing, there'd be nothing there that was supposed to be there. So for a long time, every few days I'd come down to dinner and tell my wife "You won't believe this, but home ownership rates were declining during the subprime boom, isn't that a crazy thing? That doesn't make any sense, right? Or-"

Beckworth: So was your wife engaging on this, because I may get in trouble saying this, but I can't always get my wife to get excited about nominal GDP targeting and safe asset problems. Are you telling me your wife really is like "Really, dear?"

Erdmann: Well, I think it sort of came back to bite ... She originally had suggested blogging things and by this time I was blogging these ideas, and so her original idea was "He'll tell his blog readers about it and stop bothering me about it," but eventually it sort of ended up being, now I'm in this three year obsession and now she has to hear about it anyway, right?

Beckworth: Okay. So she tolerates you.

Erdmann: Yeah.

Beckworth: Okay, continue your story.

Erdmann: Yeah, so it was a sort of accident, it was just this process, every time I would look at the data, everything was backwards or just nonexistent from the premises that all the different people arguing about what happened agreed on these premises, and the premises aren't anywhere in the data. And so eventually it sort of just, I accidentally took baby step, baby step, baby step, and a year into it here I am with this big pile of data, and sort of starting to form a new way of looking at the framework of it. And so at that point it's like, "Why am I the guy that found this?" But here I am, and so I guess this is what I'm doing now.

Beckworth: Okay. Also to complete the story, Scott Sumner, my colleague, my boss here, and many of our listeners know, the author of the Money Illusion, he started reading your blog, right, or posts that you had written.

Erdmann: Yes.

Beckworth: Somehow he found you and he found it interesting, and then he asked you to write a book about this. Is that how it sent, or did it go a different way?

Erdmann: Well, I'd say the book idea sort of, there was a longtime reader of the block that was supportive of it, and sort of encouraged me to succumb to Mercatus. But Scott and I and Tyler had sort of a preexisting relationship from the blogosphere, really this whole ... None of this could have happened before the blogosphere, before, that I could sit in my loft in my pajamas and download census data.

Erdmann: So, and really that preexisting relationship had a lot to do with it, the market monetarist sort of idea that there was recession that came out of Fed policy decisions that was separate from the housing bubble I think sort of was an intellectual foundation that I was, that all made sense to me. So intellectually I was open to the idea that the story was a little bit different.

Erdmann: And then I think just coming from that sort of tactical investor point of view that I had of sort of a trust but verify viewpoint toward efficient markets, I think gave me sort of permission to, in effect, I was approaching this market in the same way I would approach what you would normally find in an illiquid microcap stock where people's perceptions were wrong and you could find mispricings, but when you're doing that, when you're engaged in that sort of work, you have to be very disciplined, and 99% of the stuff going on you assume is efficient, and you find the little things that look inefficient, but then you have to be very disciplined about, look, most of the things that look inefficient really are efficient and it's me that's wrong, and that's sort of the process I took down this pathway of, "Okay, that's weird, the perceptions seem wrong, but I really need to confirm this." And I've spent three years trying to prove myself wrong, and here I am, I couldn't.

Beckworth: And now you have a book called Locked Out, and how it really was a story of housing shortage that drove the boom, and even what we see today. And really what's interesting about your book is it kind of weaves together, and we'll come back to this in more detail, but it weaves together a lot of the discussions today about NIMBYism, the shortage of housing in urban areas. But what your book does, it shows that this story that people are paying attention to today was really part of the story back then as well.

Beckworth: But let's go ahead and get into your book, and let's begin by kind of getting the executive summary from you and then we'll work out some of the details, but give us the overview of your argument.

Erdmann: Yeah, so what I would say is there's this housing shortage problem. I say writ large, it's basically, sort of we're in this new epoch, we're in this new era where there's a new wave of urbanization that's really required just by the state of technology, the state of culture, the state of the economy today. There's this new post-industrial, non-manufacturing based, service-based economy, and so at the core of that are the sort of innovative workers in tech fields and finance, and they're centered in trade, so they're centered in New York City and Boston and San Francisco and LA, and you can see the same thing internationally, there's London and Sydney, all these cities are sort of dealing with this unstoppable force and immovable object problem where these cities suddenly have this tremendous value for those workers, they need to be close to one another to be involved in skunkworks and relationship-building, and politically we've evolved to a position where locals have so much control over what happens in the rest of the city that we're not building tenements in Manhattan anymore.

Erdmann: So that was probably a swing too far in a direction by today's standards in terms of building housing that was unhealthy, or whatever. But the pendulum has sort of swung far in the other direction today at a time where urbanization is really valuable. And so the innovation workers are probably the core of that, but the second part of that is the service economy that sort of comes along for the ride there. And so we have this problem throughout the country of people who have, say, lost their jobs in the manufacturing sector, and a lot of labor immobility that comes from that, or that's sort of making that problem worse. So workers are sort of stuck in cities with high unemployment and they're not moving toward places that have more opportunity.

Erdmann: Well, what I think I have found is the reason that that's not happening, I think there's a lot of agreement and understanding about this is that they can't move to those ... Where they would move is they would move to Manhattan or to San Francisco and be a barista or a nurse or a teacher or something, serving these core labor pools of the innovative workers that are sort of the income source in those cities, and they're sort of locked ... the working title of the book, Locked Out. Those service sector workers can't ...

Beckworth: They can't afford to move there because of the limited supply of housing.

Erdmann: Exactly. And we call those the non-tradable sectors. By definition, those are jobs that have to be close to their customer, that's what keeps them from being bid away to other markets. And these local housing policies are the source of obstruction that keeps that transition from happening. Now, I think in today's context there's sort of a broad agreement that that's happening. I think that the data I found basically says "Well, everyone seems to also agree that we had millions too many homes in 2005."

Erdmann: Well, that's just sort of a weird thing to have to tack onto this problem that we have today, that there's not enough housing and that's the obstruction. So what I found is there actually has never been too much housing. All along our problem has been a lack of housing, especially in those particular cities. And so once we remove that, to me it's sort of a virus in the national ... we remove that virus, and then everything actually becomes much more coherent in terms of thinking about the problems that are keeping the economy from being more vibrant over a couple of decades.

Erdmann: And so we look back at, say, 2005. So Ben Bernanke in 2006, his first meeting of the Fed, he comes away, in his memoir he talks about how they felt like it was a successful meeting because they'd finally raised interest rates enough to start to sort of pull down residential investment, because we had this overhang of supply that everyone thought we had. Well, the year leading up to that meeting, there were probably, I don't have the numbers off the top of my head, but probably 200,000, 300,000 households that were piling out of those cities, out of New York City, Boston, San Francisco, and LA, the cities with the highest income potential in the country, cities that are the centers of kind of the new prosperity, hundreds of thousands of households were moving out of those cities for lack of a house.

Erdmann: And so they were sort of moving into second best alternatives in other cities, but their problem wasn't too much housing. And in fact we'll get into the details later, but even in cities like Phoenix at that time, there weren't too many houses when you really look at the details of what's happening. And five or six year later, Bernanke is still saying "Well, the economic recovery is still moving a little slow because we're still just working off this overhang of supply." And so this sort of, the mythology that we had too many houses sort of muddied our public policy decisions, it really has been for a decade. We really had a shortage of housing, we'd been managing the economy as if we had a surplus, and you can imagine how just that mistake itself, when you follow it through one policy decision after another, could itself lead to a crisis situation. This is a huge asset class that we're treating in an upside down, sort of bizarro ...

Beckworth: For most people it is the main asset class in America, right?

Erdmann: Yes, yeah.

Beckworth: For better or for worse, housing does make up most of our asset side of household balance sheets.

Erdmann: And just to follow up on that, in terms of data, just one of the basic things that I've looked at is say, just housing units, these are basic, this is not buried in the bowels of the BEA or the BLS or something, just go to the Census Bureau and look at housing units per adult or housing units per capita. There was a rise in those measures through, say, the '70s and '80s as household reorganization was sort of happening culturally, but those have been fairly measure levels from-

Beckworth: Yeah, so that's what I want to get into with you now is the metrics that you present. And your book has a lot of them, so we'll touch on just a few. But kind of your argument is that there was actually a shortage of housing that led to a number of bad policy decisions, misunderstanding, that ultimately led us to a Great Recession, and even since then, the response that's been given has been in adequate. But let's kind of build your argument up for our listeners, and let's begin with one you've just touched on. And this is, you're kind of contesting the claim there was a supply overhang. And so you go through a number of measures that do this, so let's work through them.

Beckworth: Let's start with housing units per capita, or number of homes, I guess, or housing stock per person. You just mentioned actually that it had been going up in the '80s pretty sharply, right?

Erdmann: Yes.

Beckworth: And the early to mid-2000s really wsnt exceptional based on that measure.

Erdmann: There's a little bit of a rise, but there had been quite a bit of a decline during the ... Housing starts, compared to historical standards, were pretty tame through the '90s, and so there'd actually been sort of a slow decline. So we sort of made up some of that ground, and I think partly what led people of the track there is that we tended to concentrate, it seemed like the bubble was in single family housing, and that was a lot of the focus. And so the idea was that there was this excess of lending into that market, and so whenever you'd see charts and graphs or whatever, it would be single family homes built for sale, and that, it was probably more than 50% growth over a few years in that category, and prices are rising at the same time, that looks crazy. All these measures, they looked at first glace as if, how could it be anything but a demand bubble? You've got quantity and price going up at the same time.

Beckworth: …at the same time, yeah.

Erdmann: But over that entire period, there's manufactured homes, there's multi-unit housing, there's homes built by owners or homes built by contractors for individuals. There's five or six major conduits of new housing units. Most of that growth in single family homes was just market share shift from those other units. So when you add up all those other units, and especially manufactured homes, which you have to sort of-

Beckworth: What is a manufactured home?

Erdmann: So they're sort of, you could have anything from, say, a park model or something that's more like building off the recreational vehicle category toward more just factory constructed homes that have been-

Beckworth: Prefabricated, okay.

Erdmann: ... prefabricated and brought to the site and made into a permanent location. So there's a couple different categories there, and they're sort of tracked separately by the Census Bureau, so to really get the total number of units you have to sort of grab these two different-

Beckworth: Add these all up, yeah.

Erdmann: Yeah. And when you add them all together, there's really nothing during that boom period in 2002 to 2005, we were slightly over the long term average. In fact, you could even, if you measure it, say, at a per capita basis or something, we didn't really ever ...

Beckworth: Yeah, so on a per capita basis it's very clear that there's at most a little bump. In fact, the way I would describe it looking at your figures, housing stock per capita per person, there's a little bump on a road that's going downhill.

Erdmann: Yeah.

Beckworth: There's actually, the recent few decades there's been a decline relative to the '80s, but then I think the second point you're making here is even if you just look at the absolute number of housing starts, if you don't divide by population, what the mistake was, and I'm guilty of this too, is that we looked at housing starts for single family homes. And even, you're watching CNBC, watching some news channel, the picture they throw up is an individual home, the kind of home like I live in, they put that up and you identify. You don't think of condos, apartments, you're not thinking of multifamily and these other ... But so your point is if you add up all the housing, and I went and looked you're absolutely right, your charts show this too, that it really wasn't that unusual, which is kind of a mind-blowing moment. I had to do a retake on this, of like "Did Kevin get these numbers right?"

Beckworth: Now, one question I had about your housing units per capita was the 1980s. It's a pretty stark change in the trajectory, there's huge increase in the '80s in the number of homes per person built. And one question I had, I was wondering if maybe that's tied to the baby boom, or are we past that period?

Erdmann: Well, I don't think it's so much the ... I think that that shift in the '80s probably had more to do with falling household size, sort of that. I mean, it had to do with the baby boomer lifestyle change, divorce rates rose for that period of time, there was sort of more, maybe single head of household households, that sort of thing. So the number of adults per household has been declining, and it was probably pretty strong factor at that point.

Beckworth: Okay.

Erdmann: Yeah, so that's-

Beckworth: Yeah, actually, the more I think about it, the Baby Boomers would've been several decades before, probably in the '60s and '70s when they bought their first homes.

Erdmann: Well, they were probably in the market as first time buyers at that time.

Beckworth: Some of them, okay.

Erdmann: But that's why, I kind of like the households per adult measure better for that reason, because it sort of factors in that. But just the number of adults, that is a little messier than the population data, and so I like to use both to sort of show-

Beckworth: That's right, you do show both in the graph.

Erdmann: Yeah.

Beckworth: Okay, so we have these measures that show there wasn't a boom in the quantity relative to population, and even in the absolute number. You also show the amount of expenditures in housing or on housing adjusted for inflation, and that similarly does not show some kind of alarming growth, right?

Erdmann: Yes, yeah. It's funny, when you sort of look at all this topic from a new framework and you read the other analysis on the housing market, there's a lot of sort of misplaced gravitas overlaid on the analysis, there's a lot of ... For some reason housing is treated as this sort of nasty form of overconsumption. So one of the things that I've looked at, there's sort of the education, healthcare, housing trifecta that's the three sort of sectors eating the economy, right? And the funny thing is, housing is very sensitive to income. Over decades, in nominal terms, in terms of rental value, households have spent, in terms of GDP I think it's maybe around 12% or something of GDP, if you look at income data it's maybe more like 18%, but it's, since the, say, late '70s to today, that portion of spending is flat in rental terms.

Erdmann: But education and healthcare have shot up, they're taking more and more of our incomes every year. But for some reason, when people think about housing, they think about keeping up with the Joneses, oh, people were buying houses that are more than what we need, and we have this sort of-

Beckworth: The McMansions.

Erdmann: ... yeah, we have this judgmental idea about it, but actually for the last 30 years, Americans in the aggregate have been ratcheting down their real housing consumption because rent inflation has been high. So out of those three categories, it's actually the one category that's sort of tethered to, that we're naturally unwinding our real consumption as a sort of compensation for these other costs.

Erdmann: So yeah, over that period of time, since the mid-'80s, if you compare real consumption of housing in terms of rental value, which to sort of back up a little bit, I think that's one of the really important sort of foundations at looking at all this coherently is to not confuse price and rent. I think rent, we need to reorient ourselves in this sort of analysis toward rent as being the cost of the service of housing, the thing that we build everything else on, and price is more like the price of a bond, it's more like the decision of allocating capital in a certain way. 30% or 40% of the households of the country are renters, the price of their unit is insignificant to them.

Erdmann: So in this BEA data, we're looking at it in terms of the rental value of the housing stock. And since the '80s, every year, households, their real incomes go up 2 or 3%, but their real consumption of housing only goes up 1 or 2%, but then they have an extra percent of rent inflation every year that's focused in these cities, that's focused in New York City and Boston and San Francisco and LA. And so yeah, in real terms, we've been cutting back on housing, we haven't been keeping up with the Joneses at all. In fact, what Americans are trying to do in the aggregate is to keep up with the Joneses' incomes, and the way to keep up with the Joneses' incomes is to bid your way into LA or San Francisco, and all that income goes to the landlord.

Beckworth: Yeah, so to answer your question, why do we think we're keeping up with the Joneses, part of it is because there is this perception, a wrong perception, which is what you're trying to argue, that housing is, these bubbles have occurred, and again your point is it's more of a supply-side constraint.

Erdmann: Yeah.

Beckworth: And that housing, really, I mean, the last point you just made, housing, to the extent it matters, is really a symptom of the deeper underlying problem, this is the desire to move to a city to get the income. But these cities have problems, and that's a nice transition into the next part of your work, and that is the closed access problem. So you mentioned New York City, Los Angeles, Boston, San Francisco as the key cities that have this problem. You call them closed access cities. So tell us about that, what makes them a closed access city?

Erdmann: If you're casting a wider net you could through San Diego in there probably, but it's a little bit, these four cities are 95% of the issue in terms of aggregate total value or aggregate ... Eventually if you cast the net wide enough, I think Honolulu has some of the same signatures, but every other city in the country, even Seattle in Washington managed to build more than the average number of housing units in their metro areas, and so they don't have these signatures. So the thing that makes those cities different is across the board they have much lower rates of housing starts than the average city. And really what's happened is, if you think of the drag this puts on an economy, the natural, historical sort of moderating feature of a free economy is that migration of jobs and labor to sort of moderate differences between regional areas.

Erdmann: And that was happening for decades, and in a lot of ways it's still happening in other parts of the country, in other parts of the economy, but because these four cities have, there's basically a cap on their populations, in effect, they've actually reversed that flow. So now we actually have a strong migration pattern of households with lower incomes who are in need of opportunity having to move out of those cities, and households with higher incomes moving into those cities. So what's been a centuries-long process of moderation has now turned into a process of inequality, of separation, of segregation by skill and income.

Erdmann: And so in these cities you get this very distinct signature where their average incomes are very high and growing, housing starts are very low, and then rent inflation every year just keeps ratcheting up and ratcheting up. So a good portion of those high incomes is simply just routed through wages to the real estate ownership class.

Beckworth: So it's a way of protecting the people who first were there who owned the real estate, it's kind of a rent-seeking ...

Erdmann: Yeah, I think it sort of happens by accident. I think people, every neighborhood has always sort of not wanted to change.

Beckworth: I mean, that's true for me. I think all of us kind of naturally want to keep our neighborhood looking nice. It's easy to be critical of the NIMBYism, not in my backyard mentality, but I think all of us would be guilty ... It's rational as an individual, as a society, collectively, it's harmful.

Erdmann: Yeah. And I think that's the political dilemma is how to sort of solve that community problem, and so I think it has to be ... And one of the things I go into in the book is sort of the North, Wallis, and Weingast idea of limited access orders versus open access orders. I think that's the real, in terms of this part of the problem, that's the real dilemma, is even they sort of lay out how universal access and free flow of labor and capital and all these things are important to an open society with abundance and economic growth, and even they sort of say "We know what it looks like, we don't know how to make it."

Erdmann:  And in a way, these cities have devolved back into a limited access order, and so the local politics are all about getting your part of those rents, extracting your part of the rents, and that can be by, it's already happening, a monopoly on real estate ownership, that can be through rent control, so that you get to live in the 4000 a month unit for only $1000 a month. And so in that context, everyone feels like they have the moral authority to get their part of the fixed pie, and somehow collectively we all have to pull back from that and allow it to be open again, which means everybody sort of makes compromises for the sake of an open society, but how to do that is a tough nut to crack.

Beckworth: Yeah, and what's interesting is that the problems in these cities are spilling over into other cities, so you've got that same NIMBYism. We'll get to that in a minute. But I just want to mention, the point you're raising here is these cities represent kind of the hub of innovation creative growth. Enrico Moretti story about, there's so much growth, in fact, he argued, and his coauthors, that if you could get full access, GDP would go up by a significant amount, I mean quite a large amount, and the fact that we haven't been able to allow people to move there has really restricted the amount of growth we would otherwise have experienced in the US economy. It might be part of the decline in economic dynamism.

Beckworth: All right, so we have these closed access cities, and you said you could nudge the definitions, the margins, but for sure we know New York City, Los Angeles, Boston, San Francisco are in the group, maybe a few other cities.

Erdmann: Yeah, well, I mean it really is a significant category. There's not a lot of gray area. Seattle and Washington have sort of a bit of a high cost problem, they have some rent inflation, but at least so far, we'll have to see how Seattle deals with it, because really they're sort of dealing with the same pressures, but so far, Seattle, for instance, has built more units than the national average, generally, over time. And they don't have that migration, but you don't see low income households flooding out of Seattle to other cities like you see out of the California cities.

Erdmann: I think what happens is they maybe aren't capable of building enough new units to keep rent inflation from being above general inflation, but they're able to build enough so that those households on the margin maybe have to move to a less convenient location or maybe a smaller unit, but they can remain in the Seattle labor market, they don't have to actually leave their entire life there behind. And so there's a real distinct difference you can see in those four cities.

Beckworth: All right, so those four cities are unique. And then, I don't think you used this term, but the opposite would be an open access city, right?

Erdmann: Yeah.

Beckworth: So open access city, and they're very different. I mean, open access city you have high housing starts, or homes are being built. Rents and prices are moderate, so it's more elastic housing supply. Versus the closed access city, you got very few housing starts and rents and prices are high and inelastic housing supply. So the confusion arises, if I understand you correctly, is that we aggregate all these cities together, and we look at a national measure, we see some housing starts going up, we see on average, prices going up, but it's conflating these two different types of cities. So the housing boom, the Shiller indexes, all those things are really kind of masking these very strong difference underneath the aggregate indicator, is that right?

Erdmann: Yeah, so I think that's sort of one of the things I did differently that sort of led to this new viewpoint is if you treat each metropolitan area as sort of an individual economic zone that's going to have substitutions within that housing market, and so you sort of treat it as an entity, and if you take all, say the top 50 metro areas, and there's, these closed access cities were outliers in terms of the price change.

Erdmann: But if you look at rent, they're actually, what happened is they were outliers in terms of rent. So at the national level it looked like price to rent ratios were way out of whack, and that was a sign that prices had become unmoored from rents, so that looks like it was another piece of evidence that fed this credit supply view, that credit was behind rising prices, and so it would naturally have to retract. But if you actually look at individual, the difference between metropolitan areas, rent really explains everything. They are outliers first in terms of rent, and then where prices are high is where rents are high.

Erdmann: And I think there was a little bit of a regime shift. Up until the mid-'90s, a city that had, say, a temporary rise in rents might, you'd have mean reversion. Over time, there was no sort of, there weren't these outliers that developed. So before, say, the mid-'90s, if a city sort of became an outlier in terms of rent, it would eventually move back toward the norm, and prices at that time didn't reflect an expectation of future rising rents. But as we've entered this period of the immovable object and the unstoppable force, where there's demand for urbanization and these cities won't accommodate it, part of what led to a little bit of that extra price rise in that '95 to 2005 era is now suddenly a San Francisco where rent inflation for that period is above normal. Nobody's expecting that to revert to the mean.

Erdmann: So now the price of a house in San Francisco not only reflects the rent inflation that's happened for a decade, but now it's almost like buying a growth stock versus a value stock. Now that house is actually a rent hedge for the rent inflation that everybody knows is going to happen for the next decade. And so it created a little bit of an extra price boost in those cities, and so that's what moves the national price to rent ratio above a norm, because there's these few locations where you have this more than one to one reaction.

Beckworth: Also, is part of your story, though, the contagion cities? So your contagion cities are the cities you mention where people move from New York City, from San Francisco, because they're looking for cheaper housing, but when they get to these cities, these cities are having a hard time keeping up with the influx of folks, so their prices also go up. So when we look at the national indicators and we see housing prices take off, the boom in the early to mid-2000s, it's because of the closed access cities, but it's also partly the result of the contagion cities?

Erdmann: Yeah, so that's one of the things that I end up ... So talking about these migration patterns and these housing start patterns, really everybody is sort of lumped together in the "bubble cities" that includes the Florida cities and Arizona and Nevada and inland California, and sort of lump those together with the closed access cities and it was just a bubble story.

Erdmann: But in a lot of ways, these are mirror images to one another, and the closed access cities in terms of just total size are much larger than, LA and San Francisco and New York City are much larger than Phoenix plus Las Vegas plus Riverside plus Tampa or Miami. And so firstly, just the closed access issue is, just in terms of absolute numbers is-

Beckworth: By itself-

Erdmann: ... the largest part of the story.

Beckworth: Okay, yep.

Erdmann: In those, so what I call the contagion cities, they're the mirror image, because when you had these massive outflows of migration during the housing bubble from the closed access cities, what makes Florida and Arizona and Nevada and inland California different from the rest of the country isn't that they had a different credit market, really. What makes them different is that they're catching the first wave of that out-migration, they're the main net receivers of net migration out of the Atlantic northeast and out of the Pacific coast.

Erdmann: So they were overwhelmed. For them it was a migration event, a refugee crisis, really, in a way. And so they were overwhelmed with new households that needed houses. So as part of that process, say you have an oil boom that's happening, it's sort of the similar sort of story. When oil prices jump to $120, whatever, there's always this sort of part of the story that oh, it's the speculators driving up prices, people want to tend to blame speculation, and sort of the same thing happened in housing. These other factors were actually creating this boom in those cities, and when that sort of boom happens you're going to get a lot of activity, you're going to get a lot of people transitioning in and out of that market because the context has changed, you're going to get people speculating, people that smarted out as a small-time investor and had a quick 40% return on investment and they're re-upping into that market.

Erdmann: So you're going to have a lot of speculative activity that goes along with that, but that doesn't do anything to tell you what the cause was. That's always going to be there in any market where prices are rising. But when you look at the actual numbers of what was happening, so you're in a city like Phoenix, over the course of five or six years their housing permits went up probably 50%. At the same time, their prices went up something like 75%. And again, it's another thing where any reasonable person looking at that pair of indicators can say that's a demand issue, I know that now, now I can look at other things and know that this is a demand-caused problem is sort of part of the canon now. I don't have to question it, it's like gravity. If other evidence comes to me that contradicts that, I can disregard it, because I've decided that now.

Erdmann: But if you'd noticed these migration flows, over that time maybe it's 20, 25,000 extra units a year that were being built in Phoenix, there were 20 or 25,000 extra households moving from coastal California into Phoenix at the time. They were barely keeping up with the demand, and the irony is if you're in Phoenix, it looks like speculation, it looks like you're ... I live in Phoenix, we were laughing about the price of housing at the time, and there probably were some unsustainable, you could truly call those cities bubbles in a way because there was probably some snap back that was going to happen. But it wasn't a bubble at its core caused by credit, it was a bubble caused by this mass migration event, and the irony is on the margin those households moving in from LA were moving to Phoenix explicitly and clearly to reduce their housing consumption.

Erdmann: So yeah, I basically say even by 2005 when this migration event's happening and Phoenix looks like it's a bubble and everyone is convinced that this is speculative excess, it's actually a bubble, and again, I say this as a proud resident of Phoenix, for those marginal buyers, it's a bubble in an inferior good. It's similar to-

Beckworth: Second best solution.

Erdmann: ... Yeah, so it's like if we were a subsistence society and there was a drought and the price of rice went up and everybody's consolidating into rice consumption, that's not something you tamp down, that's something you stimulate your way out of. So you think of the family living in LA, they're moving out of a little 1000 square foot condo that's renting for $3000 a month, they might be moving into a 3000 square foot house in Phoenix that's only $1500 a month.

Erdmann: So even though, again, another sort of thing that leads us off the path. It looks like there's a lot of gypsum board and lumber and all sorts of things going to Phoenix, residential investment's actually increasing as a product of this migration. So again, now residential investment confirms the excess story, that that family that just built that 3000 square foot house in Phoenix was actually making a tremendous downshift in their real housing consumption.

Beckworth: Okay, so the story again is it starts in the closed access cities where there's supply restrictions. There's a sudden rise in prices during this period, because the can't meet the demand for housing. That drives up the aggregate picture, but on top of that there's spillover into contagion cities, which also can't meet this demand. And I like the way you frame it, they're refugees, housing refugees, this image of "We can't find any homes, please, help us." And so California, Nevada, Arizona, Florida, open arms.

Beckworth: And you have a nice graph in your paper that shows migration flows for the closed access cities and the contagion cities, and they're almost, it's like a mirror image of each other. It's a very striking graph, so it's a very convincing story. Let me ask this question, though. In the closed access cities, there is kind of a surge in prices during the early to mid-2000s. So I was looking at the Shiller-Case index and there's kind of a little bit of a surge. It's growing the whole time, but is that surge atypical? Is there a reason why it goes up as much as it does during the early to mid-2000s in those key cities?

Erdmann: Yeah, I mean I think there is a credit aspect to the story here. And so you can see, especially before the Fed starts raising rates in 2004, there's a brief period of time there where the subprime boom was happening and the boom was happening, sort of starting at the end of 2003. And that was especially happening in the California cities. So during that period you can see a bit of a separation between the Boston New York City type market and the LA San Francisco market, and I think a lot of that can be attributed to that private securitization market, which was very heavy in California.

Erdmann: And so I think there's a credit portion of the story there, but I think the general view of the public of this has been that anyone with a pulse could get a mortgage right? That's the line you hear all the time. And really what was happening was a little bit more subtle. These are markets where if you're a renter, you're outside the norm. The risk in those housing markets is inherent in the close of access problem. If you move to LA or San Francisco, you're taking on housing risk full stop, period. It doesn't matter what you do to supply housing for yourself, you're in a high risk environment.

Erdmann: So we don't have agencies in Washington writing rules that you can't spend more than 50% of your income on rent, but we do have agencies that say you can't lend to somebody that's spending more than 30 or 50% on mortgage. So actually, I would say there was a dislocation in the market before the private securitization market, it was actually keeping prices below what would be a reasonable level in those cities because we regulate one and don't regulate the other. And in the data, what you really see happening in those cities is, it's implausible that households with low incomes that couldn't afford a mortgage could have had any significant effect on those markets. A very small portion of the LA and San Francisco market is owned by people with, say, median and below incomes. And to the extent that they own homes, they bought them 20 years ago and they're sitting on capital gains.

Erdmann: And so what you actually see happening is those sorts of loans were actually allowing high income households to make a rational decision for themselves between, they were basically buying those houses as a rent hedge. Now, they were probably spending 50% of their income on the mortgage payment, so documenting their income is not useful, trying to get them into a loan with Fannie and Freddie is not useful, they're not going to meet conventional norms. So the irony is, it was actually the most qualified borrowers that were feeding that closed access part of the story, and then the out-migration into those-

Beckworth: Spillover effects.

Erdmann: ... other cities is a little more messy, it's a combination of sort of tactical sellers and desperate renters looking to lower their expenses, so that's a little messier situation. But at the core, it's credit to qualified borrowers. And you can see this in the Federal Reserve Survey of Consumer Finances, in that 2004 to 2007 period, there was a boost in what you would call, say, distressed mortgages, mortgages that require more than 40% of your income. The entire growth of that problem was in the top two income quintiles.

Beckworth: Huh.

Erdmann: And it's because that was the story, it was those households bidding their way into those cities where-

Beckworth: To the better markets.

Erdmann: Yeah.

Beckworth: Yeah, so the key story again is, in these closed access cities, there's increased desire to live there because that's where the income growth is, that's where the Enrico Moretti story is being told, the creative class, and so it's a supply constraint. People want to move there, but there aren't enough housing. Now, I guess my question would be does this desire to move there really come to a head during this period? The advent of the internet's really come to fruition after the tech boom, all that over-investment in internet technology, communication technology, so maybe there is this sudden spike in real demand for that kind of living, and so the credit story is just kind of the, maybe icing on the cake in those cities.

Erdmann: Yeah, sort of along the ride. And I think there were some aspects of that private securitization mk that, like I said, probably boosted things in LA and San Francisco a little bit, but Fannie and Freddie take a lot of heat for this period, but they were highly counter-cyclical, they lost a lot of market share during that 2004 and 2005 period. So there were a lot of things happening pushing prices higher and also pushing against ... In fact, this migration issue is one of the parts of that story. So one of the things that there's been a lot of focus on is sort of the housing ATM, this idea that-

Beckworth: Pulling equity out of the house.

Erdmann: Yeah, and so there's this idea that there's this bubble sort of process, that prices rise and that encourages households to overspend and buy more and leverage up, so of course there were some households that were utilizing their newfound home equity for spending, some of that was definitely going on. Although even on that topic I would say the reframing makes a big difference on how you look at that, because if you look at it as this unsustainable credit bubble that's destined to collapse, then that just looks like people consuming out of something that's fake, that's paper profits. But if those houses are expensive because there's this persistent problem of supply that will continue, and a decade later the relative prices of houses in those cities is just as much above the rest of the country as it was then, so it's certainly a persistent value in relative terms.

Beckworth: So housing's a good investment in these cities still.

Erdmann: Well ...

Beckworth: You won't go there, okay.

Erdmann: I mean, to me it's a political risk you're taking on. But my point is just switching our mental framing of what happened, that this is a persistent issue and not a credit cycle event, even those low income households that happened to be homeowners that were using this even to spend out of home equity, effectively they're rentiers, right? They have newfound wealth that's permanent, and especially the ones that sold the house and moved to Phoenix, that's permanent. They're engaging in just natural consumption smoothing. They have newfound wealth, they'd like to spend some of it today as opposed to in the future because they're wealthier now as a result of it. There's nothing unsustainable about that, and actually for those new households, for the ones moved in and had these huge mortgage payments, for them, that's not stimulative. For them, they have these huge mortgage payments they have to make. And the people that sold that home to them and moved to Phoenix or Denver or whatever, they stuck all that money into AAA securities or whatever, they saved it.

Erdmann: So I think one of the aspects that's sort of, the focus was on all the demand side stuff, the focus was on the housing ATM, unsustainable borrowing. One of the things we missed coming out of this migration pattern is in 2004 and '05, a lot of that migration was tactical, was homeowners sitting in their house in 2004 and '05, "Jeez, this thing's going for a million and a half, that's crazy," they may have even thought of it as a bubble, but it doesn't really matter what they thought of it, they saw home equity as an unsafe asset class now, and so they're transitioning their portfolio really in a way out of that asset class.

Erdmann: And so during that period, there's about 2% a year in the closed access cities of net out migration of homeowners. That's a lot of selling pressure. So everyone's focused on all these was that a bubble feeds on itself, and since price is higher, but every year maybe, I don't know, five or 6% of the housing stock may transact? Well, in those cities, there was a 2% selling pressure of people hiking out of town, taking their gains and getting out of town. That's a huge amount of actually mitigating downward pressure on prices.

Beckworth: Let me ask this question, then, how does this come to a head when the so-called bubble burst, when things slow down? For example, in the contagion cities, you talk about Phoenix in your paper quite a bit, you kind of zero in on it as a case study of sorts. Why does the migration to Phoenix slow down? If that's what's driving prices, or what causes the slowdown? Was it the Fed raising interest rates? What was the shock that suddenly now people aren't wanting to move to Phoenix and the housing prices drop.

Erdmann: So the conventional story is that Phoenix overbuilt, there were way too many houses, they were all bought by people that couldn't afford their mortgage and so they all started to default, then those neighborhoods sort of were undermined and prices collapsed. And the data tells the opposite story. So housing starts really, one thing is, across the country, in every city, housing starts pretty much start to collapse at about the same time around early 2006, which is about when the federal reserve was reaching the highest rate level and the yield curve was inverting. And so whether it's Phoenix or Omaha or Los Angeles, housing starts were already collapsing then. And prices stayed pretty level for 18 months or so up until sort of mid to late 2007 just about in every city.

Erdmann: The interesting thing is, so in a city like Phoenix what you see is housing starts start to collapse. The other thing that happens is, again, basically in just about every city in the country, way back in 2006, as soon as housing starts start to collapse, rent inflation across the country shoots up. Now, if we had an overabundance of housing and this was a natural process, that's a strange thing to see happen. And it's a strange thing to see happen in every city at the time, but that's what we see. And so even in Phoenix where we thought there was an oversupply, rent inflation shoots up, housing starts start to decline, and what you see is, backing up before that, that 2004 and '05 period really, like I said, was these households basically seeing rate hikes coming, thinking of home equity as something unsafe, they're moving to Phoenix to get out of home equity, eventually is that process unwinds, and sentiment starts turning sour. Now people start retracting, so now that migration pattern starts to wind down even among renters.

Erdmann: So I think even by like 2006, by the end of 2006 especially, what you have are people that were migrating to lower costs now are saying "I'm getting a little nervous about the economy, we're going to stay in LA, we're going to put up with the rising rents because we're starting to feel nervous about the economy."

Beckworth: So the migration wave came to an end that was driving the-

Erdmann: Yeah, so that-

Beckworth: Okay.

Erdmann: ... so by the end of 2007, what Phoenix lacks isn't tenants. Rent inflation is going high, and tenancy vacancies were low in Phoenix until 2008. So there's no sign in the vacancy ... In fact, the funny thing is, in most cities there's never a sign of a vacancy issue in the housing stock. In the few cities that have it, it's the contagion cities that had this mass migration event and it suddenly dried up over a period of a couple years, and it's only after the migration dries up that you see that entire sort of local economy fall apart, and then vacancy rates are a lagging issue.

Beckworth: Okay. So the initial slowdown, because you know my narrative is in 2008 the Fed really made a mistake. But the story begins before then, because we know housing slows down beginning in 2006, and your story is that this migration wave came to a stop, and it came to a stop you said because people's expectations were changing about what was going to happen to the prices. My question is was that change in expectations driven by Fed policy or by something else?

Erdmann: I think there's a bit of an avalanche effect here. So by the time we get to late 2008, everything's big, everything's a crisis and fluctuations everywhere. The stuff that led to that, it starts small. So you start in spring of 2006 and the Federal Reserve is saying "Oh good, we've caused residential investment to decline," when we actually have a shortage of housing. That's a big mistake, but we're still sort of within the boundaries of fixable errors. And so we sort of move along with that idea for a while, and then the Federal Reserve trying to sort of comfort everybody starts saying "Well, you don't have to worry everybody, we've sort of run the numbers and we've decided even if home prices decline by 10 or 20%, the economy's going to be fine. So don't worry about the housing bust, this is just a natural," even in their press releases they used the word correction, which I harp on a lot, in the followup book I harp on, like that root word correct. This was actually a dislocation, and the way we were officially describing it was correct. So the upside-down sort of nature of our policy responses is just clear in that verbiage.

Erdmann: But thinking about that household around the dinner table in LA that says "Okay, we're selling at the top of the bubble and we're moving to a less expensive city," in financial terms what they're having a conversation about, we have a large portion of our portfolio in what we thought was a safe asset class, which was home equity and we don't consider that a home asset class. We're fleeing away from the home equity asset class because we don't trust it anymore. And now the Fed, thinking that they're making us feel more comfortable, says "Yeah, that asset class could use 20% of its value and we're not going to do a thing about it. We think that's correct, that's a correction." That only feeds the sense that the home equity asset class is unsafe, and I think this goes to the topic you talk a lot about of this safe asset shortage. I think actually the first step of this safe asset shortage was this change in sentiment that sort of happened as rates were rising, that suddenly this huge, multitrillion dollar asset class goes from being a safe asset to not being a safe asset, that actually ends up feeding this frenzy for AAA securities in the CDO market.

Erdmann: And so at first it's sort of just a little tweak a little bit too far in just basic monetary management, but then there's this sort of sentiment side of it, because everyone just assumed that prices were out of whack, and so as housing starts were falling to levels that are very recessionary in any other context, we though "Well, yeah, we had too many houses, it's going to take a while for that to catch up," and in the meantime we're sort of waiting like eventually prices are going to have to follow too.

Erdmann: Well, the fact that for 18 months that collapse was happening and prices were pretty stable actually to me is sort of a confirmation that prices didn't need to fall, and the fact that everyone sort of was waiting for that to happen before we could accommodate, the general public conversation or even conversations at the Fed at the time were "We probably should lower rates, but these speculators need to learn their lesson, they need to know that house prices go down sometimes."

Beckworth: So the original error was that the Fed raised rates too much early on, which began to raise questions in these homeowner's minds, and it kind of created a snowball effect. And then by the time you get to 2008, which is the story that I usually tell and Scott Sumner would tell was the Fed greatly amplified matters, made them worse. So you would put the original sin back in 2006?

Erdmann: Yeah.

Beckworth: Okay.

Erdmann: Maybe I shouldn't say this out loud, but the market monetarist view-

Beckworth: No, that's fine.

Erdmann: ... is sort of that the Fed caused the recession and doesn't have anything to do with the housing bust, I would say the Fed caused the housing bust, and then-

Beckworth: Which then fed into a broader crisis.

Erdmann: Yeah. But even by late 2007, it was all ... I think had they sort of kept rates a little lower in 2006, would would maybe even never even had a recession and we would all be sitting here having a conversation today about whose fault it is that home prices are still so high, and we would have taken for granted that we'd had nice, good economic growth for the last 10 years and we'd be looking about who to blame for housing prices.

Beckworth: Be a very different world.

Erdmann: Yeah. But even my late 2007, that's all fixable, but it would have taken a total, the Fed would've been pilloried for it. The entire public would have been upset at them for letting speculators off the hook.

Beckworth: Right, bailing out housing and people who made bad decisions.

Erdmann: Yeah. And so it's sort of a model risk thing. If your model's wrong, when the bust happens, you take that as confirmation. If your bubble says we have to have a contraction, that that's necessary, then when the contraction happens, you take that as confirmation, but the contraction was actually a result of model risk, we had the model wrong.

Beckworth: Yeah, so in the time we have left, just a couple observations and a final question. My observation is your story, which seems fairly compelling and you present lots of data to support it, and I encourage our listeners to get the book. But the observation I want to make is that this story of financial imbalances, of excesses was simply wrong. We talked earlier about, this really undermines the Austrian business cycle interpretation of the early to mid ... There's a bunch of malinvestment that occurred, and this would say "No, that's completely wrong." But also, not even Austrian, but just kind of this general narrative that these excesses, financial imbalances, and what your analysis shows is that that's completely wrong, that this is really a supply side story, a misreading of the data.

Beckworth: And I think in addition to what you show for that period, seems to me, and this is where my question comes in, that what we're talking about today is kind of confirmation of what you are saying happened back then, the fact that people on the left and the right are talking about zoning restrictions, NIMBYism, the lack of labor dynamism, the decline in labor mobility to these big, high-income earning cities, it's really just a continuation of this trend that you say actually was going on back them.

Erdmann: Yeah.

Beckworth: Okay.

Erdmann: So one of the things that I find in the data, for instance, as in inflation data, at the peak of what we called the bubble in late 2005, so for the last 20 plus years, rent inflation has been consistently above core inflation. So a good part, maybe close to half of what's been measured core inflation has been from shelter inflation. And it was only toward the end of 2005 that shelter inflation and the non-shelter part of core inflation both sort of converged at 2%. So at the top of what we were calling a bubble was actually the Only period of time in the last 20 years that we actually achieved a sort of normalcy in this second best world where you can't build the houses that you'd like to have in the coastal metropolises, our second best alternative, if we're going to have a vibrant, growing economy, whether it's nominal or real doesn't really matter. Any economic growth will lead to a bidding war to get access to the economic potential of those cities.

Erdmann: And so if there's a lot of talk about the bubble economy, and that's all put on the Fed, it's all put on the banks that they're feeding this bubble, all the assets are overpriced because they keep feeding the bubble. The source of the bubble is that there's an inevitable bidding war for assets that give you acess to those cities, and until we solve that problem, that's the sign of actually an economy that's running normal in the second best state. So what we've been doing is trying to avoid running an economy in the second best state because we're afraid of the bubble that's inevitable in that context.

Beckworth: So we're running the third best state.

Erdmann: Yeah, or ...

Beckworth: Or fourth, depends where-

Erdmann: Yeah, so that urban housing problem, that's a tough nut to crack, and I don't know how we do ... That's a complicated topic.

Beckworth: That is the key problem, though.

Erdmann: That's the key problem, but the easy thing, once you take this framework into mind, if we can get a regime shift in the framework we use to see this period, the easy problem to solve is let's just stop beating ourselves up over it to avoid the signal of a second-best economy. And the funny thing is, all the conversations in the United States are sort of built on American exceptionalism in terms of the causes of these problems, the subprime crisis and banking deregulation and the GSEs, but the hilarious thing is, the countries most like us all around the country, Australia and Canada and Great Britain, all these sort of other Anglosphere countries, they actually have more or less been running along in that second best economy. They avoided the deepest parts of the crisis, they didn't have as much of a rise in unemployment and the foreclosure crisis, and they have "bubble economies" that people keep trying to blame on their local credit markets. They're actually doing it right if you can't fix the context of these urban ...

Beckworth: So they are doing the best that can be done in a second best world where you have these supply side constraints. So you mentioned Australia, UK, all these places around the world that have "housing bubbles" are really just, it's the fundamental response to a second-best world. And so they all have to address this issue, this local housing supply issue, that's the ultimate solution. But you're right, so I'll just end on this note. I'm looking at right now in front of me a measure, a graph of global house prices in different countries and they all have these booms. New Zealand, Australia, Britain, Canada. And what's unique about United States is we're the only ones that engineered a sustained decline in housing prices. So we are exceptional, American exceptionalism, we are practicing it. But it's a problem that's more fundamental than monetary policy, than the credit excesses. And all those things need to be done right, but at the end of the day we need to address the housing supply issue. So with that our time is up, our guest today has been Kevin Erdmann. Kevin, thank you so much for coming on the show.

Erdmann: Yeah, thank you David. It's been a pleasure.

Support Mercatus

Your support allows us to continue bridging the gap between academic ideas and real-world policy solutions.Donate