More Notes on Kalamazoo

The previous post was a reaction to a New York Times piece by Conor Dougherty, titled “What Kalamazoo (Yes, Kalamazoo) Reveals About the Nation’s Housing Crisis”. My main point in the piece was that while Dougherty seems to recognize some place for tighter lending standards in this story, he doesn’t center it as much as he should - to the point that he didn’t seem to question why some of his subjects were renting when buying surely would have been preferable, and they appeared to be decent credit risks.

What is even happening?

But, the more I have thought about the article, the more the anecdotes just don’t make any sense to me.

The first anecdote involved a family making more than $100,000, who bought a home for $170,000. They say that the value of the home has doubled. I touched on this in the previous post. That’s probably going from a price/income ratio of about 1.5x to 2.5x. What an odd anecdote to lead off with in an article about unaffordable housing. That is exceedingly cheap, even after the doubling, even in normal times in normal cities that don’t have housing shortages.

The second anecdote is a renter making $65,000 a year. They had to move out of a unit that was $630/month. They had to settle for a new place renting for $1,500. Again, as a percentage of income, this is 12% and 28%. The original unit was extremely cheap. But, even the new unit isn’t above the commonly cited 30% threshold for rent stressed budgets. The unasked question on the second anecdote is why didn’t they buy, since homes renting for $1,500 in Kalamazoo usually sell at prices that would require a much smaller mortgage payment. They could have personal reasons, or they could be denied under current underwriting. At least this anecdote makes some sense in an article about unaffordable housing. Their rent jumped sharply in a way that affects their family budget. This is an example of a family whose real income for other expenses has declined because of rising rents.

The third anecdote is the strangest one. This is a family that, again, is identified as having income over $100,000. They qualified for state subsidies to move into a rental for $1,700/month. That’s rent/income less than 20%. I don’t understand anything about this. They may not qualify for a mortgage under current standards, so that could explain why they are renting. But, why are we subsidizing families with $100,000 incomes for rent to be 20% of that income? Also, there are a lot of existing homes in the area that have market rents near that value or much lower. I don’t understand why they were dependent on the subsidy. Why they weren’t shopping for market-rate homes, even if they were rentals. And, if that meant paying $2,500/month, is it really the state’s place to make sure a family with $70,000 of income remaining after rent gets to live in a nice new home when other homes are available? Why isn’t the family in the second anecdote getting the subsidy?

I can believe that screwed up underwriting is preventing this family from having more options via buying, but nothing else there makes sense.

The odd thing is that, in an article about housing affordability, Dougherty doesn’t make any attempt at quantifying the scale of these expenses in the way I have here. These numbers are just thrown around with an undefined presumption that they are unaffordable, even for families with high incomes. What was the point of this? The article is based on the presumption that families with higher incomes in Michigan now need housing subsidies, but does Dougherty even try to quantify or justify that presumption?

The Three Parts of the American Housing Donut

All that said, these three anecdotes encapsulate well the condition of the post-2008 American housing market.

We have 3 basic groups of Americans:

1. The haves: Families that still get mortgages and basically still consume housing the same as they did before 2008.

2. The should haves: Families that used to be owners, but now are renters. Since they are locked out of the entry-level single-family homebuyer market, those homes generally aren’t being built. So, in many places, they are poaching rental homes and apartments from the existing market. Or, in this case, they are poaching state subsidies to facilitate new units. Which, hey, maybe is better than nothing. It got a house built.

3. The never haves: Families that were renters before 2008 and are renters today. They get the worst of everything because the main result of the mortgage crackdown is that the “should haves” are claiming more of the existing stock of homes and driving up rents in the lowest tier of units.

Let’s think about how these groups interact. First, some background.

I think I have mentioned this elsewhere. Housing finance and public debate is littered with attribution error. Everyone else over-consumes, over-leverages, tries to “keep up with the Joneses”, and maxes out to the regulated maximum. So, we have to limit the liquidity available in housing to keep them safe.

A lot of this relates to costs. As I frequently write about, there is a lot of overstatement and confusion here. Frequently, the narrative starts with a dovish Federal Reserve, keeping interest rates artificially low and “spiking the punch bowl” to stimulate overinvestment and excessive construction. Practically everything about that sentence is wrong. It’s not the point of this post, but the Fed doesn’t control mortgage rates. It certainly doesn’t lower them with dovish rate cuts over the course of years or decades. Low rates aren’t associated with higher prices or more construction. The mortgage required to buy the median home has never been a remotely stable variable over time, so mortgage affordability is clearly a weak factor in home price trends. The idea that the GSEs or FHA and other tax and subsidy factors are a net subsidy is greatly overstated or not true.

Then there is what we would call in finance, liquidity - how easy is it to own and trade a good. Here, liquidity is mostly access to mortgage financing.

Liquidity is a public good. It makes us communally richer. It makes things more valuable. In finance, where it is most commonly discussed is stock and financial markets. A corporation that trades on a liquid stock exchange with active trading and low spreads will be worth more than a private company that must be bought or sold in chunks to a private market with few traders. In other words, for a given dollar of earnings, investors will pay more for shares of the liquid public company than they will for the private company. And that’s because liquidity is valuable. It’s worth the extra price.

Homes are already illiquid because transactions are expensive. They are hard to sell. Buyers frequently can’t secure mortgage financing. So, housing is underpriced because it is illiquid. For a given dollar of rental value, homebuyers pay less than they would if housing was more liquid.

The mortgage crackdown in 2008 was a massive liquidity shock. The prices of homes declined for a given dollar of rental value. But, since the price of homes, at the end of the day, is tethered to the cost of concrete, lumber, gypsum board, etc., making housing less liquid doesn’t, in the long run, lower its price. It raises the rental value.

And, that is basically the problem we have created for ourselves. In 2008, we collectively decided to remove liquidity value from housing markets. We made ourselves poorer. And, the way we made ourselves poorer was by raising our rental expenses. Rents went up, real housing consumption went down.

Figure 1

But, this plays out differently among our three groups.

The Haves

The Haves just kept consuming housing like they had before. The lines in Figure 1 just kept moving horizontally for them. And, the article drives home a point about housing consumption and attribution error. A price/income level of 3x or a bit more is normal in affordable, reasonable markets. Did the family that bought the home in Kalamazoo take advantage of the extreme affordability in Kalamazoo to max out their affordability and move into a big mansion? No. They bought the house that seemed appropriate for them, saved a bunch of money, and even used the leftover capital to buy an investment home.

That is how actual families act. There are always a variation of consumer behaviors, and anecdotes to reference that point to excess. But, this is why the various financial cost factors don’t move the macro-level numbers around that much. Families generally buy what they need. They don’t systematically max out their liquidity. Not remotely.

It seems to me that the modal housing analyst treats a measure like mortgage affordability (the percentage of the average family’s income that it takes to buy the average home at current mortgage rates) as a key driver in home prices and housing activity. If you’re the modal analyst, keep this family in mind. You will do yourself a favor to remember that they are the modal homebuyer.

So, these families don’t have that much of an effect on the rest of the market. They are just buying the homes they would have before. Some of them buy new homes. A few also buy investment homes.

The Should Haves

The should haves would mimic the haves, but we don’t allow them to. They actually are poorer than they used to be. We didn’t actually have a mass sit down in 2008 and discuss how we had decided to be poorer. (A common refrain about the period was that we weren’t as rich as we thought we were. That wasn’t really the case. I suppose we decided to be poorer because we thought we were already poor.) So, families don’t sit down at the kitchen table and say, “Well, we will have to rent, because we all agreed to be poorer. So, we each had our own bedroom. Our parents shared a room with their siblings when they were young. Our kids are going to have to go back to that arrangement… Because we’re poorer. Because we decided to do that illiquidity thing, and so now we can’t reap the benefits of homeownership.

Additionally, they see the haves, who they naturally benchmark to, and the haves are still living like they used to live. Their kids each have their own bedroom, etc. It doesn’t seem like we agreed to be poorer.

So, a Should Have family intuits that the home a family like them should own should be similar to one a family like them would have lived in a generation ago - maybe a home that would rent for $1,700. But, being poorer, they are stretched a little further than they would have been a generation ago, and they feel the need for public help to rent that house.

Or, maybe in another market, they might compromise. They might spend a little more than they would have, but for a place not quite as nice. Maybe the bedrooms will just be a little smaller. Maybe it will be a place that rents for $1,500.

The Never Haves

The never haves have been made the poorest. The poverty comes first through rent inflation, eating up any additional income, and then some. They don’t have so much to compromise away. They may already be sharing rooms. So, they frequently put up with high rent inflation. In the aggregate they are losing housing. Their apartment might be sold and upgraded for some Should Haves, and so they might find themselves in a bidding war for that $1,500 unit.

Dougherty writes eloquently in places about these sorts of interactions in his book “Golden Gates”.

One of the major motivators for the Erdmann Housing Tracker is the data pattern created by these trends. We were poorer in the 1970s. We consumed less housing per capita in the 1970s. But, in the 1970s, this didn’t typically cause families to spend 30% or more on rent, and there weren’t markets where homes sold for more than 10x the typical local income, as they do in some cities today.

The reason they do today is because we initiated this poverty so quickly that families haven’t had the time to adjust. They are getting poorer faster than their expectations can keep up.

Productivity in housing construction has been notoriously lackluster. That sort of effect would lead to stagnation in housing. Families might stop living in nicer, larger homes as a result of that, but they wouldn’t start spending more of their incomes on housing.

Our illiquidity event was more acute than stagnant productivity. We are so much house poorer than we were that many families would need to downgrade in order to “be as poor as we are”. Families don’t downgrade easily or as a continuous function. In fact, when downgrading has to come in the form of regional displacement, many families will spend most of their incomes on housing, or even will live without a house to avoid downgrading by moving away from family, friends, jobs, and social supports. The high rent inflation and the decline in incomes after rental expenses are de-growing pains. It will take decades to grow our way back out of it. It will probably take generations to adjust out of it without growing our housing stock.

The regressive decline in incomes after rent. The regressive rise in housing costs within cities. These aren’t just signs of some difficult supply conditions. These are signs of a housing economy that is devolving so quickly that families have to experience real decline, en masse, to an extent that difficult transactions are required. They must choose between material devolution in their housing quality or paying more, and those choices must happen in chunks rather than continuously.

You can see all of this happening in Dougherty’s anecdotes.

There really was one switch that started all of this moving, and that was the exclusion of families like family 2 from mortgage access - the creation of the Should Have class.