What kind of good is housing?

In a previous post, I walked through some of the value that comes with the consumption package we call “housing” in a given metropolitan area, in an effort to understand the peculiar price patterns that are caused by inadequate building. What are we buying when we buy a house? It includes:

1. Shelter

   Price settles at cost, variable across metro

2. Neighborhood Amenities

   Price determined by income, variable across metro

3. Metropolitan Area Scarcity

   Price determined by income, uniform across metro

4. Endowments

   Idiosyncratic, determines inter-metro migration

In this post, I will attack the same question from a slightly different angle.

Let’s start with definitions of different types of goods.

Inferior goods

“As consumers' income rises, they buy fewer inferior goods. A typical example of such a type of product is margarine, which is much cheaper than butter.” (Inferior goods have a negative income elasticity of demand.)

Normal goods

“As incomes rise, more goods are demanded at each price level.” (Normal goods have a positive income elasticity of demand.)

We can further divide normal goods into two categories.

Necessity goods

Necessity goods are “products and services that consumers will buy regardless of changes in their income levels. Examples of necessity goods and services include tobacco products, haircuts, water, and electricity. As income rises, the proportion of total consumer expenditures on necessity goods typically declines.” (Necessity goods have an income elasticity of demand less than one.)

Luxury goods

Consumers will buy proportionately more of a particular good compared to a percentage change in their income. Consumer discretionary products such as premium cars, boats, and jewelry represent luxury products that tend to be very sensitive to changes in consumer income. (Luxury goods have an income elasticity of demand more than one.)

Let’s work backwards.

Luxury Goods

Urban housing as a luxury good is frequently the most noticeable, salient form of housing demand: complaints about foreign-owned vacant units in the “superstar” cities, second homes, gentrification, Airbnb. All of those uses have a luxury element - either outsiders using housing in a less intensive or less deserving way than simply as necessary shelter, or bidding up home values when local amenities change to match upper income sensibilities.

These forms of housing demand exist. They aren’t made up. But, they are exaggerated in scale. And one reason they are exaggerated is that when supply is constrained, luxury demand is especially galling. But, it is galling because of the supply problem. A point I have made before is that Phoenix has tons of seasonally vacant units. Whole parts of the metro area are basically empty in the summer because they are filled with second homes. If luxury demand was inherently problematic, no city would be more vocal about it that Phoenix. But, until recently, Phoenix built amply, so there is enough housing to meet all types of demand. When housing is adequately supplied, all housing consumers can coexist symbiotically, even luxury and less intensive consumers.

In fact, I would argue that as a general rule, inadequate supply creates an ironic perception. Housing can be a necessity good or an inferior good, but inadequate supply turns housing into a luxury. So, common reactions to inadequate housing are to insist, “No more homes unless they are affordable homes” or to scold millennial first-time homebuyers for thinking the world owes them opulence.

This is just a generalizable perception when supplies of any important item are constrained. When you’re the next one out in a game of musical chairs, it is tempting to blame the big, fast dude that snuck in front of you for the last chair. But there are specific ways in which this plays out in housing.

When housing is a necessity good, it looks like a luxury good.

It is the case that some metropolitan areas have so thoroughly shut down housing production region-wide that they have become exclusionary locations - luxury locations - “superstar” cities, as it were. Luxury is the result of both value and scarcity. This creates some confusion, narratively, since “luxury” is associated with high status. But, the association with high status comes from the competition for scarcity rather than the value of the good itself.

It’s like the iconic scene from “Wall Street” of Michael Douglass on the beach with an early mobile phone. That’s high status. Today, even homeless people have phones that are infinitely better than the “Wall Street” phone. The “superstar” cities (or, more accurately, the “Closed Access” cities) have obstructed progress. They have actually obstructed value. And so the scarcity they have created appears to be high status. American urban housing seems like a luxury because it is stuck on the beach with Michael Douglass in the 1980s. It’s “luxury” because of the stagnation and obstruction of value and progress.

But, certainly, the result of that is that residence in some cities is a luxury good, in the technical sense. The richer you are, the more likely you are to consume it. And, in cities with richer residents, those residents spend more of those high incomes on housing, either renting or buying (Figure 1). Seems like the definition of a luxury good.

Figure 1

But, there is a twist. The median home in expensive cities is only expensive because the cheapest homes in the expensive cities have become very expensive, and median is halfway to cheap. The pattern in Figure 1 is reversed inside cities. Here, we can see the mechanics of a necessity good becoming a luxury good. One generalized way to describe economies in crisis is that “necessity goods became luxury goods”.

In human history, when necessity goods have become luxury goods, the result has usually been something like famine and starvation. Before 2008, in the US housing market, it mostly meant that families with lower incomes moved from place like Los Angeles to places like Phoenix. Mortgage suppression has taken that option away in many ways, so now the housing famine is nationwide, and we can literally see the results in the numerous tent cities that have popped up across the country.

And the tent cities pop up the most in the cities that have most turned housing from a necessity good to a luxury good.

The facet of housing that makes it look like a luxury good at the metro area level, but like a necessity good inside those metro areas is its endowment effect, which I discussed in the earlier post where I wrote about types of housing demand. The endowment effect is just a fancy way to say that people like what they have and don’t like to give it up. In housing, this can include neighbors, friends, and family that live close by, a job, or a host of other local amenities, experiences, and memories that connect us to a place. That is what turns housing a city like LA into such a necessity good that, on average, it looks like a luxury good. If people didn’t mind being displaced, and just moved to other cities as soon as costs were a bit high, then no city would look like a luxury good. Rich people would prefer some cities, and they would pay more to live there, but they wouldn’t spend a higher percentage of their incomes to live there than poor families typically spend on housing. Or, let me put it another way: Luxury goods aren’t luxury goods because rich families spend more on them than poor people do. They are luxury goods because poor people spend less on them.

Los Angeles doesn’t have above average income, rent, price/rent ratios, and prices because it is a luxurious place to live. (Careful! It could be a luxurious place to live! I’m not saying that it isn’t! The fact that it might be is not a confirmation of the mistaken causality about housing demand! Was it less luxurious 50 years ago when it was affordable?)

We can confirm this quantitatively! Luxury goods are goods that rich consumers spend more of their incomes on than poor consumers do. Figure 2 compares price/income patterns across Atlanta and Los Angeles in 2016. Atlanta housing was a normal good. Not particularly “necessity”, not particularly “luxury”. Just right in the comfy middle. That is how many cities looked before 2008. Los Angeles housing is deeply into “necessity” territory. Housing is further from a luxury good in LA than housing is in any other city in the country. The typical rich family in Los Angeles spends a much smaller portion of their income on housing than the typical poor family in Los Angeles. This is the technical definition of a necessity good. This is basically what the Erdmann Housing Tracker is measuring. How much has the lack of adequate building turned housing in each metro area into a necessity good?

Figure 2

The fact that LA housing, in general, looks like a luxury good is because it is a necessity good in shortage. The only reason LA residents have higher incomes than the average city and spend more of their incomes on housing than the average American is because so many Angelinos have been “starved” of housing, as it were, due to the shortage. You’re just looking at the survivors.

One sign of a poor economy is that its residents have to spend more on necessity goods. Residents of the Closed Access cities, like Los Angeles, spend more on necessity goods. The higher nominal incomes of the remaining residents is a misdirection.

Unfortunately, all American cities are looking more like Los Angeles now. They all have increasingly downward sloping price/income lines. Housing everywhere is becoming less like a neutral normal good and more like a necessity good. And, predictably, as a result, the calls for limiting financing, lowering nominal incomes, and blocking market rate (“luxury”) housing have ironically increased with it.

This means that some portion of democratically elected representatives will passionately oppose the only solution to the problem. When a necessity good is in shortage, everyone can unite against a shared enemy: other people with more money.

There is some truth to the “luxury good” perception. For a potential newcomer, LA or New York City or San Francisco can be like a luxury good. The price of the luxury good of a Closed Access city is the bidding war on the necessity good of housing you have to embark on when you arrive.

As the price/income slope steepens - as housing becomes more of a necessity good in a city, the necessities families are bidding to hold on to become more critical. It might start with having an extra bathroom or a yard. Eventually it’s neighborhood character, avoiding crime. Finally the necessity is increasingly difficult choices about location or commute time, ending with regional displacement.

When housing is an inferior good, it looks like a luxury good.

Large cities should be a normal good. Very broadly speaking, the larger they are, the more value they can provide. This is the notion of agglomeration effects - again the “superstar” city idea. The idea is that Bigger cities are better, and to get bigger they must eventually become denser. Being denser is, then, associated with being better, and being better surely must be associated with being more costly.

As I have discussed before, there is a nugget of truth to this. Building in dense areas is more costly. But, I don’t think it is nearly as simple as some models make it out to be.

Figure 3 shows a scatterplot of ZIP codes in New York City, the quintessential American density story. Figure 3 displays median home prices in ZIP codes arranged by population density. More dense ZIP codes in New York City have only been more expensive than less dense ZIP codes within the last decade. Figure 3 compares 2002 and 2021. In 2002, the densest ZIP codes, on average, were less expensive than the least dense ZIP codes.

Figure 3

I think there are a few things going on here.

You don’t have to spend very much time watching housing markets to realize that density comes with a lot of perceived negative externalities. People really don’t like letting other people live near them.

In the earlier post, I speculated that having rich neighbors is itself an amenity. People dependably go to the city council to complain about new housing and they dependably pay more for housing in order to avoid living near people with lower incomes than them. This creates a positive feedback loop to higher rents and higher price/rent ratios in ZIP codes with higher average incomes. This becomes especially important where population density is high. More people near you means more potential trouble.

So, when a neighborhood is gentrifying, locals worry that they will be priced out. This is related to the points I made above. When there is a shortage of a necessity good, it looks like a luxury good, and we tend to take offense at the winners and losers the bidding war creates.

There is some subset of high earning newcomers in places like New York City who are seeking dense, high-amenity neighborhoods to live in. It seems like we can infer from Figure 3 that density in New York City has become more valuable over the past 20 years, and that the primary trend here is that dense housing is a luxury good being bought up by outsiders.

Figure 4 shows adjusted gross incomes in New York City in 2002 and 2021. There are a handful of ZIP codes where average incomes have soared (the orange dots at the top of the range where the average income is much higher than it was in 2002). But, those rising ZIP codes are not concentrated in the dense parts of New York City. There isn’t really any correlation between density and income growth across New York City over the past 20 years. Income growth is pretty similar across ZIP codes of all densities. In fact, while home prices have risen more in dense ZIP codes, incomes have risen more in less dense ZIP codes. (Compare the trendlines in Figures 3 and 4.)

Figure 4

Figure 3 shows that the average price of NYC housing isn’t that correlated with density, but income is highly negatively correlated with density. So, price/income ratios across New York City, in both 2002 and 2021, are higher in the densest ZIP codes, which are mainly inhabited by the poorest residents. Residents with the lowest incomes spend a larger portion of their incomes for dense housing.

In other words, density is an inferior good.

Figure 5

As you can see in Figure 5, the cost associated with density in New York City has risen tremendously. As I discussed above, all of these complicated interactions under the duress of inadequate supply become misinterpreted as markets in luxury goods. I think that leads to an inference from Figure 5 that the price of density is being bid up because density is a luxury good. And, surely, the handfuls of local neighborhoods that are idiosyncratically rising in socioeconomic status are taking in a lot of luxury-seeking newcomers. Those newcomers prefer to be surrounded by other luxury-seekers. The anecdotes are easy to see.

But, many of the dense neighborhoods aren’t gentrifying. And, yet in all those non-gentrifying dense neighborhoods, price/income ratios are soaring. Do the poorest residents of New York City suddenly value density more than they did in 2002? Not likely.

The correlation is spurious. The poorest residents of New York City are mostly being outbid for a necessity good that is in short supply, and it just so happens that the neighborhoods they are holding on in are neighborhoods that are dense.

Figure 6 compares Houston, New York City, and Los Angeles - the price/income ratio by ZIP code, arranged by income. Houston and Los Angeles don’t have the extensive high-density areas like New York City does. They don’t really have as much of the inferior good component. They differ in that Houston builds a lot of homes, so homes in Houston are mostly just a regular normal good. Los Angeles doesn’t approve enough homes, so homes there are necessity goods in short supply.

New York City is generally a housing market, like LA, which is now a necessity good in short supply. In New York City, the low tier areas have always been inferior goods.

Don’t make too much, by the way, of the term “inferior” good. In many cases, like with margarine compared to butter, it can refer to a lower quality and cheaper alternative. But, the technical definition is simply that consumers with lower incomes consume more of it. In the case of dense neighborhoods in New York City, inferior goods usually do contain some negative externalities, like crime. But, they also are neighborhoods where you can depend on public transportation and other public services. These neighborhoods have a lot of positive aspects that are especially valuable for residents with lower incomes. Density has value as an inferior good, in the technical sense. That doesn’t mean it’s inferior, in the colloquial sense. Dense cities provide a lot of value to residents who need the most public services! This is one big reason why growth in our dense cities of any kind tends to benefit the neediest residents. Millions of Americans have been denied the “inferior” value of dense urban living. The social costs of Closed Access housing have been considerable. Think of it instead as “progressive” value - housing that poor families value more than rich families.

Also, I must mention that if metro area size tends to add value, then Houston is, by a long stretch, the only metro area in Figure 6 who added much of it. The only way that the size-to-density-to-value story makes sense in New York City is if the pre-existing dense parts of the city became more valuable, as is, because of changing preferences about the density it had. Outside of some of the New Jersey municipalities, it hasn’t added much density.

It would be interesting to dig into the demographics of New York City a bit more. There is so much focus on the effect that new developments have on nearby rents, but it seems probable that much of the cost pressure in dense New York neighborhoods is happening to neighborhoods that are demographically stable, either because they don’t have a lot of residential turnover or because poor residents from other parts of the city are moving into them - not because of demand for “luxury” density, but because of demand for “inferior” density as less dense parts of the city become more expensive.

Conclusion

Large cities can be valuable. They are a normal good. And, where housing supply is constrained, the city itself resembles a luxury good. But the housing within the city becomes a necessity good. This is all tangled up in the same process, driven by scarcity.

Adding a layer of complication: in order to become larger, a city must eventually become more dense. Density is an inferior good. However, having a lot of rich neighbors is a luxury good.

Existing residents of dense neighborhoods value them for the amenities like public transportation. New residents value them as the entry point into an exclusive metropolitan area. The clash between these groups is the dramatic center of the housing drama in housing deprived cities. When housing supply is inadequate, that drama is amplified.

All of these processes tend to get misinterpreted into a broader simple story of “luxury” consumption. It is a much more complicated story than that.

The story is complex, and in practical terms, the solution is complex, in a city that has given itself permission to make it complex. But, the solution simplifies everything, because ample housing isn’t particularly inferior, or necessity, or luxury. Where housing is amply supplied, it’s just pretty normal. Not particularly dramatic, after all, when you just do it. We have given that broad easy story up for the complications of the impossible goal of neighborhood preservation.

In the complications we have created, there are so many stories one can concoct from the different types of demand housing has to start filling as supply becomes distressed. The worse it gets, the more anecdotes there are with villains and scapegoats, and the more committed people become to missing the point.