You’re familiar with the blind men and the elephant parable, right? There are many variations of the parable, but it basically goes like this: several blind men are asked to approach an unidentified object (the elephant), and based on what they can touch, provide an answer for the object. One man touches the trunk, and believes he is touching a snake. Another touches an ear and thinks it’s a fan. One more man touches a leg and is convinced he is touching a tree trunk; yet another man touches the tail and believes it to be a rope. The man who touches the tusk believes he is touching a spear. Each has enough information to make a judgement, but none have the perspective to make the right call.
It’s a useful story because it points out how people have a tendency to project a personal or partial experience as a whole truth — and ignore the experience of others. One of the lessons we an infer from the parable is that another person’s experience, no matter how contradicting it is to yours, can be just as valid as your own.
I raise this point because the housing challenges facing our nation’s large coastal cities — staggeringly high prices and rents, rising fast, and with precious little new housing production — have become the dominant narrative virtually for all cities throughout the country. The housing constraints on the coasts are absolutely real challenges, and threaten their future as livable communities for a wide spectrum of people. However, their challenges don’t apply to all cities. Believing that the coastal city experience is universal is a projection of one’s experience as a whole truth. And if we universally apply policy prescriptions to all cities because of the experience of a select group, we may ultimately be doing cities more harm than good.
I decided to put that hypothesis to the test. I recently pulled a bunch of U.S. Census American Community Survey data to evaluate just how similar, or different, the housing markets are for our largest metro areas. My purpose was to see how housing markets, at the metro level, were faring in the aftermath of the Great Recession. I pulled data from two periods — 2008, arbitrarily designated at the start of the Great Recession, and 2016, the most recent year available for ACS data. I zeroed in on the 51 metro areas with more than one million in 2008 (two metros, Grand Rapids, MI and Tucson, AZ, eclipsed the million mark since then, but weren’t included in this analysis). The data points I sought out for 2008 and 2016:
- Total housing units
- Total vacant housing units
- Housing vacancy rate
- Total owner-occupied units
- Total renter-occupied units
- Median rent
- Median home value
Here are the findings at the national level:
- Average median home value grew from $254,000 in 2008 to $266,000 in 2016, an increase of 5.3%.
- Average median rent value grew from $757 in 2008 to $923 in 2016, an increase of 21.5%.
- On average, the largest metros increased their number of housing units by 7.4% from 2008 to 2016.
- Metro areas reduced the number of housing unit vacancies by an average of 9.6% over the period.
- Renter-occupied units increased by 19.3% over the period, while owner-occupied units increased only 1.1%.
However, those numbers are virtually meaningless at the national level. There are no metros that directly mirror the national housing market averages. But when you look at things from a regional perspective, trends begin to emerge:
Northeastern metros have the lowest rate of housing unit growth, but it’s the only region with increasing numbers of housing vacancies. Midwestern metros have a similarly low rate of housing unit growth, but have the lowest increase in median rent and is the only region with stagnant to falling home prices. Southern metros are adding the most housing units, but have the greatest increase in the number of renters. Western metros show a dramatic drop in the number of housing vacancies, and have not only the highest absolute prices and rents, but rising at the fastest rate.
Data at the regional level begins to confirm many perceptions that individuals have about housing markets in specific metros, but is still incomplete. As I analyzed the data, however, I saw several possible categories that shed greater light on metro housing markets. Focusing exclusively on prices and rents above or below the average for the 51 largest metros, and the rate of change in total housing units above or below the average for the 51 largest metros as well, a clearer picture emerges on where metros rest on the national housing market spectrum. I developed four distinct categories:
xpensive Metros. These are metros with median home values above $266,000, median rents above $923, and adding housing units at a rate less than 7.4%. Infill development, whether in the core cities or in suburban areas, is the dominant method of adding housing units.
Expensive/Expanding Metros. These metros are just as high-priced as the ones in the first category, but generally are adding more housing units annually than above. This group exhibits a mix of infill and greenfield development as a method of adding housing units.
Expanding Metros. Metros in this category have median home values and rents below $266,000 and $923 respectively, and are adding new units at a rate greater than 7.4%. The metros in this group have been able to keep prices down by continuing to add units on the metro periphery.
Expended Metros. With median home values and rents below $266,000 and $923 respectively, and adding new units at a rate of less than 7.4% annually, the metros in this category could be considered as the slow-growth counterparts to the fast or moderate growth metros in the first three categories, resulting in fewer units added overall.
Here’s how they stack up in a table:
The expensive and expensive/expanding categories contain virtually all of the dense coastal cities whose urban rebirth has driven the national debate on housing, with a couple notable surprises. Sacramento, Riverside/San Bernardino and Salt Lake City, at to a lesser extent Baltimore, were surprise (to me) inclusions on this list. I wonder if geographical constraints might play some role in their relative lack of housing production.
The expanding metros are the largely Sun Belt metros of the South and West that are still wedded to the suburban growth model of growth at the edges, even if they exhibit some elements of successful infill development within their core cities. A possible exception here is New Orleans, which in 2008 (and since) was rapidly reclaiming housing units lost in the wake of Hurricane Katrina. Reclamation might be giving the Big Easy the profile of an expanding metro as it continues to recover. The expended metros, so named because they appear to have economies that lack the firepower that fuels super-charged growth and high prices, are a mix of Rust Belt, Midwest and non-superstar East Coast metros, Old South metros that never really saw explosive Sun Belt growth, and Sun Belt tourism draws that may be trying to find the appropriate housing balance in the wake of the Great Recession. Atlanta is an interesting fit in this category — perhaps the first Sun Belt metro to mature and enter a new, slower growth phase.
There’s one notable metro absence from any of the categories — Chicago. Incredibly, Chicago stands alone. Median home values dropped from an above-average $269,900 in 2008 to a below-average $229,900 in 2016, a drop of 14.8% and the largest such drop among the metros evaluated. The home value decline puts Chicago in the same category as hard-hit Sun Belt metros like Las Vegas, Orlando and Tampa/St. Petersburg. However, median rent values rose dramatically, going from $772 in 2008 to $1,050 in 2016, an increase of 36%. Although the absolute median rent value is not as high as some of the high-priced metros, Chicago’s rate of change is: only Denver, Portland, San Francisco, Seattle and San Jose saw their median rent values rise at a faster pace. And Chicago’s dearth of housing unit growth, just 1.1% between 2008 and 2016, puts it on par with Rust Belt brethren Detroit, Buffalo, Pittsburgh, Cleveland and St. Louis, which all had housing unit growth rates less than 1.5%.
There you have it. There are four (or five, if we count distinctive Chicago) metro area housing market types, with each presenting a varied set of challenges. How should metros the challenges by metro housing market type? I’ll address that in a follow up.