Asia stocks climb tracking Wall St rally; Nikkei hits record high, China GDP beats
In the mid-20th century, in a typical year, the US added a bit more than 1.5 million new homes. About 200,000 were new vacant units, the number of which scaled with the size of the occupied housing stock. Of the others, about 1 million were occupied by homeowners and about 500,000 were new renter households. And, generally, for each new owner household, there was a new single-family home and for each new renter, there was a new apartment.
As the national willingness to permit apartments was pushed down to about 300,000 annually, some of those half-million new renter households, on net, started to move into older single-family homes. By the turn of the century, the number of renter households had stopped growing. Since then, the growth of owner and renter households has moved in opposition to each other because the housing shortage means that more of one requires less of the other.
The one-time deep change in mortgage access in 2008 temporarily put owner families on their feet, but eventually the shock of that receded, and owner families returned to the driver’s seat in housing consumption.
For most of the past decade, we’ve been back to the condition where all the new housing was claimed by owners, and renter household formation was limited by the housing shortage. And, rather than adding 200,000 new vacant units, we have been harvesting more than 300,000 annually, just to keep the number of renter households from declining further.
It’s really a testament to the human ability to tell stories that there are still skeptics of the supply shortage issue. People can believe anything they set their sights on.
At this point, we have harvested all the vacant units we can functionally harvest. But the trend in household formation marginally changed in the past couple of years. New homes have topped 1.6 million the last couple of years, and household formation has returned to a somewhat balanced ratio, about 1 million owners and 700,000 renters.
Harvard’s Joint Center for Housing Studies forecasts (under their “low immigration” forecast) about 600,000 new households annually for the next decade. I’m going to assume that that includes little or no catch-up growth since their estimate of a supply shortage is much smaller than mine.
Immigration had been high in recent years. With Trump in the White House, much lower growth is likely. But, in any case, between vacancies and additional households waiting for housing so they can form, there will be 15 million or more in addition to the 600,000 annual growth from population and demographic trends.
Maybe that’s where we are now. Currently, market rent inflation is running about 1% below non-shelter core CPI inflation. That implies that production is about 750,000 above neutral household formation trends. Maybe we are already working off the 15 million-plus shortage.
If that is the case, and 600,000 new households is the new normal, then at something like 2.3 million units annually, rents nationally would be flat while core non-shelter inflation would run at about 2%. Or, since the Fed includes rent in their target inflation, maybe non-shelter core inflation will run closer to 3% while rents rise at 1%, and we spend a decade facilitating an extra 1.5 million households annually.
I have been expecting the current trend to be a bit of a head fake, and for rent inflation to have one last bump up before new supply starts to permanently put us in this condition, but I might be wrong.
In any case, let’s accept this as the scenario going forward, with 600,000 annual household growth plus 1.5 million new households from pent-up demand, and rent inflation 2% below non-shelter inflation. (If that is the case, and we are already a year or so into that condition, the timing for this garbage report based on an ahistorical, made-up claim that builders don’t increase new home construction when rents are moderating couldn’t be more delicious.)
That would amount to about 400,000 new owner households annually, on net, and more than 1.5 million renter households. New apartments briefly rose to about 600,000 in the last couple of years, but they are settling back down at the 400,000 range. YIMBY reforms might push that up a bit, but we are a long way from seeing 2 million new apartments a year.
In the late 20th century, the lack of adequate apartment supply meant that the supply was met through existing homes transitioning into rentals. Of course, there is a backlash against that now, with policymakers like Elizabeth Warren pushing to obstruct that source of rental housing. That just goes to show that sincerely observing the world and having an opinion about it can be much worse than just being ignorant and indifferent. Or, maybe she isn’t sincere. Maybe it’s soulless demagoguing.
In any case, I think the state we will find ourselves in as new home construction continues to increase is that there aren’t enough transactions within the existing stock of homes necessary to transition 1.5 million owner families into new homes and existing homes into the rental market. Some owner families will prefer existing homes or high amenity infill locations over new homes in the exurbs.
And, so, I think there will be practically limitless potential for build-to-rent single-family homes in the coming years… that is if enemies of renter families like Senator Warren allow them to remain legal. At the same time, there may be an unprecedented transition of existing homes into the rental market, also.
If we don’t screw it up, we should be entering a phase where home prices continue to be flat and an increasing number of both exurban single-family and infill housing will be available to buy. The “investors are pricing families out” story will be increasingly untenable. Though, I suppose, it is already untenable and that doesn’t stop it from being popular and dangerous.
When new construction reaches 2 million units+ annually, the official homeownership rate will start to drop by something like 0.5% a year. That will probably lead to more freakouts. And, if you suggest that we make the one change that will fix that - returning to 20th century lending norms - you’ll get “Oh! This joker wants to bring back NINJA loans!”
It’s really a fascinating situation. The private market will fix these problems, to the extent that they can be fixed, so quickly that there will be panic about it. And we just have to hope that state capacity reflects enough incompetence to let it happen. As far as the homebuilders and developers are concerned, they will either make returns via quantity or land inflation. Let’s hope, for our sake, we let them earn returns via quantity.