What if the housing shortage era is ending for some metros?

The Changing Landscape

“This is not a forecast. A forecast is a prediction, the validity of which my ego and I are professionally responsible for. What I offer here is speculation – something that is likely enough to write about but not so likely that my ego hangs in the balance.” – George Friedman

“When the facts change, I change my mind. What do you, sir?” – attributed to John Maynard Keynes

For over a decade I have operated under the belief that we will never build enough single-family housing in our growing metros. I came to this conclusion watching entitlements become ever more difficult, time-consuming, and costly – a pattern I first observed in California that has since spread to most markets, from different starting points but on the same trajectory.

I have seen nothing to change my mind about supply constraints. But I have begun to question the demand side of the equation.

For most of the post-war era, the homebuilding industry operated with the wind at its back. Population grew steadily. Women entering the workforce created two-income households and expanded the buyer pool for a generation. The Baby Boom produced sequential demand waves – entry-level, then move-up, then luxury – that kept rolling for three decades. Globalization held input costs in check. And a 40-year secular decline in interest rates, from 18 percent in 1981 to below 3 percent in 2021, compounded purchasing power at virtually every step.

Most of those tailwinds are now reversing.

Population growth is slowing sharply – natural increase turns negative around 2030, leaving immigration as the only demographic backstop, and immigration is now a policy variable rather than a reliable input. Women’s entry into the workforce – which created a generation of two-income households and expanded the buyer pool – has run its course as a tailwind; the labor force participation gap between men and women has narrowed from 46 points to 11, leaving little remaining lift from that source.

The generational demand wave is fading as Boomers age out. Supply chains are being rebuilt for resilience rather than lowest cost, and that resilience carries a price. Rates have normalized; sub-3 percent was a once-in-a-generation anomaly, not a floor to return to.

These are structural changes, not cyclical ones. And when combined with affordability that is already stretched, they point toward a potentially materially different operating environment.

The demand picture is weaker than the headline numbers suggest

Slower immigration means significantly reduced household formation projections – in time, down to levels where, for the first time in decades, it is at least conceivable that supply could eventually meet demand without heroic production assumptions. That alone would represent a regime shift.

But there is a second demand-side problem that receives less attention than it deserves. When we measure affordability, we almost always use median household income against median home price. That ratio captures the existing stock of households well. What it misses is the income profile of households forming at the margin – the incremental buyers that new construction actually depends on.

New household formation is increasingly concentrated in non-family households – singles, roommates, unmarried partners – whose median income is substantially lower than married-couple households. The overall median household income is sustained by decades of established, higher-earning married couples. The marginal formation income is meaningfully lower. Run a standard price-to-income affordability calculation using marginal formation income rather than overall median, and the market looks significantly worse than the headline number implies – and the gap is structural, growing a little each year as the household mix continues its long-running shift.

Geography: net domestic migration becomes the game

If the demographic picture is broadly correct and immigration is low, net domestic migration becomes the dominant demand driver for growth markets. International migration, when it was running at elevated levels, added net new households to the national count.

Domestic migration does not – it is a zero-sum redistribution. One market’s gain is another’s loss, and the national housing need is unchanged – although it can create the need for more housing, since houses can’t move with people.

International migration historically did two things for domestic migration: it added households directly, and it pressured prices in gateway markets over time, widening the price differential that makes other markets attractive to NDM (net domestic migration) movers.

Significantly reduced international migration may slow both mechanisms – meaning the price gap that drives NDM from gateway to other markets could compress rather than expand, reducing the advantage to moving.

The supply-shortage era may be ending

The industry has spent the better part of a decade focused on supply – and rightly so. The shortage was real and the urgency of solving it crowded out attention to anything else. But there is a risk that this frame is now working against clear thinking.

The timing of any transition is genuinely uncertain – existing pent-up demand, the slow play of demographic effects, and the unknown new baseline on international migration and household structures all resist precise forecasting. What we can say is probably not in the next few years, but certainly within a time frame that matters to long-term business plans.

In the intermediate term, land shortages in specific markets will persist, and the supply-constrained framing will remain correct. But the trajectory is toward a regime shift: from structural undersupply in most markets to a mix of undersupply, equilibrium and oversupply.

Builders and developers who are still calibrating their land pipelines, capital allocation, and product strategy to a shortage-era playbook will find themselves badly positioned for that transition in some locations.

The uncomfortable truth is that the supply problem we spent years trying to solve may be on its way to solving itself in many markets – through reduced demand rather than increased production.

What would it take for this analysis to be wrong?

Demographics are largely fixed at this point. Even a dramatic increase in birth rates would take decades to reach the housing market, and there is no evidence of developed countries reversing this kind of trend despite a variety of proposed solutions. A marriage boom would produce more dual-income households but fewer households on net. Net domestic migration is by definition a zero-sum game.

So, it comes down to international migration policy. Historically the country has moved between relatively open and relatively closed postures, with those cycles tending to be long-lived. But a policy reversal is possible, and its effects on household formation would be meaningful. That is a political projection everyone will have to make for themselves.

Does this mean everyone should be short land? No. There will be metros – large and small – that will grow through natural advantages, pro-business policies, and favorable price differentials to feeder markets, where demand is likely to outstrip supply for years to come.

But there will be fewer of them than in the past. Identifying markets with durable NDM will be more important than ever – especially for longer-term land plays where the consequences of getting it wrong compound over time.

What this means for builders and developers individually is a longer discussion. But the central conflict is already visible: smaller households with less purchasing power need smaller, cheaper product – while land use policy, through density restrictions, per-unit fees, and entitlement friction, systematically produces the opposite. The builders who find ways to resolve that conflict will be best positioned for what comes next.

It’s not easy to let go of an idea I’ve held for such a long time.

But the data says it’s time to reconsider.