While the housing challenge is national, it is important to note it’s not uniform. It is particularly severe in expensive, supply-constrained coastal markets, where the relationship between home prices and household incomes has become increasingly stretched.
In cities like Boston, New York, San Francisco and Los Angeles, the median sale price of a home is more than six times the median family income, and that ratio climbs to almost 10 times in some markets. In major markets with less density and fewer constraints, such as Dallas, Cleveland and Pittsburgh, the ratio actually falls to around three times median family income, below the national average.3
The divide between households is also widening, regardless of region. Homeowners who bought before 2023 and secured low fixed-rate mortgage rates have largely been insulated from the increase in financing costs. At the same time, prospective buyers are facing elevated prices, higher mortgage rates, larger required down payments and rising expenses for insurance, taxes and maintenance.
Younger Americans are feeling the effects most acutely. Nationally, 65% of families owned their home as of the second quarter of 2026, but only 35% of Americans under 35 did, and the trend for the younger cohort is downward as rising costs force them to save up for longer before buying. The median age of first-time homebuyers hit 40 years in 2025, a record high and up from 33 in 2021.4
The root of the problem
The housing affordability issue is inextricably tied to the broader economic landscape. Mortgage rates, which dipped below 6% early in the year, quickly reversed course and have risen steadily since the Iran conflict began, as inflation fears drove yields up across the board. They now stand above 7%, near multi-decade highs. To put this in perspective, an equilibrium mortgage rate implied by J.P. Morgan Asset Management’s Long-Term Capital Market Assumptions is far lower, near 5.8%.5
Higher mortgage rates increase monthly payments for new buyers and reduce purchasing power. They also discourage existing homeowners from moving: As we’ve noted, many owners have locked in much lower rates and would need to surrender them to purchase a new home.6 This “lock-in effect” limits the number of existing homes reaching the market, further constraining inventory. Illustrating the problem, pending home sales fell to their lowest level ever in September.7
Mortgage spreads, or the gap between Treasury yields and the mortgage rates being offered, have also widened as interest rate volatility soared. Additionally, the difference in mortgage loan providers has pushed spreads wider. In 2008, around 60% of mortgages originated with banks. However, many banks left the mortgage market after new regulations and capital requirements were imposed in the wake of the GFC. In today’s market, nonbanks predominate, and only 35% of mortgages originate with banks.8
This has reduced the number and types of institutions competing for mortgages. Although this may be adding tens of basis points to mortgage rates rather than full percentage points, even a modest increase compounds over the life of a mortgage.
Artificial intelligence plays a role
The ripple effects from the artificial intelligence (AI) boom are making housing less affordable. With the economy near full employment, capital is limited. This can cause surges in investment to “crowd out” other areas of the economy by pushing rates higher, suppressing other investment.
Crowding out is usually discussed in relation to government borrowing, but a private investment surge can produce a similar effect—and that’s exactly what’s happening. As the below chart shows, artificial intelligence spending has remained relatively resistant to higher rates, while more rate-sensitive parts of the economy, including housing, have felt the pressure.