The price of pristine credit

America's housing market is heading for its worst year in three decades. Existing-home sales fell again in August, to an annualised 3.98m, and analysts at Capital Economics project the full year will average about four million — the weakest annual showing since 1995. The usual suspects get the blame: mortgage rates near 7%, prices that never fell, sellers locked into 3% loans. A study published this year by the Pew Charitable Trusts argues that a quieter culprit deserves an equal share — the lending system itself, which now demands that homebuyers present a credit history and has drifted so far from its post-crash purpose that it routinely turns away borrowers who would almost certainly repay.

A score becomes a wall

The churn in America's mortgage market, Pew says, is now organised around a single number. The average new borrower in 2024 had a credit score of 742, 29 points higher than the typical American consumer. Between 2005 and 2024 the share of loans to borrowers with scores of 600–699 fell by 13.3 percentage points, to 22.3%. Denial is the default below 620 — two-thirds of those applicants are turned away — and remains stiff even for the respectable: one borrower in five with a score of 680–719 is rejected. The contrast with the anarchy of the boom is stark. Before the crash, loans requiring little or no income documentation made up more than a third of the market; today that share is close to zero. A borrower is now judged, in effect, by a number that rewards age, income and accumulated wealth — the very attributes that the youngest and least credit-rich buyers do not yet possess.

The risk that no longer exists

The strictness was, on its own logic, a triumph. After the subprime disaster, the Federal Reserve and the government-backed mortgage firms Fannie Mae and Freddie Mac rebuilt underwriting around the fear of another collapse. The payoff is delinquency at a quarter-century low and default rates so small they verge on the theoretical. This is where the analysis bites: the borrowers being frozen out are, by the market's own recent record, essentially safe. Those with scores of 620–700 defaulted at a rate of 10.1% a generation ago; on the loans Fannie and Freddie actually purchased in 2020–2024, the figure was 0.1%. The catastrophe that justified the crackdown has, for precisely this cohort, practically vanished.

To be sure, prudency deserves its defence. The 30-year fixed-rate mortgage is a vast, implicit government subsidy, and a single default scandal can erase a decade of good behaviour; the system's overseers have every reason to ration it carefully. The trouble is that the incentives are one-sided. A bad loan is a scandal with a named culprit; a borrower quietly rejected is an abstraction. Of the purchase mortgages made in recent years, about nine in ten were shaped by federal rules or the automated underwriting of Fannie, Freddie and the federal agencies. Lending standards are thus a policy choice rather than a market outcome — and the politics point only one way. No regulator is applauded for widening access, while many a lending official has been sacked over the one default that follows.

No one gains from letting go

That one-sidedness has casualties with names. The excluded skew young, rural, minority and self-employed — precisely the groups that federal housing policy claims, in its founding language, to serve. The consequence is not an abstraction that can be shrugged off. Sales are squeezed from both sides at once: sellers cling to cheap loans they will not surrender, while buyers meet a credit-score wall that no interest-rate cut can move. Rates, for what it is worth, are climbing rather than falling — the benchmark 30-year fixed touched 6.76% in September, its highest in over a year, with some forecasts for something above 7%.

For an investor this reframes the housing cycle. The conventional read — that a turn by the Federal Reserve will unlock pent-up demand — describes only the supply side of the standoff. Pew's evidence suggests demand is separately capped by lending standards that no one in Washington is rewarded for relaxing. Housing-adjacent equities are therefore priced for a long, quiet freeze, and that is a defensible bet for structural reasons that have little to do with the central bank. The variable that would genuinely change the equation — a loosening of access — is the one with no political patron.

America has bought safety with access. It is a coherent bargain, but its ledger has become lopsided: the marginal borrower, excluded by points on a score, would almost certainly not have defaulted. A market that cannot turn over to its next generation congeals. The price of pristine credit is paid in the quiet permanence of a frozen housing market, and the young pay it first.