Why One National Rate Hurts Some Markets More

The mortgage-affordability mechanism becomes a sharper test when the same national rate is observed across cities with different loan sizes and incomes. This chapter examines whether nominal-rate increases are associated with a larger reduction in new mortgage issuance in high-loan-to-income markets, and how the paper uses that comparison to distinguish a payment-constraint channel from a simple national correlation. The result is a relative effect, not a claim that every market loses the same number of loans.

Why compare cities instead of rates alone

A national time series creates an immediate identification problem. Nominal and real mortgage rates often move together. A monetary-policy shock can change both rates, while also changing employment, house prices, expectations, and the desire to buy a home. If mortgage originations fall when the nominal rate rises, the decline could reflect the higher nominal payment, a higher real borrowing cost, a weaker economy, or several forces at once. The rate movement itself does not tell us which explanation is doing the work.

The paper takes a different route. The mortgage rate is broadly common across U.S. cities at a point in time, but its consequences need not be. If the nominal rate matters because it raises a payment that is tested against current income, the effect should be larger in places where loans are large relative to incomes. Those places should have more borrowers close to a debt-to-income boundary. The test is therefore not simply whether originations move when the rate moves. It is whether they move differently across cities in the direction predicted by payment constraints, after accounting for the real rate.

The paper summarizes local exposure with loan-to-income, or LTI. LTI is the loan amount divided by annual income. Debt-to-income, or DTI, is the ratio of scheduled debt payments to income and more directly represents the underwriting constraint. LTI is used as a proxy because a larger loan relative to income generally implies a larger payment relative to income when term and rate are comparable. It is a way to locate markets where the constraint is more likely to matter, not a statement that every loan in a high-LTI city has a high DTI.

The Houston and Los Angeles examples show why this comparison is plausible. In the paper’s illustration, a Houston household borrows about $500,000 on an income of about $125,000, an LTI of four. At a 3 percent mortgage rate, the annual payment is roughly $25,000, or 20 percent of income. At 7 percent, it rises to about $40,000, or 32 percent. That is a major increase, but it remains below the 45–50 percent lender cutoff used in the example before other debts are counted.

The Los Angeles example starts closer to the boundary. A household earning $200,000 takes out about $1.2 million, an LTI of six. At 3 percent, the annual payment is about $61,000, or 31 percent of income. A rate above roughly 6.4 percent pushes the payment ratio above 45 percent.