The Mortgage Payment Is a Contractual Schedule
Residential mortgage affordability depends on more than the real cost of borrowing. This chapter focuses on the payment schedule created by a fixed-rate, multi-period mortgage: why a higher nominal rate can raise the payment tested at origination even when the real rate is unchanged. Understanding that schedule lets readers tell when the nominal rate is the active borrowing constraint and when other factors, including the real rate, still determine the purchase.
A fixed mortgage has two clocks
A nominal interest rate is the percentage written into the loan contract. It is applied to dollars: the dollars the borrower receives, and the dollars the borrower must return through scheduled payments. A real interest rate adjusts that nominal rate for expected inflation. As a rough approximation, if the nominal rate is 7 percent and expected inflation is 4 percent, the expected real rate is 3 percent.
The real rate answers a familiar economic question: how much purchasing power is transferred from the future to the present? If the real rate rises, borrowing usually becomes less attractive because the borrower must give up more goods and services later for a given amount of spending today. This is why standard macroeconomic models often make the real rate the central price of intertemporal choice.
A fixed-rate mortgage introduces a second clock. It is not a one-period loan settled by one real payment in a distant future period. It is a long contract with a sequence of payments, commonly 360 monthly payments over 30 years. The payment is fixed in nominal dollars when the rate is fixed. The borrower therefore faces a stream of dollar obligations, while the lender and underwriting process must decide whether the early obligations are manageable against income available now.
That creates a distinction between the payment’s real profile and the payment’s contractual amount. Suppose the nominal rate and expected inflation both rise by the same amount. The real rate can remain unchanged, but a mortgage written at the higher nominal rate still requires a larger nominal payment for a given principal. In real terms, the fixed payment may be worth more early in the loan and less later as inflation accumulates. The later decline in purchasing-power terms does not change the amount of dollars due at the beginning.
This timing difference is easy to overlook because the same mortgage can look less burdensome in the future. A payment of a fixed number of dollars may consume a smaller share of a household’s future income if wages rise with prices. But a lender evaluating a new application is not free to replace the first payment with its future, inflation-adjusted equivalent. The current payment is compared with current income and with other current obligations. A long-run real calculation and an origination-time qualification test can therefore point to different constraints.
The Federal Reserve Bank of San Francisco working paper Real Effects of Nominal Interest Rates builds its argument around that institutional fact. The authors identify three features that are common to fixed-rate mortgages: the debt lasts for multiple periods, the nominal payment is constant across those periods, and lenders resist loans whose payment is too large relative to current income. The last feature can arise from formal underwriting rules, regulation, asymmetric information, or a borrower’s own reluctance to commit too much income to debt service.
A mortgage payment is not simply the interest rate multiplied by the balance. It is the scheduled amount needed to service a balance over a specified term. At the beginning of an amortizing loan, the payment includes interest on a large outstanding principal and a smaller amount of principal repayment; as the balance falls, the composition changes. For a given term and principal, changing the nominal rate changes the payment stream. For a given payment budget, it changes the principal that the stream can support. That is the financial link between the quoted rate and the borrowing limit.
The payment-to-income constraint need not be a single universal rule printed on every loan. The paper describes both explicit lender limits and implicit limits created by a borrower’s other commitments or risk aversion. A lender may reject a payment that leaves too little room for other debts and ordinary income shocks; a household may make the same judgment even if a formal rule would allow more. Either way, the constraint is assessed in current dollars against current income. The mechanism is therefore about the interaction of a contract and a financing boundary, not about the interest rate in isolation.
Expected inflation belongs on the real-rate side of the comparison, but it does not rewrite the nominal amount printed on a fixed-rate payment schedule. If expected inflation rises while the contract’s nominal rate does not, the expected real burden changes. If the nominal rate rises alongside expected inflation, the real rate may stay constant while the nominal payment for a given principal rises. Keeping those two statements separate is necessary: one describes purchasing power across time, and the other describes whether a borrower can meet a dollar obligation at origination.
The resulting mechanism does not require anyone to misunderstand inflation. A borrower may know exactly that a 10 percent nominal rate combined with 10 percent expected inflation leaves the real rate near zero. A lender may know it as well. Nevertheless, the borrower may be unable to obtain the same principal because the first nominal payment is higher. The effect is real in the practical sense that the household’s feasible purchase changes: a house that could be financed under one contract cannot be financed under another, even though the real rate comparison is held constant.
The mechanism can be stated as a sequence:
- The quoted nominal rate helps determine the payment schedule for a given principal.
- A higher nominal rate raises the payment required to carry that principal.
- A payment-to-income constraint limits the share of current income available for the mortgage and other debts.
- If that constraint binds, the maximum principal falls.
- A lower maximum principal reduces the price of the home the household can finance, unless the household changes its down payment, income, debts, or other choices.
The conditional phrase “if that constraint binds” is essential. A rate is not automatically the active constraint for every borrower. If a household has substantial room below the payment ceiling, a higher nominal rate may make the loan more expensive without preventing the same purchase. If the household’s desired purchase is already limited by a real-rate calculation, the payment constraint may not determine the outcome. The paper’s claim is that actual mortgage contracts create an additional nominal channel, not that the channel overrides every other determinant of housing demand.
The same budget buys less principal at a higher rate
The paper’s numerical illustration shows how sharply the constraint can operate. Imagine a household earning $100,000 per year. For the example, assume that the lender will allow no more than 50 percent of current annual income to go to the first-year mortgage payment, after considering other household debt payments and property taxes. The allowed mortgage payment is therefore $50,000 per year. The 50 percent ceiling is a simplifying assumption, not a claim about a universal underwriting rule.
With a zero nominal rate and zero expected inflation, a 30-year mortgage with a $50,000 annual payment supports roughly $1.5 million in principal. The arithmetic is intuitive in this limiting case: with no interest, 30 annual payments of $50,000 repay $1.5 million. Now keep the real rate at zero but set expected inflation and the nominal mortgage rate to 10 percent. The same $50,000 first-year payment supports only about $470,000 of principal in the paper’s example.
The contrast is not produced by changing the household’s income or the payment ceiling. It is produced by changing the nominal rate that prices the stream of future payments. At the higher rate, part of each payment is needed to cover interest on the outstanding balance, leaving less of the fixed payment to retire principal. A given payment budget can therefore carry a smaller loan. Over a 30-year term, this is the present-value consequence of exchanging the same current payment for a stream of 360 contractual payments at a different nominal rate.
A rough first-year check makes the scale intuitive. At a 10 percent annual rate, interest on a $1.5 million balance would be about $150,000 before any principal repayment, already far above the example’s $50,000 annual ceiling. At a principal near $470,000, first-year interest at the same rough rate is about $47,000, leaving only a small amount of the capped payment for principal. The actual mortgage calculation is monthly and amortizing, so these figures are intuition rather than a replacement for the paper’s calculation. They show why a higher nominal rate can compress the supported principal so sharply.
The lender’s concern is not merely that a payment looks large in the abstract. A payment that consumes most of current income leaves less room for other debts, ordinary income variation, and unexpected expenses, increasing the risk that the contract will not be serviced. The paper allows for this constraint to be formal, regulatory, or implicit in household behavior. That flexibility matters: the mechanism does not depend on one exact cutoff, but it does depend on some current-payment boundary being relevant to the borrowing decision.
The example also separates two ideas that are often collapsed into the word “affordable.” One meaning is whether the household can pass a current payment test and obtain a specified principal. The other is whether the household considers the total real cost acceptable over the life of the loan. The San Francisco Fed example is designed to isolate the first meaning. It shows that the quoted nominal rate can change the amount of housing financeable under a payment-to-income constraint, even when a real-rate comparison alone would not predict a change.
The comparison is a controlled thought experiment, not a claim that nominal rates historically move without real-rate movements. In an actual rate cycle, a higher nominal rate may also reflect changing inflation expectations, monetary policy, growth prospects, or risk. The paper’s conceptual question is narrower and more useful: after holding the real rate fixed in the comparison, is there still a pathway from the nominal rate to the feasible loan? If the answer were no, then the nominal rate would matter only as a proxy for the real rate. If the answer is yes, the payment schedule supplies an additional channel.
This channel changes the household’s feasible set before it changes the household’s preferences. A borrower may still want the same home and may understand that the future real cost is not higher, yet be unable to choose that home under the new contract because the first payment fails the current-income test. The available responses—borrowing less, bringing more cash to the purchase, reducing other obligations, or postponing the transaction—are changes in financing possibilities. They are not evidence that the borrower suddenly values housing less. At the market level, many such constrained choices can reduce new mortgage issuance even if potential buyers’ underlying desire for a home has not changed by the same amount.
The real payment profile helps explain, rather than contradict, the result. In the zero-rate, zero-inflation case, the annual real payment is flat at $50,000. In the 10 percent nominal-rate, 10 percent-inflation case, the nominal payment is still fixed, but its value in the household’s initial purchasing-power units declines over time. The early real payment is relatively high and later payments are relatively low. The payment-to-income rule bites at the start, so later relief cannot make the original application pass retroactively.
This is also why a nominal-rate effect should be concentrated among marginal borrowers. Consider two households facing the same national mortgage rate. The first proposes a modest loan relative to income and remains well below the lender’s ceiling after the rate rises. The second proposes a much larger loan and is already close to the ceiling. The first may absorb the higher payment or make a small adjustment. The second may have to borrow less, choose a less expensive home, increase its down payment, reduce other debt, or postpone the purchase. The rate movement is common; the constraint exposure is not.
The direct timing of the mechanism matters as well. A household applying for a new fixed-rate mortgage receives a payment schedule based on the rate available when the loan is originated. A household that already has an unchanged fixed-rate mortgage is not described by this mechanism as receiving an immediate payment reset merely because market mortgage rates rise. Existing borrowers can still be affected indirectly through employment, house prices, refinancing opportunities, or other debts, but those are different channels. The supplied argument is strongest at origination and at points where a contract is reset or replaced.
The stylized calculation leaves out many features that matter in actual underwriting: taxes and insurance beyond the example’s treatment, loan limits, down payments, credit risk, fees, changing income, adjustable rates, and the borrower’s other obligations. Those omissions do not undermine the comparison, because the paper is not presenting $470,000 as a typical borrowing limit. They do set the boundary of what the numbers establish. The supported lesson is comparative and conditional: holding the real rate and real wage fixed, a higher nominal rate can reduce the principal supported by a fixed current-payment ceiling.
There is a corresponding boundary on the role of the real rate. Suppose a household would voluntarily choose a smaller home because the real rate has risen, even before the payment-to-income limit is reached. In that case, the real rate can determine the purchase directly. Or suppose the household has enough income and a small enough loan that the payment test never binds. Then a nominal-rate increase may change the loan’s cost without changing the approved principal. The additional nominal effect becomes most visible when the payment constraint is the bottleneck.
The contract therefore supplies the missing link between a quoted nominal rate and a real-world purchase decision. “Nominal rates matter on their own” does not mean they move independently of inflation or the economy. It means that, conditional on the real rate, the payment schedule can create an additional borrowing limit. The next question is who sits close enough to that limit for the same rate change to have a visibly different effect.