When Mortgage Constraints Become Macroeconomic
The mortgage payment constraint does not stop at the household or city boundary. This chapter examines how the same contract-level friction changes aggregate demand in the paper’s modified New Keynesian model: what happens when a nominal-rate increase reduces credit-financed purchases, shifts the mix of spending, and affects output and inflation. The model supplies a disciplined extension of the mortgage mechanism, but its results remain conditional on a stylized economy rather than a direct forecast of the U.S. housing market.
A credit good turns the payment constraint into aggregate demand
A standard macroeconomic model can make the mortgage mechanism disappear by allowing households to borrow through a single real debt instrument. To keep the mechanism visible, the paper divides consumption into two goods. The cash good is paid for when it is purchased. The credit good is purchased today but paid for later. A mortgage-financed home is the natural real-world reference, even though the model’s credit good is deliberately more general than housing.
The model imposes a payment-to-income constraint on the credit good. In plain language, the household may commit only a specified share of current income to the payment due on credit-financed purchases. The payment includes the nominal interest rate. When the nominal rate rises, the amount of credit-good consumption compatible with the current-income ceiling falls. This is the same timing found in the mortgage example: the relevant question is not only how much the purchase costs in real terms over its life, but how large the current nominal obligation becomes when the contract is written.
The authors use a one-period nominal bond in the baseline model to keep the mathematics close to the familiar New Keynesian framework. That simplification does not remove the key feature. It allows the credit good to be constrained by a payment that rises with the nominal rate, so the model can isolate the extra channel created by a nominal debt obligation. The purpose is not to reproduce every underwriting rule, mortgage term, or housing-market institution. It is to ask what changes in a macro model once a credit-financed good is subject to a current payment boundary.
The usual intertemporal substitution channel remains. A higher real interest rate makes current consumption relatively expensive compared with future consumption, reducing demand in the standard way. But the payment constraint adds another route. Even if the real rate is held constant, a higher nominal rate raises the payment attached to the credit good. The household must either reduce the amount of that good, shift spending toward the cash good, or run into the constraint.
That shift in composition is important. Aggregate output is the sum of the cash and credit goods. If credit-good demand falls because the nominal payment becomes harder to carry, the household does not simply vanish from the economy; some spending moves toward the cash good. But the model’s goods are not perfect substitutes in utility. As demand shifts toward the cash good, its marginal utility changes. That affects the household’s labor-leisure choice and therefore the supply side of the economy.
The linearized demand relation in the paper makes the additional channel explicit.