When a Reputation Changes the Future
This chapter follows a credit assessment after it has been formed. The specific question is how a lender’s belief changes the price of borrowing and how that price changes saving, debt, default, and the distribution of consumption and wealth. The distinction matters because tracking a borrower can improve repayment incentives and credit access while also making the consequences of different histories more unequal.
Two forces inside a changing loan price
A lender’s price for unsecured credit reflects the expected probability of repayment. The ordinary force behind that price is limited commitment: a borrower may choose not to repay, so a larger obligation can be riskier than a smaller one. Even if the lender knew the borrower’s type, the possibility of default would make the price depend on the amount borrowed and the borrower’s state.
Hidden information adds a second force. The amount a person saves or borrows can change the lender’s assessment of the mix of high- and low-patience types asking for credit. If high-patience types tend to save more and low-patience types tend to borrow more, a request for a larger loan can be informative about who is seeking it. The lender then prices not only the risk of a larger obligation conditional on type, but also the possibility that the action has revealed a riskier composition of borrowers.
These two forces can point in the same direction without being the same mechanism. The first says that more debt is harder to repay. The second says that the choice of debt can change what the lender believes about the borrower. A change in the interest rate can therefore contain both a leverage effect and a reputation effect. Separating them is important when asking whether a score merely describes risk or changes behavior by changing the terms attached to it.
Imagine two borrowers who arrive with the same current type assessment. One asks for a small loan and the other asks for a larger one. Even if the lender learned nothing new about their underlying types, the larger obligation could be harder to repay. But if the size of the request is also associated with the behavior of the less-patient type, the larger request changes the lender’s assessment as well. The second borrower may face a worse price for two reasons: the loan itself is riskier, and the request has changed the expected composition of the borrower pool. In the opposite direction, saving can reduce exposure to default while also signaling a type that the lender expects to repay more reliably.
The borrower also cares about the belief itself. In the paper’s two-period model, reputation has value because a higher assessment can preserve access to other markets or future opportunities. Saving may then have a return beyond the resources it carries into the future: it can signal patience and improve the next assessment. Borrowing may provide more consumption today but weaken the reputation that supports better terms later.