Working Paper: NBER ID: w21942
Authors: Alejandro Justiniano; Giorgio E. Primiceri; Andrea Tambalotti
Abstract: The surge in credit and house prices that preceded the Great Recession was particularly pronounced in ZIP codes with a higher fraction of subprime borrowers (Mian and Sufi, 2009). We present a simple model with prime and subprime borrowers distributed across geographic locations, which can reproduce this stylized fact as a result of an expansion in the supply of credit. Due to their low income, subprime households are constrained in their ability to meet interest payments and hence sustain debt. As a result, when the supply of credit increases and interest rates fall, they take on disproportionately more debt than their prime counterparts, who are not subject to that constraint.
Keywords: subprime borrowers; credit growth; housing market; Great Recession
JEL Codes: E21; E44; G21
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Expansion in credit supply (E51) | Increase in mortgage debt for subprime borrowers (G21) |
| Expansion in credit supply (E51) | Increase in house prices for subprime borrowers (G59) |
| Higher fraction of subprime borrowers (G21) | More pronounced surge in credit and house prices (E39) |
| Relaxation of lending constraints (G21) | Reduction in interest rates (E43) |
| Reduction in interest rates (E43) | Increase in borrowing capacity of subprime households (G51) |
| Increase in borrowing capacity of subprime households (G51) | Increase in mortgage debt (G21) |
| Increase in borrowing capacity of subprime households (G51) | Increase in house prices (R31) |
| Cumulative credit growth (E51) | Share of subprime borrowers (G51) |