Working Paper: NBER ID: w22878
Authors: Daniel Green; Brian T. Melzer; Jonathan A. Parker; Arcenis Rojas
Abstract: We evaluate the Car Allowance Rebate System (CARS) by comparing the vehicle purchases and disposals of households with eligible “clunkers” to those of households with similar, but ineligible, vehicles. We find that CARS caused roughly 500,000 purchases during the program period and that the liquidity provided by CARS was critical for generating this large response. CARS provided less liquidity for households owning clunkers securing loans, since participation required loan repayment. The participation rate of these households was low, which we attribute to liquidity constraints and distinguish from the effects of other indebtedness, household income, and the size of the program subsidy.
Keywords: Cash for Clunkers; Household Liquidity; Aggregate Demand
JEL Codes: D14; E62; G18; H24; H31
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| CARS program (R48) | vehicle purchases (L62) |
| liquidity provided by CARS (E44) | household vehicle purchasing behavior (D10) |
| CARS program (R48) | average spending on new vehicle purchases (L62) |
| CARS program (R48) | likelihood of purchasing a vehicle (L62) |
| binding liquidity constraints (E51) | lower treatment effect (C22) |
| CARS program (R48) | aggregate vehicle purchases (E20) |
| CARS program fiscal cost (H56) | vehicle purchases (L62) |