Working Paper: NBER ID: w7366
Authors: Emmanuel Saez
Abstract: This paper investigates whether taxpayers bunch at the kink points of the US income tax schedule (i.e. where marginal rates jump) using tax returns data. Clear evidence of bunching is found only at the first kink point (where marginal rates jump from 0 to 15%). Evidence for other kink points is weak or null. Evidence of bunching is stronger for itemizers than for non-itemizers. Theoretical models of behavioral responses to taxation show that bunching is proportional to the compensated elasticity of income with respect to tax rates. These models are used to perform simulations of bunching and calibrate the key parameters (the behavioral elasticity and the extent to which taxpayers control their income) to the empirical income distributions. Except for low income earners, the behavioral elasticity consistent with the empirical results is small.
Keywords: taxpayer behavior; income tax; kink points; bunching; elasticity
JEL Codes: H31
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| tax rate changes (H29) | taxpayer behavior (H26) |
| first kink point (D43) | taxpayer behavior (H26) |
| absence of bunching at second kink point (D10) | taxpayer behavior (H26) |
| compensated elasticity of income with respect to marginal taxes (H31) | size of bunching (D20) |
| ability of taxpayers to control their income (H26) | taxpayer behavior (H26) |