Working Paper: NBER ID: w29045
Authors: Morris Davis; Jesse M. Gregory
Abstract: In recent location choice models, households randomly vary with respect to their utility of living in a location. We demonstrate the distribution generating this randomness is not identifiable from location choice data and the optimal allocation chosen by a social planner is not identified. We propose an algorithm for setting the distribution generating the random utility across locations that implies a planner will optimally choose no redistribution in the absence of externalities or equity motives between different types of people. Our algorithm preserves a planner's motives to redistribute due to equity considerations between types of people and efficiency in production.
Keywords: location choice; redistribution; equity; efficiency
JEL Codes: H0; R38
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| social planner motives for redistribution (D63) | across-type equity (G12) |
| social planner motives for redistribution (D63) | efficiency (D61) |
| social planner motives for redistribution (D63) | within-type transfers (F16) |
| across-type equity (G12) | reduce consumption inequality among different household types (D12) |
| efficiency (D61) | incentivizes households to internalize external impacts (H23) |
| within-type transfers (F16) | equalize average marginal utility of consumption (D11) |
| location attachment draws (Y91) | complicates identification of optimal within-type transfers (F16) |
| across-type equity or efficiency motives (D63) | predicted redistribution (D39) |
| adjustment to planning problem (P21) | eliminates within-type transfers (F16) |
| adjustment to planning problem (P21) | preserves motives for redistribution due to equity or efficiency (D63) |