Working Paper: NBER ID: w23497
Authors: James J. Heckman; Rodrigo Pinto
Abstract: This paper defines and analyzes a new monotonicity condition for the identification of counterfactuals and treatment effects in unordered discrete choice models with multiple treatments, heterogenous agents and discrete-valued instruments. Unordered monotonicity implies and is implied by additive separability of choice of treatment equations in terms of observed and unobserved variables. These results follow from properties of binary matrices developed in this paper. We investigate conditions under which unordered monotonicity arises as a consequence of choice behavior. We characterize IV estimators of counterfactuals as solutions to discrete mixture problems.
Keywords: unordered monotonicity; instrumental variables; causal identification; discrete choice models; treatment effects
JEL Codes: C16; C93; I21; J15
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| unordered monotonicity (C69) | identification of causal parameters (C20) |
| instrument satisfies conditions (rank and exogeneity) (C36) | causal effects can be identified (C22) |
| additive separability of choice equations (D10) | establishment of causal links (C22) |
| binary matrices characterize choice sets (C25) | necessary and sufficient conditions for identifying counterfactual outcomes (C62) |
| unordered monotonicity (C69) | identifiable treatment effects (C22) |
| identification of causal effects (C22) | accurate policy analysis (D78) |