John Rust

Overview

John Rust is an economist known for foundational contributions to the structural estimation of dynamic discrete-choice models. His 1987 Econometrica paper on Harold Zurcher's bus engine replacement (the "Rust bus problem") is the canonical example of dynamic programming estimation. His work on Social Security and Medicare (Rust and Phelan 1997, Econometrica) is the basis on which Benítez-Silva et al. (1999) extends DI application modeling. He was at Yale during the 1999 DI paper and is now at Georgetown University.

Key Contributions

• Rust and Phelan (1997) — Structural dynamic programming model of retirement behavior and Social Security/Medicare claiming, showing how institutional program features shape retirement timing under incomplete markets. Foundational reference for the B-S et al. DI game-tree model.
• Benítez-Silva et al. (1999) — Corresponding author; conceptualized the DI application, appeal, and award process as a multi-stage dynamic game; contributed the marginal probability model derivation; established the HLIMPW sufficient-statistic result. See SSA Sequential Determination Process.
• Benítez-Silva et al. (2004) — Co-authored the formal test of the RUR hypothesis; GHK simulation estimator for the bivariate probit draws on Rust's simulation methods. See Rational Unbiased Reporting.
• Rust (1997) — "Using randomization to break the curse of dimensionality" (Econometrica 65(3): 487–516); developed randomized algorithms that reduce computational complexity of DP models from exponential to polynomial.

Related

• Hugo Benítez-Silva
• Rational Unbiased Reporting
• SSA Sequential Determination Process
• DI Application Costs and Take-Up

Sources

• Benítez-Silva et al 1999 — An Empirical Analysis of the Social Security Disability Application, Appeal, and Award Process
• Benítez-Silva et al 2004 — How Large Is the Bias in Self-Reported Disability