economics Practical Magic

The most effective way to fix elderly care is to stop micromanaging caregivers and just pay them if the patient stays alive.

March 19, 2026

Original Paper

Survival Incentives in Dynamic Care Contracts: Optimal Payment Design for Thailand's Long-term Elderly Care System

Gu Luoluo

SSRN · 6302138

The Takeaway

This model of Thailand's care system reveals that 'survival incentives'—where pay is deferred until the next period—outperform direct monitoring of caregiver effort. Surprisingly, the research proves that perfect monitoring is actually less efficient for the state than allowing a small amount of oversight 'tolerance,' as it lets the survival incentive do the heavy lifting of motivation.

From the abstract

We develop a discrete-time dynamic principal-agent model for long-term elderly care in which the care recipient's health evolves stochastically across four states-functionally independent, home-bound, bedridden, and dead-and the care provider's effort is unobservable. The model features a logarithmic barrier cost function, two compliance tiers with switching costs, and ε-approximate incentive compatibility under limited liability. A central mechanism emerges from the absorbing nature of death: t