Working Paper: CEPR ID: DP8925
Authors: Robert Rowthorn; Flavio Toxvaerd
Abstract: This paper fully characterizes the optimal control of a recurrent infectious disease through the use of (non-vaccine) prevention and treatment. The dynamic system may admit multiple steady states and the optimal policy may be path dependent. We find that an optimal path cannot end at a point with maximal prevention; it is necessarily zero or at an intermediate level. In contrast, an optimal path must end at a point at which treatment is either maximal or minimal. We find that the comparative statics of the model may radically differ across steady states, which has important policy implications. Last, we consider the model with decentralized decision making and compare the equilibrium outcomes with the socially optimal outcomes. We find that steady state prevalence levels in decentralized equilibrium must be equal to or higher than the socially optimal levels. While steady state treatment levels under decentralization are typically socially optimal, steady state prevention (if used) is socially suboptimal.
Keywords: Economic Epidemiology; Hysteresis; Nonconvex Systems; Optimal Policy; Equilibrium Policy; Mix of Treatment and Prevention
JEL Codes: C73; I18
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| level of prevention (I12) | optimal endpoint of the control strategy (L21) |
| level of treatment (I18) | optimal endpoint of the control strategy (L21) |
| decentralized decision-making (D70) | suboptimal prevention levels (I12) |
| suboptimal prevention levels (I12) | disease prevalence (I12) |
| steady state prevalence levels in decentralized equilibrium (C62) | socially optimal levels (H49) |
| treatment levels under decentralization (H77) | socially optimal levels (H49) |