Optimal Lockdown Policy with Virus Mutation

Abstract

We examine the implications of virus mutation for optimal lockdown policy in an epimacro model. We consider three ways of modelling virus mutation—one deterministic setup and two stochastic setups featuring a two-state and three-state Markov process. We find that the effects of virus mutation are asymmetric. In particular, a future reduction in the transmission rate increases lockdown intensity by more than a future rise in the transmission rate lowers it. As a corollary to this asymmetry, an increase in uncertainty about future mutation is non-neutral and reduces lockdown intensity under the optimal policy.