Covid and social distancing with a heterogenous population

Abstract Motivated by the Covid-19 epidemic, we build a SIR model with private decisions on social distancing and population heterogeneity in terms of infection-induced fatality rates, calibrate it to UK data understand quantitative importance these assumptions. Compared our model, calibrated benchmark version constant mean contact rate significantly over-predicts rate, death toll, herd immunity prevalence peak. Instead, counterfactual endogenous but no massively under-predicts statistics. We use how impact mitigating policies epidemic may depend responses induce across various segments. find that shut down some essential sectors have stronger toll than infections compared non-essential sectors. Furthermore, there might not be an after-wave after are lifted. Restrictions can generate welfare gains relative case intervention. Milder longer restrictions less activities better stricter shorter restrictions, whereas opposite for more activities. Finally, shutting larger