ABSTRACT
This paper develops a methodology for the assessment of the short-run effects of lockdown policies on economic activity. The methodology combines labor market data with simulation of an agent-based model. We apply our methodology to the Santiago Metropolitan Region, Chile. We recover the model parameters from observed data, taking into account the recurring policy adjustments that characterized the study window. The model is used to build counterfactual scenarios. We estimate an 8 percent output loss in the first 5 months of the pandemic from the policy that was put in place, achieving a 56 percent reduction in the total number of infections. During this period, with an output loss to 10.5 percent of GDP, the infection rate would have decreased 92 percent, significantly delaying the spread of COVID and spike in infections. Our methodology applied to real data provided results that could be valuable in guiding policies in other lockdown situations in times of disaster, pandemics or social upheaval.