The public debate on the effectiveness of lockdown measures is far from being settled. We estimate the impact of not having implemented a strict lockdown in the Bergamo province, during the first wave of the COVID-19 pandemic, despite observing an infection rate in this area similar to the one observed in nearby municipalities where a strict lockdown was instead promptly implemented. We estimate the causal effect of this policy decision on daily excess mortality using the synthetic control method (SCM). We find that about two-thirds of the reported deaths could have been avoided had the Italian government declared a Red Zone in the Bergamo province. We also clarify that, in this context, SCM and difference-in-differences implicitly restrict effect heterogeneity. We provide a way to empirically assess the credibility of this assumption in our setting.