Economy wide COVID-19 lockdown policies, which included stay-at-home orders, school and business closures, along with mask and other non-pharmaceutical interventions, were a radical and untried social policy. Over the two years that followed the initial declaration of a pandemic an exceptional amount of research was done on all things related to COVID-19, and this included the costs and benefits of lockdowns.
The benefits of lockdowns were originally expressed in terms of mitigating the rush to hospitals and preventing the health system from being overrun. Later many thought that the virus might actually be eliminated by lockdowns (so-called “zero-COVID”). Initial benefit estimates were based on simple models that predicted the number of hospitalizations and deaths without lockdowns. Initial estimates of the costs of the lockdowns were based only on lost GDP from reduced labor-force participation. This led to grossly inaccurate cost/benefit estimates.
Examining the research over the past two years reveals several robust and clear findings: Epidemiological models that do not include human behaviour changes in response to a novel virus drastically over-estimate the number of hospitalizations and deaths. All of the early models made death predictions that were off by factors of 10 or more. The infamous Imperial College of London model, led by Neil Ferguson, predicted that with full lockdowns in place there would be 132,687 COVID-19 deaths in Canada by July 30, 2020; in fact, by that date there were 9,019 actual deaths.Changes in people’s behaviour in response to the arrival of the virus were immediate, and around the world and in every country infected by COVID-19 changes in behaviour meant that an endemic state was reached in the spring of 2020.Behaviour effects were not limited to acting cautiously. Other behaviour changes including incomplete compliance with mandates—and compliance levels varied over the course of the pandemic. These changes in behaviour meant that deaths and hospitalizations were not substantially different in jurisdictions with different degrees of lockdown when other factors were controlled for.Ultimately, estimates of the benefit of lockdowns in terms of lives saved were made based on data. Analysts used many different procedures in an attempt to identify the causal effect of lockdowns. Over and over, findings showed only small positive effects on death rates. The most recent and thorough meta-analysis found that after combining all lockdown effects there was only an average reduction in mortality of 3.2 percent. All of the lockdown efforts amounted to almost nothing.The costs of lockdown go far beyond the lost GDP. In areas like worldwide food insecurity, international trade reductions, reduced travel, increased domestic violence, increased drug/alcohol/mental health issues, and employment disruptions, we are only aware of the costs and no estimates have yet been made of the level of these costs. Much work has been done on the effect lockdowns have had on children’s physical well-being, lost education, early development, IQ, and social abilities. Again, no widespread estimates of the actual size of these losses have been made, but it is generally acknowledged that children and youth have suffered under the lockdowns.Lockdowns created collateral deaths. Behaviour changes in the face of COVID-19 and lockdowns included forms of self-protection that often ended up increasing mortality. These behaviour changes included missing regular medical checkups out of fear of contracting the disease. Estimates in the US show that there were 171,000 excess non-COVID-19 deaths through to the end of 2021. By that time the US had recorded 825,929 COVID-19 deaths. However, if lockdowns only reduce deaths by 3.2 percent, then only 27,303 lives were saved by lockdowns. Just on collateral deaths alone the cost/benefit ratio of lockdown is 171,000/27,303 = 6.26.
Widespread, economy-wide lockdown policies were a disaster. They had only marginal effects on the ultimate number of deaths, but imposed enormous costs.