People are getting familiar with the concept of R0 amid the coronavirus pandemic. Defined as the average number of new infections caused by each infected person, R0 indicates how fast a disease spreads in a community and how big the outbreak will be. To reduce effective R0 and flatten the infection curve, countries around the world have implemented lockdown measures of varying degrees. With case growth rates slowing, some countries are gradually exiting lockdowns, raising fears of a rebound in effective R0 and a second wave of infections.
Is it true that only harsh lockdown measures can contain effective R0? Or put it another way, do harsh lockdowns always lead to lower effective R0? Our analysis on the evolution of mobility and R0 across 42 countries suggests “no” to both questions. Instead, we have found that the key factors for controlling R0 seem to be the robustness of a country’s healthcare system and social-economic strength.
Clustering Countries on Strictness of Lockdowns
To investigate how different degrees of lockdowns influence effective R0, we divided the 42 countries into three groups: Hard Lockdown, Medium Lockdown and Soft Lockdown. We allocated the groups by performing K-means clustering with Google mobility measures as variables. Specifically, we included the latest levels of Google mobility as well as mobility changes since end-March for each country across five segments: retail & recreation, grocery & pharmacy, parks, transit stations and workplaces. As shown below, the separation between identified clusters is quite clear.
The Soft Lockdown cluster mostly consists of developed markets (DM) countries, like Sweden, Norway, Denmark and Germany, or relatively strong Eastern European emerging markets (EM) countries, like Poland and the Czech Republic. The Medium Lockdown cluster contains a mixture of DM and EM countries, including the US, UK, Japan, Australia, as well as Brazil and Egypt. The Hard Lockdown cluster mostly consists of EM countries, like India, South Africa, Mexico and Turkey, with a few DM exceptions, including Italy, Spain, France and New Zealand.
Lack of Clear Relationship between Effective R0 and Lockdown Strictness
We then looked at how effective R0 differs between each country cluster. Surprising we found that the latest average effective R0 is actually the highest for the Hard Lockdown group at 1.00, and the lowest for the Soft Lockdown group at 0.86, though with big variations in the Hard Lockdown group. We also compared the declines in effective R0 since the start of lockdown. Again, surprisingly we found that the average R0 decline for the Hard Lockdown group does not differ too much from the Soft Lockdown group, at -0.42 and -0.43 respectively.
On the other hand, there seems to be a division between DM and EM countries. As shown below, the countries with higher real-time effective R0 or slower decline in R0 in each group tend to be EM countries. This is especially pronounced in the Hard Lockdown group.
We also looked at the evolution of effective R0 over time against the changes in average mobility (see appendix). It seems a handful of EM countries struggle to get their effective R0 below 1, the level needed for infections to die out, even with drastic declines in mobility.
Health Sector Robustness and Overall Country Resilience Seem to Influence R0 More
Given the observed division between DM and EM, we investigated the relationship between countries’ real-time effective R0 and various measures of country strength. For the latter, we used subcomponents of the 2019 Global Health Security Index. From the analysis we found that two measures have clear correlations with countries’ effective R0, namely Health Sector Robustness and Overall Country Resilience. In other words, richer and stronger countries can more effectively control the spread of Covid-19 with less disruption to their social and economic activities.
The finding is intuitive: countries with stronger healthcare systems and social-political institutions are likely to be more efficient in mass testing, contact tracing and quickly isolating infected cases, i.e. measures that can more effectively influence effective R0. To illustrate as an example, the number of Covid-19 tests done per 1,000 people in general is higher in DM countries than EM countries, as data compiled by Our World in Data shows.
Weak EM Countries Are Likely to Be Bigger Losers in the Post-Covid World
The pandemic has accelerated history and exposed fragilities accumulated from the past decades of debt cycle. Winners and losers will emerge in the post-Covid World. With weaker medical/institutional strength to manage the spread of the disease and less fiscal/monetary headroom to limit the economic fallout, EM countries are likely to fall behind in the recovery cycle and become the bigger losers.