Mortality · Post

A pandemic of three giant outliers

Brazil, India and the USA are the giant outliers of this pandemic: they have huge populations and at the same time deviate significantly from their income classification peers.


The US, Brazil and India – in this order – are the world’s 3 largest contributors to COVID-19 mortality. They are giant outliers : not only do they have very large populations, their per capita mortality rates have also surpassed those of their peers.  

The income classification lens

In the panel chart below, we track per capita mortality and show the daily progression of the weekly totals. We do so for Brazil, India and USA as well as their peer groups. Also shown is the same for weekly cases (click on the image carousel to zoom in).

The peer groups are taken to be the income groups these countries belong to in the World Bank income classification of countries. To make a clean comparison, we exclude the three countries of interest from these groups. So we compare the US with 60 other high-income countries, Brazil with 49 other upper-middle-income countries and India with 46 other lower-middle-income countries. We’re also showing the full group of 28 low-income countries for comparison purposes. 

(That makes a total of 186 countries. The income classification list of the World Bank actually includes more entities, but most of them are territories and not countries and others are places that are not reporting COVID-19 mortality data.)

Outliers in COVID-19 mortality

It’s immediately obvious that the US, Brazil and India are indeed outliers. They can be distinguished by differences in levels and dynamics with respect to their peers. Per capita mortality levels are higher, except for India since September. 

The dynamics also differ. The US appears to be an amplified version of its peers: the swings in mortality outcomes are far greater. The directions of mortality in Brazil and India, on the other hand, seem to be much more disconnected from their peer groups.

And of course the three countries are very large population-wise. 

Digression on peer groups

Peer group mortality rates are calculated as population-weighted averages. So the mortality rate for the peer group of Brazil (all UMICs except Brazil) is obtained by dividing total deaths in the peer group by the total population size of the peer group. 

Let’s consider the cumulative mortality rate (total fatalities since the start of the pandemic by population). This is currently 103 per 100,000 people for Brazil, 29 for all UMICs including Brazil and 23 for all UMICs excluding Brazil. 

The simple point here is that we compare Brazil with the rest of the world without regard for borders. This is an appropriate comparison given that we want to take into account Brazil’s population weight. By excluding Brazil, the UMIC average drops from 29 to 23 per 100,000. This makes the contrast greater and more relevant.

Good and bad news

This is both good and bad news. The bad news is that these 3 outliers are both population giants and outliers when it comes to pandemic mortality outcomes. As a result, they exert a major influence on the global outcome of this pandemic.

The good news is that the 183 other countries, in the aggregate, are doing a whole lot better. I realize this is a tautology, but it is also a reminder that there are huge differences across countries that go beyond the variations related to the income classification. Policy will be one key explanatory factor among many others.