Excess mortality in the BRICS

Several BRICS show strikingly similar excess mortality outcomes in spite of several structural and other differences

Brazil, Russia, India, China, and South Africa – also referred to as the BRICS – have in common the fact that they are major emerging market economies. But apart from that they exhibit large differences in terms of development level, economic structure and socio-political context. Does this diversity carry through to pandemic outcomes? In what ways do the BRICS differ? And why are several of them remarkably similar? 


Two perspectives

Diverse outcomes on reported COVID-19 mortality

The reported data on confirmed cases and deaths suggest enormous differences in country experiences across the BRICS. The chart below shows cumulative COVID-19 cases per capita since the start of the pandemic. Brazil and China occupy the extremes of the spectrum, with India, Russia and South Africa in the middle. 

We see not only large differences in levels but also in the shape of the curve, where the curviness reflects the effect of successive waves. The Brazilian and Russian curves are particularly wavy. India’s huge wave during the major outbreak in May 2021 also comes through clearly. All countries, except China, saw a major steepening of the curve related to the recent Omicron/Delta outbreak.

In terms of reported COVID-19 mortality, the picture is equally diverse. The BRICS adopt the same ranking for cumulative mortality rates as for cumulative case rates: Brazil and China at opposite extremes and India, South Africa and Russia in the middle. The shape of the mortality curve however looks quite different from the cases curve. Most notable is the fact that the recent Omicron/Delta wave has not led to a major spike in mortality despite the very considerable increase in cases. The exception here is Russia, where reported mortality has continued to climb quite considerably.

More similar outcomes for estimated excess mortality?

Excess mortality captures the gap between the total number of deaths that occur for any reason and the amount that would be expected under normal circumstances.  Given the massive undercounting of the mortality toll both directly and indirectly attributed to COVID-19, excess mortality provides a useful way to get a glimpse of the true mortality toll.

Below we show the cumulative estimated excess mortality rate, where the solid lines are actual data supplemented with estimates based on the excess death model produced by The Economist. The shaded areas represent the range of alternative possibilities within the 95% confidence interval (for the part of the curve based on estimates as opposed to actual data). The estimates taken into account over 120 variables that include seroprevalence, demography and urbanization, access and quality of healthcare, political regime and media freedom, connectivity and mobility and government responses to COVID-19. The model is discussed here

Three BRICS countries – Brazil, India and South Africa – have remarkably similar excess mortality rates. China and Russia remain worlds apart, with China at the absolute bottom of the spectrum and Russia registering rates that are well above the Brazil-India-South Africa cluster. Even when taking into account the uncertainty, the 95% confidence intervals suggest that China and Russia are indeed remarkably different. As to the Brazil-India-South Africa cluster, we have high confidence in Brazil and South Africa having similar rates. But it may well be the case that India’s rates are significantly larger or smaller.  


As far as the mid-point estimates of excess mortality go, mortality outcomes are surprisingly similar for Brazil, India and South Africa, even though they clearly are not for China and Russia. What could explain this convergence in outcomes even if structurally these countries are very different?

On the infection side, the stringency of containment measures will have played an important role. We can see in the two charts below that proxies of stringency vary a lot over time and across countries. Still there are some systematic differences. In the first chart, we see that, for most of the pandemic period, the indicator for China exceeds the one for Russia, which confirms the observation that China has followed a zero-COVID policy which few other countries around the world have. 

The stringency of government response has varied widely across Brazil, India and South Africa, with the indicator for India most of the time exceeding the one for South Africa, and Brazil most of the time somewhere in the middle. Of course, the comparisons are complicated by the seasonality of viral outbreaks, the fact that we are comparing across hemispheres and other idiosyncratic factors that desynchronize the timing of outbreaks across countries. 

One other factor that varies considerably across the BRICS is urbanization. Note the contrast between Brazil and India. Urbanization may be conducive to infection keeping all else equal. The “all else equal” proviso is important though. For example, Brazil’s high degree of urbanization may have facilitated the spread given that high urbanization is combined with high population density in urban areas. This seems to be consistent with the high cumulative case rate for Brazil even though cases are of course a very imperfect proxy of underlying infections. 

India on the other hand has a low urbanization rate, meaning most of the population lives in rural areas, which should dampen the spread. It should be noted however that rural population density in India is very high high, which may limit the beneficial effect of lack of urbanization. (This by the way stands in sharp contrast with Sub-Saharan Africa, which is similarly urbanized but has a much lower rural density). 

The high excess mortality rate for Russia may be partly explained by the age structure of its population. Russia has by far the highest share of people aged 60+ and 70+ among the BRICS. That need not be automatically lead to high mortality as the experience of China with its zero-COVID approach illustrates. Another example is Japan – the world’s second-oldest country after Monaco – which has kept excess mortality very low. But the combination of a large elderly population with other factors that seem to have contributed to the spread provides in the case of Russia a plausible explanation. 

Conversely, India and South Africa have much lower elderly age shares, which means that its demographic structure will afford the country a level of protection against the severe consequences of COVID. Given however that excess mortality rates are estimated to be relatively high at similar levels, this must indicate that the demographic advantage is offset by other factors. This could include higher infection prevalence (the reported cumulative case rates, for what they are worth, are lower than those for Brazil and Russia but higher than China’s). Most likely it reflects also higher age-adjusted infection fatality rates, which relates to the pattern of co-morbidity, access to health care and quality of health care, among many other host-specific and environmental factors, which together are likely to correlate with life expectancy at birth. 


The BRICS are highly heterogenous. Not just in terms of the way the countries differ in terms of institutions and structures, but also in terms of pandemic outcomes. That comes through very clearly when we look at the reported data on COVID-19 cases and fatalities. 

Curiously, however, when it comes to excess mortality outcomes, the various factors that seem to differentiate the BRICS are working together to produce a more homogenous outcome, at least for Brazil, India and South Africa, which all have experienced considerable excess mortality, even on a per capita basis. The exceptions are China and Russia, which continue to stand out at the opposite extremes within this sample of countries.