Emerging findings

  1. Based on numerous random antibody studies conducted all over the world, the Covid-19 infection fatality rate (IFR) is in the region of 0.1% to 1.2%.  
  2. The daily death rates across the Northern Hemisphere have plummeted from their peaks and continue to fall. There are three leading contenders to explain this trend in our view: the massively increased use of masks and social distancing, the weather (including temperature) and inherent/cross immunity pre-existing in populations. All are likely to play a role, but the question is whether one driver is dominant.
  3. There is no data to suggest that the failure to lockdown creates an exponentially worsening situation in a country. The lives saved from lockdown therefore look incremental, and would need to be carefully evaluated against a) the lives lost from lockdown especially in developing countries (see our article on this here) and b) the economic, social and human-rights impact of the lockdown.
  4. The data on the risks faced by different age and health profiles is now increasingly robust (see our analysis of the New York City data here). The burden of this disease is heavily skewed to the elderly and those with existing comorbidities. This raises the need to use triage-type ethical weightings when evaluating the net-benefit or loss from lockdowns and other social distancing regulations.

South Africa remains barely impacted by the virus. Its death rate is 10 times less than even Germany, which has emerged relatively unscathed by the virus. The South African Government's models, to the extent that they have been revealed, show that regardless of whether we lockdown or not, the same amount of people will get infected in a year or two. The benefit of the lockdown is that flattening of the (daily active cases) curve prevents our health care system from being overwhelmed. But how many lives does that actually save over the next year, given our latest understanding of the fatality rates and other variables? This has not been provided.

Our opinion remains that the failure for the South African Government to produce a reasonable quantification of the lives saved by the lockdown renders their lockdown improperly imposed.

Key contextual facts

  • Total global deaths from Covid to date are 374,323 (as of June 1).
  • Annual global deaths from seasonal flu according to the WHO: between 290,000 to 650,000
  • Annual flu deaths in South Africa: 10,000 according to United States CDC and between 6,000 and 11,000 according to SA Department of Health.
  • In an average year, countries tend to lose (through deaths) and gain (through births) around 1% of their population. Annually, that's 10,000 per million deaths and 10,000 per million births. Its a rough estimate and it obviously differs by country, but it's a reasonable number to get a sense of the data.

Overall fatality counts per million

As of 1 June 2020, Belgium has been the hardest hit country in the world, followed by Spain, and the UK. Italy has improved to 4th position. Sweden is the 5th worst performer. South Africa remains, relatively speaking, within the league of countries that have claimed to have nearly eradicated the virus from their borders, e.g. New Zealand.

The current UK death rate of 565 per million people is roughly 5% of their average annual deaths, using the back-of-envelop 1% annual deaths number highlighted above.  

South Africa has lost 611 persons (10.3 per million). This is well below their annual flu deaths of around 10,000 per year (175 per million). So far the country has hardly been scratched by the virus. Of course, the impending winter, and modestly rising daily death curve suggest that caution remains warranted (see graphs below).

Please note that not all of these deaths will be "additional deaths" when comparing to yearly average. Given that the age profile of fatalities is heavily skewed towards the elderly (which you can read about on our Covid snapshot of New York), a portion of these deaths can be expected to have anyway occurred within the year.

This is crucial when applying triage-type ethical weightings.  

Death rates in Europe and the United States plummet

This graph, updated 1 June 2020, shows the number of daily deaths per capita since a country first recorded 0.1 per million deaths.

In all cases, the daily death curves have fallen sharply from their peaks. There are three likely causes for the continued to decline in death rates:

  1. the weather (flu/ winter season is over),
  2. the continued practice of social distancing and especially mask wearing, and
  3. inherent/corss immunity

If masks were the driving reason behind declining rates, we would expect some uptick as nations emerge from lockdown: masks are less effective than a state of complete lockdown (where masks were worn anyway). This has not occurred. On the other hand, the consistent downward trends are consistent with temperature changes as the North moves towards full summer.  

Sweden did not lockdown its country: the population began to observe significant social distancing, though this did not include face masks.

Yet the decline in daily deaths has also been true of Sweden until arguably very recently. We find it surprising that the divergence should show sharply in the last few days. After all, more and more countries are coming out of lockdown, and therefore are beginning to behave more and more like Sweden. Why would a sharp difference start to show in the last 7 days? We therefore anticipate that the recent uptick in Swedish Covid-19 deaths will revert to trend, but we may be wrong.

Overall, Sweden argues a) that Covid is not so serious as to warrant complete lockdown and b) that in the medium term (over the next few winter/flu seasons), death rates per capita will even out. In this light, we note Sweden's approach has not created an exponential fatality curve. Sweden's daily death curve is coming down (except for last week), and its overall fatality rates are average, though at the higher end.

This is exactly what Sweden has argued its citizens could anticipate after the first "winter" season has passed.

Death rates in Southern Hemisphere are increasing

The daily fatality graph below, updated as of 1 June 2020, focuses on the Southern Hemisphere. The US is kept in the graph to provide a reference point. As above, we focus on days since 0.1 per million deaths were recorded.

Brazil has had arguably one of the least restrictive social distancing approaches. Its President continues to argue that lockdowns are not necessary. Peru, on the other hand, had an extremely strict lockdown - as did Argentina and other parts of South America. It is too early to tell whether or not the Brazilian approach led to significantly worse outcomes given the experience in Peru and Mexico.

The daily death curves of South American countries have not increased as rapidly as occurred in Europe and the United States in the beginning stages of their respective pandemics. As such, total per capita deaths in the south are well below those in the north (see first graph above). On the other hand, by the time the United States and other European countries reached this point in the pandemic, their death curves had already turned downwards.

South Africa's daily fatality rates remain low at under 50 deaths per day. Its fatality curve is still following a low trajectory, even as it now moves into the heart of winter. On June 1, the country lowered its lockdown restriction level (from level 4 to level 3).