Some questions about Chinese Covid-19 data

China’s official coronavirus data suggests that the country was incredibly successful in not letting the virus spread beyond its place of origin: Wuhan and the neighbouring Hubei province.
As of today, China confirmed a total of 14,402 cases of Covid-19 outside Hubei with 120 deaths⁽¹⁾.
A basic analysis of these figures is sufficient to conclude that the figures are wildly understated. The difference against the lowest reasonable estimate of the true spread of coronavirus across China excl. Hubei is so great that this cannot be simply due to difficulties in testing and data management that other countries also experience.
On the face of it, there appears to have been a systemic effort to understate the spread of the epidemic across Chinese territory. This may have lulled Western decision-makers into a false sense of security as it appeared that the virus was largely confined to a single province in central China with only limited spread elsewhere.
Once the smoke settles, we will need to seriously consider to what extent China is responsible for the slow Western response to the onset of the crisis in late January and February.
We should be very sceptical of claims that China’s management of the virus provides the roadmap out of the crisis, given the dubious quality of data coming out of the country.

Iceland has arguably done the best job in the West in containing the virus. It has tested about 10% of its population, by far the largest share in the world. It is fair to assume that Iceland’s coronavirus data is the most reliable currently.
- As of today, Iceland has recorded 1,711 cases of Covid-19 with 8 deaths⁽²⁾. The recorded cases represent about 0.47% of the Icelandic population.
It makes sense to assume that coronavirus has spread at very least as widely across China (outside Hubei) than in Iceland.
- Clearly, the rest of China has far stronger links to the origin of the virus in Hubei than Iceland and there has been plenty of opportunity for Covid-19 to spread from Hubei to the rest of China.
- Prior to Wuhan being locked down on 23 January, there has been high levels of intercity travel in China ahead of the Chinese New Year on the 25th.
- There are also reports of a large number of people fleeing Wuhan just before the lockdown was announced, with estimates ranging from 100,000 to the millions.
- Iceland, on the other hand, is not particularly plugged into global travel, especially at the depths of winter.
However, shockingly, China reported that just 0.0011% of its population outside Hubei tested positive for the virus – 435x less than in Iceland (Chart 1). Clearly this impossible.

reported china vs iceland

Assuming that China outside Hubei must have at least the same % of its population infected as Iceland, there would need to be at least 6 million cases, instead of the 14,402 cases reported (Chart 2).

cases reported vs estimate

Death counts are likely hugely understated as well. Assuming a 0.66%⁽³⁾ mortality rate against the low-end estimate of 6 million cases, there would have been about 41,000 deaths in China excl Hubei, compared to just 120 deaths reported (Chart 3).

reported vs estimated deaths

While every country is struggling to find out the true number of coronavirus cases within their territory, death counts are generally reliable. At very least, they are not understated by a factor of 345x!

Notes:

hvk