How reliable is WHO coronavirus data?

And Europe has now been declared the epicentre of the virus. But viruses do not follow a political logic.

Instead, the numbers that the World Health Organisation (WHO) publishes, the numbers that journalists and governments around the world refer to, are contaminated with politics.

They are not useless, they tell us something, but they paint a skewed picture. How so?

The WHO receives its case numbers from governments around the world. It cannot verify these numbers and it cannot oblige governments to provide accurate numbers.

A cursory glance at the numbers provided, shows that some of them are implausible.

The five countries of Central Asia, some of them neighbouring China and enjoying close trade links, until some days ago did not report a single case in a population of 72 million people.

These countries are ruled autocratically and have no free media or effective opposition. If their governments do not want to admit that they have coronavirus cases, they won't do so.

Myanmar is another neighbour of China, with many Chinese investments and lively traffic between the two countries. As of now, it claims not to have any cases. Myanmar has no robust media and no effective separation of powers.

Turkey, a country of 80 million people, with Istanbul's airport being a major hub between Europe and Asia, until a few days ago claimed it had no cases. It now reported 47 cases, indicating that they were traced to Europe. In Turkey critical media have been silenced and president Recep Tayyip Erdogan rules with few checks and balance.

Why would some government not report on the coronavirus?

They have many reasons. They may not want to deter investment or tourists. They may not want to close down the country and hurt the economy. They may not want to spread a panic. They have seen how Chinese people started questioning their government and may fear a rise in domestic opposition.

Or they may want to wait for everybody to have a problem, before admitting their own.

Furthermore, some countries may simply not have many testing kits. The entire edifice of WHO numbers is based on administered test. But the WHO updates do not tell how much testing is done.

Myanmar's ministry of health reported that it tested around 100 persons. In South Korea 12,000 people are tested every day.

Obviously only South Korea's numbers are meaningful.

This is one explanation for why high-income democracies look most affected now. They are the ones who have health services that test people and report truthfully.

They do not arrest doctors who tell the bad news, as China did in the crucial first weeks of the epidemic. In short, Europe is the epicentre of transparent reporting and therefore appears to be the epicentre of the virus. The WHO won't tell you this, because it cannot criticise its member states.

US an outlier

The US is the most curious and opaque case of a high-income country. By 9 March it had tested less than 5,000 people out of a population of 330 million.

Compare this with the 35,000 people who have been tested in Germany in March.

The result is that Germany looks like a bigger problem with almost 6,000 cases compared to less than 3,500 in the US. As the doctors say: "If you don't take a temperature, you don't find a fever."

Obviously, many other factors play into the equation. High-income countries are more integrated into the global economy, their populations travel more and further than those in poorer countries.

One only needs to look at the global flight patterns on a normal day.

Most of them move between parts of Asia, Europe and North America. Also, European leaders badly underestimated the problem, doing too little, too late.

Thus, it is plausible that more Germans or Italians are infected than Kazaks or Uzbeks.

But it is not plausible that there are zero cases in sone of these countries and that the US has such a low number of infections compared to its population. Little or no testing or lack of political will to report explain these numbers.

Whenever a government talks about its coronavirus numbers you should ask how many people were tested. It is the essential variable that is missing in most reports.

In an ideal world everybody would publish these number in easy graphics as Iceland does. The WHO should ask its member states about their level of testing and publish these numbers too. The EU should collect the statistics on testing across its member states and publish them.

The coronavirus is not a neutral subject of scientific debate. It is as politicised as anything else in a world where many governments try to shut down critical thinking, independent media and checks and balances.