OPINION

Of Chinese statistics and the Keqiang Index

Is there reason to believe China is covering up new infections?

Evaluating the accuracy of official data is a challenge for journalists in every country. When that country is China, the problem is especially acute. Over my nine years reporting from there, figuring out how to use official numbers was a constant dilemma. Every three months, journalists would shuffle in to the State Council Information Office in Beijing for the quarter’s GDP numbers. The news wires would compete madly to be the first to file the figures. Having done so, reporters would later shrug that the numbers were probably massaged, but that was all we had to go by.

What may come as a surprise is that this view is widely shared by both the Chinese establishment and public. Chinese Premier Li Keqiang was quoted as saying in a WikiLeaks cable that GDP figures were “man-made”. He made these candid remarks in 2007 when he was a provincial leader. Rather than GDP numbers, Mr. Li turned to three more reliable indicators: electricity consumption, railway cargo volume and bank lending, now famously called the ‘Keqiang Index’. As the cable put it, “By looking at these three figures, Li said he can measure with relative accuracy the speed of economic growth. All other figures, especially GDP statistics, are ‘for reference only,’ he said smiling”.

Leaving aside sceptics who would insist even this approach is pointless because all Chinese data are “made up” (that’s neither entirely accurate nor an argument one can engage with), a more helpful question to ask is, how do you mitigate for the fudging?

Most of the fudging happens at the local level by politicians with an eye on furthering their careers. The National Bureau of Statistics (NBS) in Beijing has been finding ways to use technology to bypass them.

Economists find it useful to assess indicators that officials have either less incentives or ability to fudge, which include freight volume and energy data. The general consensus is official data are actually getting more reliable in suggesting trends, even if actual figures are massaged. Consider, for instance, that on March 16 the NBS itself said industrial production shrank a record 13.5% in the first two months of the year with the COVID-19 shutdown, an estimate worse than most economists’ forecasts. This would have been unthinkable a decade or two ago.

I thought of the ‘Keqiang Index’ amid the questions surrounding China’s most recent COVID-19 numbers, with authorities saying for several days now the country had zero local infections — a claim that made headlines around the world, even if greeted by scepticism by many both outside and inside China.

Is there reason to believe China is covering up thousands of new infections every day? Let’s apply the Keqiang Index test, for instance, by looking at the number of medical workers still working in Wuhan, or the number of makeshift hospitals still functioning. On March 10, Wuhan closed the last of its 16 makeshift hospitals (which the authorities wouldn’t have done without good reason considering the stakes), while on March 18, 49 medical teams comprising 3,787 medical workers were sent home. More than 42,600 personnel had been parachuted into Wuhan during the peak of the crisis. Both indicators strongly suggest a trend of declining infections, even if the official numbers might be, to borrow a phrase, “for reference only”.

Recommended for you