Bad Chinese Statistics: Panda Census Edition

China has a complicated relationship with statistics. The country has in recent years been widely accused by executives and academics of faking economic statistics. But the phenomenon of massaged data isn’t limited to economics, and now some observers are questioning whether figures showing an increase in China’s panda population are being exaggerated.

On Saturday, Chinese scientists unveiled a long-anticipated survey of the country’s panda population, finding that the number of the adorable bears had increased by 17 percent to 1,864, compared to the last such survey.

Sounds like great news, right? Not so fast, argues Nature in a blistering takedown of China’s official panda statistics. The area surveyed by the Chinese authorities, the journal notes, is some 72 percent larger than the area examined in the last census. A rising panda population would be good news for Chinese authorities, who have in recent years taken steps to try to overhaul its environmental policies. A documentary about pollution that aired over the weekend generated a firestorm of discussion in China and has led many observers to believe that the government will make environmental issues a centerpiece of a series of high-level meetings in Beijing next week. A booming panda population would certainly be good news for officials trying to claim a positive environmental legacy.

For that reason, some scientists see another plot to juke the numbers in the latest panda statistics. “It’s a fine balancing act, so officials can claim the credit for rising panda populations but the number is not too high to diminish conservation funds,” a researcher who works with the Forestry Ministry and requested anonymity when discussing the surprisingly-sensitive issue of panda numbers, told Nature. Chinese state media reported that the rise in the panda population was attributable to “legislation, state-funded protection, and research programs.”

The panda survey can be seen as a microcosm of China’s difficult relationship with statistics. Counting China’s panda population is an enormously difficult task. The population is concentrated in three provinces — Sichuan, Shaanxi, and Gansu — in often very remote, mountainous locations. This has led scientists to develop various sampling methods for counting the bears. According to Nature, the latest panda survey combines two such sampling methods but declines to explain the difference in counts resulting from different methodologies.

Indeed, the problems of counting pandas are similar to the issues found in Chinese economic statistics. Panda researchers can’t reach the bears in their mountainous, remote territory, just as Chinese economists find it difficult to capture data from all parts of the gargantuan Chinese economy. At all levels of the system, there exist incentives for officials and businessmen to fudge their data, even as Chinese economic statistics are much improved since the dark days of the Cultural Revolution.

In both pandas and economics, official Chinese data can’t be completely written off, and it is simplistic to say that the country’s statistics are falsified across the board. Chinese air pollution statistics, for example, are better and more readily available than in years past, even if they remain flawed. A recent report on soil pollution broke ground in acknowledging the problem but provided little information on how the data was collected.

The problem lies in not knowing the degree to which this data deviates from the truth. If Chinese authorities were more transparent, scientists might be able to construct better estimates to fill the gap between official speculation and hard fact. And in the absence of good data, bad numbers are left to inform important policy decisions.

MANDEL NGAN/AFP/Getty Images