How China Cooks Its Books
It's an open secret that China has doctored its economic and financial statistics since the time of Mao. But could it all go south now?
In February, local Chinese Labor Ministry officials came to "help" with massive layoffs at an electronics factory in Guangdong province, China. The owner of the factory felt nervous having government officials there, but kept his mouth shut. Who was he to complain that the officials were breaking the law by interfering with the firings, he added. They were the law! And they ordered him to offer his workers what seemed like a pretty good deal: Accept the layoff and receive the legal severance package, or "resign" and get an even larger upfront payment.
"I would estimate around 70 percent of workers took the resignation deal. This is happening all over Guangdong," the factory owner said. "I help the Department of Labor, and they’ll help me later on down the line."
Such open-secret programs, writ large, help China manipulate its unemployment rate, because workers who "resign" don’t count toward that number. The government estimates that roughly 20 million migrant factory workers have lost their jobs since the downturn started. But, with "resignations" included, the number is likely closer to 40 million or 50 million, according to estimates made by Yiping Huang, chief Asia economist for Citigroup. That is the same size as Germany’s entire work force. China similarly distorts everything from its GDP to retail sales figures to production activity. This sort of number-padding isn’t just unethical, it’s also dangerous: The push to develop rosy economic data could actually lead China’s economy over the cliff.
Western media outlets often portray Chinese book-cooking as part and parcel of a monolithic central government and omnipotent Beijing bureaucrats. But the problem is manifold, a product of centralized government as well as decentralized officials.
Pressure to distort or fudge statistics likely comes from up high — and it’s intense. "China announces its annual objective of GDP growth rate each year. In Chinese culture, the government has to reach the objective; otherwise, they will ‘lose face,’" said Gary Liu, deputy director of the China Europe International Business School’s Lujiazui International Financial Research Center. "For instance, the government announced that it wanted to ensure a GDP growth rate of 8 percent in 2009, and it has become the priority for government officials to meet that objective."
But local and provincial governmental officials are the ones who actually fiddle with the numbers. They retain considerable autonomy and power, and have a self-interested reason to manipulate economic statistics. When they reach or exceed the central government’s economic goals, they get rewarded with better jobs or more money. "The higher [their] GDP [figures], the higher the chance will be for local officials to get promoted," explained Liu.
Such statistical creativity is nothing new in China. In 1958, Chairman Mao proclaimed that China would surpass Britain in steel production within 15 years. He mobilized villages throughout China to establish backyard steel furnaces, where in a futile attempt to reach outrageous production goals, villagers could melt down pots and pans and even burn their own furniture for furnace fuel. This effort produced worthless pig iron and diverted enough labor away from agriculture to be a main driver in the devastating famine of the Great Leap Forward.
Last October, Vice Premier Li Keqiang said in a speech after inspecting China’s Statistics Bureau, "China’s foundation for statistics is still very weak, and the quality of statistics is to be further improved" — a brutally harsh assessment coming from a top state official.
Indeed, China has predicated its very claim of being the healthiest large economy in the world on faulty statistics. The government insists that even though China’s all-important export sector has been devastated — contracting about 25 percent in the past year — a massive uptick in domestic consumption has kept factories producing and growth churning along. A close examination of retail sales and GDP growth, however, tells a different story. China’s domestic retail sales have risen about 15 percent year on year, but that does not really translate into Chinese consumers purchasing 15 percent more televisions and T-shirts. The country tabulates sales when a factory ships units to a retailer, meaning China includes unused or warehoused inventory in its consumption data. There is ample evidence that state-owned enterprises buy goods from one another, simply shifting products back and forth, and that those transactions count as retail sales in national statistics.
China’s retail statistics seem implausible for other reasons, too. They would imply an increase in salaries among Chinese people, allowing them to purchase that extra 15 percent. To be sure, the Statistics Bureau reported salaries had increased 12.9 percent in the first half of 2009. But Chinese netizens complained such numbers were hard to believe — as did the bureau’s chief.
A look at GDP growth also raises serious questions. China’s economy grew at an annualized 6.1 percent rate in the first quarter, and 7.9 percent in the second. Yet electricity usage, a key indicator in industrial growth and a harder metric to manipulate, declined 2.2 percent in the first six months of the year. How could an economy largely dependent on manufacturing grow while its industrial sector shrank?
It couldn’t; the numbers don’t add up. China announced a $600 billion stimulus package (equal to about 14 percent of GDP) last fall. At that point, local governments started counting the dedicated stimulus funds in GDP statistics — before finding projects to use the funds, and therefore far before the trillions of yuan started trickling into the economy. Local governments keen to raise their growth and production numbers said they spent stimulus money while still deciding on what to spend it, one economist explained. Thus, China’s provincial GDP tabulations add up to far more than the countrywide estimate.
Alternative macroeconomic metrics, such as the purchasing managers’ index (PMI), which measures output, offer a no more accurate reflection. One private brokerage house, CLSA, compiles its own PMI, suggesting a sharp contraction in industrial output between December 2008 and March 2009. Beijing’s PMI data, on the other hand, indicated that industrial output was expanding during that period.
Unfortunately, such obfuscation means China’s real economic health is difficult to assess. Most indicators that would help an intrepid economist correct the government numbers — progress on infrastructure projects, end-user purchases, and the number of "resigned" workers — are not public.
Still, it is possible to infer the severity of the gap between economic reality and China-on-paper by looking closely at monetary policy. China’s state-owned banks dramatically increased lending in the first half of 2009 — by 34.5 percent year on year, to more than $1 trillion. This move seems intended to keep growth artificially high until exports bounce back. Most analysts agree that it is leading to large bubbles in the stock, real estate, and commodity markets. And the Chinese government recently announced plans to raise capital requirements — an apparent sign it sees the need to reign in the expansion.
For the long term, China is banking on its main export markets — in the United States, Europe, and Japan — recovering and starting to consume again. The hope is that in the meantime, rosy economic figures will placate the masses and stop unrest. But, if the rest of the world does not rebound, China risks the bursting of asset bubbles in property and stocks, declining domestic consumption, and rising unemployment.
That’s when the Wile E. Coyote moment could happen. Once Chinese citizens no longer believe that the economy is doing well, social unrest and more widespread worker riots — already increasing in scope and severity — are likely. That’s something that China will have a harder time hiding. And then we’ll know whether China’s statistical manipulation was a smart move or a disastrous mistake.