China: Fuzzy Numbers No More?

Statistics could improve if Beijing's new economic census is a success

Beijing this winter is festooned with orange banners and billboards that look like Communist agitprop. But instead of slogans exhorting hero-workers toproduce more rice or steel, the message reads: "The economic census needs your support!" Time was, China's centrally planned economy didn't need such a census. Planners would have been expected to know how much of everything Chinese companies were churning out. But now, China is marshaling millions of T-shirted recruits to ask business owners across the country dozens of questions about their sales, profits, and production.

Will China's new capitalist roaders reveal such sensitive information? It's too early to tell, since the census only began a month ago and the first results won't come out until August. But the census is so important for the government that it has pledged to keep the data confidential and not use them to track down tax evaders. Beijing hopes the census will help it better understand China's $1.4 trillion economy -- and improve the quality of Chinese statistics, which many experts have long decried as grossly inaccurate. "We consider the criticism from home and abroad an impetus to speed up our statistical reform," Li Deshui, commissioner of China's National Bureau of Statistics (NBS), said at a conference on Jan. 17.

Just about everybody would benefit from better numbers out of China. Over the past decade, China's share of global output has doubled, to roughly 4%. The country sucked in some $65 billion in foreign direct investment last year, overtook Japan as the world's No. 3 trading nation behind the U.S. and Germany, and amassed more than $600 billion in foreign currency reserves. The price of oil, steel, aluminum, copper, and iron ore all turn on the economic figures China releases.

Until recently, though, China's statistics were about as reliable as a $19 DVD player. During the 1950s statistical fakery ran rampant; everyone from steel producers to hatmakers pumped up figures to meet government quotas. After the 1997-98 Asian financial crisis, economists say, Beijing exaggerated its expansion to show how well it was withstanding the fallout. UBS (UBS ) economist Jonathan Anderson estimates that the official 1998 gross domestic product growth figure of 7.8% was nearly double the real level. "There is a smoothing exercise that goes on," explains Anderson. "The net result is that China's numbers are much less volatile than reality."


These days, China is getting more heat for underreporting its growth than for exaggerating it. Some private economists think the country grew by more than 10% in 2004 despite the 9.5% figure released on Jan. 25. Shoddy recordkeeping is partly to blame. But there's also an incentive for local officials to meet -- precisely -- the official growth target set annually by Beijing. Once enterprises in a given municipality reach the desired figure, any excess production may be moved to the following quarter. "There is political pressure when growth is running hot to err on the negative side and when running slow to err on the positive," says Arthur Kroeber, managing editor of the Hong Kong-based China Economic Quarterly. He adds, however, that China's numbers aren't any worse than those from most developing countries.

It's not just the headline numbers on growth, inflation, and investment that are questionable. Some experts say China's banks may hold $600 billion in nonperforming loans -- or 35% of all outstanding debt. The official number is just $200 billion, in part because the statistics bureau isn't as strict in classifying dud loans as it might be, and also because debt from many smaller regional lenders and credit cooperatives isn't included. Nor do government finance data include unfunded pensions or likely bank-bailout liabilities.

Some economists are taking matters into their own hands. Anderson oversees his own growth index that he compares with the official figures. He looks at hard-to-fake data including external trade, passenger traffic on planes, and energy consumption. Official figures show Chinese consumer spending last year was robust (fueling 13% growth in retail sales), rural incomes rose sharply, and investment jumped 26% for the whole economy. Put all together, "it is very difficult to get an economy growing at just 9.5%," he says, guessing the real number was 10.6%.

That's not to say there has been no improvement in statistical reporting from China. The World Bank, International Monetary Fund, and private economists all see huge gains. In 2001, for instance, China doubled to 700 the number of items it includes in its monthly survey of consumer prices, leading to a more accurate assessment of spending patterns. To bypass local party officials who might be inclined to fudge economic performance to move up the ranks, the NBS now collects data directly via the Internet from 5,000 large and midsize companies -- accounting for nearly half of China's industrial production -- and from 3,000 real estate companies and 1,200 wholesale and retail outfits. The NBS is making more random checks of data, which discourages cheating. And the census should help shed light on exactly what is happening in the private sector. Chinese statistics "are light years ahead of where they were," says Kent Kedl, executive director of Technomic Asia, a Shanghai-based market strategy consultant.


Two areas, though, continue to concern China watchers. There has long been a suspicion that Beijing has at various times either over- or underreported numbers such as consumer spending, retail sales, and capital outlays to leave room for massaging the final GDP calculation when needed. Another mystery is the actual size of China's service sector. While China is widely known as the world's manufacturing workshop, its health-care, retail, leisure, tourism, and education sectors are booming, too. The government says services accounted for about a third of China's output last year. Yet many private economists believe the census will show that services are much bigger -- perhaps as much as 40% of output. The Chinese "have always recognized the need to improve their data on the service sector," says China watcher Nicholas R. Lardy, a fellow at the Institute of International Economics in Washington, D.C.

The push to add more integrity to Chinese economic data is a work in progress -- and the census is a good start. But the real test will come if China's growth bubble pops and economic expansion slows dramatically. If Beijing comes clean about what's really happening in its economy even when the numbers are embarrassing, China will prove it has matured. But if the government continues to hide the truth, all the sloganeering around the economic census will ring as hollow as the old-time propaganda.

By Brian Bremner in Hong Kong and Dexter Roberts in Beijing

    Before it's here, it's on the Bloomberg Terminal.