BEIJING — When China announced better-than-expected trade numbers last month, the statistics were met with outright suspicion from international powerhouses such as Goldman Sachs, Swiss financial firm UBS and Australian bank ANZ. The disbelieving scoffing only mounted days later, when the government unveiled numbers showing yet another positive trend — a narrowing income gap between China’s rich and poor.

Numbers in China have long faced suspicion, from optimistic recordings of visibly hazy air to the age of its Olympic gymnasts. But the credibility of its economic data is now coming under particular scrutiny, at a time when China’s growing global role weighs on investors, analysts and governments worldwide, even as the country’s economy is slowing after years of unbridled growth.

“The issue is less important when everything’s rosy,” said Patrick Chovanec, a business professor at Beijing’s Tsinghua University. “It gets more sensitive when people are trying to assess just how deep this downturn is and whether China’s facing a hard landing.”

For foreign economists, suspicions about why such numbers are off — and by how much — often boil down to a Rorschach test of sorts for how they view the Chinese government: as a developing nation honestly struggling with statistical challenges or as a sinister actuarial force always prepared to stretch the truth to meet its goals.

Regardless of motivation, some experts say the biggest danger in all this inaccurate data may be domestic, especially if it leads to bad policy decisions by Chinese leaders wrestling with how to manage a complex, dynamic economy.

“It’s like driving a car on a road where the traffic lights don’t work,” said Wei Yao, an economist at Société Générale bank in Hong Kong. “How do you make monetary policy for example if you can’t accurately tell if inflation is too high or too low? It all relies on statistics.”

The latest round of accusations over trade data began in January after analysts compared China's reported December exports to other available information, including corresponding import numbers from surrounding countries. Little of it matched up. Analysts’ characterizations ranged from the gentle, noting “anomalies,” to insulting cries of “ridiculous” and “obvious errors.”

Amid such uncertainty, a booming cottage industry of analysis has sprung up, mostly produced and sold by by foreign analysts. They sift through obscure though often more reliable government data — looking at things such as electricity usage, rail freight traffic, air travel and the footage of floor space being built.

Even the man taking over China’s economy — incoming Chinese premier Li Keqiang – supposedly disparaged his government’s Gross Domestic Product figures, calling them “man-made,” according to a leaked U.S. diplomatic cable from 2007, and saying he too relies on rail-cargo, bank-loan and power-consumption data.

But that information can also be manipulated. Once power consumption became a favored indicator, for example, news reports began circulating last year that tallies of coal usage and other elements of energy production were being tampered with.

Part of the reliability problem is the government’s near-total control over numbers, in some cases meaning an outright ban on all others collecting competing data, experts say.

“You take something as innocent as weather data,” said Anne Stevenson-Yang, co-founder of J Capital Research, a Beijing-based equities analysis firm. “It’s treated like a national security issue. . . . No one can collect or release numbers but the government.”

Another problem is the lack of transparency for statistics that are released. Methodology is never disclosed, and the underlying data used to arrive at the final numbers are equally rarely revealed.

For that reason, there are some who believe China’s leaders deliberately manipulate the numbers with firm goals in mind: They push the GDP up to meet pre-announced targets and keep unemployment and income gap numbers down to avoid embarrassment and public rage.

Question of interpretation

Others have a more charitable view — that China’s long-ridiculed National Bureau of Statistics is trying its best, in a chaotic market transitioning from a socialist economy dominated by monolithic state enterprises to a more modern and market-driven one, to provide numerical disclosures closer to international standards.

Take the GDP, for example – long a politically fraught number in China. For decades local leaders have sent unrealistically high figures to Beijing to look good and secure promotion. But the national GDP figure is always lower than the combined total of the provinces’ numbers.

Does that mean that someone in power is trying to get a more accurate reading by somehow adjusting for their exaggerations? Or does it simply prove the arbitrary nature of China’s GDP as it’s adjusted to follow prescribed targets?

Similarly open to interpretation is the latest rich-poor gap measurement. The mid-January release of the figure, called the Gini coefficient, came as a shock to many because it’s been more than a decade since authorities have posted new estimates. Many believe officials feared such data would provoke a populace already angry about inequality and the opulent lifestyles of government officials.

When China unveiled a new figure for last year showing improvement over the past few missing years, there was no shortage of skeptics. The government pegged China’s Gini coefficient at 0.474 on a scale of 0 (for a perfectly equal society) to 1 (for utter inequality). Critics quickly pointed to a study released just one month earlier by a Chinese university research center estimating the 2010 Gini at 0.61.

“I can’t say why their result was so much lower than ours since the national bureau didn’t make their sampling procedure public,” said Yin Zhichao, vice director of the research center at Southwestern University of Finance and Economics.

Obstacles to accuracy

Even skeptics of the government, however, acknowledge the inherent difficulties that its statisticians face in China.

Determining the rich-poor gap, for example, requires knowing the real wealth of China’s uber-rich – a Sisyphean task given the large amount of “gray income” that goes unreported because it’s earned through shady or outright illicit means.

Between the black and white views of China’s numbers, however, is a popular middle-road assessment: that China’s statisticians are reforming over time, and they’re deliberately smoothing out the data to avoid jarring comparisons.

“Say they do calculate one year’s stats perfectly,” said Yukon Huang, former China director for the World Bank. “It might make last year’s look horribly wrong, completely incongruous, because to correct your calculation you’ve changed definitions, measurements. So what do they do? Adjust it gradually, smooth out the trends.”

For economists in this camp — the stats are bad, but getting better — all the attention on China’s numbers can only lead to good.

“I think they’re under a lot of heat right now,” Huang said. “But because of the scrutiny, the hope is you’re going to get better numbers as a result.”