TOKYO/SYDNEY/NEW YORK (Reuters) – When TAL Education Group, a mid-sized tutoring services firm in China, reported a three-fold increase in its 2018 third-quarter net income earlier this year, few people made much of it.

Rather, investors fretted over a weak performance in China from Apple Inc – a classic yardstick for measuring demand and the health of the world’s second-largest economy.

For Brendan Ahern, the New York-based chief investment officer at Krane Funds Advisors, TAL’s results showed that, while China may be showing signs of pain from a long-running U.S.-Sino trade war, parts of its economy are still doing well.

“With some study of China, you can get results that are very contrary to this view of China as one entity,” said Ahern.

Ahern is not alone. Foreign investors keen to get a grasp of China’s economy have for years distrusted official data.

Many have models built on inputs such as auto sales or air traffic and even indicators such as commentary from food and beauty firms.

“Many people look for alternative data sources in China as proxies for growth, rather than the just the official figures,” said Ross Hutchison, a fund manager at Aberdeen Standard Investments in Edinburgh. “I find it helpful to consider ‘micro’ stories, where you can get a better sense of actual activity.”

In 2018, even as the trade war sapped exports and investment at home, China’s official GDP growth was 6.6 percent – as always, unerringly within official targets.

That was a 28-year low for China, but still almost double the global average of 3.7 percent.

“There are always questions about it and there are always doubts raised about it,” said Michael O’Rourke, chief market strategist at JonesTrading in Greenwich, Connecticut.

“U.S. investors for the most part put a little haircut on it. If the number is X, they suspect the real number is X minus something, and then it is more about trying to identify the trend of numbers rather than the absolute levels.”

What’s more, for an economy that exceeds $10 trillion and a population of nearly 1.4 billion, analysts says China’s economic statistics are released too swiftly for investors to have faith in their accuracy, leading them to read the economic tea leaves elsewhere.

China’s National Bureau of Statistics did not respond to a faxed request for comment.

FAKES, FILTERS

Around the same time that Apple’s disappointing iPhone sales were making investors nervous about China, Aberdeen’s Hutchison was looking at athletic apparel firm Nike Inc’s estimate-beating results and double-digit sales growth in China.

“To me, it suggests that the worries around Apple are not a canary in the coalmine for faltering Chinese growth, but that increasingly expensive iPhones just aren’t as cool as they used to be,” said Hutchison.

Peter Bye, a portfolio manager on the U.S. equity team at UBS Asset Management, agrees some filtering of data is needed.

Bye cites the example of the 2013 crackdown by Chinese authorities on gift-giving by officials, which skewed demand for luxury brands.

“Therefore, China growth numbers out of some high-end brands again weren’t that indicative of China consumer health.”

His discretionary list of data points includes travel commentary from booking platforms as well as commentary from beauty and food companies around the region pandering to Chinese travellers and customers.

For U.S. hedge fund manager Teddy Vallee, founder of Pervalle Global, the correlations between China’s credit growth and other markets such as global crude and copper are material, so he looks at China’s base money or M1 growth.

“The issue today is we haven’t seen M1 turn higher, so it’s quite difficult to be constructive until this turns,” Vallee said.

For a graphic on Proxy economic indicators for China, see – tmsnrt.rs/2Ch1a4i

BIRD’S EYE VIEW

The Economist magazine’s Li Keqiang index was inspired by comments from the Chinese premier a decade or so ago that the official GDP was ‘man-made’. Li based his preferred measures of economic growth on bank loans, rail freight and electricity consumption.

But as China weans its economy off a reliance on manufacturing and heavy industry, analysts have found the Li Keqiang index needs additional measures.

Unfortunately, China’s modern economy, more represented by internet payment platforms and sales systems such as WeChat and Alipay, isn’t yet openly sharing data on online traffic.

Jian Shi Cortesi, an Asian equities portfolio manager at GAM Investment Management in Zurich, tracks a China Satellite Manufacturing Index, developed by San Francisco-based SpaceKnow.

It analyses industrial facilities in China using imagery from space and algorithms measuring the level of manufacturing activity.

Qinwei Wang, a London-based senior economist at European asset manager Amundi, uses in-house models incorporating freight and passenger traffic data, floor space under construction and electricity consumption.

“These data are more independent from official GDP and headline data, of high frequency, difficult to manipulate by officials, and cover the relatively broad economy,” said Wang.

For many analysts like Sarah Shaw, a Sydney-based global portfolio manager at 4D Infrastructure, nothing beats boots on the ground.

“If you’re wandering around in Beijing and you can’t get into a restaurant or you can’t get on a train because they are all full, that’s an indication that things are still relatively okay.”