How analysts calculate China’s true — and huge — burden of bad loans


Corporate investigator Violet Ho never put a lot of faith in the numbers on bad loans that are reported by China’s banks.

Crisscrossing provinces from Shandong to Xinjiang, she has seen too much — from the shell game of moving assets between affiliated companies to cover-ups by bankers who are loath to admit that loans they made won’t be recovered.

“If I have one piece of advice for people worrying about the financial status of Chinese companies, it’s this: It’s right to be worried,” said Ho, senior managing director in Hong Kong for Kroll Inc., a U.S. risk consultancy. “Often a credit report for a Chinese company is not worth the paper it’s written on.”

As China’s banking industry persists with publishing delinquent-debt numbers that few have faith in — a survey in 2014 indicated that even lenders didn’t believe them — some financial analysts have turned detective to try to work out the real numbers.

The amount of bad debt piling up in China is at the center of a debate about whether the country will continue as a locomotive of global growth or sink into decades of stagnation like Japan after its credit bubble burst. Bank of China Ltd. on Thursday reported its biggest quarterly bad-loan provisions since going public in 2006.

While the analysts interviewed for this story differ in their approaches to calculating likely levels of soured credit, their conclusion is the same: The official bad-loan estimate of 1.5 percent is way too low.

Charlene Chu, who made her name at Fitch Ratings making bearish assessments of the risks from China’s credit explosion since 2008, is among those crunching the numbers.

While Ho relies on her observations on the road, Chu and her colleagues at Autonomous Research in Hong Kong take a top-down approach. They estimate how much money is being wasted since the nation began getting smaller returns on credit in 2008. Their assessment is informed by data from economies such as Japan that have gone though similar debt explosions.

While traditional bank loans are not Chu’s prime focus — she looks at the wider picture, including shadow banking — she says her work suggests that nonperforming loans (NPLs) may be at 20 to 21 percent, or even higher.

The Bank for International Settlements cautioned in September that China’s ratio of credit to GDP indicates an increasing risk of a banking crisis in the coming years.

“A financial crisis is by no means preordained, but if losses don’t manifest in financial-sector losses, they will do so via slowing growth and deflation, as they did in Japan,” said Chu. “China is confronting a massive debt problem, the scale of which the world has never seen.”

Some analysts, such as Judy Zhang at BNP Paribas SA and Patricia Cheng at CLSA Ltd., start micro by trawling through the balance sheets of thousands of listed companies, assessing how many can’t afford to service their debt. Zhang estimates a bad-loan ratio of about 9 percent for Chinese banks listed in Hong Kong — a group that includes Industrial & Commercial Bank of China Ltd., the world’s biggest lender. Cheng’s model spits out a ratio of at least 6.1 percent for the entire industry.

Slicing and dicing the official loan numbers, Christine Kuo, a senior vice president of Moody’s Investors Service in Hong Kong, focuses on trends in debts that are overdue for at least 90 days, rather than those classified as “nonperforming.” Another tactic some analysts use is to add nonperforming debt to “special mention” loans — those that are overdue but not yet classified as impaired, yielding a rate of 5.1 percent.

Banks’ bad-loan numbers are capped by “evergreening,” the practice of rolling over debt that isn’t repaid on time, according to experts including Keith Pogson, a Hong Kong-based senior partner at Ernst & Young LLP. Pogson was involved in restructuring debt at Chinese banks in 1998, when their NPL ratios were as high as 25 percent.

The issue of unreliable data is one that the industry itself is well aware of. In a 2014 survey of bank executives by China Orient Asset Management Corp., one of the nation’s bad-loan managers, only 17 percent believed that the official bad-loan data were in line with reality, while 38 percent saw the risks as “severely” underestimated and 43 percent as “slightly” underestimated.

A clearer picture may yet emerge. A shift in 2018 to accounting standards called IFRS9 will force Chinese banks to provide better numbers, according to partners at Ernst & Young and PricewaterhouseCoopers LLP. It will be a “night-and-day” change, because of requirements for a forward-looking assessment of loan quality, Pogson said.

The looming deadline means banks will likely keep accelerating disposals of bad credit to avoid an embarrassing surge in nonperforming loans when the new rules take effect, he said.

For Ho, the best way of assessing the financial health of opaque private companies is to focus less on the numbers they report and more on face-to-face checks with industry peers, suppliers, partners and customers.

She credits that approach for uncovering, among other murky practices, firms trying to import commodities just so they could use them as collateral for borrowing more — even as they were cutting staff and production and having trouble paying bills.

One thing she is sure of: Demand for her investigative skills will stay healthy for years to come.

  • Karly Johnston

    Nazis burn books, communists cook the books.