How serious is China about overhauling the banking system?
At the beginning of the reform process, China was very big on overhauling the SOEs, but now the emphasis seems to have very definitely turned to the banks. The first tangible step came in 1999 when the government disposed of a huge chunk of legacy through four AMCs (Asset Management Companies) set up under the four state-owned banks. These NPLs, amounting to Rmb1.4 trillion, all related to pre-1995 loans when bank lending had been completely state-directed. At that point, the banks kept two types of NPLs on their books - those that had been were generated through their own mistakes post 1995 and the very worst category, which had to be written-off.
What do you think of the government's 'official' NPL figures?
For 2001, the Chinese government reported total NPLs of 25.3% and then said the figure had dropped by 2% by the end of the first quarter of this year. However, we think that the true figure is somewhat higher and have applied our own analysis using those banks listed in Shanghai as a useful barometer for the system as a whole. These four Shanghai-listed banks - Shanghai Pudong, Bank of Communications, Shenzhen Development Bank and China Merchants Bank - all have to prepare their accounts to both Chinese Gaap and international accounting standards. What we found was that there was a roughly 25% understatement in NPLs between the two accounting systems.
This is largely because the Chinese system has four classifications for loans and the international system has five. Under the Chinese system, loans are typically classified as either normal, overdue or loss-making. However, until 2000, overdue loans were classed on a 180-day rather than 90-day basis and very few banks willingly put loans into the loss-making category. To do this, they also had to secure Ministry of Finance approval as the government was very concerned about moral hazard towards the weak SOE sector. As a result, the government limited banks from writing-off loans to individual borrowers that amounted to more than 1% of the their total portfolio.
The other main problem with the Chinese system is that loan classification is backward-looking rather than forward-looking. This means that there's no analysis of a borrower's general health or ability to pay. So to get an absolute total for system-wide NPLs, it's not only important to take into account the 25% understatement between the two accounting systems, but also factor in the lack of truly clean borrowers on the Mainland. We applied a 50% haircut and came out with a total NPL figure of 46%.
Having said all this, the Central Bank has been steadily applying more stringent reporting requirements and from 2002 all banks have to classify their loans using the international system.
What about provisioning requirements?
Some of the better banks might be coming up to about 40%, but the central bank hasn't really spelt out its general provisioning requirements yet.
Some people have been suggesting that the solution to China's banking problems is for the government to step and inject new capital. What do you think?
If the government lets the banks transfer out a whole chunk more of NPLs and then simply injects new capital, what incentive would there ever be for them to reform their lending practices? We think the government will continue to hold the purse strings very tightly, but this doesn't mean it wont come in and help when necessary.
Do you think the NPL problem might be mitigated by strong loan growth?
There is good loan growth in China at the moment, but credit is very tight and not everyone gets those loans. The central bank continues to control the loan margin and deposit rate, so it's difficult for banks to compete and attract customers.
The SMEs (Small and Medium Enterprises) and joint-ventures are often the most clean from a lending perspective, but they find it difficult to get funding. In Korea, the Industrial Bank of Korea was set up to cater for a sector which the big banks were never that interested in devoting resources to while there was little initial return . But China has never set anything similar up for ideological reasons. So banks have not really developed a client relationship with this sector at all and they still dont know to. The SMEs also have a problem because most Chinese lending is usually done on a collateralised basis rather than a credit or secured basis.
It's also difficult getting realistic valuations of the collateral isnt it?
Yes, although there are signs that the rules on asset transfers to foreigners are being relaxed. The big stumbling block was the fact that the government didn't want the market to determine the value of assets. It didn't want to lose face or be accused of selling assets cheaply to foreigners.
How successful do you think the government has been in trying to overhaul the system so far?
The government's aim is to get the banks to act responsibly on their own. In terms of understanding all the problems, the government certainly adheres to principles of best market practise and has been well coached by the likes of the World Bank. Everyone knows the talk. The problem is how far this has reached down to the local level. ICBC, for example, has 30,000 branches. I'd say it's very questionable how far anything has permeated down.
So you're not really that optimistic then?
No I'm not, although I don't think the Chinese banking system is on the verge of immediate collapse. Over the course of the next 10-years, there are a number of scenarios that could play out. The worst case scenario would be total collapse and of course it goes without saying that this would be very unfortunate. The best case scenario would be a completely clean banking system, but I don't think this is very likely either.
There's a gradation of different outcomes, depending on how quickly the banks can reform themselves and how committed the government remains to reform. I think this is where the real danger lies. There tends to be a lot more talk than action in China. The domestic banking system is very very hierarchical and it's going to be hard turning them into commercial entities. There's a danger that once they get to a certain point, they will start to think, "ok we can relax now and go back to the old ways."
