Will NPLs haunt Bank of China HK and a future US listing?

Pages 213-215 in the Bank of China HK prospectus are definitely worth reading.

One interesting aspect of the Bank of China (Hong Kong) IPO is the manner in which it has transferred the bulk of the non-performing loans to its parent. It may well be that complications in this process were one of the primary reasons for cancelling its US listing.

The Bank of China has readied its Hong Kong subsidiary for IPO by taking much of the rubbish off its balance sheet and putting it into vehicles it owns.

This process began in 1999 when HK$26.9 billion ($3.45 billion) worth of loans were injected into Zhong Gang, a wholly-owned subsidiary of Bank of China. It paid its Hong Kong subsidiary HK$21.8 billion for these dud loans.

In June this year it followed with a further purchase, this time paying HK$8.7 billion ($1.1 billion) for loans with gross book value of HK$11.4 billion ($1.46 billion). These have been injected into BOC Cayman.

The loans in question are reckoned to be China-based, and have left the Bank of China (HK) as something of a pure-play Hong Kong bank.

But what does this mean for Bank of China?

The big four Chinese banks have set up asset management companies to recover NPLs. The average recovery rate last year for Huarong, Cinda and Great Wall was 30.5%.

This suggests that Bank of China will recover HK$11.8 billion from the (book value) of HK$38.3 billion of loans. It paid HK$30.57 billion for these loans and so will be down HK$18.89 billion on the deal.

That equates to $2.42 billion. This, it can be argued, was the cost to the parent of getting the IPO off the ground.

Bank of China will sell 25% of its subsidiary and if it does so at the mid-point of the range (HK$8.21) that would equate to raising $2.41 billion.

Bank of China, by this reckoning is flat. It has lost $2.42 billion on buying NPLs and has been given $2.41 billion by equity investors.

One way of looking at this is that it has eliminated an NPL problem in exchange for 25% of its Hong Kong subsidiary.

Things get more interesting when you ask whether the bank will list in the US any time soon. Previous IPOs from China have viewed a New York listing as de rigeur and Bank of China (HK) would view an ADR as the logical next step after its Hong Kong listing.

But here's where the NPLs could come back to haunt the bank. The prospectus is not entirely clear on this issue, but it is clear enough for someone to smell a problem will exist down the line.

Page 213 of the prospectus deals with the first sale - to Zhong Gang in 1999 - and states: "Although this sale resulted in the derecognition of the loans from our Hong Kong GAAP financial statements from the transfer date, we believe that similar accounting treatment would not be achieved under US GAAP. In addition, we can provide no assurance that this sale would result in a derecognition of the loans under the generally accepted accounting principles prevailing in other jurisdictions."

On page 215 it deals with the sale in June to BOC Cayman. It states: "This sale will result in the derecognition of the loans sold from our Hong Kong GAAP financial statements and we believe that a similar accounting treatment would be achieved under US GAAP".

This sounds like rather a contradictory situation. The first sale is not recognized by US GAAP, and the second is. Why? Well, under US GAAP if you sell loans to a vehicle controlled by the major shareholder, they cannot be derecognized. However, if you sell the loans to a vehicle that is structured as a partnership then the loans can be derecognized. This technique was honed by Enron to a fine art, and for lack of any better explanation in the prospectus is reckoned to be the way BOC Cayman is structured.

This would tend to suggest that while the most recent sale has succeeded in getting the classified loan ratio down to 9.48%, the much larger HK$21.85 billion of bad loans would return under a US GAAP treatment. That is to say, they would return at the time of a US listing.

What then becomes important is just how much of the HK$21.85 billion has been recovered and written off. In the worst case scenario - ie zero achieved - a US listing would take Bank of China HK's classifed loans back up above 14%.

A happier scenario would be that Bank of China has recovered 30% (HK$6.5 billion) but then has to write off the remaining HK$15.35 billion in the months before the US listing of its subsidiary.

Whatever it doesn't write off will come back onto Bank of China HK's US GAAP balance sheet like a boomerang.

Everyone is in the dark about this as the prospectus gives no guidance whatsoever as to how successful the Zhong Gang vehicle has been in recovery, or of the parent's write offs. In the new age of transparency we allegedly live in, the prospectus is a throwback to the past in this respect. The information has no bearing on the Hong Kong listing, but it will (to repeat) have a bearing on a future US listing and has the potential to negatively affect the stock price at that future point.

It is difficult to get anyone to talk about Bank of China (HK). A large number of top analysts are on business trips or convenient holdays (ie those from Merrill, CLSA and Fox-Pitt Kelton). One analyst mentioned that the leads had done a good job of tying up a lot of analysts in the deal and preventing them speaking.

One exception is Macquarie, whose analyst Simon Ho just put out a report yesterday. Titled 'More dream than reality', the report assigns a fair value to the IPO of HK$7.50 a share (below the mid-level of the bookrunner's range) which equates to 1.5 times book value.

Ho is cautious on Bank of China HK on a few fronts. He worries about conflicts of interest arising between parent and subsidiary; and he notes the merger of the 12 banks to create BOCHK is "arguably one of the most complex mergers ever globally".

He concludes of the loan portfolio: "There are few assurances that all the problems of the past have been fully dealt with."