Wednesday, September 03, 2008
My latest contribution to the RGE Monitor
Bundling in China
Email PrintShare
Delicious Digg Facebook reddit Technorati
Victor Shih | Sep 3, 2008
Due to strong political pressure at the highest level and seemingly declining inflation, the State Council caved and increased the credit quota by some 200 billion RMB. Well, that only goes so far, and much of it still goes to larger firms. So, how are they dealing with the continual liquidity problem? Bundling!! Local governments, including Sichuan, Chongqing, Henan, Beijing, Liaoning, Zhejiang, and Shenzhen, are all planning to issue tranches of corporate bonds whose cash flow comes from a group of small and medium enterprises (SMEs). Each province will issuing 1 to 2 billion RMB of notes for the approved SMEs. The local governments will guarantee these notes, which have 3-5 years maturity!! This is a familiar scheme of borrowing to fulfill current policy needs and leaving bad debt for future leaders of a province or city. This is why the central government banned local governments from issuing debt, but it is coming back in a latent form. Granted, it is on a small scale now, but it can really take off. I think the NDRC is backing this effort, though I am not sure if the financial regulators in Beijing like this. This will also create good business for domestic investment banks, especially those with local government ties. It might also give a boost to state owned asset management companies which are trying to transform themselves into investment banks.
Great, everybody wins, right? We must revisit the question of whether bundling a few highly leveraged companies together makes it better than one highly leveraged company issuing debt? We eventually found out in the US that such bundling, even if done in a very clever way, does not reduce fundamental risks. But I think in China, they know of the risks-- they just don't care because in three years' time, it will be someone else' problem. Perhaps those more familiar with this kind of product can comment further on its risks.
Generally, the recent decision to increase credit quota has made the signal from the central government much more ambiguous and subject to local interpretation. Instead of fighting inflation and reining in investment, growth and survival of SMEs are also paramount policy goals now. I have always argued that ambiguous signals are very dangerous in the Chinese context. The local regulators are already on the side of the local governments and businesses. If the center gives even a hint of loosening, local governments will encourage local regulators to forget all about the rules. In Shenzhen, the local CBRC office is thinking up all kinds of "methods" of financing SMEs, including healthy measures like allowing city commercial banks elsewhere to do business in Shenzhen. Less healthy suggestions include using government money to set up a fund to invest in SMEs and allowing lower level branch banks to operate side investment and small loans companies. Matt Miller, a former Bloomberg reporter who has researched the Kaiping case extensively, can tell you all about side investment and loan companies operated by branch banks. We basically saw an explosion of NPLs and corruption scandals as local bankers siphoned billions from the banking system.
The CBRC in Beijing is also exploring "flexible" policies, such as allowing commercial banks to serve as quasi investment banks to provide funds and guarantees for companies that merge. Basically, instead of allowing bankruptcies, which would increase NPLs, the CBRC would rather that banks force good companies to take over the bad ones. Sounds familiar? That is what they did in the 90s with SOEs--now the same formular is applied to teetering real estate companies and SMEs. Why are they doing this? Isn't everything fine? The first half reports of banks all show spectacular profit and falling NPL ratios. Well, if we look a bit more carefully at overdue loans (definition varies, but typically lapsed interest payment over 90s days), we see that they are on the rise. Overdue loans for Industrial Bank, for example, has gone up over 20% in the first half. Accounting tricks are holding the NPL figures low for now, but by first quarter next year......
Bundling in China
Email PrintShare
Delicious Digg Facebook reddit Technorati
Victor Shih | Sep 3, 2008
Due to strong political pressure at the highest level and seemingly declining inflation, the State Council caved and increased the credit quota by some 200 billion RMB. Well, that only goes so far, and much of it still goes to larger firms. So, how are they dealing with the continual liquidity problem? Bundling!! Local governments, including Sichuan, Chongqing, Henan, Beijing, Liaoning, Zhejiang, and Shenzhen, are all planning to issue tranches of corporate bonds whose cash flow comes from a group of small and medium enterprises (SMEs). Each province will issuing 1 to 2 billion RMB of notes for the approved SMEs. The local governments will guarantee these notes, which have 3-5 years maturity!! This is a familiar scheme of borrowing to fulfill current policy needs and leaving bad debt for future leaders of a province or city. This is why the central government banned local governments from issuing debt, but it is coming back in a latent form. Granted, it is on a small scale now, but it can really take off. I think the NDRC is backing this effort, though I am not sure if the financial regulators in Beijing like this. This will also create good business for domestic investment banks, especially those with local government ties. It might also give a boost to state owned asset management companies which are trying to transform themselves into investment banks.
Great, everybody wins, right? We must revisit the question of whether bundling a few highly leveraged companies together makes it better than one highly leveraged company issuing debt? We eventually found out in the US that such bundling, even if done in a very clever way, does not reduce fundamental risks. But I think in China, they know of the risks-- they just don't care because in three years' time, it will be someone else' problem. Perhaps those more familiar with this kind of product can comment further on its risks.
Generally, the recent decision to increase credit quota has made the signal from the central government much more ambiguous and subject to local interpretation. Instead of fighting inflation and reining in investment, growth and survival of SMEs are also paramount policy goals now. I have always argued that ambiguous signals are very dangerous in the Chinese context. The local regulators are already on the side of the local governments and businesses. If the center gives even a hint of loosening, local governments will encourage local regulators to forget all about the rules. In Shenzhen, the local CBRC office is thinking up all kinds of "methods" of financing SMEs, including healthy measures like allowing city commercial banks elsewhere to do business in Shenzhen. Less healthy suggestions include using government money to set up a fund to invest in SMEs and allowing lower level branch banks to operate side investment and small loans companies. Matt Miller, a former Bloomberg reporter who has researched the Kaiping case extensively, can tell you all about side investment and loan companies operated by branch banks. We basically saw an explosion of NPLs and corruption scandals as local bankers siphoned billions from the banking system.
The CBRC in Beijing is also exploring "flexible" policies, such as allowing commercial banks to serve as quasi investment banks to provide funds and guarantees for companies that merge. Basically, instead of allowing bankruptcies, which would increase NPLs, the CBRC would rather that banks force good companies to take over the bad ones. Sounds familiar? That is what they did in the 90s with SOEs--now the same formular is applied to teetering real estate companies and SMEs. Why are they doing this? Isn't everything fine? The first half reports of banks all show spectacular profit and falling NPL ratios. Well, if we look a bit more carefully at overdue loans (definition varies, but typically lapsed interest payment over 90s days), we see that they are on the rise. Overdue loans for Industrial Bank, for example, has gone up over 20% in the first half. Accounting tricks are holding the NPL figures low for now, but by first quarter next year......
Comments:
Post a Comment