Considering how such companies operate in countries with economic challenges similar to those in Thailand could give a broader understanding of how the Thai Asset Management Corporation (TAMC) may soon operate.

AMCs have been around for many years and in many countries, notably in the United States. Its Resolution Trust Corporation (RTC) was set up following the collapse of about 750 Savings & Loans Associations in 1989.

The RTC oversaw the collection of the underlying loan portfolios comprising mainly property loans. Mexico, Sweden, Korea, Malaysia and Indonesia have adopted a similar approach. Models differ, but the overall aims of an AMC are to provide an efficient debt-collection agency, and remove problem loans from banks.

Details of the proposed structure for the TAMC can be found at the Finance Ministry's web site (www.mof.go.th) and are summarised below. There will be three managerial layers:- Policy level: a supervisory committee with members of the government, private sector and experts. It will set policy and approve major restructuring cases.

About 1.1 trillion in non-performing loans will be transferred to the TAMC from Bangkok Metropolitan Bank, Siam City Bank, BankThai, Sukhumvit AMC and Bangkok Asset Management (the AMC dealing with the NPLs of Bangkok Bank of Commerce). Another 250 billion baht will come from private-sector banks. The loans will be transferred at net book value, meaning the amount of the loan minus provisions made. Assuming that the loans have all been NPLs for at least a year and that the collateral is generally land and buildings, this effectively means that the loans will be transferred at 90% of the underlying collateral value. Providing the underlying collateral has been properly valued at market value, then there should be no further losses unless collateral values decline further. Many commentators believe, however, that banks have generally overstated collateral values.

Gains and losses will be split as follows:- Losses up to 20% of net book value will be the responsibility of the bank;- Losses between 20% and 40% of net book value will be shared evenly between the bank and the TAMC;- Losses in excess of 40% will be borne by the TAMC;- Gains up to 20% of net book value will be shared evenly between the bank and the TAMC; and- Gains in excess of 20% will be credited by the bank.

The TAMC will issue notes to participating banks to pay for the loans. The Financial Institutions Development Fund will guarantee the notes. The TAMC will redeem the notes over a 10-year period starting in the second year but not exceeding 10 years. Settlement of gains and losses with banks will occur during the fifth and 10th years.

The proposed structure for the TAMC probably most closely resembles that of Danaharta in Malaysia. NPLs are transferred to Danaharta without the debtors' consent and at market value. Funding is also through the issue of zero-coupon loan notes issued to institutions selling the NPLs.

Malaysia, Indonesia and Korea all adopted a centralised AMC model in response to the fallout from the 1997 regional crisis. Thailand, by contrast, began with a decentralised model for banks and a centralised model, the Financial Sector Restructuring Authority (FRA), for closed finance companies.

One of the key issues now under discussion is the price at which NPLs are to be transferred to the TAMC. While for state-owned banks the transfer price is a zero-sum game, there is an important issue for privately owned banks that could either lose or gain.

In Korea and Malaysia, NPLs have been transferred at market value so that the state does not ultimately end up bearing the loss. In the case of Danaharta, an external review is done before the transfer price is agreed.

Once the assets are transferred they are dealt with very differently in each country (see chart). Korea's Kamco has adopted a rapid disposal strategy with assets bundled into smaller portfolios and auctioned. Purchasers have generally been foreign specialists such as Lone Star, Lendlease, GE Capital and Lehman Brothers.

In contrast, Danaharta and Ibra have longer-term strategies. Given the long-term nature of these workouts it is difficult to assess which is the most effective approach, especially as the quality and nature of portfolios varies significantly.

The TAMC should result in more efficient collection for many reasons:- Cost savings through having only one organisation dealing with the debtor rather than several banks each having to allocate resources.

- A shorter negotiation process with only one bank representing the banks participating in the TAMC structure.

- Combining bank claims should put the TAMC in a position to force through a restructuring on its terms, either through the central bank's Corporate Debt Restructuring Advisory Committee (CDRAC) or business reorganisation processes. Again, this will save time.

- Special enforcement powers will enable the TAMC to foreclose on the assets of unco-operative debtors in a short time, expected to be a few months compared with several years for banks using the existing legal processes.

The shortened foreclosure process may be a clear recognition by the government that this is a very effective weapon for creditors to use in resolving non-performing loans. Many people believe that if the foreclosure process had been shortened immediately following the baht devaluation in 1997 then Thailand would have made much more progress in resolving NPLs.

The need for legislative change in the foreclosure law is not unique to Thailand. The Korean government revised the foreclosure process so that banks can now foreclose on a debtor's assets three months after the debtor defaults.

A review of the foreclosure law is required, not for the benefit of the foreign banks that have largely dealt with their problems, but to help Thai banks sort out the remaining NPLs. An effective foreclosure process is also an encouragement to lend. Banks want to lend, but there are few creditworthy customers and they are reluctant to take on the more risky propositions.

But transparency is a key issue. The success or otherwise of the TAMC will depend very much on the expectations created. The follow-up to the Mexican financial crisis in 1994-95 is a warning as to how matters can go wrong. Following its financial crisis, Mexico's government set up a centralised agency, the Bank Fund for the Protection of Savings (Fobaproa), to help bail out the banks. A few years later the cost of the bailout had risen from initial estimates of US$15 billion to closer to $65 billion and there were widespread allegations of cronyism and fraud. The need for transparency and accountability is a key issue in the operations of the TAMC in order to cull support from long-term investors.

The TAMC alone will not resolve many of the problems facing the banking sector, such as the lack of creditworthy customers and the risk of re-entry NPLs. However, experiences in other countries suggest that provided the creation of a national AMC is accompanied by other reforms in accounting, legal and regulatory frameworks, then significant progress can be made.

Many changes have already been made to bring Thai accounting standards in line with international accounting standards, strengthening bankruptcy laws and upgrading banking supervision so much of the groundwork has already been completed.

Much of the restructuring in the private sector in Thailand to date has been superficial with a significant amount of financial rescheduling. The restructuring of NPLs should involve significant changes to these businesses as they dispose of non-core assets, remove outdated working practices, bring in new management and generally seek to become more competitive. These changes will challenge many vested interests, and the willingness of TAMC to impose these changes will be closely watched by many analysts.