III. Design and Implementation of The Thai asset Management Corporation1
1. To help alleviate weaknesses in the bank and corporate sectors (Chapter II), the authorities established the Thai Asset Management Corporation (TAMC) in June 2001. The TAMC was designed to consolidate the management of distressed assets in the public sector, and provide an impetus to the restructuring of large multi-creditor corporate loans. It is expected to purchase up to one-half of the financial sector’s distressed assets, of which a large majority (some 80 percent) is expected to come from state-owned financial institutions.
2. The establishing law provides the TAMC with a flexible and powerful framework for the management and resolution of distressed loans. The TAMC can choose among various loan resolution strategies, including debt restructuring, which aims to improve a debtor’s balance sheet, and more intrinsic business reorganization, which aims to revive a debtor company so that it can recommence servicing its debts. To support these strategies, the TAMC is endowed with special legal powers to bypass weaknesses in the existing court-based framework for debt restructuring. Particularly notable are TAMC’s powers to foreclose on the collateral of recalcitrant debtors, and to place cooperating debtors into a streamlined business reorganization proceeding.
3. However, the TAMC’s success will crucially depend upon the implementation of its mandate. Experience in other countries illustrates that the success of a centralized agency is determined by several crucial factors. These include: (i) maximizing the use of private sector expertise in the management of bad loans; (ii) applying commercial criteria and the principle of value maximization to restructuring terms; (iii) using special legal powers in a consistent and even-handed manner; and (iv) minimizing political interference and maximizing transparency of operations.
4. This chapter describes the TAMC’s key features and objectives, the conditions under which it could be successful, and its likely financial impact. The next section discusses the TAMC’s main features as laid out in the establishing law. Section C then draws on the experience of other asset resolution entities in the region to determine the conditions under which the TAMC could be successful (Annex I compares and contrasts country experiences in more detail). Section D provides a preliminary assessment of the impact of the TAMC on banks and government debt, and Section E concludes.
B. Key Features and Objectives of the TAMC Law
5. The TAMC’s establishing law was formally enacted in June 2001.2 The law is comprehensive and highly detailed, containing both general principles and specific operational provisions. About one-third of the law’s articles focus exclusively on the management of nonperforming loans (NPLs) and the administration of special legal powers. However, the law also mandates the separate issuance of operational guidelines to spell out more fully the TAMC’s approach to debt and business restructuring. These guidelines are currently under preparation.
6. The objectives of the TAMC are defined somewhat broadly. Emphasis is put on the revival and continuation of businesses—to enable them to repay their debts—with a view to fostering national economic recovery. In addition, the law gives prominence to the speedy resolution of NPLs, containment of operating costs, and supportive treatment of cooperating debtors.
Eligibility and Pricing of Loans
7. The TAMC is to acquire only large, multi-creditor loans from private banks, while it is expected to acquire a potentially large share of state bank distressed assets. The TAMC is given statutory powers to buy assets with certainty of title (by-passing the need to obtain borrowers’ consent), and the law mandates that the transfer of assets be a onetime operation. Eligibility for private banks is limited to collateralized loans of at least B 5 million ($110,000), and involving more than two creditors. State-owned institutions are expected to transfer both multi-creditor loans and uncollateralized single-creditor loans. Thus, loan transfers from state institutions could reach over 80 percent of total asset transfers to the TAMC (see Box III.1). Participation by private banks is voluntary. However, the law provides a strong incentive for participation by requiring banks that fail to transfer eligible NPLs to reappraise their holdings of collateral (by an independent appraiser appointed by the Bank of Thailand), and to top off any resulting provisioning shortfalls.
8. The purchase price of loans by the TAMC is to be set equal to the value of collateral, with future gains or losses shared with the originating institutions. Collateral appraisals are to be carried out according to procedures set by the TAMC.3 The price received by private banks is not to exceed book value minus regulatory provisions.4 Meanwhile, the purchase price of uncollateralized NPLs of state institutions has not yet been determined. Gains and losses from debt restructuring are to be shared over time with the originating institutions according to a loss-sharing formula. The formula implies that banks’ exposure to losses realized in the TAMC is capped at no more that 30 percent of the transfer price (see Box III.2 for details). Financial institutions are expected to refund, pro rata, their portion of the TAMC’s operating and funding costs. In the event the TAMC realizes gains, most of these will accrue to the originating institutions. Losses will be settled at the end of the 5th and 10th years of TAMC’s operations by reducing the face value of the note it issued to purchase the NPLs (see below). Private banks also have the option of settling losses by issuing ordinary shares to the TAMC.
9. The FIDF will provide the TAMC with the initial capital, and will guarantee bonds issued to purchase NPLs from financial institutions. In addition to receiving an initial capital injection (of B 1 billion, or $22 million), the TAMC has the ability to issue new shares, borrow funds from the market, and extend loans and guarantees. The TAMC will purchase NPLs by issuing FIDF-guaranteed (ten-year) nontradable, callable bonds, with a yield comparable to bank funding costs.5 Losses incurred by the TAMC are initially to be absorbed by the FIDF, and later refunded by the Ministry of Finance.
Loan Management and Restructuring Strategy
10. The law enables the TAMC to manage the NPLs flexibly. It is expected that multi-creditor loans acquired from financial institutions will be managed with bank participation. The TAMC can also outsource the management of NPLs in separate tranches to the originating institutions, or to private asset managers with priority given (by law) to Thai legal entities. Finally, the TAMC can bundle distressed loans and use them as collateral to issue asset-backed securities to investors.
11. The TAMC is authorized to use three different loan resolution strategies:
Debt restructuring, including debt and debt-service reductions, debt-for-equity swaps, and settlement through transfer of collateral or other debtor assets;
Business reorganization with a view to revive debtor companies and enable them to repay their debts;
Foreclosure and disposal of loan collateral to settle the debtor’s obligations.
The law permits the flexible use of these three approaches, so that, for example, elements of debt restructuring, which only affect a debtor company’s balance sheet, can be combined with more fundamental business reorganization. The latter allows for management or staff changes, disposal of non-core assets, and closure or reorganization of business lines. The TAMC will decide on the loan resolution strategy to follow, based on a preliminary assessment of the financial position of the debtor, its level of cooperation, and overall business prospects.
Box III.1.Loan Portfolio Eligible for Transfer to the TAMC
The TAMC could acquire up to VS of system-wide distressed loans, though only a small fraction is from private banks. About B 1.3 trillion ($30 billion) in book value of distressed loans, consisting of over 100,000 individual accounts, are eligible for purchase by the TAMC.1 Of this, over 80 percent in value (B 1.1 trillion, or $25 billion) consists of the NPLs held by various state-owned institutions, including state banks and state-owned loan work-out units. The remainder (B 230 billion, or $5 billion) consists of around 2,500 multi-creditor loan accounts of private banks eligible for transfer (that is, those with an outstanding balance greater than B 5 million and with collateral). The pool of loans slated for transfer from private banks is equivalent to only about 6 percent of their total assets, and about ¼ of their distressed loans. This relatively small share is partly due to the fact that over ½ of all NPLs currently held by private banks is to single creditors. It could also reflect progress achieved thus far in restructuring larger, multi-creditor exposures.
Bank Distressed Assets Transferred to TAMC
The corporate loans from private banks that could be transferred to the TAMC are dispersed across a wide variety of sectors, with industrial loans as the largest share. Data on the distribution of private bank NPLs eligible for transfer to the TAMC indicates that industrial loans account for about 40 percent of the total, with real estate loans accounting for another 16 percent. Moreover, even within the industrial sector, NPLs are fairly disparate, though the steel sector accounts for about of eligible industrial loans. These data highlight the fact that the TAMC will be charged with managing a fairly diverse asset portfolio, making it more difficult for the TAMC to specialize in a particular kind of asset resolution strategy.
The loan portfolios of state financial institution that could be transferred to the TAMC appear more fragmented than those of private banks. In particular, the size per borrower of state institution loans eligible for transfer is much smaller, on average, than is the case for private banks. The exception being for the loans transferred to the SAM (the AMC set up last year to resolve the NPLs of the state-owned Krung Thai Bank), which are of comparable size to those transferred from private banks. Data on the sectoral distribution of exposure by state institutions are not available, although it is likely that their NPLs are also spread across a wide array of retail and industrial type loans.
Size Distribution of Private Banks Assetes to TAMC
Size Distribution of State Banks Assets to TAMC
Box III.2.A Closer Look at The TAMC’s Gain-Loss Sharing Arrangement
Loans from commercial banks will be sold to the TAMC in the context of gain-loss sharing arrangements. Such incentives and risk sharing mechanisms are commonly used internationally in contracts for the sale of distressed financial assets, including recently in Thailand in the context of the sale of the two smaller intervened banks, Nakornthon and Radanasin.
Initial Pricing: In the case of the TAMC, eligible assets from private banks will be sold at a transfer price equaling the lesser of the value of the collateral securing the loan (excluding personal guarantees), or the loan’s net book value. Net book value refers to a loan’s original on-balance sheet value (not including unrecognized accrued interest in excess of three months, or other penalties) less required regulatory reserves as per BOT regulations. From a regulatory perspective, net book value should therefore reflect the currently assessed economic value of the underlying loan obligation.
Gain-loss sharing: The loss-sharing contract between the bank and the TAMC is non-symmetric. In the event that the TAMC realizes a gain, i.e., when the ultimate value realized from a loan is higher than the transfer price (after including operating and interest costs), the originating bank and the TAMC will share equally the first 20 percent of the gains relative to the transfer price, with the remainder accruing to the bank (though the share of total gains accruing to the originating bank cannot exceed the transfer price; any such excess gains will accrue to the TAMC). However, in the event of a loss, i.e., when the value realized from a loan is less than the transfer price (after including operating and interest costs), (i) the originating bank will assume the first 20 percent of additional losses relative to the transfer price, (ii) any losses above 20 percent but less than 40 percent of the transfer price will be shared between the bank and the TAMC, and (iii) losses in excess of 40 percent of the transfer price will be borne entirely by the TAMC. This effectively caps banks’ losses at 30 percent of the transfer price (the first 20 percent plus half the next 20 percent of losses).
Thus, individual banks’ maximum exposure will vary depending on the transfer price which, in turn, is related to the value of collateral and/or existing provisioning for the loan portfolios transferred to the TAMC. As illustrated in the Text Table, banks that transfer their loans at relatively lower prices (line C) are exposed to lower additional losses to be realized in the TAMC (equivalent to the sum of lines D and E). However, transferring at a lower price means effectively that a relatively larger share of losses has already been shouldered by the bank prior to the transfer (in the form of balance sheet provisions—i.e., line B of the Table). Thus, the banks’ maximum loss depends directly on the transfer price: the lower the transfer price (line C), the higher is the ultimate loss to the bank (line G)—and the lower the ultimate cost to the TAMC. That is, banks with a higher appraised value of collateral are exposed to smaller overall losses compared to banks with lower appraised values of collateral. This is why, as in any typical loss-sharing arrangement, the seller (i.e., the bank) has an incentive to sell at the highest price possible. Therefore, fairness in a loss-sharing agreement is achieved when the ex ante transfer price is aligned as closely as possible with the ex post recovery value. In the case of TAMC, this means, putting particular emphasis on ensuring that collateral is appropriately valued.
|C.||Net book value||55||50||45|
|D.||Banks 20 percent solo loss||11||10||9|
|E.||Banks 10 percent shared loss||6||5||5|
|F.||Max cost to TAMC (C - D - E)||39||35||32|
|G.||Max cost to Bank (B + D + E)||62||65||69|
12. The law provides incentives for debtors and guarantors to cooperate in the debt restructuring process. For example, guarantors are to be released from their obligations whenever debtors pledge additional collateral sufficient to cover their residual indebtedness to the TAMC. Moreover, in cases where the debtor has limited ability to repay its debts, the guarantor is discharged of any obligation provided it pays at least two-thirds of the debtor’s outstanding obligations to the TAMC.
Special Legal Powers
13. The law provides the TAMC with the power to foreclose on collateral and liquidate debtor businesses with minimum court involvement. To speed up negotiation with recalcitrant debtors, the law provides the TAMC with enhanced foreclosure powers, enabling it to by-pass the existing slow court and administrative procedures. The TAMC may itself directly seize collateral, and dispose of it by public auction. If the recovery from the sale of the collateral is less than the original obligation, the debtor and guarantor remain liable for the outstanding debt. Moreover, if debtors do not cooperate in the debt restructuring process and are deemed to be fraudulently concealing their assets, the TAMC has the power to liquidate the assets of the debtor or the guarantor by obtaining a final court order to declare the debtor (or the guarantor) bankrupt.
14. The TAMC is also empowered to administer its own business reorganization proceedings outside the current court-based insolvency system. The TAMC may place consenting debtors directly into a five-year business reorganization proceeding, with the role of the bankruptcy court limited only to the approval of the business plan. The TAMC can initiate business reorganization provided it is a majority creditor, it obtains debtor consent, and the company is considered viable or of strategic importance. While the business reorganization plan is implemented, execution against the debtor’s assets is stayed. If the plan fails, the debtor may agree to liquidate its business—in which case the debtor and guarantor are released from their obligation to pay any shortfall from the liquidation receipts.
15. The TAMC is established as a government agency (i.e., not a state enterprise) with a limited lifespan. Therefore, it is not subject to laws pertaining to state enterprises. In particular, as the State Employee Liability Act does not apply, the TAMC’s staff is immune from lawsuits for losses resulting from bona fide debt restructuring.6 Moreover, the TAMC’s operations are further insulated as decisions taken by the TAMC cannot be disputed through the filing of a petition with the Administrative Court. The TAMC is also granted tax exempt status, thus freeing it from paying taxes on asset transfers, and on cash or other asset recoveries. A sunset clause requires the TAMC to cease its operations within ten years of establishment, and to liquidate itself within two years thereafter.
16. Policies and operational guidelines are set by a Board of Directors. In addition to providing policy direction, the Board has the authority to oversee broad implementation. The Board consists of twelve members, appointed by the Finance Minister with Cabinet approval. The Chairman of the Board and at least three other Directors are selected from the private sector. Politicians are not permitted as Board members. Two of the private sector members must be selected from industrial and commercial associations (representing debtors), while the third must come from financial institutions (representing creditors).
17. Operational decisions are taken by a four-member Executive Committee (ExCom). The law entrusts the ExCom with the authority to approve debt restructurings, asset disposals, business rehabilitations, and to take any decision on the management (and outsourcing) of NPLs. Two members of the ExCom, which is appointed by the Board, are from the private sector—with the other two chosen from amongst government officials. The ExCom is headed by a Chairman (CEO), who is concurrently an ex-officio member of the Board, thus ensuring coordination between policy setting and executive decisions within the TAMC. The law also envisages the appointment by the Board of a professional Managing Director, in charge of managing staff and taking routine operational decisions, including with regard to the preparation of debt restructuring plans.
Supervision and Disclosure
18. The law envisages Board appointment of an Audit Committee responsible for internal auditing and supervision. General oversight is to be exercised by the Ministry of Finance, which may delegate to the Bank of Thailand the task of examining the TAMC’s operations and reviewing its financial performance. The law also mandates a performance review of the TAMC after two years of existence. The review, to be carried out by a committee appointed by the Ministry of Finance, is to be submitted to Cabinet which, based on the findings, can decide on whether to overhaul or dismantle the TAMC’s operations.
19. The TAMC is required to produce and publish semi-annual audited financial statements. The Minister of Finance is to appoint the Auditor General, or any other entity, to audit TAMC’s financial statements on a semi-annual basis. The law also requires the TAMC to prepare a business report, including an assessment of past performance and elements of a forward-looking business plan. Prior to publication, audited financial statements and business reports are to be submitted to the Ministry of Finance and, for information, to the Cabinet.
C. Success Factors for the TAMC Derived from International Experience
TAMC Mandate and Incentive Framework
20. As noted above, the TAMC has been provided with a flexible operational framework. It includes the ability to resort to a wide range of approaches for the management of distressed assets, such as outsourcing and securitization. The TAMC will also be able to choose among different asset resolution strategies, including debt restructuring and business reorganization. Importantly, the law provides the TAMC with special legal powers to bypass weaknesses in the legal framework for debt restructuring. And finally, TAMC employees are granted immunity against losses from debt restructuring.
21. The TAMC framework aims to balance incentives for participation of private banks against the objectives of containing taxpayer costs and minimizing moral hazard. For example, by setting the initial transfer price of bad loans equal to the appraised value of collateral, bank shareholders are insulated from taking losses up-front when selling bad loans to the TAMC. Over time, the loss-sharing arrangement allows the TAMC to charge the private banks a portion of any realized losses. The existence of an effective cap on total bank exposure, the long-lags (5 to 10 years) on the settlement of losses, and the provision of a minimum yield on the transferred assets allow bank shareholders to “buy time” while strengthening their banks’ financial position. If banks’ positions deteriorate over time, to the point that they are unable to refund their portion of losses, the TAMC has the ability to obtain compensation by acquiring share ownership in the banks. A key incentive for banks to participate in the TAMC is the provision of special legal powers to leverage debtor cooperation and assist in recovery efforts. The TAMC framework also includes other features to curb moral hazard, including with respect to the terms for loan transfers.7
22. The TAMC law also includes elements of protection for cooperating debtors. The TAMC has the discretion to place consenting debtor companies into out-of-court business restructurings, effectively providing debtors with a “second chance” to repay their debts while retaining their businesses. Other elements of debtor protection include: (i) the need for debtor consent to commence bankruptcy proceedings; (ii) the inclusion of cooperating debtors in the design and implementation of restructuring plans; (iii) the provision of an automatic stay against the assets of the company; (iv) the option for debtors to self-liquidate their companies upon failure of the restructuring plan; and (v) the release of debtors (and guarantors) from the obligation to top off any shortfall from the liquidation receipts.
Lessons of Regional Experiences
23. The TAMC framework shares features in common with other centralized asset resolution entities in the Asia region (Table III.1). Such entities include Malaysia’s Danaharta, Indonesia’s AMC unit of IBRA, and Korea’s KAMCO (see also Annex I). The overall design of the TAMC is most similar to Danaharta, although, in terms of size and scope of operations, the TAMC resembles IBRA more closely. More specifically, the TAMC draws on Danaharta’s three-pronged loan resolution strategy, encompassing debt restructuring, business reorganization, and liquidation, and on its design of special legal powers, including the ability to foreclose and to place debtors under a streamlined business reorganization proceeding (through so-called “special administrators”). In terms of size and complexity of operations, the TAMC draws a closer parallel to IBRA’s AMC unit, which manages a differentiated portfolio of loans, amounting to some 25 percent of GDP in book value and accounting for more than 130,000 individual debtors.8 In contrast, Danaharta’s operations are more focused, with loans under management amounting to less than 15 percent of GDP in book value, and just over 3,000 accounts.
|KAMCO (Korea)||Danaharta (Malaysia)||IBRA (Indonesia)||TAMC (Thailand)|
|Objectives||Freeing banks from NPLs and supporting corporate restructuring while maximizing recoveries||Freeing banks from NPL problem; maximizing recoveries||Assist banks in recapitalizing; aim at maximizing recoveries||Supporting recovery of the corporate sector|
|Structure—Policy Board||Professional Management Committee, with 5 out of 10 members from the private sector||Professional and independent board of directors, with 6 out of 9 members from the private sector||Professional Oversight Committee, with most of the 7 members from the private sector||Non-professional Board of Directors, with 3 out of 11 members from the private sector|
|Disclosure of financials (FS)||FS: independently audited, published semi-annually||FS: independently audited, published semi-annually||FS: independently audited, published annually||FS: audited by Auditor-General Thailand, published semi-annually|
|Criteria for Asset Transfer||Classified loans (Cat. IV and V) from closed, state-owned, and jointly recapitalized banks|
|Purchase/Transfer Price||Purchased at zero value but bank losses from the sales were effectively shouldered by the state (see below)|
|Pays with||Interest bearing KAMCO bonds guaranteed by the government||Zero coupon Danaharta bonds guaranteed by the government||Bank losses from selling NPLs at full discount were made up with issuance of interest bearing government bonds||Interest bearing TAMC bonds guaranteed by the FIDF|
|Asset disposition and management||Auction; public sale; equity partnership; and securitization||Private auction; tenders; securitization; special administration (business restructuring)||Debt and business restructuring of larger loans; outsourcing of medium sized loans; auctions of smaller loans; foreclosure||Debt and business restructuring foreclosure; outsourcing|
|Special powers||n.a||Power to seize debtor assets (PP17)|
24. The experience with centralized asset resolution agencies in other countries suggests factors needed to ensure success.9 Among such factors are:
Keeping the acquired loan portfolio to a manageable size to allow for more focused operations and specialization. The most successful centralized asset management vehicles have been those designed to acquire a relatively small share of financial system loans, with a small number of individual accounts and concentrated exposure to a limited number of sectors or industries.
Narrowing and better focusing the statutory objectives enhances prospects of achieving them.
Making full use of private sector expertise, and employing a variety of methods for the management of distressed loans. A successful example of this is KAMCO, which invited private equity partnership in the management of NPLs. Private sector expertise has also been used in Danaharta (e.g., by appointing outside experts as “special administrators” of companies under restructuring), and to a lesser extent in IBRA (through its outsourcing program of medium-size commercial loans).
Applying commercial criteria and value maximization in designing the terms for debt and business restructuring. In all three of the other Asian AMCs, value maximization is a key objective (if not explicit in the enabling laws, reference to this principle is made in operational guidelines). This highlights the need for the TAMC Board to clarify in its operational guidelines the key principles for the preparation of debt and business restructuring agreements. It would also be beneficial to clarify the conditions under which foreclosure or business liquidation would be preferred over the objective of preserving the existence of debtor businesses.
Leveraging the use of special legal powers. Danaharta provides an example of how the provision of special legal powers, and willingness to exercise them when appropriate, can facilitate the transformation of (“strategic”) NPLs into performing loans.
Minimizing political interference, and conducting operations transparently. Although difficult to quantify, the operations of Danaharta and KAMCO have enjoyed a relatively high degree of independence, which is likely to have contributed to their success. In these cases independence was provided “by design,” in that, for instance, the management boards of KAMCO and Danaharta were designed to have large representation from the private sector. Given that the risk of political interference increases directly with the degree of operational discretion, and inversely with the degree of disclosure, one way to shield government entities from political interference is to limit management discretion, enhancing accountability, and imposing a high standard for disclosure.
25. Finally, experience elsewhere illustrates the importance of the design of the governance framework. In this context, it would be useful for the TAMC Board to draw up and implement TAMC’s governance framework, specifying elements of accountability, disclosure, and oversight in the operations. This is particularly important in light of the limited private-sector participation at the Board and executive level (especially when compared with other Asian AMCs), which could raise questions about TAMC’s degree of independence. Related to the governance framework, the Board will also need to clarify the respective roles of various bodies (including the MOF, Bank of Thailand, the Auditor General, and TAMC’s Audit Committee and Board of Directors) in overseeing and reviewing the performance of TAMC operations.
D. The Impact of the TAMC
26. While it is too early to draw conclusions about the broad economic impact of the TAMC, information is available to assess the probable impact on private banks. The TAMC is expected to have only a modest positive up-front impact on private banks, given their fairly small share of the total loans eligible for transfer (see Box III.3). After the sale of loans to the TAMC, private banks’ NPL ratios will fall and, given the resulting decline in risk assets, capital adequacy ratios will automatically increase. Cash flows are also likely to improve, as interest bearing bonds replace non-performing assets in bank balance sheets. The beneficial impact will likely differ across banks, depending on the relative share of NPLs transferred and their net book value (i.e., their transfer price). Banks that sell a relatively high share of distressed loans at comparatively high net book values would benefit the most.
27. Ultimately, the extent to which banks benefit will depend on the TAMC’s ability to recover value. If recovery is less than the transfer value, some of the up-front financial benefits to banks (discussed above) would be eroded over time, because of the loss-sharing arrangement. However, banks could still benefit, as the loss-sharing arrangement effectively acts as a stop-loss on the loan portfolios transferred to the TAMC. In particular, the agreed formula implies that private banks’ total exposure will be capped at an estimated 60–65 percent of original book value (this includes exposure from provisions set aside by banks prior to the loan transfers). In the event the TAMC recovers more than the transfer value, however, private banks may not reap the full benefits from selling NPLs given that gains are also to be shared. Ultimately, the benefit to banks depends on the extent to which the TAMC has a greater ability to recover value than the banks themselves.
Box III.3.Financial Impact of the TAMC on Private Banks
The TAMC is not expected to have a large up-front impact on the private banking sector, though some individual banks will likely benefit more than others. An analysis of the impact of the TAMC on the private banks is presented in the following Table, and this highlights key features of the overall low, but somewhat varied impact of the TAMC on the private banks:
The aggregate share of private bank loans eligible for transfer is not large, but the ratio of eligible loans varies across banks. The TAMC could take on NPLs worth about 6 percent of private banks’ total assets on average. However, this masks some variation, with three banks in particular having a larger ratio of eligible assets. These include one large private bank, due to its significant exposure in syndicated/multi-creditor industrial lending, as well as two medium-sized private banks, which will all be transferring NPLs worth up to 9 percent of their total assets.
Given the small absolute size of the transfer, overall private bank distressed loan ratios are not much affected by the establishment of the TAMC. Both headline NPL and distressed loan ratios fall by about 8–9 percentage points, but distressed loans remaining on private banks books will still account for about 25 percent of all loans (compared with 50 percent at the peak of the crisis). As a result, over ¾ of private bank distressed loans will remain to be resolved on their own books.
Distressed loans eligible for transfer appear to have higher collateral value and less provisioning than those that would remain in private bank books. Very preliminary information (not presented in the Table below) on the provisioning coverage for NPLs eligible for transfer indicates that net book value (NBV) of these loans (i.e., their acquisition price) is on average just over 60 percent of book value. However, based on information on the actual overall provisioning rate, the NBV of the overall NPL portfolio is lower (at about 50 percent of original book value). A number of banks will transfer NPLs at a higher price (NBV) than average. In particular, preliminary information on the NBV of loans eligible for transfer shows that two medium-sized private banks could sell their NPLs at a price close to 65–70 percent of original book value, while the average price for all banks could be just over 60 percent.
The up-front financial impact of the TAMC on private banks is small. Private bank CARs would on average be increased by about ½ of a percentage point by the end of the first year of TAMC operations, while the cash flow from receiving the TAMC note in lieu of NPLs should add about 0.1 percentage point to banks’ net interest margins (currently at about 2 percentage points). Given the small gain in annual margins, it is unlikely that banks could afford any reduction in lending rates after transferring loans to the TAMC.
While loss-sharing will partly insulate bank capital from future large losses, the overall impact on capital is not large. In particular, assuming a conservative loss of 60 percent of original book value on the overall portfolio transferred to the TAMC (which implies a recovery rate for the TAMC of 65 percent of acquisition value), private banks would bear additional losses of about 0.4 percent of their total assets (B 16 billion), while the TAMC would take on losses of about 1 percent of total bank assets (B 36 billion). However, given the high level of pre-existing provisions, banks would still take the bulk of total loan losses, in this case about 3 percent of their total assets (B 100 billion).
|Book value of NPLs eligible for transfer 1/|
|(share of gross assets)||9%||3%||3%||6%||9%||7%||2%||6%|
|Transferred NPLs/Total Current NPLs||34%||10%||12%||25%||24%||22%||10%||24%|
|Pre-transfer Distressed Asset Ratio 2/||36%||34%||31%||29%||42%||36%||23%||33%|
|Adjusted Distressed Asset Ratio||24%||30%||27%||22%||32%||29%||21%||25%|
|Share of distressed assets remaining||66%||87%||88%||75%||76%||78%||90%||76%|
|Pre-transfer CAR 3/||5.2%||7.1%||9.9%||7.2%||8.5%||10.0%||10.1%||7.4%|
|Post-transfer CAR 4/||5.6%||7.3%||10.3%||7.6%||9.2%||10.9%||10.3%||7.8%|
|Implied net book value of NPLs eligible for transfer (NBV/B V) 5/||37%||31%||50%||65%||60%||55%||6%||47%|
|Impact of cash-flow from TAMC note|
|on net interest margin 6/7/||0.2%||0.1%||0.1%||0.1%||0.2%||0.2%||0.1%||0.1%|
|Adjusted CAR 8/||5.9%||7.4%||10.4%||7.8%||9.5%||11.1%||10.4%||8.0%|
|Loss Sharing Scenario9/|
|Loss, net of existing provisions|
|share of NBV||–24%||–26%||–56||–40||–42||–60||–54||–36|
|Loss sharing borne by bank|
|(share of gross assets)||–0.3||–0.1||–0.9||–0.9||–1.5||–2.4||–0.6||–0.4|
|Loss sharing borne by TAMC|
|(share of gross assets)||–0.8||–0.6||–0.6||–0.6||–0.1||–1.6||–0.4||–0.9|
|Total loss borne by bank|
|(share of BV of assets transferred to TAMC)||–51.0%||–50.3%||–39.5%||–49.2%||–48.6%||–36.0%||–40.9%||–44.2%|
|(share of gross assets)||–4.6%||–1.7%||–1.1%||–2.9%||–4.3%||–2.4%||–0.9%||–2.7%|
|Excess provisions at end-2000 as share of RWA 10/||6.0%||3.2%||3.1%||1.4%||2.3%||1.2%||1.9%||3.6%|
Impact on the Government’s Fiscal Position
28. The effect on the fiscal position is limited by the fact that much of TAMC’s activities involve impaired assets already within the public sector.10 Shifting the responsibility for the resolution of the assets to the TAMC does not in and of itself add to the long-term government debt burden. In fact, the incremental impact on long-term government debt is confined to the additional potential losses stemming from the resolution of the private bank loans purchased by the TAMC. The fiscal impact of these operations is capped given the loss-sharing arrangement and the principle that banks will ultimately also bear TAMC’s interest and operational expenses (TAMC losses are capped at 70 percent of the acquisition value of private bank loans, which should not exceed 2 percent of year 2000 GDP). However, given that private banks’ responsibility for their share of losses is back-loaded in time, the incremental effect on government debt may initially be above the maximum estimated losses, but not larger than the acquisition value of private bank loans (3 percent of year 2000 GDP).
Wider Impact on the Economy
29. The establishment of the TAMC could also have ripple effects on the economy through its impact on the corporate sector. Provided the TAMC is successful in catalyzing debt restructuring and improving its quality, corporate indebtedness could be reduced over time to more sustainable levels. This would in turn increase the pool of performing borrowers in the economy. It is unlikely, however, that an improvement in corporate balance sheets alone will be sufficient to accelerate lending and revive investment demand, given the current depressed rates of return on capital and the sizable overcapacity in the economy. Over time, however, investment demand can accelerate and contribute to faster economic growth provided that more fundamental rationalization of enterprises and liquidation of nonviable businesses is implemented—both are areas where the TAMC could play a role. The above considerations, however, hinge on the assumption that the TAMC will be able to identify and restructure viable enterprises, and is decisive in liquidating unviable enterprises.
30. The Thai authorities have moved rapidly to set up the TAMC as a vehicle to centralize the management of unresolved distressed assets. The TAMC will soon begin purchasing the first tranches of distressed loans, most likely starting with the largest multi-creditor exposures from private and state institutions.11 The smaller, individual-creditor loans from state institutions are expected to be transferred later.
31. The framework has been designed to overcome existing obstacles to debt restructuring. By centralizing loan management, the TAMC will minimize creditor coordination problems and facilitate the restructuring of large multi-creditor loans. The provision of immunity to TAMC staff will remove the main obstacle thus far to progress in debt restructuring by state banks. The combination of these features, together with the provision of special legal powers to circumvent existing weaknesses in the legal framework, could considerably speed up and improve the quality of debt restructuring in Thailand.
32. However, lessons from international experience show that for TAMC’s chances of success to be maximized, a number of conditions will need to be met. These include, in particular: (i) making full use of private sector expertise in managing the assets under TAMC’s purview; (ii) applying consistently the principle of value maximization in drawing up terms for debt and business restructuring; (iii) using the provided special legal powers in a consistent and even-handed manner; and, finally, (iv) minimizing political interference and conducting operations transparently.
33. It is premature to assess the full impact of the TAMC on the economy and much will depend on implementation. However, it is likely that the near-term beneficial impact on private banks will not be large, on account of the fairly small share of private bank distressed assets eligible for transfer to the TAMC. For the same reason, the incremental long-term impact of the TAMC on the public debt is limited. The key issue is whether the TAMC will be able to improve the quality of debt restructuring, thereby accelerating the much-needed process of de-leveraging of the corporate sector. In this regard, as the TAMC has control over half of system-wide distressed assets, it is in a unique position to act as a catalyst for debt and corporate restructuring, particularly given its special legal powers.
1. Reflecting the magnitude of the financial sector dislocation in the Asia region, most affected countries established some form of centralized national A VIC. Peak distressed assets rose to above 20 percent of all banking system loans in most Asian countries during the crisis. However, the severity of the financial sector crisis in Thailand—where distressed assets peaked at close to ½ of banking system loans—was surpassed only by that in Indonesia.
Impaired Assets, 1999
2. Asian centralized AMCs have generally been given broad mandates for bank and corporate sector restructuring. This is in part a reflection of the systemic nature of the financial crises they have been designed to address. They have also varied in terms of size and efficacy. Some key features of regional AMCs with regards the size of assets acquired, the purchase price, and the current rates of asset disposal and recovery by AMCs across the Asia region are as follows (see below for a broader international experience):
Regional AMCs have been large relative to international comparators, and have acquired assets with value ranging from between US$10–70 billion, but more importantly, ranging from between 15 to 40 percent of GDP. The TAMC will be among the largest regional AMCs as a share of GDP, and large also in absolute terms.
Most regional AMCs have purchased assets at a heavy discount. The Indonesian Bank Restructuring Agency (IBRA) and Thailand’s Financial Sector Restructuring Authority (FRA) acquired NPLs at essentially zero direct cost—die resulting recapitalization needs generated in the counter party financial institutions were effectively met by other government entities. Both the Korean Asset Management Company (KAMCO) and Malaysia’s Danaharta have acquired assets at a steep discount, generally reflecting a rigorous market-based valuation. The TAMC is expected to acquire assets on average at fairly low overall discount, reflecting the low provisioning (and consequent high net book value) of NPLs held by state financial institutions. The discount on acquired private bank NPLs is instead larger, though the resulting purchase price of just over 60 percent of original book value is still above others in the region. Nevertheless, because of the loss-sharing arrangement the ultimate cost to the banks may be higher than those reflected in the transfer price.
The speed of asset disposal has varied depending on the mandate and method of asset resolution. The FRA was designed to focus on rapid asset disposal of the loans acquired from closed finance companies. KAMCO disposed quickly of a large share of assets reflecting in part its innovative approach to bundling and securitization of loans for rapid sale on secondary markets. Danaharta’s relatively high disposal rate reflects in part a willingness to combine asset sale methods with slower business restructuring, but more importantly that its special powers have resulted in a large proportion of up-front payments and settlements of loans. This raises the issue that a large fraction of Danaharta’s assets under continued management are those in business restructuring, such that the disposal rate will likely slow. This contrasts with the situation in KAMCO, which has adopted more across the board asset sale and auction methods.
Recovery rates on disposed assets have tended to be low across the region, including relative to best international results. The U.S. Resolution Trust Corporation recovered 87 percent on original book value of assets under management. In comparison, recovery rates in the Asia region have so far averaged around 50 percent. Danaharta is the best performer with recoveries to date averaging just over 60 percent. However, it is important to bear in mind that recovery rates can be expected to drop as harder to resolve assets are addressed last. Moreover, expressing recovery on remaining assets in appropriately discounted (NPV) terms would likely depress ultimate recovery rates further. Also, country-specific measurement issues further complicate comparisons of recovery rates.
AMC Face Value of Assets Purchased
AMC Face Value of Assets
AMC Purchase Price
AMC Asset Disposal Rate
AMC Recovery Rates
3. International experience suggests that successful central AMCs have tended to be relatively small and with focused operations.1 Centralized AMCs have been used in many different countries as vehicles for addressing distressed asset problems arising from financial sector crises. The size, scope and modalities of AMCs have however varied significantly. Most importantly, the success of AMCs in an international context has been highly variable. Indeed, most “successful” AMCs have tended to be small relative to the economy and have been fairly focused in the nature of their operations (for example on managing largely real estate assets).2
International Experience with Assets Transferred to Central AMCs
(source: Klingrbiel (2000)
4. AMCs have been set up both as business restructuring and rapid asset disposal vehicles. A review of 7 international asset management companies, set up prior to the Asian crisis and for which data are available, indicated that 3 were established mainly as restructuring vehicles. Further, it is argued that 2 of the 3 restructuring vehicles did not achieve their narrow goals, with the successful Swedish experience reflecting in part a concentration on real estate asset management. Meanwhile, only 2 of the 4 rapid disposal vehicles were considered successful (in Spain and the U.S. RTC), reflecting their relatively small size, easily sellable assets, professional management, political independence, as well as a strong overall legal framework. These characteristics are shown in the Text Table across the range of countries considered.
Creditor Rights 1/
|Peak Level of NPLs|
(percent of financial
Prepared by Lorenzo Giorgianni, with input from Sean Hagan and Vikram Haksar.
The law was first issued as a Cabinet Emergency Decree and later ratified by Parliament.
For real estate collateral, the TAMC law requires financial institutions to use the Land Department’s valuations, which are typically lower than banks’ own appraisals.
The book value is calculated by subtracting from the original principal claim any principal re-payments made by the debtor, and by adding up to 3 months of overdue interest payments.
The yield is paid annually and capped at the average of the deposit rates of five Thai banks.
Moreover, the TAMC law states explicitly that employees and staff are not liable when acting with “care and professionalism.”
The law mandates that the transfer of assets from the private banks be a one-time operation and subject to meeting predefined criteria—reducing banks’ discretion to transfer only their worst NPLs. The law also incorporates a sunset date, limiting the duration of its activities.
Interestingly, in the case of the TAMC, large corporate loans account for less than half of the value of the total portfolio of eligible loans for transfer (in terms of number of debtors, the share is estimated at about 5 percent). In the case of IBRA’s AMC, instead, corporate exposure accounts for a much larger share (close to 80 percent in value, and just over 1 percent in number of accounts) of total exposure.
See Klingebiel, D. (2000), “The Use of Asset management Companies in the Resolution of Banking Crises: Cross-Country Experiences,” mimeo.
See Chapter VI for a detailed analysis of the public debt dynamics and the impact of financial sector restructuring costs.
According to the latest announcements by the authorities, the TAMC would initially buy from private and state financial institutions close to B 400 billion in book value of the largest multi-creditor loans (roughly equivalent to 300 accounts out of an estimated total of over 5,000 accounts comprising the universe of multi-creditor loans eligible for transfer).
This discussion draws on Klingebiel, D. (2000), “The Use of Asset Management Companies in the Resolution of Banking Crises. Cross-Country Experiences,” mimeo.
Success could include the achievement of broader social goals, but here refers to the ultimate recoveries achieved by central AMCs.