The cross-country experience in management of non-performing assets (NPAs) suggests that the establishment of a bad bank proves most effective when it has adequate government and legal backing. In the Indian context, the setting up of the National Asset Reconstruction Company Ltd. (NARCL), with a clear mandate and an explicit government guarantee, will help in alleviating the financial stress on commercial banks. NARCL will be an additional mechanism for resolution of large stressed assets, complementing the activities of existing asset management companies. Going forward, continued commitment, professional staff and transparency in operation will help in making the exercise cost and time effective.

1. Introduction

The establishment of the National Asset Reconstruction Company Ltd. (NARCL) as a ‘bad bank’ and the recent announcement by the Government to extend guarantee of ₹30,600 crore for its security receipts (SRs) have been received with enthusiasm as well as scepticism in equal measure¹. While some have hailed this development as a panacea for all ills, other less sanguine analysts point out the pre-existence of a multitude of avenues for resolution of distressed assets and the challenges faced by them.

Such conflicting views represent the trade-offs inherent in establishment of bad banks. The present paper analyses cross-country practices of such banks to understand the characteristics that helped in their success. The features of the NARCL are then juxtaposed against the lessons learnt from international experience to gauge the possibility of its success. The rest of the article is structured as follows - Section 2 explains the concept and the evolution of bad banks and Section 3 elaborates upon cross-country design elements. Sections 4 and 5 outline the advantages in their creation and challenges faced in their operation, respectively. The next section evaluates the features of the NARCL vis-à-vis cross-country learnings, while Section 7 concludes.

2. Concept and Development of Bad Banks

Globally, the policy options to deal with the overhang of non-performing assets (NPAs) include loan restructuring, harnessing and reforming insolvency and resolution frameworks and operationalisation of bad banks. Although these alternative mechanisms have been tried and tested over a period across countries, the bad bank remains one of the more popular concepts, especially while dealing with system-wide stressed assets.

The term ‘bad bank’ may be used to represent any structure which enables a segregation of performing assets from the non-performing, either on or off-balance sheet. On-balance sheet models, where bad assets are placed in a separate internal unit, help increase transparency and work as a signalling mechanism for the market, showcasing the bank’s commitment to clean its balance sheet. The off-balance sheet models, on the other hand, can take the form of a special-purpose entity structure, wherein bad assets are offloaded, securitized and sold to a diverse set of investors. Alternatively, the bad assets can be shifted to an external asset management or reconstruction company (AMC/ARC). The latter model, though complex and expensive, ensures maximum risk transfer in comparison to the others (Martini, et al., 2009).

The present study focuses on centralised asset management companies (CAMCs) - characterised by varying degrees of government involvement and discusses their design features, while outlining factors that led to the success of some CAMCs over others. Pioneered in the 1980s in the US, the idea of CAMCs received traction when several other countries adopted it in the aftermath of the East Asian Crisis in the late 1990s, and more recently, after the global financial crisis (GFC) (Annex I). A list of CAMCs reviewed for this study, along with their dates of establishment, is presented in Table 1.

3. Design Elements of Bad Banks: Cross-Country Evidence

Centralised bad banks can be differentiated from each other based on several criteria, such as their capital funding, methods of acquisition of assets and the price paid for the same, mechanism for disposal of bad loans, as also the bank’s winding down strategy on completion of its objective. An analysis of cross-country experiences suggests that political consensus, efficient legal processes, adequate statutory powers to the new institution, well-developed financial markets and use of private sector expertise are crucial ingredients for the success of bad banks (Martini, et al., 2009). In the case of some Asian AMCs, a clear mandate, well-defined lifespan, realistic pricing and explicit government financial support have also been touted as reasons for their success (Fung, George, Hohl, and Ma, 2004).

3.1 Financing of CAMC

A bad bank can be financed through partial or full funding by the government—either through issuance of bonds or through equity contribution, special loans from the central bank, or public offering of shares. In the case of Sweden and the United States, Securum and Resolution Trust Corporation (RTC), respectively, were funded fully by their respective governments (Bhagwati, Khan, & Bogathi, 2017). In many East Asian countries, CAMCs have been financed through issuance of government guaranteed AMC bonds, which allows the government to keep the potential liability off its balance sheet. For instance, the Korean Asset Management Company (KAMCO) issues government-guaranteed bonds to raise funds and other financial institutions contribute about one-fourth of its paid-up capital (He, 2004). Explicit government guarantee ensures that these bonds are often characterised by zero risk weight in capital adequacy calculations, thus providing an incentive to banks to invest in them.

3.2 Acquisition of Bad Assets

CAMCs have historically bought toxic assets from financial institutions, mostly from banks and investment trust companies, at varied discount rates, ranging from 20 per cent to 80 per cent. Typically, the seller is compensated with cash and security receipts (SRs), the latter usually guaranteed by the government. NPAs are also screened based on whether their collateral and transfer rights are legally executable. Some of the other factors that are considered in the acquisition process are as follows:

• Depending on the degree of government involvement and the mandate of the bad bank, transfer of stressed assets can be voluntary or compulsory. In Thailand, for instance, while TAMC acquired assets from both government-owned and private-owned financial institutions, transfer of qualifying assets was obligatory for the former and optional for the latter.

• In some cases, bad banks are mandated to buy only a part of the toxic assets. In South Korea, for example, half of the loans were required to be disposed off by financial institutions themselves, and the remaining half were to be purchased by the CAMC. In Malaysia, Danaharta concentrated only on non-performing loans (NPLs) above MYR 5 million, as smaller NPLs would be best handled by the financial institutions themselves. Moreover, the sheer number of accounts relating to small loans made it cost and time ineffective to be dealt with by a centralised AMC.

• There are wide variations in terms of the volume of assets taken over in different countries depending on the scale of the problem and the capacity of the bad bank, among other factors. Danaharta acquired 80 per cent of the total NPLs, while the Irish NAMA took over assets worth 44 per cent of the country’s GDP in 2009.

• In some cases, there are geographical restrictions on acquisition of assets; for example, in China, the local and provincial AMCs being set up since 2013-14 can only acquire toxic assets originating from the same region in which they are located, although no such regional restrictions exist while disposing them off.

• Bad banks can also be distinguished based on whether their mandate is sector-specific or general purpose. General purpose bad banks are those which do not make a distinction in the sectoral composition of the bad assets they acquire. In China, for example, the four state-owned AMCs took over diverse assets, ranging from the manufacturing sector to the farm sector. IBRA has concentrated more on corporate loans, which constitute 84 per cent of its acquired assets. On the other hand, some of the European AMCs, such as NAMA and SAREB, have land and commercial real estate as their target assets and acquire these from different banks.

3.3 Pricing and Recovery

The viability and success of the bad bank crucially depend upon pricing that is fair to both the originator bank as well as the CAMC, irrespective of the method of transfer. Overpriced NPAs dampen the CAMC’s profitability and ability to achieve the mandate, in addition to serving as a disguise for the government to bail out troubled financial institutions. Alternately, acquiring NPAs at very low prices defeats the purpose of the entire exercise, affecting the financial gain that banks stand to make. Countries adopt different mechanisms to decide on the transfer price depending on various factors like the availability of data, nature and quality of assets acquired, and the market for stressed assets (Table 2). In the European countries, transfer price is typically set between the estimated (current) market value and the estimated real (long term) economic value, which in turn is based on projected future cash flows, book value, market value, value of collateral, and probability of recovery, amongst others.

3.4 Management and Disposal of Assets

Acquisition and aggregation of assets by a centralised agency does not in itself guarantee their effective or profitable handling, as the two functions may require separate sets of skills and expertise. In Malaysia, for example, Danaharta was accompanied by a Corporate Debt Restructuring Committee, as well as Danamodal, a banking recapitalisation agency, to aid the completion of the resolution process. Joint venture partnerships are also used to access specialised knowledge on asset management and bridge the lack of expertise in a government-owned AMC.

Techniques employed by CAMCs for asset disposal are tailor-made to suit the type of assets being dealt with. Viable loans may be restructured, while the non-viable businesses, assets, or the underlying collateral, are sold off. For example, homogenous assets could be pooled and securitised through issuance of asset-backed securities. Cases in which liquidation of individual assets is required, foreclosure and public auctions of collateral are undertaken for price discovery.

3.5 Wind-down Strategy

Two views have emerged on the ideal lifetime of the AMC. The ‘European view’ is dominated by moral hazard concerns and advocates a clearly defined, realistic lifetime of the AMC, with asset transfers from banks preferably being completed in a single round. SAREB, UKAR and FMS were all set up with clear mandates of winding up after making profits or selling off all the loans on their books, or after a pre-specified time period. SAREB aims to make a profit of 15 per cent over its 15-year life, while the UKAR was set up to sell off the government’s stake in the troubled institutions. The ‘Asian view’, on the other hand, usually believes in making AMCs a part of its financial crisis toolkit, retaining the infrastructure and expertise, while requiring safeguards to prevent moral hazard problems (Martin, 2020). While the Chinese and Korean AMCs had no clear sunset dates, Danaharta was set up with a finite timeline and wound down its operations in 2005.

3.6 Super Legal Powers

In countries where the extant legal system caused impediments or delays to debt workouts, restructuring negotiations or foreclosures, the CAMCs were bestowed with super legal powers to overcome the hurdles. For instance, Danaharta was conferred with two special powers: ability to acquire NPLs via statutory vesting with certainty of title; and power to appoint independent special administrators to manage financially troubled company borrowers. Similarly, SAREB enjoys a special status as a preferential creditor for subordinated debt over other creditors, and other legal advantages that do not apply to all Spanish limited liability companies. NAMA, under its Act, has been allowed to take over commercial real estate collateral from its insolvent debtors, a facility not allowed for other lenders in Ireland.

4. Advantages of a Centralised Bad Bank

Evidence world-wide suggests that a centralised bad bank can help in substantial stress reduction and has various systemic benefits, complementing and enabling the pre-existing stressed asset resolution mechanisms.

• For selling banks, the clear segregation of weak assets from the rest grants an occasion and rationale to restructure its balance sheet and reshape the business model, while providing transparent insights to customers and investors into the bank’s core performance. By freeing up provisions and thereby, boosting capital for lending, CAMCs enable banks to support the economy. Scarce human resources can also concentrate on more important banking operations such as new lending, rather than managing distressed assets and corporate restructuring.

• The CAMC model can be relied upon when banks do not have sufficient resources to resolve large amounts of NPAs through individual departments within banks or through subsidiaries. As opposed to internal restructuring, a CAMC provides a swift vehicle to get stressed assets off a bank’s balance sheet and reduces the need for fresh recapitalisation either from the government or from markets.

• Leveraging this mechanism, banks can avoid fire sales of assets which result in lower recovery values. Gradual sales through CAMCs that have expertise in the field can help in optimal recovery rates.

• CAMCs are usually driven by other concerns in addition to value maximisation, such as preserving the economic value of assets. For instance, RTC timed the sale of its real estate assets so as to not cause further deterioration in prices.

• Consolidation of stressed loans in bad banks and centralised ownership of collateral can effectively address the problems faced by individual banks in attracting investors for these assets. For instance, public AMCs in Asia provided a market for NPLs by giving banks an option to either sell their NPLs or through forcing them to offload the problematic assets (Lee, 2020).

• Centralised AMCs and public or private ARCs can complement each other, like in Thailand, and need not be substitutes in stressed assets management. Bad banks can hold assets which the private AMCs may not, due to their long gestation periods, or the assets that might not be adequately profitable in the medium term but are economically important and have significant positive externalities (Acharya, 2017)

• CAMCs are usually set up through special acts of law or through government orders. As such, they can be bestowed with special legal powers to accelerate loan recovery, as seen in some successful experiences such as Danaharta.

• The use of cash and/or coupon-paying government-guaranteed securities to purchase NPAs, apart from being more attractive than the SRs issued by private ARCs, improve banks’ balance sheets and add to the income of the financial institutions. It facilitates more efficient price discovery through creation of a vibrant market for distressed assets and a uniform valuation criterion as compared to private forces working in isolation.

5. Challenges of a Centralised Bad Bank

Economists and policymakers have long acknowledged that the establishment of a bad bank is fraught with many trade-offs. These include – how to relieve banking stress without encouraging moral hazard; how to minimise the cost to the government exchequer and how to arrive at a valuation that is fair to both the acquiring entity as well as the seller. In view of the following serious challenges, utmost care needs to be taken in their design and structure.

• Bad banks, essentially formed to ease the burden off banks’ balance sheets, may encourage further build-up of risky assets. Knowing that they have the bad bank option to fall back on, banks may become less vigilant in giving out loans, leading to the problem of moral hazard.

• The design of CAMCs needs to be carefully crafted to suit country-specific and/or sector-specific requirements. Furthermore, there is also a tendency to deviate from the primary objective as in the case of China’s big four bad banks, wherein investments in financial institutions—such as banks and insurance companies—rather than in bad loans, now constitute a major part of their business (Ingves, Seelig, & He, 2004). There is also a need to provide proper incentives to their governing bodies and employees, as staff recruitment and retention are problems that often arise.

• Financing remains a dominant issue in the context of a central bad bank. Unless carefully designed and efficiently executed, the cost of setting up a CAMC falls inevitably on the resource-constrained government, and ultimately, on the taxpayers. Even in countries where the CAMC has succeeded in reducing the NPLs, it has come at a cost to the government exchequer (Fung, George, Hohl, & Ma, 2004).

• Another challenge that emerges is keeping the bad bank funded enough to tackle a major share of NPLs. International experience suggests that in certain cases, the bad bank’s resources accounted for a miniscule share of the bad assets, thus rendering the whole exercise redundant.

• Market creation for distressed assets and their fair valuation at both the buying and selling stages is another operational challenge that arises, as it determines the viability and sustainability of the solution. Asset disposal by state-owned AMCs may be disincentivised by challenges such as political pressures against selling of certain assets and the fear of audit scrutiny relating to sell-offs (Fung, George, Hohl, & Ma, 2004).

• The CAMCs, by virtue of being financed by the government, are required to take a macroeconomic perspective rather than focusing on asset value maximisation. Thus, their decisions relating to either sell-offs or retaining the assets may be constrained by the need to help stabilise asset markets.

• Lack of transparency in operations and lack of reliable data can hinder the objective valuation of assets and creation of optimal portfolios for resale (Deloitte, 2020). In countries such as Ireland and Spain, insufficient or missing documentation created hurdles in smooth functioning of their CAMCs (Cas & Peresa, 2016).

6. Stressed Assets Management in India

The Indian banking system has been weighed down by mounting NPAs since 2012. Various mechanisms like Lok Adalats, Debt Recovery Tribunals, SARFAESI Act and the Insolvency and Bankruptcy Code (IBC) are available to banks to deal with the stress, while maintaining these assets in their books till resolution. An internal restructuring unit-like model has also been experimented in India, but only with partial success. While the Specified Undertaking of Unit Trust of India (SUUTI) was set up as a special purpose entity with the primary function of managing the bad loans of UTI and administering redemptions, it is yet to be wound up. Other assets that were transferred to SUUTI have subsequently increased in value, disincentivising their disposal and making their retention and continuation of SUUTI as a going concern, beneficial.

Alternately, banks can exercise the option of outright sale of bad loans to private ARCs against the issue of SRs, albeit at a haircut. The ARC model has proved to be constrained by several factors, like vintage NPAs being passed on to them, lack of debt aggregation, non-availability of additional funding for stressed borrowers, difficulty in raising funds by the ARCs, etc. Also, they lack the focus on recovery and acquiring necessary skill sets for holistic resolution of distressed borrowers (RBI, 2021). Further, the uncertainty surrounding recovery makes the secondary market for SRs illiquid, causing these SRs to remain on the books of the subscribing entity, which are often the original selling banks themselves (Yadav & Chavan, 2021).

In the Indian case, majority of the bad assets lie with public sector banks (PSBs) and their top management have always faced a dilemma while taking decisive action on stressed assets: what is the right amount of haircut that could be assumed without the fear of raising eyebrows now and the possibility of vigilance action in the future?² Deciding the right price for a bad asset could be tricky and these commercial decisions can go wrong with evolving circumstances. In such a situation, not taking bold actions for resolution now and kicking the can down the road may always be perceived as a pareto superior option.

The experience so far suggests that due to repeated litigations, the resolution processes drag on, leading to asset value erosion, which adversely affects the health of the banking sector in the meantime. The ineffectiveness of these mechanisms manifests itself in the large outstanding stock of bad assets, a major share of which (72.1 per cent as at end-September 2021) are in the balance sheets of PSBs, resulting in the burden of recapitalisation falling on the public exchequer. Moreover, the bad loan problem has dampened the risk-taking ability of the banks, affecting their credit growth.

Design Features of NARCL

Certain features of NARCL closely resemble successful models adopted elsewhere and may prove helpful in reducing the NPA stress.

• Structure: The establishment of NARCL for acquiring and consolidating stressed assets, along with the India Debt Resolution Company Limited (IDRCL) for managing these assets by engaging market professionals and turnaround experts, is similar to the mechanism followed in Malaysia, which proved highly successful in resolving its stressed assets.

• Ownership: NARCL, has an initial capital base of ₹6,000 crores, of which PSBs own a majority (51 per cent) stake while the remaining shareholding vests in private sector banks and non-bank finance companies³. This structure, thus, does not put immediate strain on the government’s limited resources. The majority private shareholding of the IDRCL and the resultant professional and expert handling of bad assets is expected to ensure maximum and timely recovery.

• Acquisition: The NPAs will be acquired from banks by paying up to 15 per cent of the agreed or discounted value of the loans in cash, while issuing government-guaranteed SRs for the rest, following successful models such as Korea and Malaysia. Given that in phase I, assets worth ₹90,000 crores (out of total planned acquisition of ₹2,00,000 crores) that have already been fully provided for are expected to be acquired, recovery will instantly strengthen banks’ balance sheets. Further, by concentrating on legacy large value accounts of more than ₹500 crores, the NARCL may lead to faster resolution of overall stress⁴.

• Government guarantee: The guarantee of up to ₹30,600 crores may be invoked to make good the shortfall between the face value of the SR and the actual realisation. The time-bound nature of the guarantee, valid for 5 years conditional on resolution or liquidation, and gradual increase in guarantee fee payable to the government by NARCL, are expected to disincentivise any delays in resolution. The structure, ownership pattern and the government guarantee backing the SRs is expected to impart credibility to the institution and allay fears of banks regarding scrutiny by various regulators about their sell-off decisions.

• Security Receipts: Apart from reducing upfront capitalisation requirements of the bad bank, guarantees by the government are bound to improve the liquidity and tradability of SRs, helping in development of a secondary market for them.

• Complementarity with existing ARCs: Under the proposed mechanism, NARCL is required to go for the “Swiss Challenge method”, where the 28 existing ARCs in India will be invited to make a better offer for the stressed asset. Rather than being substitutes or rivals as buyers in the market of stressed assets, the nationalised entity will act as a complement to the existing companies. They will help in debt consolidation, minimising the time taken for aggregating the bad loans, and avoiding the inter-lender litigations.

7. Conclusion

Creation of a bad bank at the current juncture may prove helpful in reducing banking stress and kick-start the credit cycle. The cross-country evidence suggests that if the logistical and financial challenges are carefully navigated, experiments of centralised bad bank can have more hits than misses. While it may be unfair to hail them as a universal antidote to deal with financial stress, they have proven to be a worthwhile exercise when armed with conducive institutional frameworks. Experience of international best practices suggests that the NARCL in India is likely to serve as a time-efficient mechanism, while reviving investor interest in primary as well as secondary markets for stressed assets and SRs, respectively. Going forward, continued commitment, professional staff and transparency in operation will help in making the exercise cost and time effective. At the same time, care needs to be taken to ensure that fresh slippages are arrested, and bank balance sheets are strengthened to avoid future build-up of stress.

References

Acharya, V. (2017). Some Ways to Decisively Resolve Bank Stressed Assets. Indian Banks’ Association Banking Technology Conference. Mumbai.

Bhagwati, J., Khan, M. S., & Bogathi, R. R. (2017). Can Asset Reconstruction Companies (ARCs) be Part Solution to the Indian Debt Problem? Working Paper No. 338, ICRIER.

Cas, S. M., & Peresa, I. (2016). What Makes a Good ‘Bad Bank’? The Irish, Spanish and German Experience. European Commission Discussion Paper No. 036.

Deloitte. (2020). Bad Banks in India.

European Commission. (2018). AMC Blueprint. Brussels.

Fung, B., George, J., Hohl, S., & Ma, G. (2004). Public asset management companies in East Asia- A comparative study. Occasional paper, Financial Stability Institute.

He, D. (2004). The Role of KAMCO in Resolving Nonperforming Loans in the Republic of Korea. IMF Working Paper.

Ingves, S., Seelig, S. A., & He, D. (2004). Issues in the Establishment of Asset Management Companies. IMF Policy Discussion paper, PDP/04/3.

Klingebiel, D. (2000). The Use of Asset Management Companies in the Resolution of Banking Crises: Cross- Country Experience. World Bank Policy Research Working Paper No. 2284.

Lee, J. (2020). Country Case Studies on Resolving Problem Loans in Asia. SEACEN Centre Seminar. Manila: Asian Development Bank.

Martin, R. (2020). Asset Management Companies and NPL Resolution. SEACEN Webinar. Joint Vienna Institute.

Martini, L., Stegemann, U., Windhagen, E., Heuser, M., Schneider, S., Poppensieker, T., . . . Brenna, G. (2009). Bad Banks: Finding the Right Exit from the Financial Crisis. McKinsey Working Papers on Risk.

McCauley, R. (2003). Unifying government bond markets in East Asia. paper presented at the Hong Kong Institute of Monetary Research, Hong Kong SAR.

Reserve Bank of India (2021): Report of the Committee to Review the Working of Asset Reconstruction Companies (Chairman: Shri Sudarshan Sen).

Terada-Hagiwara, A., & Pasadilla, G. (2004). Experience of Asian Asset Management Companies: Do They Increase Moral Hazard?- Evidence from Thailand. ERD Working Paper No. 55, Asian Development Bank.

Yadav, A., & Chavan, P. (2021, April). ARCs in India: A Study of their Business Operations and Role in NPA Resolution. RBI Bulletin.

* The article is prepared by Dr. Snehal S. Herwadkar, Arpita Agarwal and Sambhavi Dhingra from Banking Research Division of Department of Economic and Policy Research under the guidance of Ashok Sahoo, Adviser. The views expressed in this article are those of the authors and do not represent the views of the Reserve Bank of India.

¹ Press Bureau of India, press release dated September 16, 2021, available at https://pib.gov.in/PressReleseDetailm.aspx?PRID=1755466.

² ‘CVC finds many flaws in sale of bad debt by banks’, Times of India dated March 18, 2019. Available on https://timesofindia.indiatimes.com, accessed on February 15, 2022.

³ Accessed on February 14, 2022, from https://www.moneycontrol.com/ article dated October 5, 2021.

⁴ At the time of publishing of this article, assets worth ₹50,000 crores are scheduled to be transferred to the NARCL by end-March 2022. Accessed from https://www.thehindubusinessline.com/ article dated January 28, 2022.