Asset Reconstruction Companies – Making Good Misuse of the Law

Headline news in financial press today [here] carried a seemingly-sensational story about one of the leading asset reconstruction companies (ARCs) working under stress of its major shareholders, practices about breaches of law and regulations by the ARCs and so on. For those several people, including professionals, who would have some experience of dealing with ARCs, this is hardly surprising, because many would know that ARCs have been engaged in variety of practices including arm-twisting troubled companies into accepting loans at steep rates of interest from private lenders, allowing private lenders to enforce security interests on assets without satisfying the claims of banks and financial institutions, getting allotment of shares in troubled companies, forcing change in management in violation of RBI norms on takeover of management, etc. Many, however, many not know that ARCs have been vested with special legal powers that need to be deployed in strict compliance with the law, and therefore, many things that ARCs are doing currently are outright illegal.

In fact, in a larger context, the very institution of ARCs, vested with special powers of recovery, is as much a misconceived idea as the parent legislation under which the ARCs are created.

Models of ARCs: Indian model and international models:

While ‘asset reconstruction company’ may be a typical Indian expression, possibly due to the heritage of the Narsimham Committee referring to an asset reconstruction fund, the global model of dealing with non-performing assets is called asset management companies (AMCs). The first most important point to note is that AMCs are created to resolve the problem of NPAs that result from a systemic crisis. That is, if a systemic crisis leaves the banking system infested with bad loans, there has to be a one-time, central remedy to resolve the problem. AMCs are not envisaged, and intuitively cannot have been envisaged, to resolve the problem of loans that go bad due to bad lending.

Therefore, most countries brought about AMCs as a one-time measure with a sunset clause. The classic example is that of the Resolution Trust Corporation of the USA that acquired assets of savings and loans associations in the 1990s. Closer home, the Danharta of Malaysia is an example that took over loans that went bad due to the 1997 SE Asian crisis, and once the loans were resolved, it wound up its affairs.

Whether one-time or continuing, AMCs in most countries have taken the centralized AMC model – that is, one AMC formed to resolve a systemic crisis.

In India, our unique ARC model has lot of differences from the global models. First, there has not been any systemic crisis that the ARCs seek to resolve. Most of the bad loans in India are a product of bad lending – they represent what is called the flow problem, rather than the stock problem. Second and a key difference is that we have not envisaged a sunset clause for the ARCs – the SARFAESI Act envisages ARCs to be an ongoing business model. Hence, it could not have been a response to a systemic crisis. On a policy plane, it sounds completely counter-intuitive for an agency armed with special powers conferred by law to come and rescue a loan that goes bad due to bad lending. What are the insolvency laws, winding up laws and civil laws on recoveries are doing, if a special law had to come to resolve every loan that goes bad? Hence, the very concept of ARCs in India seems to be a paradox. It is easy to attribute the concept of ARCs to the Narasimham Committee, but the Narsimhan committee had not thought of profit-seeking companies fitting the bill of ARCs in the country.

That brings us to the third significant difference – a single AMC versus multiple ARCs. In India, over the years, several ARCs have come into existence. In fact, ARC has become a business model.

Vesting a profit-driven entity with special powers:

From a legal policy perspective, how does not envision a profit-driven, shareholder-wealth-maximising entity resolving the problem of bad loans? Sure enough, bad loan resolution is a business model, but such a business model has to fit into the overall regime of recovery of loans, enforcement of security interests, and so on. It would be hard to think of entities armed with special powers of the law that buy bad loans and resolve them. If power corrupts, then a special power would corrupt especially. Sure enough, there is no equity and justice on the part of a borrower who does not pay a loan, but then one cannot close eyes to the fact that lenders who foreclose loans quite often commit excesses. Assets are sold in opaque manner, at prices that do not represent fair values. Sure enough, one cannot expect a borrower-centric fair deal from an entity that has to focus on shareholder wealth.

Trust route: the easy escape route to regulation:

Another very unique feature of the Indian ARC model is that virtually all the NPAs are bought in the name of trusts, of which the ARC becomes the trustee. This is a simple device to wish away the regulation of the RBI. RBI regulations require capital adequacy, NPA treatment and income recognition norms in case of ARCs almost in the same tone as applicable to NBFCs. However, if the assets are bought in the name of ARCs, other than the mandatory investment requirement, much of the regulation is not applicable. This may sound real strange, as the legal privileges of the special powers are presumably available even where the assets are sitting in the books of the trusts.

In fact, this question – whether the powers of the ARCs are exercisable where the assets are sitting on trust books – is a significant question that has not been discussed adequately in legal forums. In order for exercise of the special powers, the asset must be an NPA in accordance with the directions of the RBI – the directions which are not applicable in case of trusts. So, the issue is, if the directions are not applicable to trusts, are the trust assets NPAs are per directions of the RBI?

ARCs – traveling much beyond their limited business:

ARCs have a very limited sphere of powers allowed by law – their powers are limited by the SARFAESI Act. Whatever else they do has to be incidental or ancillary to what they are permitted to do under the law. An ARC is not a normal business entity that can buy equity shares, takeover management of businesses, engage finance companies to acquire loans of the defaulting companies, and so on. However, in real world, ARCs are being run exactly as bania shops.

Sale of assets the opaque way:

Also, the sale of assets by ARCs, in spirit, is no different from the sale of assets by state finance corporations and others vested with special powers of recovery. Courts have, over the years, ruled that there is no scope for any lack of transparency in causing a sale of the borrower’s assets, because every single penny realized from the sale goes to the credit of the borrower. It is also clear in SARFAESI Act that the ARC acts as the trustee of the borrower in exercising the rights of sale of the property. Hence, any opaqueness in the sale is completely intolerable in law. There have been numerous instances where ARCs have sold assets by way of so-called private auctions, and simply served a notice on the borrower giving credit to the extent of sale proceeds. No account of how much was the sale proceed, who was the buyer, who were the competing bidders, how was the asset sold, how much were the expenses on sale, etc., have been provided to borrowers. ARCs have gone on record saying they are not obliged to disclose the particulars of the sale – this is completely erroneous.

Do trusts formed by Asset Reconstruction Companies have powers under SARFAESI Act? - Vinod Kothari

The asset reconstruction companies in India is a curious case in itself – the concept was conceived as a device for resolution of bad loans that arise out of a banking crisis, and in fact, it has become a business model by itself. I have discussed this issue elsewhere – see my note http://securitizationdossier.blogspot.com/2010/05/asset-reconstruction-companies-india.html. Recently, there has been a spurt in formation of asset reconstruction companies as there is apparently keen interest in buying distressed NPLs.

ARCs in India commonly buy the NPLs in the name of trusts, of which ARCs act as trustees. This practice sprung because RBI Directions and Guidelines on asset reconstruction companies exempted the assets acquired in the name of the trusts from several requirements of the SAFAESI Act and the Guidelines. Specifically, paragraphs 4, 5, 6, 9, 10(i), 10(iii) 12, 13, 14 and 15 of the Guidelines and Directions do not apply to trusts formed by the ARCs. This is quite strange of a regulator – imagine someone writing a list of rules, and in the next breath, writing an exception line that if you form a trust, then the list of rules that I just wrote will not be applicable. So, this is a clear invitation to exploit the exemption to the hilt because most of the rules that apply to assets bought by the ARCs on their own books do not apply when the NPLs are bought on the books of the trust.

And truly, asset reconstruction companies have exploited the trust route to the hilt. The 31^st March 2009 balance sheet of Arcil shows investments in security receipts of 344 trusts of which Arcil is the trustee. That would mean substantially all the NPLs that Arcil would have bought would be on the books of the trusts, with Arcil simply holding a fraction of the securities issued by the trusts.

And these trusts have been exercising powers under sec. 13 of the SARFAESI Act. Until recently, section 9 of the SARFAESI Act that empowers ARCs causing a takeover of the management of the borrower was lying in a dormant state as the RBI had not notified the guidelines on takeover, which it has now done.

Significant legal questions remain on the power of the trusts. Needless to say, the trusts are distinct from the ARC. The trust is not an asset reconstruction company – it is not a company at all. The books of the trust are different and those of the ARC are different. The assets of the trusts are not reflected in the balance sheet of the ARCs – if that were so, the whole purpose of forming the trusts would go away.

So, if the trust is not an ARC, neither a bank, does the trust have the right to exercise any of the powers granted to an ARC or a secured lender under the SARFAESI Act? People quickly look at the definition of “secured creditor” in sec. 2 (1) (zd) which includes exercise of powers by an ARC managing a trust set by the ARC.

But then, one should not overlook sec. 13 (2) which uses two significant words – “default” and “non-performing asset”. In order to invoke section 13 (2), there must have been a default, and the asset in question must have been characterized as a non-performing asset in the books of the secured creditor. The word “default” is defined in sec. 2 (1) (j) and “non performing asset” is defined in sec. 2 (1) (o). Reading both the provisions, it is clear that in order for constituting a default within the meaning of sec. 13 (2), the asset in question must have been classified as non-performing asset in the books of the secured creditor seeking recourse to the rights under sec. 13 (2).

Now, let us realize what happens in case of ARCs. There was originally a bank that held that asset. The asset was an NPA in the books of the bank. The bank sells the NPA to the ARC. In the books of the bank, the asset disappears – there in no question of the asset still being an NPA in the books of the bank. It cannot be said that the asset continues to be an NPA in the books of the buyer as well, since the RBI norms in case of sale of NPAs to financial system actually provide that where an NPL is sold by a bank, the buyer may treat it as a performing asset and observe the track record of recoveries versus expected recoveries. It is based on such track record that the asset may once again become an NPA in the books of the buyer, but surely, the NPA tag does not go with the asset. What is even more important is that the NPA classification norms are explicitly excluded in case of trusts of ARCs.

So, if the trust is not covered by the NPA norms of the RBI, it is clearly not possible for the trust to claim that the asset in question is characterized as an NPA in the books of the trust. If trusts use their discretion in doing so, it is still not an NPA as per directives of the RBI – which is mandatory to fall within the meaning of “default” under sec. 2 (1) (j).

That leads to the conclusion – the trusts have no power to exercise any of the rights under sec. 13 (2). And note that the powers under sec. 9 are not applicable to trusts at all, as the extended definition of “secured creditor” including the trusts is not applicable to sec. 9 – section 9 is limited to asset reconstruction company only.

Resultantly, neither sec 9 nor sec 13 (2) are available to trusts of ARCs. This may be quite a steep view to take – as thousands of crores of NPAs are sitting in these trusts. But when it comes to interpretation of law, it is not for the interpretor to cure the defects in language of the law – that is the job of the parliament. No one can contend that there was a great philosophy, discussed thread-bare, when the SARFAESI Act was drafted - actually, the very fact that clearly incompatible provisions of securitisation and enforcement of security interests were merged into a common law is a clear indication of lack of thinking. Nor can one contend that the powers that ARCs are enjoying in India is a globally accepted principle. So, if a fundamental challenge to the working of the ARCs forces a rethinking or rewrite of a very crude law, it should not be seen as legal vandalism.

Whether SARFAESI Act is applicable to derivatives dues of banks - Vinod Kothari

The Securitisation Act (SARFAESI Act) marks the swinging of the pendulum – law was against the lender and for the borrower prior to 2002; the SARFAESI Act swung the pendulum in favour of banks. Unfortunately, as the pendulum swings for one side to the other, it is still not a case of a balance.

The passing of the SARFAESI Act, banks and legal fraternity is taking a highly optimistic, and in the opinion of the author, a wrong view that all the debts of the bank against the borrower are covered by the SARFAESI Act. This is not a correct view as the Act is applicable only where the dues arise out of a “financial assistance”.

A derivatives contract is a trade between the bank and the borrower. Note that in accounting parlance, all derivatives deals are accounted for as trading assets/trading liabilities of the bank – they are not recognized as a part of the “banking book”. There is no question of the bank granting a “financial assistance” to the “borrower” when the bank enters into a derivatives deal with an entity. There is no question of an any “assistance”, as it is not that the bank is assisting, and the counterparty is assisted. No one knows for whom the derivative deal with turn out of the money. Hence, it is beyond any banking sense to take a derivatives deal as a borrowing-type transaction.

However, here is a curious case where not only has a derivatives deal been treated by a court as a “financial assistance”, the court has even invoked sec. 34 of the SARFAESI Act to block the action taken by the borrower in injuncting the bank from declaring the borrower as an NPA.

In State Bank of India v Sharda Spuntex P Ltd [1], an entity had been granted limit for purchase of forward contracts in foreign exchange. Upon the company failing to pay the dues of the bank, the bank characterized the account of the borrower as NPA. The borrower moved the civil court for a suit of injunction, which the lower court passed. The bank appealed to the High Court, citing sec. 34 of the SARFAESI Act as the reason to exclude the jurisdiction of the civil court in the matter. According to the bank, the amount owed by the borrower to the bank, albeit for purchase of foreign exchange, was debt owed to the bank, and that in the matter of recovery of debts, the jurisdiction of civil courts was excluded by sec. 34 of this Act. Analyzing the definition of “borrower”, the Rajasthan High court came to the conclusion that the grant of a limit for forward purchase of foreign exchange was a “financial assistance”, and the amount owed by the borrower was a “debt” and hence, the civil court had no jurisdiction in the matter.

With respect, the case is wrongly decided, on two counts. First, SARFAESI Act is limited to recovery of debts of the bank. In the instant case, the bank was not proceeding against the borrower for recovery of debt – instead, the borrower went against the bank claiming that the characterization of the debt as non-performing asset was wrongful. It is not even clear if the so-called forward purchase of foreign exchange by the borrower was backed by security interest – if not, the question of the bank using the SARFAESI Act for recovery of a derivative transaction did not arise at all. In any case, there was no proceeding under SARFAESI Act – hence, the question of bar of jurisdiction did not arise. Secondly, the forward purchase of foreign exchange is a derivatives transaction. The bank and the entity are two parties to a commercial transaction of purchase and sale. The bank is acting in its capacity as a derivatives dealer – it is not a case of grant of a financial assistance akin to a loan or credit facility. The Court went by the meaning of the term “debt”, without carefully analyzing the meaning of the term “borrower” and “financial assistance”. It is important to understand that the word “borrower” has no relevance unless the case is one of “financial assistance”. The word “financial assistance” has to be read ejusdem generis with other forms of financial assistance enumerated in sub-section (1) (k).

---

[1] I(2010) BC 562 (Raj)

Asset reconstruction companies: India lacks thinking or reasoning

ARCIL, the first asset reconstruction company in India, started its operations in financial year 2003-4. As on date there are more than a dozen asset reconstruction companies registered with the Reserve Bank of India and most of them seem to have commenced operations. The Indian model of asset reconstruction companies differs substantially from the global model of companies formed to resolve systemically impaired loans. Briefly, the chief differences are:

Globally, asset management companies were formed to resolve loans that went bad because of a systemic crisis, not loans that went bad because of bad lending. In India, there is no finding that the loans that have gone bad in the past have suffered a systemic crisis. In fact, there is no evidence of a systemic or market crisis in India at all. Hence, the approach has had nothing to do with managing the evils of a system breakdown.
India is the only country where ARCs have sprung up as a business model, with special statutory powers granted by the lawmakers. There is no doubt that dealing with distressed assets is a global business, but in no other case have governments come forward to grant special incentives or special legal powers to profit-oriented asset management companies. In other words, where special powers were granted, as in case of Danaharta, Malaysia, the vehicle a singular brief – resolution of loans that went due to a system crisis. It is arguable whether asset reconstruction companies, which were envisaged as tool of resolving the problem of bad loans, could actually be a business model, and if it is a business model indeed, is there any justification for granting special statutory powers to them.
India is the only country where asset reconstruction companies do not have a sunset clause. In most other countries, such as USA, Malaysia, etc., asset management companies came in response to a crisis. Crisis resolution measures cannot be everlasting – they lose their meaning if they last forever.

Clearly, there has been no central thinking on the role and powers of ARCs. The Committee reports on whose recommendations the ARC model was initially mooted may have actually never envisaged an asset reconstruction business boom that India currently seems to have.

RBI's move on securitisation

The Reserve Bank of India, in its credit policy announced on 27th Oct 2009 (view the credit policy here) has put up two requirements for securitisation – a minimum 10% originator retention, and a minimum seasoning of 12 months. Obviously, the RBI is meaning to be overly stringent when it comes to securitisation. In the wake of the subprime crisis, at some forums, RBI officials self-complimented saying that India had not suffered due to securitisation as RBI was sufficiently prudent. The fact, actually, is that if we are crawling, you cannot feel happy at the fact that the other guy who was riding a horse fell.

Even while international regulators are imposing a minimum of 5% risk retention for securitisation (see EU regulation here and US regulation here), the RBI just doubles that number. Of course, it is not clear from the credit policy whether the 10% slice is a vertical slice or horizontally lower slice. If it is vertical slice, it would make no difference at all, and in fact, would defeat the very purpose of laying down risk retention. If it is lower horizontal 10%, it would be one of the most impractical stipulations to lay down, as, if there is 10% risk of losses in a pool of assets, then the very idea of securitisation, nay, the very idea of banking, should fail. Banks globally require 4% Tier 1 capital, and 8% total capital, with the understanding that it is unlikely that unexpected portfolio losses should exceed 8% of assets. Note that capital requirements are only for unexpected losses – as expected losses are anyway taken care of by credit spreads. If originators were to keep a 10% first loss piece in a pool, they are keeping more risk to themselves than there, whereas a securitisation should result into at least some extreme loss risk.

There is simply no basis to explain that number 10%. It is clearly the product of over-enthused regulation, unmindful of the concept of economic capital.

What makes the 10% risk stipulation acute is that this is over and above the excess spreads, which are usually anyway retained in securitisation pools. The excess spreads should logically take care of expected losses. If unexpected losses exceed 10%, there is no point in bankers creating loan pools, not to speak of securitizing them.

Equally illogical is the minimum seasoning requirements.

Though the fine print of the RBI stipulation is yet to come, one would not be surprised, if the guidelines remain limited to “securitisation” and not a transfer of a loan. Past guidelines have made a stupid reference to “securitisation” meaning a transfer of financial assets to SPVs, thereby excluding from its scope such transactions as do not involve SPVs.

In short, the RBI move will simply kill the securitisation market in India.