4 min read.Updated: 12 Nov 2020, 08:46 PM ISTRohit Prasad,Yogesh B. Mathur
There exists a superior way for Indian banks that must clean up their balance sheets to sell off their non-performing assets
The problem of accumulation of non-performing assets (NPAs) on the balance sheets of banks doesn’t appear to be going away anytime soon, with current moratoriums only providing temporary respite.The deleterious effect on the operation of the banking sector, in particular the revival of credit, is only too well known. Asset reconstruction companies (ARCs) were created via the promulgation of the Securitisation and Reconstruction of Financial Assets and Enforcement of Security Interest (SARFESI) Act in 2002, and were intended to be a vehicle for mopping up NPAs through the injection of private capital. ARCs pay for NPAs using a combination of cash and securities. Although NPA sale transactions between banks and ARCs have resulted in an accumulation of assets of approximately ₹1 trillion (about $15.4 billion) under the management of ARCs, this represents only 10% of the country’s total stock of NPAs. Further, the growth of assets under management of ARCs is slowing while the growth of NPAs is rising.
This year’s Economics Nobel Prize was awarded for the subject of “market design". An optimal design of the market for NPAs can play an important role in minimizing public investment, triggering transactions, optimizing the transfer of risks out of the banking sector, and determining the “true value" of assets. The Swiss Challenge auction currently used in India is a form of public procurement that requires a bank which has received an unsolicited bid for an NPA to publish the bid and invite third parties to match or better it. The aim of this format is to provide incentives to private parties to initiate deals and provide initial estimates of market value, while at the same time allowing “challengers" to compete with the price of the original proponent. Apart from the mode of initiation of the bid, the Swiss Challenge is a sealed bid, first price, reverse auction.
The Swiss Challenge is incongruous with respect to the Indian NPA market for a number of reasons. First, the Indian NPA market is a forward auction, in which buyers bid for an object put up for sale by the seller, while the traditional Swiss auction is a reverse auction where sellers bid. Second, given the high burden of NPAs on the balance sheets of banks, the auction is usually initiated by banks rather than by an unsolicited bid from a potential buyer.
However, the most flawed aspect of the current auction design is that it omits the crucial variable of the cash proportion of the deal from the ambit of the market, and does not build in safeguards to prevent the artificial inflation of deal values. The deleterious effects of these omissions can be seen in the various phases of the Indian NPA market. For many years, deals took place at very low cash proportions, reflecting insufficient risk transfer. Of late, after the Reserve Bank of India (RBI) put in place extremely onerous guidelines on cash proportions, going up to 90%, deals have taken place at close to 70% of book value, reflecting an artificial inflation of value. High cash proportions are favourable for banks, but RBI has further sweetened the deal for them by providing concessions on provisioning requirements against NPAs in case a deal takes place at or above the stipulated proportions. Indeed, the deal has been sweetened so much at the expense of ARCs that the market may have been rendered entirely infeasible.
Our recommendations are as follows: First, RBI should stipulate a much lower minimum cash proportion, somewhere around 30%. The concomitant accounting gains should not get triggered in a binary fashion with no accounting gains below and 100% gains above the stipulated minimum. Instead, there should be a sliding scale of gain—a 30 % cash proportion should imply the accrual of 30% of the accounting gains, and so on. This would reduce the possibility of situations in which the bank is potentially earning high profits from a deal but the deal does not go through on account of the expected losses of the ARC.
Second, the bid should move from unidimensional bids on deal value to two-dimensional bids consisting of a proposed deal value, and a proposed cash proportion that is greater than or equal to the stipulated minimum. This two-dimensional bid should be converted into a cash equivalent by using a discounting factor for the security receipts component. This factor should be derived from the past trends of the value of security receipts held by the ARC.
Third, the sealed bid “static" auction should be replaced by a dynamic auction where bidding takes place over a number of rounds. This format is known to reduce the chances of a “winner’s curse".
Finally, given that these measures would improve market conditions for ARCs, one could also incentivize a higher level of risk transfer by weakening the recourse provision available to them that allows them to return NPAs to banks at no cost. This weakening can be two pronged—first, the period of recourse can be increased from 5-8 years to 10-15 years, and, second, there could be a clawback of part of their management fees on the invocation of the recourse clause. This measure would dampen the ARC’s incentive to bid higher than the true value of an NPA.