The assumption that 15:85 structure would enhance ARC’s cash commitment is erroneous. RBI needs to urgently allow banks' provisioning, resulting from sale of assets to ARCs, to be spread over three years as against two years
The 15:85 structure introduced by Reserve Bank of India (RBI) in August 2014 raising asset reconstruction companies (ARCs) minimum Security Receipts (SR) subscription to 15% of the cost of acquisition of non-performing assets (NPAs) from banks was designed to enhance cash commitments by ARCs. The underlying assumption was that the acquisition cost was sacrosanct, and with 15:85 structure, the ARCs’ 15% investment in the acquisition would mean “skin in the game” for ARCs.
Was the assumption correct? The returns to ARCs, and banks under the 5:95 and 15:85 structures were based on the assumption that the acquisition cost, management fee, upside sharing, ARC’s required return, expenses, and back-ended recovery profiles are the same for a portfolio over a 5-year horizon.
Over a five-year horizon, with back-ended recovery profile, the ARCs internal rate of return (IRR) in 5:95 structure varies between about 20% to 50% for recovery ratios of 30% and 150%, respectively (line “AB”). However, in 15:85 structure, the ARC’s IRR varies between around minus 13% to 21% for the same recovery levels (line “CD”). Hence pricing of the portfolios under 15:85 and 5:95 structures cannot be the same. Assuming that in the above scenarios, if the required IRR of the ARC is 20%, the acquisition cost works out to 2.88 times the estimated recovery under 5:95 structure due to management fee exceeding ARC’s investment, and only 0.68 times under the 15:85 structure due to limited impact of management fee.
However over the same time horizon if the recovery is front-ended, for the same 20% IRR, the ARC will price the portfolio at 2.32 times and 0.80 times under 5:95 and 15:85 structures, respectively. As a result, contrary to the expectations of the banks, under 15:85 structure, ARC’s cash investment is lower than in 5:95 structure for the same portfolio. Thus the assumption that 15:85 structure would enhance ARCs’ cash commitment was erroneous. This is logical since recovery from a portfolio is independent of the structures adopted.
In a 15:85 structure, even with a recovery ratio 100%, the ARCs’ return is less than 8% with back-ended recovery. Hence with a target return of 15-20%, the portfolio pricing by ARCs ensures 100% redemption of SRs with likely upside share -- unlike in 5:95 structure, which sees loss in SR value over years. However, pricing in 15:85 structure also tends to cause notional losses to the banks when the asset sale price being lower than net debt outstanding. If such notional loss exceeds the normal provisioning during the year of sale, the banks have no incentive to sell the asset to ARC, particularly the new NPAs and Special Mention Accounts (SMA). In other words, the 15:85 structure results in front-ending of the provisioning by the banks whereas 5:95 structure resulted in back-ending of the same.
Interestingly the banks’ IRR remains almost unchanged in both the structures (line "EF") confirming that the 15:85 structure cannot offer any advantage to the banks. This has not been comprehended by the banks who have been fixing reserve prices at the levels, which leave nothing for the ARCs. Consequently, acquisitions by ARCs have shrunk drastically. Out of about 94,000 put for sale by the banks, in FY-2015, the ARCs acquired only about Rs20,000 crore with major acquisitions done prior to introduction of 15:85 structure.
In sum, ARCs impinge cost on the banks by virtue of returns required from investment in risk-proven assets. The banks, therefore, need to accept this cost and weigh it against the benefit of freeing large manpower for productive asset building. Nevertheless, the banks’ anxiety to avoid additional provisioning at a time when the profitability is under stress is understandable. This needs to be urgently addressed by RBI by allowing the banks’ provisioning resulting from sale to ARCs to be spread over three years as against two years permitted now. Further, this facility should be extended for four years so that banks are enabled to offload major part NPAs to the ARCs.
Need for pro-active action
As on 31 March 2015, of the total SRs outstanding of Rs61,364 crore, while ARC held SRs of Rs8,819 crore (14.4%), the banks held SRs of Rs50,305 crore (82%). Holding by qualified institutional buyers (QIBs) and foreign institutional investors (FIIs) was low at Rs2,120 crore (3.4%) and Rs120 crore (0.2%), respectively. Redemption of existing SRs has been an issue with cumulative redemptions being low at 22-25% till May 2015 (ASSOCHAM / CRISIL). It has to be recognized that the portfolios acquired under now defunct 5:95 structure will generally result in part SR redemption.
Alternate models with incentive for recovery under 15:85 structure will not raise ARCs’ bids significantly. For ARCs’ bids in 15:85 structure to equal the same in 5:95 structure, banks will have to raise the management fee to over 5%, which is not feasible as it will invite swift regulatory action. Hence acceptance of ARCs’ highest bids under 15:85 structure by the banks is the best way forward as it results in efficient price discovery and releases SRs with 100% redemption characteristic, with likely upside distribution. Such SRs might also generate premium, improve the demand for SRs and provide a market for profitable exit to the SR holders. It is time RBI took a pro-active action and rationalized provisioning norms for banks for NPA sale to the ARCs. This will also help the ARCs to grow by attracting capital and enable them to acquire even SMA assets for reconstruction.
(Rajendra M Ganatra is Managing Director & CEO of India SME Asset Reconstruction Co Ltd-ISARC. He had over 25 years of experience in project finance, asset reconstruction and financial restructuring. The views expressed in above article are personal