Here's how govt should structure recapitalization bonds

Latha Venkatesh

CNBC-TV18

Now that the government has announced that Rs 1.35 lakh crore will be infused into public sector banks through recapitalization bonds, a lively debate has sprung up among bankers and fixed income traders on how to design these bonds. Here are some thoughts:

1. Firstly, it makes sense for the government to more-or-less follow the path taken in the nineties: Sell Rs 1.35 lakh cr of these special recap bonds to the banks, and use the money from that sale to provide share capital to the banks. What’s different this time from the nineties is: the bond market is evolved, and the banks are listed. So the bonds have to be designed keeping in mind their impact on and reception by bond and equity markets.

2. It makes eminent sense for the government to constitute a special undertaking (a la SUUTI) which in turn can issue the bonds and then hold the equity stake in banks. With or without an explicit government backing for these bonds, they will be regarded as sovereign bonds. Creating a special undertaking will not just distance the bonds from being part of fiscal deficit, it will also prevent accounting confusion, when the bonds are sold and the equity divested. There is a school of thought which believes the RBI should issue these bonds. That idea needs to be demolished. The regulator can’t be owning shares in the regulated. (This idea partly springs from those in government wanting to put their hand in RBI’s till. Not a great idea, but that discussion, on another day) For the moment creation of an SURB (Special Undertaking For Recapitalization of Banks) seems like a good idea.

3.Should these bonds be part of the SLR (statutory liquidity ratio) and LCR (liquidity coverage ratio)? Without getting into the prudential principles behind SLR and LCR, suffice it to say these bonds shouldn’t have SLR, LCR status. For one, it will lead to a jump in bond yields and secondly, remember we are still trying to keep them at an arm’s length from the fiscal deficit.

4. If these bonds do not have SLR or LCR, it stands to reason they should not be accepted by the RBI at its repo window. Repo-ability will mean higher liquidity in the system, and doesn’t square with the desire to keep the bonds liquidity neutral. More important, these bonds should not be allowed to interfere with the RBI’s and MPC’s goals as a monetary authority.

5.The next question is if these bonds should be held-to-maturity (HTM) or be tradable in the inter-bank market or indeed even with mutual funds and insurance companies. HTM for the entire tenure of the bonds, can make them a dead weight, but making them tradable will bring with it the problem of marking them to market. An elegant way to reduce the MTM burden, according to a market veteran, is to float the interest rate on these bonds. If the yield is kept as as a spread over the 364-day t-bill, the MTM burden in any year may be limited.

6.Fixing the tenure of the bonds will require all the ingenuity at RBI. One way to reduce the risk of redemption pressure on the fisc or on the special undertaking will be to issue very long tenor bonds as were issued in the nineties. But a better idea is to probably keep the bonds between 10 and 15 years. If the plan is for the Special Undertaking to divest the equity shares and bring government stake down to at least 51%, if not lower then, 10-15 years is a reasonably good enough time frame. However, the tenors have to be staggered so that there is no bunched up redemptions.

7. The hope is this recapitalization is going to be followed by some meaningful reforms. The government hasn’t clearly committed to anything, but scrapping the Banking Companies Act, making bank managements more insulated from political influence and even reducing the number of banks are steps that can’t be postponed any more. Keeping the bond tenures to around 10 years may keep up the pressure to reform, so that the SURB can sell its shares at a profit and create the kitty for the bond redemption.

8. A related idea would be to make the bonds callable. If bank reform indeed succeeds, the SURB can divest shares at a profit and call back at least some of the bonds. On the flip side the bonds should by no means be perpetual. They will look too much like a sleight of hand to side-step fiscal discipline. As well, if the issuing authority intends to hold equity shares and sell them, a “perpetual” weight on the banks balance sheets is uncalled for.

Equity Infusion

The next set of questions are on how to infuse the equity. This can be done through a preferential issue or a rights issue or additional tier 1 bonds (or AT1). A QIP by the banks is also assumed along side.

1. A rights issue may bring some reality check to the current rally in PSU bank stocks. One will know how many investors are willing to put their money where their mouth is. But a rights issue takes too long.

2. If the idea is to give minority shareholders a chance to participate, then the SURB may be given shares preferentially, and a QIp could be launched along side.

3. Clearly for the weaker banks neither a rights nor a QIP issue will work. Here, giving survival capital through a preferential issue is the only way.

4. The AT1 is not a good idea for the government or for the banks or for the markets. For the money it is spending and the risk it is taking, government has a right to expect equity holding. Banks have long hated AT1 bonds because their dividend is a drag. Equity needs to be serviced only when there is a profit.

5. It may make sense to give more capital to those banks who are expected to (or whose boards are willing to) acquire one of the weaker banks.

6. Finally, the infusion of share capital by the SURB will almost definitely breach the 75% maximum limit that promoters can hold in a listed company. A special SEBI forbearance needs to be arranged. Forebearance from the creeping acquisition rules will also have to be arranged. These may have to come with a lot of legal vetting and ring-fencing.